Jump to:


Quality of Scientific Evidence in FDA Regulatory Decisions

Date: 04 Oct 1978
Length: 298 pages
Jump To Images
snapshot_lor 03590037-03590334


Date Loaded
05 Jun 1998
Named Organization
Ama, Ama
American Assn for the Advancement O
American Journal of Obstetrics and
American Public Health Assn
American Statistical Assn
Appropriations Comm
Ash, Action on Smoking & Health
Atomic Bomb Casualty Commission
Biometric & Epidemiological Methodo
Boston Drug Epidemiology Unit
British Medical Journal
Bureau of Drugs
Bureau of Drugs Biometric
Bureau of the Buget
City College of Ny
Duke Hospital
Duke Memorial Hospital
FDA, Food and Drug Administration
Federal Register
Fertility & Maternal Health Advisor
Harvard School of Public Health
Hew, Dept of Health Education and Welfare
House Comm on Government Operations
Inst of Mathematical Statistics
Intergovernmental Relations and Hu
Journal of Ama
Journal of the American Medical As
Ladies Home Journal
Man Resources Subcomm
Natl Academy of Sciences
Natl Interagency Council on Smoking
Natl Research Council
NCI, Natl Cancer Inst
NC State Univ
NIH, Natl Inst of Health
Oak Ridge Natl Lab
Obstetrics & Gynecology Advisory Co
Oxford Research Lab
Planned Parenthood Federation of Am
Population Council
Research Triangle Inst
Rockefeller Foundation
Royal College of General Practition
Stanford Univ
State Dept
Ta Budne & Assocs
TI, Tobacco Inst
Univ of Tn
Univ of Wi
Usda, U.S. Dept of Agriculture
Wa Post
Who, World Health Org
Named Person
Anrudh, J.
Aspin, L.
Beral, V.
Blouin, M.T.
Brooks, J.
Brown, C.J.
Budne, T.A.
Burton, J.L.
Carlson, J.P.
Collins, C.
Conyers, J.
Copenhaver, W.H.
Corcoran, T.
Cox, G.M.
Crout, R.
Cunningham, J.E.
Dinan, B.
Doll, R.
Drinan, R.F.
Dubey, S.D.
Duncan, J.M.
English, G.
Erlenborn, J.N.
Evans, D.W.
Fascell, D.B.
Fithian, F.J.
Fountain, L.H.
Fuqua, D.
Gardner, S.
Gibbons, J.D.
Goldberg, D.C.
Goldhammer, S.
Gori, G.B.
Gray, R.
Harrington, M.
Higginbotham, L.
Hightower, J.
Horton, F.
Inman, Whw
Jain, A.K.
Jenrette, J.W., J.R.
Jick, H.J.
Jones, W.M.
Jordan, B.
Kastenbaum, M.A.
Kasten, R.W., J.R.
Kennedy, D.
Kindness, T.N.
Kostmayer, P.H.
Levitas, E.H.
Litt, B.D.
Maguire, A.
Mann, J.I.
Mccloskey, P.N., J.R.
Mintz, M.
Moffett, A.
Moore, J.E.
Moorhead, W.S.
Moss, J.E.
Naughton, J.R.
Oneill, R.T.
Ory, H.W.
Preyer, R.
Quayle, D.
Rosenthal, B.S.
Rothman, K.J.
Ryan, L.J.
Segal, S.
Strangeland, A.
Thompson, R.L.
Thone, C.
Thorogood, M.
Vessey, M.P.
Walker, R.S.
Waxman, H.A.
Weiss, T.
Welch, P.H.
Wydler, J.W.
Wyndler, J.W.
Master ID

Related Documents:
UCSF Legacy ID

Document Images

Text Control

Highlight Text:

OCR Text Alignment:

Image Control

Image Rotation:

Image Size:

Page 1: azo61e00
Page 2: azo61e00
Page 3: azo61e00
I ) CONTENTS Page Hearing held on October 4, 1978------------------------------------- 1 Statement of- Budne, Thomas A., consultant, applied industrial statistics, T. A. ~ Budne & Asssociates-------------------- ---------------------- 132 Gibbons, Dr. Jean Dickinson,, professor of statistics, chairman of the applied statistics program at the Graduate School, the University of Alabama------------------------------------------------- 35. Kastenbaum, Marvin A., Ph. D., director of statistics, Tobacco Institute--------------------------------------------------- 2: Kenned v, Dr. Donald, Commissioner, Food and Drug Administration. Department of Health, Education, and Welfare; aceompanied by Dr. J. Richard Crout, Director, Bureau of Drugs; and Dr. Howard William Ory, Center for Disease Cont'roU----------------------- 138' Letters, statements, etc., submitted for the record by- Budne, Thomas A., consultant, applied industrial statistics, T. A. Budne & Associates: Curriculm vitae------------------------------------------- 133 References------------------------------------------------ 135 Cox, Dr. Gertrude M., professor emeritus of statistics, North Carolina: Stat'e University: Prepared statement and! curriculum vitae------- 15-29 Fountainj Hom L. H.,, a Representative in Congress from the State of Nbrth Carolina, and chairman, Intergovernmentali Relations and Human Resources Subcommittee: April 4, 1977, statistical review of A. Jaini paper entitled "Ciga- rette Smoking, Use of Oral Contraceptives and Unpublished Attachment II and III"-------------------------------- 213-219. Dr. Jain's written comments concerning the December 7„ 1976, F'ederal Register proposals for revisions in the OC labeling- 175-204 Label and brief summary patient package insert required by FDA for OC's----------------------------------------- 235-237 September 26, 1977, FDA memorandtrm of statistical ieomments ofl 6 articles concerning the association between, cigarette smoking and myocardial infarction among users of oral contraceptivesL 223-227 September 30 and October, 18;, 1977, memorandums of FDA statisticians re OC labeling----------------------------- 229-231 Gibbons, Dr. Jean Dickinson, professor of statistics, chairman of the applied statistics program at the graduate school„ the University of Alabama:. Curriculum vit'ae------------------------------------------ 36-40 February 23, 1978, letter from~ Dr. Kastenbaum concerning statis- tical evidence used by the FDA in reaching its decision to warn oral contraceptive users against smoking cigarettes------ 41 Material accompanying statement'~-------------------------- 44-128 ~ Kastenbaum, Marvin A., PhL D., director of statistics, Tobacco Institute • Curriculum vitae------------------------------------------ 3-10 December 5, 1978, letter to Chairman Fountain with enclosures: Response: to statement of Dr. Donald Kennedy and' article r from World Health entitled "How Safe Is the Pill?-------- 140-~-147
Page 4: azo61e00
IY Letters, statements, etc.-Continued - Kennedy, Dr. Donaldi Commissioner, Food and Drug Administration, Department of Health, Education, and Welfare: Page Charts relative to the hearing------------------------------ 150-151 FDA comments on testimony given by Dr. Kastenbaumy Dr. Gibbons,an& Mr. Budne--------------------------------- 138-140 Information concerning basis forst'atement by Secretary Califano- 242 Information concerning FDA failure to give notice of warning-__209-210 Comments concerning proposed oral contraceptive guideline labeling----------------------------------------------- 205-209 Paper entitled "Patient Package Inserts: The FDA Approaeh"-250-254 Prepared statement--------------------------------------- 158-168' Relative risk of nonfat'aI MI among smoker and OC'user, t'able__ 172-173 APPENDIX Memorandum from counsel of Tobacco Institute---------------------- 2.57 Q G.~ GJ"1 Cd ~ 0 wy r
Page 5: azo61e00
. 0 QUALITY OF SCIENTIFIC EVIDENCE IN' FDA REGULATORY DECISION'S (The Adoption of an Antismoking Warning in Oral Contraceptive Pill Labeling) WEDNESDAY, OCTOBER 4, 1978 HOUBE OF REPRESENTATIVE$,, INTERGOVERNMENTAL RELATIONS AND HUMAN REsOURCEs SIIBCOMMITTEE' OF THE COMMITTEE ON GOVERNMENT OPERATION$, ZVash,znqton, D.C. The subcommittee met, pursuant to notice, at 9:45 a.m., in room 2247, Rayburn House Office Building, Hon, L. H. Fountain (chairman of the subcommittee) presiding. Present: Represent•atives L. H. Fountain, Henry A. Waxman, and John W. Wydler. Also present: Delphis C. Goldberg, professional st'aff member; Gil- bert S. Goldhammer, consultant; and John M. Duncan, minority professional staff, Committee on Government Operations. Mr. FOUNTAIN. The subcommittee will come to order. The principal purpose of this hearing today is to examine the sci- entific basis for the Food and Drug~ Administration's order published in the January 31, 1978, Federal Register requiring an antismoking warning in the labeling of oral contraceptive pills. I should! like to say at the outset' that I don't believe anyone could or should seriously object to the required warning if there is adequate scientific evidence that it is both warranted and necessary for the pro- tection of oral contraceptive users. Indeed, under those circumstances, I think the law would mandate it. However, it has been brought to the attention of the subcommittee that the scient'ific evidence relied upon by FDA is inconclusive and of questionable quality, and that FDA has based its decision wholly on this questionable data. • The January 31 Federal Register, volume 43. No. 21, at page 4223, identified two recently reported studies on which the antismoking warning was based. These studies are :('1) A report entitled "Mortality Risk Associated With the Use of Oral Contraceptives," by A. K. Jain, " published in "Studies in Family Planning," 8:5Q-54, 1977; and (2) a report entitled! "Mortality Among Oral Contraceptive iJsers" by V. Beral, published in Lancet. 2:727-731,1977.. (1) G O O .P OWN~
Page 6: azo61e00
2 FDA's order, as published in the Federal Register, became effective on April 3, 1978. Several days prior to that,, representatives of the Tobacco Institute complained to the subcommittee that the two studies relied upon by FDA did not provide a scientifically valid basis for the required warning. They expressed concern that the FDA regulation was based on inconclusive data and Door science. A subcommittee investigation of FDA's files was undertaken and according to the staff memorandum prepared by Dr. Goldberg_ and Mr. Goldhammer, the results of that investi' zation appear to lend sup- port to the concerns expressed bv the Tobacco Institute. Accordingly, this hearing is being held to examine the facts in this matter. If the Tobacco Institute's allegations are true, then this must be regarded as a serious matter. As a regulatory agency. I think FDA must base its decisions on scientificallv sound evidence. Society cannot tolerate regulatory decisions that are biased or otherwise do not meet' t:his test, even though the decision relates to a legal commodity unpopu- lar in the eves of'many people, including,doctors and scientists. The data used bv FDA as the basis for its order have been analvzed by a number of independent, nongovernmental statisticians, as well as by the Tobacco Institute's statistician. This morninp-, three of these experts will give us their findings,, after whichi we will hear fromi the Food and Drug Administration. «'e are very happy to have as our first witness this morning, Dr. Marvini A. Kastenbaum. director of'~ statistics. Tobacco Institute. Dr. Kastenbaum, we would be glad to hear from you at this time. STATEMENT OF MARVIN A. KASTENBAUM, PH. D., DIRECTOR OF STATISTICS, TOBACCO INSTITUTE Dr. KASTExsAIIas. Mr. Chairman, I ami Marvin A. Kastenbaum, a professional statistician. Currently I am, the director of statistics at the Tobacco Institute, Washington, D.C: I received my bachelor's degree in mathematics at the City College of New York and both my master's and doctor's degrees in statistics at North Carolina State TIniversity. I have held a number of professional positions in my field. I served as a biostatistician for the Atomic Bomb Ca.sualtv Commis- sion in Hiroshima, Japan. and subsequentlv as a biometirician for a 14-year period at the Oak Ridge National Laboratory. I was appointed for a 4-year period to the Biometric and EnidemioloQical Methodol~ ogy Advisory Committee of the Food and Drug Administration. I have had faculty appointments at Stanford TIniversitv and~ the Uni- versities of Nisconsin and Tennessee. I am a fellow of both the Amer- ican Statistical Association and the American Acsociation for the Ad- vancement of Science. I have written over 50 reports and scientific articles. My complete curriculum vitae has been submitted' to the committee. Mr. FouxTAir. Without oblection, your curriculum vitae will be in- serted into the record at this point. [Dr. Kastenbaum's curriculum vitae follows :] . ., r
Page 7: azo61e00
3 0 ya,.~: .SarTin A. Zasteraaum Janua_-y1fi, 1926 ?taw York, YewYork wAattT_.4t. S':A7^SSt :far-ied: 4ife-Halen;c.4i:=sn-,ioan (._5) 3crert II.S.A=j, '.--antry. 1'944-46, Z=Zia. 3ur3a,. G..'_na Ca~ai~:.s 327IIUTT0"t:. 4orth CarolinaStats Oniv..?h.D. in Statistics 1950 ftor`.h Carolina State L1^iv.... Y.S, in Statistics 1950 City College of VewYor'.o 3.5. i.nKathematits1943 ?4OF:SS_CYAi. ="DLC!^~?'`1:'" 25C.^.RM:. The Tobacco :nstitute Director o! Statistics Washi.gtcn,7.0. 1970 - Btan'or3 C,ni'ersity Visiting ?7Cfeasar OaK3i_ge Yationai uaborato-r. Stan-or3, Ca1i_ornia Cak3:_qe.. Ternassae 1~969 Soecial adviser on,Statistics 11968-7J Chaa! - 3iometrica Section 3iometrician Mathematics :~esearc?:C:.ater IIniversityo-` aisconsin, Visiti.g 2rc'_essor Had'_son. '1isccns_'i L960-57 1956-60. :noc-oo iastitute of Suman 3iology Consultant Rnn arbor,, xichIgan Suamer.195~ Atomic 9cmID Casual_[Covmisaion 3iostatistieian 3i=oshina. Janan 1953-54 Oun anc' 3rads`aec„ Tr.e. Chie? Stac:st:c:an - 3usinessZm'o=aticn Divisicn gaw Yor't. Yew' York 1352 II.S.. 3ureauof Ceasus 'r7ashizgton, J.C. Su=ers StudenC7aat. Stati3--4¢can n 1948 i?.SC 11 Y
Page 8: azo61e00
4 T'jBnSHI? =~:1?ROE'ESSICNaLSCCI_^_T_7S: amer:can Statistical 2.ssociaticnFxecative Coa6sittee 3iomet=icsSection 196i-53 Govera-_:.q Ccu:cil li965-66c^e111bw 1963- 3iametric Societv Interaational' GaneralTraasurer 1960-63 Council'uem6er 195e-c6 7ssociate Ecimr9 IC.LzT'tICS 1957-72 2astarn vortz ame_ican Zegion Treasurer 1959-60 Regional acv:sory 9oard 1959-71 Qhairsian 1965-68 T-astitate of )lathemacical St_tlstics 1955- 9nva1 Stat:st'_cal Society Felicw 19 6 9- The 3eaou=li. Society (icr...er1v. Iate~a;.i.onal~ associa*ion~for Statistics in 2hysical' Sciences) 1962- Internaticaal Statistical Iastitnta 1976- ?merican.dssociation .or the advanca.:.ent of Science .°ellaw 1965- Saciety of `-'ze Siy-.aa Xi 1956- }7ew ?ert academ_vof Sciences ?a11ow. 1976 Cosacs C1u5 11971- 3CI_ITT=IC ']'REC'"CRI°_S: A.ieri~c.a? uen anZ Women of ScienceAmerican Statistical Association Hernoull!i~ Soci~ety eicmetr.ic Saciety Insat:xte ofMatiema_cal, S`atiatics Iaternari onal Statlst=cal: Ins ` tlute Royal Statistical Society !%
Page 9: azo61e00
5 o^r°'3aCaD:.'ilc. ?SID PeOFy'SSIONAL dCTIVIT?2s: NorthCarolina College,. Lecturer 1955 Oa.k?rd'ge National Laboratory - Travel'_ngLectures1'958-64 Dniversity of Tennessee, Lacturer 1960-62 Professor.Hiomedical'Sci~ences 1963-69 Nationai Science Foundation - Visiting Lecturer 1966-71 Q.S. Public ?eal:h Service Radiological animalAesearc!i Advisory Committee 1968-7,0 e Food,s Drug Adainistzation - Bureau of Drugsadvisory Commi~ttee National academvof Sciences Division of Medical Sciances - 1dvisor, Ccmimittee on Hazards of U:anium DL•:ziag ~ 19fi2-72 1969-70 Climat~ic iaoact Co®ui.ttee - Consul^_ana 1972-Z4 AdCoc Panel on qesearch Needs for -sstiaating, the Biological Haza=Ps of LowDoses of Ioniziag Radiation 1973 Co-uaicatcons:a Statistics - :ditorial 3oa=d 1972- CIVIC ACTI7ITIPS :. Oak Ridge Civic ^.Sisic ?ssociat_.~on,. Iac., Oa.k ~ZSe, Tennessee 195'--54 President 1962-6.3 OakRidge Arts Council. 1962-64 1968-70 Oak Ridge Festival 9oard 1962-b.4 .
Page 10: azo61e00
6 ?II3LiCA.~ICVS k`ID :t°.?CA':'S 1~. "Cn. sa=pling for stateestiis..ates o?~he'i-a oocul'_tion ::. `1or=.: Ca=olina.usi.q the tcwnsh~i2 as aprimar,r sammliag unit."'Gn- puclished tl:esis, Depar sent of Stati~stics,. Ycz-:.: Caroliaa State 2. Gniversiny, 1950l. "a si:riey of ha.^vasting methods, fi-s nac:iaerv and produ cticn, and sales of ce-^a:n agricc.it.:ral items in :70zt., Carolina. "' of Statistics, Vcrth Carolina Ststa Oniversity„ 1351. Depar-3er.t 3. "A stufly of factors a_fecting =esmcnse, and ad;ustment of.:ata_or acn-response ia mail iaeuiries." Deoar-,.ae.nt of Statistics, Vor--`'Caro>ina State IIniversitv, 1952. 4. "Statistiea:: analyses of 8.3.C.C. studies, 1950-1953." Atem'_c3omb ¢asualty Commission 3::os:.ima, Japan> 1954'. 3. "Leu.Femwgenic effects of ion'_z~g rad_ation on atooaie bcmo su=7ivr- ors i. Firosniaa City". (with W. C. aloloney). Science,Vol_ 121 Vc. 3139, rebraa_-y, 1955. 5. "A generi;izaticn of analysis of va=ianca and' muitiviiate analyais todatacased on :rec_sercies in quaSitat_ve categories.or class iaervazs. ^ ('rith S. :T. Roy.) L^.stitnte o=Statistics, Gn'_versitv afNOrth~ Carolina, a:,aec Series vo. 131 ,. June, 1955. i ^Analysis of :rec_uency data in mniti^.ray cont_.genct tables." Gn- pulis: ed'. czasis. Depa_^CSent ef Statist;cs,. Carol:aaState 3. Gr.iversi y, 195a_ "Cn the cy~athesisois no-i..teractian' in a au1'ti«ay conti=genc_r table." iWi=h S. :1. Hov).. Ann. a= Stat_ ___s,. val. 27, Yo. 3,SeLtemLer, 1956. 9. "a ncta an the =equency..~ist=ibutior_o° _-traZs by=aa:ti drosopni~la melaaogaster."' (•+i.thL. Saadler). ce::etics,. vo1. 43,. Tio. 2,. riarGh 1938. . 1O. L9ti`.Sat].on of relative freQ_IIenciesof four 3per3 tV?es i3 ..rosap"r_laaelanoqaster." 3ioffiet_ies, Va1. '-4. Vo. 2, Sune 195a.. a.t _. -3C1.3=3 ta test --''.e no_ =._x_C_-_n gctw.esis." ( D.,C. LanoiisacJl. 3•:amec-r s, ':01. 13, sc. 1, :6arc :~1959.
Page 11: azo61e00
7 _~... : ccn=_dence cn the abscissa__ _:e nc_= c_ -_=xsee'_o._ ti=-+e -_ _ad 7c1.. _5,. Ve 2,. Ji:ne 1959. '"4e seoa_azicn o: molecalar cc=cu,_d5 by ecL ===c'~=>^c 44al_,sa : i scochasc'_c oracess." 3icmec=L<a, 7e1. 47, , ?a= s'_-and 2,. 1360. ': note cn =he ad'.itive aa:-secze i» _~._=genc.; . :a~Dies," 3_cmet-_as„ vo1.. 16, No. 3, Seoc-ber, 1360. .c "Z'_'~~''ose_^_s_'ti+,i!=_z os spleenn ce?1s 4--aa nc=al =zd g,a~.u:.ized a?~ce and a szqa=.cc__ce a « -act =Z_a:s.'•(•.r'.-~` '^ u'_i_.cdac.. and >t: 7. ?ece_scr.)'.. Soc^z`t o`" Vol. 98,. `ic. 1 Saaua=,,. L362_ _eath _anes of mice _. =rosed cc _oaizz. acen=s." .(tr-'j a.C.:?toa and J.N. Cenk__:).. C'-1'.-___ 3as~s ar_d -ke^'oLocv o' Late Samac_= -` -, - -= on_z_:ic 3edia=_en.- =cad~c ?ress, ::ar.dcnand :Iev '_.._'<,. _363_ _,... "Statlst-Ca_ w^.al'_Fs~sc= _-._ ..____ .•,ri~: `L G. _anna). ~aa'_•~es ' ?ac.~cLbc..j/Vo_.,79~, Aay, 1965. _3 ?actorsao-a'_i^,. =_emm seccnda_-y d_seasa -. ^_ocse. _adiaa_ona.53e^as.'Conqdor r.d D~2. 0a=;.~`. ._r1 . Jac-a_ yatic-a_ Cancer '-- - vcl~~. 35,. Vo. 1'. July 1963, _3. :e s , _:cest c~.C:er o_'4 s..__esses. __ a ..Jnccia'_ sa.:.1s." r..r-h {. C. 3e-r¢ar_5.. C3aL-39C9,. Cak Z'_ce `7aa_cna: Oa:< ._a=a, _s',^.nessee, Con__.cenc,=acles: A raviaw." _ - _cr- ycM f95, uac`. 3esea=c: Cancet,. `_ad_son,. Nacor.s'_:, 3e?ca.:ler, :'365. C 2'_. ^A a~s1vs:s sv.scas.: :r_th aae atsc.-i.c =__ ':^_e sem.:-e`1'-=-'-s -'---• ^ec`_ `c°' 7eoc^ 5_9„ Ya-.h. 3eseic: Can:a=, :4adiscn. W_sexs~_^., ?ebr_ar:,li66 .isee _tam 24 be'_cvl.. 22..: Ccacu_er ~_31',rs_s o= _~-__ _ac=c= -=-=ac_:ons ~.a ccn=~cenc: =ab_es.' °eck'=1 ?enc_- ac. ^36 3esaa=a- Can=_. -`lad'_scn,. a_scncsir, 1'36c .. . aae ==am. 3 3 beica,).. 23.: ....=:dence :ec_cns :oz mean =easn_= e^cs .._ __:.a^_ -.cc-cscmes. ' 7ech=- - ° _c- =6, ua=:_. 'Rese== --- -=, -'-discn., n_scar.s_... an.__,_~6c 0
Page 12: azo61e00
8 24. "adia.l.ysis system :ritb. one a:osar_i:s andone =eflact=q state.:' Jou=. 1cc1_ad?rcbac_1_tv,. VoL.. 3, 196e. 25.. "9educed secondi_-yd_sease Wcrtali-.r. in acuse radiation, chimeras.." (with C.C. ConSdon andD.A. Gardiner).Jcur. Vational C_.cer h^stitute, Val. 38, No. 4. April L967.. 25. "Chromcsome dnalysis.^ (•.:th 3s_a. 3ender and J.3.-Davidson). vse of Comcuters ?ttalvsis o:Exzer_rtal Oa_a and the o=yuclea= ?'ciIlIties. U.S. Atomic F.neryy Ccamission.,Diw. o= Tecar nical in:ormaticn Yay,. 1967.., 27. "'Late effects o: fast aeu==ans and gaffiaa rays jzaice as _zflu- enced'b_vthey dose_ata of i~_--adi~ation: L'_ashorten_zg." (cr_th A.C. IIpton, at al). Fadiation 3esea:ch, Vol. 32, No. 31, November, 1967. 29. "Late effects of fast neutrons and qasffia rays in mi.ce as iz°'_ueneed by the dose _ate of i-adiat_on: methodoln5y of izradiation and dosimet_'y." (with..Y. L.4andolphec al) . 7ad:atien 3esearch~ Vol. 32, Yo. 3. :tovember, 1967. 29. "Yinety-day moz-_a~~; ty iz i=radiated' mice treated '+ith rat bone mar=-_,w."(z+ith T. . J.. :Li~tchell,. Di?. Ga_di.er, and C.C.. Conqdon)i. ?eder=_tion ?=oceed'_acs.VaL.,25, 8'o. 2: 1967. 30. "FactoriaA desiqn-response surc-ace studv of mortality :=om secanda_rydisease in mouse :adi'at±~on ch=eras."' ('.r1th,T. S. wltchell, C.C. Conqdon, and D.a-. Gar.d.'iz_r)i.~ ?ederat'-on.?_oceed- ics, Vcl: 27, No. 2, 196e3. 31."COntinqenoz Tables." Zect*sre Serias -Stataez~cs,. OR.T^L-4347,. OaY2i~dge Vaticaal Laboratosy, Cak3idqe.Tarnessee, aarc7, .969. 32. "CCniidecse 3egions." Lectu_e SeriesonStatistics, OBNL-4347.,. Ca_< Ridge i7atienalLa'co_atory,. Oak Ridge, Tennessee, )4a=cS, 1969. 33, "GOaouter analysu of t''srae-°.actor iatersctior.s ia:.ant=gency tab- las."' 0@7L-TH-2569, Oak Zdqa Vat_,^nal Laboratorf, Oak Ridqe, Tennessee, Yay> 1969. 34. "Statistica.: ana'-ysisof t`.ee ncraal =umrn ka--yorne." (•.rithX_ A 3ender).. Amer. :our. o'?uaanGer.etics, 'Tol.-..21~,, `io. 4,. July1969'. 35.. "The classi:ication oi' human c::romosemes. "(arith Y_ a 3ender and. Cl~aud_a S. Lever)i, ?_oceed_zcs of tie 3r"^ Sess:v o_' the :zterzatca-- al Statisticali :nstit_ London, =nq'-and, Sact_1?69.,
Page 13: azo61e00
9 36. "Samale si2e reaui-7ementse tests of eeual;cv o' several Gaussian means." (with D.G. 3ce1 and' R.O.3oMiman) .. OHPIL-4468,. Oak. 2idte vational~ Laboratory, Oah. ~dge, T.ennessee„ Yovesner,. 13'09'~ .~. 37. "a ger_eralization of analysis of v_iance and multivz=iate analysis to catsgorical data inmultidimensional ..-....=gen~c_Ftables."' T_saa_'_ea'_ 3enor: ?7e. 41, Deoa=-_.ent. of Statistics,. 3tiZford. Gniversitv, Aug'.s_, 1969. 38. "The consulting.statistieian:. who needs t:a?` Rev:aw., Oak Ridge ftational Laborator_r, Oak 3idge,. ;ennessee,. ;.al'-, 1969. 39. "'Quantitative e.r2erimenti study of 1ov-Level radiation ca*c±^o- 4enesis." (}+i.th a.C. IIpton et al)~. Prcceedios of I.A.S.a.. S,4m- cosiums 3adiaticn Czt_.eee:_esis„ 1969. 40s ^'Studiesia ia:zalaticn carcinogenesise ac='-tiqae." ?roceed__cs c'-L5e Conference an LZk:a1~S_tior. Cart__ocenesis> Gatli:burg„ Tennessee,. OctcOer '-963.• 41."Arsview of ecntiager.cy taales:"S. N. QovMemor'_al Vo1ume,. Chapter21, Un?versi^F,ot aorch Caroliaa ?rass, 1970. 42. "Adequate SampDe Sizes :or 3radomized'31oakDesigns." (wi:th D, G. Hcel and it.. 0.' 3ouman).. OA:TL-4527, Oak Ridge vat_onal Lacorazc_y, Oak <t'.dge,. T~ennessee,. °ebruary, 1970. 43, "Latee?:ec sof fast neutrons and gamma-raysiz.mice as in=iuenced bv_ the dose rate oE i-adiation: inducticn of r.ecplas:a" (vith, ;L. C. IIoton st alq. 3ad_atio[: Researc-. Val. 41, yo.,3..:!a=c^.. 19 44. Tacles for deter*~~ "g _:e sta`.istical s_an:fieanceo= su_aticn freenenciss". (with K.0} 3owman)'.. Mucat'on, Reseic4.. Vol.. 9„ Ka_v,. LB7.^. 43. "Same1!e size requirsments: one-way analysis of variance^,. (wi'_hD.G.Hoel and 4. 0. 2cwman). aT_OM="^4`TZ•l, Vol. 57,, No. 2,. auSz:st,. 1970. 46.. "Secandz.^y disease aortali_[ _3t-mcuse :aziation c.~eras". C. C. Congdon, T_'. J. ~iitchell,1andD. A. Gar3ic°=1. Jcu^ al Yat_oral Cancer s stit,ts, Vol. 45, Vc. 5, Vovemner_ 1'9'0. 47. "Sz=pLes=ze recl'-rement3: randcm=zd bloc'c desiy:s". ('ri`_= ... G. Hoel andZ. 0. 3cvman). 3ICNJ°_'R_'LA,VoL. 37.,. vo. 3, Deceaaer. .979. ~. "'C=8C7.~1~-T and i1'fer8nce in ex-.ertzlentation". ,•Sutacen SceiatcFVewsLet-er,. Vc... 4,. Harch.. 1971.
Page 14: azo61e00
10 49., "Research ia smckiaq aad 'nealth _?ast, ?rssazt, and _;:c::e",. Czngresszar.al~ 3ecVo1. __7, ve: 63, kay3,197... 50. Discussion of "Demo9--aphy s?ollution? by S. Landau and R. A. ?__:Z_e. ?roceediiss of t=e 38th Session..of se I ter.:ational Staz:sti~ca_ Iastitute.. aashinqton, D. C., aug-ust 1971, 9. '_04'. 51. "Ch`-omosome 15; . A szeci`ic chromosomal gat.hwav.for the o r_q_n a` human maligr.ancy7." (wit3 Y. A 3ender and'.C. S. laver).3ri_ish Jour:nal af Cancer,, Val. 26, 2ebrca.'v 1972. 52. "Lunq caacer: An sxamination o? some.vital statistics",.Concesscona_ 4ecord,Va1 118,. Ho. 108,. June 30, 1972. 53. "MOrtalityy in rat-mouse radiation chimeras"', (with T...J.. N.d:tcael1,. C. C...Conqdon, H. E. Dova, and D~ A. Gardiner). Hev..Zu=on.. ' tndes Clin. et 3ic1., %v_S, 1972. 34. "Combinaticn 3ayesian and logical acarvacha analysis of norna_ and abnermal.chromoscme szseads" (wi;3..X. A 3end'er at a1)... Ccmnut. 3i'.o,lYed., Vol. 2, oa. 1511-io"6, 1972. 55. "31na1ysis ef categarical data: soffie we'_1-'czownanaloques and sone new eoncepts",. Commcaica_ion :n Stat-stics, Vo1. 3,va.,i.,., 1974. .. 56. "?otential nit:al1lis af._cortacle oawer" (with, Z.. 0. Bowman), Techr.omecrtcs,. 'loli.. 15, No. 3,. J.ugust 1974. 57. "Samnlesize remu~_exent: Si-.qle and doubleclass.__caties ax•:eri- aents"(wit:7 Z. 0. Scwman)i. Selered.7a:Les ~a-~erna::cal. Sta_isc_es,.. VqLrlee 3,. Ame:ixam. uaraeman•-ca: Soc.a__r, -r...:_eencs.. RLI,ode Is anc.,.. _97.5. !„
Page 15: azo61e00
11 Dr. KASZExBamK. My present responsibilities include monitoring and analyzing medical and scientific articles which examine the claimed relationship between cigarette smoking and various diseases. Early last year I was asked to evaluate an article by Dr. Anrudh Jain; which dealt with the risk of nonfatal heart a.ttack-myocardial infarction-associated with, t.he use of oral contraceptives and smoking. At that time I concluded that the evidence which implicated cigarette smoking in markedly increasing this risk was weak; in fact, statis- tically unreliable. Consequently;,I was surprised'by the Food and Drug; Administration's announcement in the January 31, 1978; issue of, the. Federal Register that a new warning would be required on the patient insert in all' packages of oral contraceptives. This u°arning was to be placed within a'bog in an apparent attemptt to emphasize its impartance. It read as follows : Cigarette smoking inereases the risk of serious adverse effects on the heart' and blood vessels from oral contraceptive use. This risk increases with~ age and witYh heavy smoking (15 or more cigarettes per day) and is quite marked in women over 35 years of age. Women who use oral contraceptives should not smoke.. After analyzing,the scientific studies that the Federal Register indi- cated had been used' by the FDA in reaching its decision, I concluded' that these papers did not statistically support the language in the warning. I then asked a number of distinguished statisticians to review these same studies and to submit written reports to me containing their objective analyses. These reports confirmed my own judgment. Because of this subcommittee's oversight responsibihties for FDA decisions, these reports, with the consent of the authors, were submitted to Chair- man Fountain to assist him in evaluating the justification for this major governmental policy decision. These statisticians have been invited, as I have, by Chairman Fountain to present their statistical evaluations to this subcommittee. According to the Federal, Register, the FDA relied mainly on two articles in making its decision to require the boxed warning. One article, an interim report of the "Oral Contraception Study of the Royal College of'General Practitioners in the United Kingdom" whose principal author is Dr. Valerie Beral, was publishedin the "Lancet," a British medical journali on- October 8, 1977. In this study a total of 29 women, died from diseases of the circulatory system-24 users of oral contraceptives and 5 controls or nonusers. The second article, which was authored by Dr. Anrudh Jain,, whom I mentioned previously, was published in the March 1977 issue of "Studies in Family Planning"-a publication of the Population Coun- cil, a private research organization. This article contains no data col- lected by the author but consists essentially of a reanalysis of data reported principally in two earlier British studies whose first named author was Dr. J. I. Mann. I
Page 16: azo61e00
12 Mann's first study was a retrospective study of 63 women who used oral contraceptives and were discharged' from hospitals with a diagno- sis of myocardial infarction. The second Mann study was discussed in two articles in the "British Medical Journal." In this second study data were collected on British women who had used oral contraceptives and had died from myo- cardial infarctions. N'o information on cigarette smoking was avail- able for these women. A third article was also cited as support for the boxed warning in one section of the Federal Register announcement. I will' briefly dis- cuss this article,,even though it did not appear in the list of referenced articles. It was written by Dr. Howard Ory,, a physician, who was also a member of the Obstetrics and Gynecology Advisory Committee which recommended the new warning requirement to the FDA. This article, like Dr. Jain's, contained no data collected by the author but was essentially a review and reanaliysis of the work of other investiga- Lors including Dr. Mann. In my professional review of statistical articles, I always begin by asking several elementary but criticaI questions : One: Is the study group large enough and'sufficiently representative to allow conclusions to be made with respect to the entire population from which these persons come ? Two : Is the information drawn from the data described accurately by the authors ? Three : Are the statistical methods used to analyze the data appro- priate? Are the statistical assumptions made by the authors justified? Four: Do the authors discuss any statistical shortcomings of their analy ses ? Have they ignored any other statistical problems a At this time, I would like to briefl,y describe the conclusions of my own evaluation of these three studies cited by the FDA. First, in the Beral article, careful attention should be paid to the following state- ment which appears in the "'Discussion7' section. This (similarity in~ death rates: from circulatory diseases~ for smokers and nonsmokers): suggests that the relative increase in mortality associated with oral-contraceptive use Is Independent of smoking habit. In this statistical context, independence means that there are no synergistic effects of cigarette smoking and oral contraceptive use. In other words, the increased risk of death from circulatory diseases among users of oral cont'raceptivea is essentially the same for both smokers and nonsmokers. In order to verify this stat'ement, I performed the required'stat'isti- cal analysis and confirmed Dr. Beral's conclusion. In my analysis of the number of deaths observed in the Beral! re- port, I was quitie concerned about the small number of persons in- volved in the sample. This is quite important because the numbers of deaths observed are used to calculate mortality rates per 100,0W women-years. If the number of deaths is small, then any slight,
Page 17: azo61e00
13 change-for example, an additional one or two deaths observed shortly after the report is published-would have drastic conse- quences on the conclusions made about mortality rates. Specifically, Beral reported only five deaths among nonsmokers who used orali contraceptives--clearly an additional two deaths would have a statistically significant effect on the mortality rate for this group. Beral recognizes this severe limitation when she writes: "These estimates (death rates) are based on small numbers and are neces- sarily approaimate." Consequently, her report does not, in my pro- fessional judgment', support the language used in the warning. My critical reaction to the Jain paper is best described in Jain's own words; namely : It Is essential to point out that the mortality estimates used In this paper are based on small numbers and may be subject to large sampling errors. It is the author's responsibility to recognize and point out any statistical weaknesses in his analyses. However, this recognition does not cure the weaknesses. It strikes me as scientifically inconsistent for Jain to base firm conclusions on such an admittedly weak statistical foundation. Further, in calculating mortality rates from myocardial infarction for smokers and nonsmokers, Jain had no smoking information avail- able for these deceased women. He conveniently overcame this diffi- culty by assuming without justification that the smoking habits re- ported from an entirely different population of women who survived myocardial infarctions would be the same as those in the population of women who died. In additiony there was no information available on the proportion of smokers among the younger women and the proportion of smokers among the older women. Agamy with no justification, Jain assumed the proportions were the same. This is a glaring weakness in Jain's work because these assump- tions are really guesses. Further, it is totally unclear to me how the language of the FDA warning, which specifically alerts women over 35 years of age to their increased risk of cardiovascular disease, could be based on the Jain article when Jain had no data related to smoking and age. The review paper by Dr. Ory contains a brief discussion of a statis- tical test performed by the author to determine whether or not oral contraceptive users who smoke have a greater risk of heart attack- myocardial infarction-than nonsmoking users. The test results indi- cated that there is no statistically significant difference in the risks. Then the author states : ... we cannot say for certain that oral contraceptive users w ho also smoke have a greater risk of Infarction than nonsmoking users ; however the data suggest that this Is the case. 35-724 0 - 79 - 2 r i k
Page 18: azo61e00
14 In other words, after performing a statistical test which failed to confirm the hypothesis under consideration, Dr. Ory writes as if this were totally irrelevant. This represents a severe departure from ac- cepted scientific methodology. In my opinion, there is simply no ex- cuse for this type of statement in a scientific article. Furthermore, all three of these articles are severely deficient in their failure to ascertain adequately and to assess other risk factors for car- diovascular disease such as diabetes, high blood pressure, and blood cholesterol which, incidentally, have been listed in the earlier patient labeling for oral contraceptives. Based on the deficiencies and inconsistencies described above, I can- not imagine how my fellow professional statisticians at the Food and Drug Administration could have arrived at conclusions about thee statistical reliability of the data in these papers which seriously differ from the ones I have discussed today. Congressman Fountain, as you know, Prof. Gertrude M. Cox, whom you invited to testifyhere this mornang, is at this moment hospitalized with a grave physical illness and will not be able to present her analy- sis of the scientific evidence leading to the FD~ decision. `Vhen I saw her at Duke Memorial Hospital last Thursday, her greatest disappointment and pain,, in spite of her own terrible dis- comfort, was the realization that she could not appear before you in person today. With~ your permission, therefore, I would like to take 2 or 3 minutes to summarize her statement for those who are in~ attendance. Mr. FouxTaix. Yes; that will be fine. Without objection, we will insert a copy of Professor Cox's testi- mony int'o the record. [Dr. Cox's prepared statement and curriculum vitae follow :] O ~.t CJ U1 C O C!1 140k . ,
Page 19: azo61e00
15 Statement for Intergovernmental and'Human Resources Subcommittee October 3, 1978: My name is Gertrude M. Cox. I am Professor Emeritus of Statistics at North Carol~ina State University and re- ceived my undergraduate and graduate education in Statistics at Ilosra State University where I was awarded my D.Sc. -- Doctor of Science. I am the founder of the Institute of Statistics at the University of North Carolina and North Carolina State University where I had taught for thirty years. The building 1'n which the Department of Experimental Statistics is located at North Carolina State University has been named Cox Hall. Other stati'stica1 appointments include Consultant to the Rockefeller Foundation and Senior Statis- tical Advisor at the Research Triangle Institute. I am a member of the National Academy of Sciences of the USA. I have been elected a Fellow of the American Association for the Advancement of Science, the American StatisticaL Asso- ciation, the Institute of Mathematical Statistics and the American Public Health Association. I have served as a consultant to many U.S. government agencies including the National, Institutes of Health,, the Bureau of the Budget, the National Research Council and H.E.W. In response to other consulting requests, I have traveled to many countries in
Page 20: azo61e00
16 the Far East, Near East, Central' America, Africa and Eu- rope. I have written numerous scientific and statistical articles and an advanced statistical text titled Experl_ mental Designs. A complete curriculum vitae is attached to my statement which wi'11 be submitted for the record~. In the eonsi'deration of the possible interrelationships of cigarette smoking and the use of oral contraceptives on the health of women, the compl'exity of this problem strikes me as greater than suggested'in the reports used by the FDA. It appears almost impossible to secure objective and unbiased~ evaluations of research resul'ts when the possible health effects of cigarette smoking are involved'. As a statistical consultant, I was concerned~ regarding the FDA's actions as reported in the January 3'1, 1978, Federal ReRister. Now, I'm even more concerned'and agitated tiy the statements in the IIadies Home Journal, May 1978, under "Your Family's Health," page 54. It states, "Women who use oral contraceptives should not smoke," and the reason as explained by FDA Commissioner ponaldi Kennedy is, "If you smoke while on the pill, you increase dramati- cally your chances of suffering a heart attack or stroke."
Page 21: azo61e00
17 Reading further, "Dr. Kennedy says that research shows a 'synergistic' relationship between the risks of smoking and oral contraceptives." I've reviewed again the articles referred to in the Federal Register, January 31, 1976, Part II, entitled "Oral Contraceptives." I've spent more time reviewing the three articles listed under "References" below. Mann's (3) data on myocardial infarction, (T~ab1e VIII), when analyzed by Chi-square, show a decidedly not signi- ficant interaction (synergistlc effect). Also, the pill effect is larger than the smoking effect. These results are shown as "relative risks" in the Jain (1) 1977 article whichh uses Mann's data. In the Beral (2) article, the table on mortaIityy rate per 100,000 women-years from diseases of the ci!r- culatory system, (1) the mortality ratio (ever-users of pTl1s/controls) is approxi'mately the same for smokers as for non-smokers, and (2): the mortality ratio (smokersL' non-smokers) is approximately the same for ever-users of pi1Ls as for controls. This research does not show 0 ~ ~ ~ 0 ~.1 ~ a=
Page 22: azo61e00
18 that there is a dramati~e increase in health hazards if you smoke and use the p111'. There are several questions that should be asked,, and clarifi'ed,, about the FDA Federal Register release,, dated January 31, 1978, and arti!clles which were used as the basis of the release. First: How reLiabLe are the conclus1ons which were drawn? A clearer assessment needs to be made of the analysis and interpretation of the data used. This is critical, as stated above. Second: The numbers of sample cases used in the three key articles were small; thus, the data are subject to a high degree of variability. Third: In the Beral (2) article, the estimates of mor- tality rate per 100,000 women-years were based on samples often consisting of much less than 50 women-years. Such an inflation of estimates usually does not have the relia- biLity which is implied'when testing inflated data. .®
Page 23: azo61e00
19 Fourth: There are the problems of the appropri'ate choice of controls and the possible confounding effects of varia- bles such as hypertension, obesity, diabetes, migraine, etc. Covariance analysis, using related variables as age, duration of the pill use and extent of smoking habits, is an appropriate tool which apparently has not been usedd in the analysis of these data. Adjustments for some va- riables were made but not clearly described. Some matching of subjects was done but there is no ind'Scation how this information was used in the analysis. Fifth: The assumptions made by Jein (1), that the smoking habits of the fatal myocardial infarction patients are the same as those recorded for the non-fatal myocard'Sal infarction patients and that.smoking habits are the same for women under 40 as for those 40 and over, lead one to wonder about the accuracy and'reliability of his subsequent calcu- lations of mortality rates. The recorded data on death certificates as well as questionnaire responses are apt to be biased due to pre- conceived opini'ons of physici'ana or survey personnel.
Page 24: azo61e00
20 Sixth: Data that someone else has collected are apt to be misused and~ misinterpreted by a second researcher. The person who analyzes and interprets the data must know the details about how the sample was selected, the accuracy of the measurements and the possible biases. It is my Considered judgment that the FDA has not correctly evaluated some of the statistical analysis information availabl'e when it reached its decisions on new labelling requirements in regard to oral contracep- ti've pills and smoking. This probabl~y is a good example where sound professional statistical advice should have been secured and used before government pollicy d'ecisions were made. There are several capablie statisticians in both HEW and FDA, but were they used'to help with the analysis and interpretation of the research results? There i's little doubt that properly designed studies could provide much sounder information regarding these questions which concern FDA and the public. ~ ~
Page 25: azo61e00
21 References 1. Jain A. K., Mortality risk associated with the use 0 oral contraceptives. Studies in Family Planning, B:50-54, 1977. 2. Beral V., Mortality among oral contraceptives users. t~he Lancet, 2:727-731, 1977. 3. Hann J. I. and M. P. Yesse Mar aret Thoro ood' r c ar o, yocar a n anetion n I young women with special reference to oral contraceptives, British Medical Journal, 2:241-245-, 1975. A't~- /-~7- Gertrude Mary Cox Professor Emeritus of Statistics North Carolina State University W ~Z C!I ~ 0 0M W ' ' ~
Page 26: azo61e00
! 22 PF_RSO.aa1.. HISTORY RECORD Miss GcrtrudeM1Cox 1116 Blenheim Drive, Raleigh~, N:C: 27612 Telenhoae: 919-787-6538 PLace of Birth: Dqyton, Webster County, Iowa Date of Birth:: January 13, 1900 Eduoation:. Perry, R.S.., Perry,, Iowa 1914~1918Diploma Iowa State University 1925-1929B.S, . 1929-1931 M.S. March 1958' University of California Berkeley 1931-1933 Iowa State University, Ames 1933-1938' D.Sc..(Honorary) Graduate work Graduate work Positions:. Name of Oraanization Title ofPositYon Dates Self-employed Consultant in StatisticsJune 1971-Present Rockefeller Foundation (IGasetsartand Mahidon Consultant December 1, 1970- June 1, 1971 Universities - Thailand) Research Triangle Institute Senior Statistical Advisor 1966-JuJy 1, 1971 Teaching at N.C. State Univ. Spring 1966-1970 University of Cairoj Egypt Program Specialist 1964-1965 Research TriangleInstitute Director, Statistics Research Division 1959=1964 Inst'itutEof Statisti'cs Director andProfessor 1944-196or North Carolina State College Head ofDepertment and Professor 194o,19k9 Iowa State College Assistant Professor 1939,1940 Iowa State College Assistant' 1933-1939
Page 27: azo61e00
23 BIOCR4?SiICAL LISTIIICS (CCGLi'ITAT7ofIS) Ameriean N.en ofScienee - 1944; 1999,. 1964, 1966~k liorld Biography^ - 1948, 1952 17tw Knows and What - 1949 National Register of Scientific Personnel - 1954, 1956, 1966 (NationalScience.Foundation and National~Research Council). Who's Wbo in the South and Southeast - 1955, 1960, 1964 European Raadbook, Biographical Section - 1956 Leaders in Ameri'can Science - 1956, 1958 Who's Who of American Women - 1957, 1960, 1964, 1966 Dictionary of International Biograpty - 1962 Nbo's Who in,America - 1977Diiectory of Statistical.Personnel~ in the American NationsiTho'sitho in American Education -
Page 28: azo61e00
24 Professinnnl and Nonornrv. Soci'rties: American Public Eenlth Assocl'ation Amtirican Association for the AdvaneementoftScience American Statisticel Association Biometric Society Institute.of Pnthematica1 Statistics International StatisticallInstitute Psychometric Society Royal Statistical Society Inter-American,.Statistical Institute Gamma Sigma Delta Phi Xavma ?hiSigma Xi(Scien ce). 1931 Psi Chi (Psychology). 1930 Delta Phi Delta (Art)',1930 Sigma Delta Epsilon (Women in Science)~1931 Alpha Lambda Delta (19'TZ Nonorae:ndAwerds.: Fellow, American Association for the Advancement: of Science, Member 1943Fellov 19k7Fellow, American St'atistiealAssociation„ 191+4 Fellow, The Institute of Mathematical Statistics, 1944 Fbnorarg Vice-President,.The South African Statistical Association, 1953 Fellow, The American Public Health Association, 1954 RHonorary Member of Societe Adolphe :tietelet,. Brussels,.Belgiumy.1954' HomoraryFellow of the RoyaL St'atistieal! Society.,.1957HonoraryDoctor of Science, ttieIova State College, 1958'The Oliver 4Enx Gardner Award, 1959' - The Distingu±shed~Serrice Award, Gaa'aaSigma Delta, 196GBonorary Life Nescberof The Biometric Society, 1964 ^Cox Ha11",.building atNorth CarolinaStat'e University, named by vote ofBoerdof Trustees, February. 1970, dedicated Oct. 19, 1970 NonoraryM!rJber of Thai Statistical Assoaiation, 1973 National Acadeapr of Science of the USA, 1975 Delta Kappa Gamma (Education) December19k3OazmmaSigma Delta(Agricultural Sciences)~1929Pi MoEpsiloa~(bat~hcaatics). December 1927Phi Kappa.Phi (Scholastic) 1929 The Watauga Medal - Awarded by the Board of Trustees of N.C. State University Presented~at Foundets Daydihner, March 17, 1977 Gertrude M. Cox Fellowship Fund - Started April 1977
Page 29: azo61e00
25 Consultnnt or Advi.or Southern Regional Education Board _ Statistical Consultant.,,1952 ~ M?mtier,,Advisory Committeee on~Statistics 1952, 1953, 1954„ 1955 Panel, FLR7'Office of Education 1963-64 National Science.Foundation Panel, Teacher Education Section 1966 Advi'sory Committee on Epidemiologyand Biometry of the National Institutes of Health (NL`I)i, U.S: ?ublicHealth Service - Sept. 1959-Jwly, 1964 Chairaan, SubCommittee on BiomathematiesTratiaing. 1960-61 Collaborator,.Agricultural Estimation Division, Agricultural Marketing Serrice,..Department.of Agriculture 1941-1950, .1954-1960NationalDefense Ekecutive Reserve, Bureau.of the Census, Department.of Co®erce 1957-1963 Advi.soryCommitiee on Statistical Polityfor Office of Statistical Staadards,. Bureau of the Budget 1956-1961 Consult'apt, World Health Organization„ IlTCAP, Guatemala 1951-1953Consultrnt, Northeastern Region of Association of Land Grant Colleges - Consultant, Federal Security Board Consultant,.Ar=y FieldForees,. Human, Resource Researc&.Office, Fort Knox Arapr DA-4k-lo9 am.-65D 'Consultent, Pineapple Research Institute, Honolulu, Hawaii (1/1-3/31): 1947 National Research Council, Advisory Coarmittee on Behavioral Research, o®rf .1961-62 'Statistic:l Education Advisor to the Government of Thailand, 7/25-9/4, 1964 Advisor, Pan Amer3caa Health Organization 1969 Consultznt.to Nat~ional Ecucation Council and the University Development. Commission on Statistics, Thailand 12/6/68-1/10/69 American Statistical Association Cosmittee on Preservation of Documents of Disting•:ished Statisticians 3/24/70 - Advisory Committee to Secretaryy ofH0i (Elliott L. Richardson),, Washington, D.C. - Jan. 1972-Anrili 1973. Report: "Automated Personali Data Sy.stems,'^ with emphasiss on invasion ofprivacy' Board.of P.aman ResourcesData.and,Anaiysesy Commission of.Fhffian Resources, National Research Council 1976-1979. Task Force onEducrtion in Agriculture end;Renewable Resources, Coffiaission . of NaturaliResourcesy National Research Council. 1976- Joint ASA-AAEA Advisory.Coamittee on Agri'culturol~.:tatisticsy 1977-1979
Page 30: azo61e00
26 Society Cfficen Hcld: . InternationaiStntistical Institute ~ Eleeted momber Seyt. 5, 1949 - First voman;.on7y 28.memDers in U.S. of 200 in the NorldBureau of ISI,.Trea..^urer - 1955-1961 - Delegate of U.S. to 25tb-ession~of ISL in Ylashington, D.C. -9/6-18/47 Delegate of U.S. to 26thsess.ion~of ISI,in Bern, Switzerland Advisor on U.S. Delegation to 30th session of'ISI in Stockholm, Sweden - 8/8/51 Chairman, Education Committee - 1962-1968 Ameriean St'etisticrll Association Electedd member 1933 Elected Fe11oc,- December 28, 1944! President 1956 Director 1949-1951 EditorialCommittee, Index.Comcittee,.NationalCeffia.ission on Training of Statisticians(19471). NorthCarolina.Chaoter of Ameriean Sta'.isticel Association President 1942-19LL, 1956- The Biometric Society (International) Cbarter member 1947Councii 1948-1950, 1955-1957, 1973-1975 . Editor,„ Biemetrics JouuaaI 19b5-1955Fresident 1968-1969 Delegate of National Research Council to 3rd International BiometriesConferenceBellagioi. Italy, 9/1-5/1953 The.Institute of Mathematieal StatisticsPkmber1941 Fellow August29, 1944 Council 1949-1950 Psychmatt:ricSociety Council of Directors, Oct. 1, 195l+-Sept. 30, 1957 Member 1935 Inter-4hericcn Statistical Associat'ion. Member 1963
Page 31: azo61e00
27 4 Seiccted ?'ublieetions: The use of the individual pa.rtsofthe aptitude test for prediMing snccessof students. Iowa Acadcmy of Sci. Proc: Z8:225•227. 1931 (with George W. Snedecor). Disproportionate subclass numbers in tables of moltiple classificatii'on, Iowa Agr. Exp. Sta. Res. Bul. 180:233-272. 1935 (Witb~George W. Snedecor) Covariance used to analyze the rela.tion between corn yield and acreage. Jour. Fkrm. Econ. 18:597-607'. 1936 (vith George W. Snedecor) Analysis of covariance of yield and time to first silks in maize. Jour. Agr. Res. 5k:449-459• 1937 (vith,Willi.m,P. Martin)~ Use of discriminant function for differentiating soilsuit~h different asot:abacter pomLtations. Iowa State ColSege Jo„^. Sci. 11:323-332. 1937 The multiple factor theory intermsn of common elements. Psrchoaetrika 4:59-68. 1939 (with Martin G. Weiss) Balanced incomplete block and lattice square.designs for testing yield differences among large.numbers of soybean va ieties.. Iowa Agr. Ekp:Sta..Res. Bul. 257:290-316. 1939F]mmeration and construction of balanced incomalete.block configurations. Ann. ofidath. Stat.., 11:72-85. 1940 Statist'ics as a tool for research. Jour. of Rome Econ. 36:575-580. 1944 (with W. G: Cochran) Designs of greenhouse experiments for st'atistical analysis. Soil Sci...6P:87-98. 1946 (with W: G. Cochran) FTPERIMENTAL DESIGNS. John Wiley and Sons Inc., N.Y. 1950 Statistical frontiers. Journal Amer. Stat. Asso. 52:1-12: 1957(vith W. G. Cochran) Revision of ESPLRI,+Q+TkL.DESIGI7S. John WL'ey and Sons, . Inc. N.Y. 1957, . (with W. S. Connor) Methodology for estimating reliability. Ann. Iasti...ofStatis. Math... The Twentieth Anniversary 16:55-67... 1964. The Biometric Societyrthe firsttwenty-five years(19k`fr1972). Biometrics 28(2)',:285-3111. 1972 (xith.Pau1,G: Aomeyer), Professional and person.=:%limpses of George W. Snedecor.. Biometrics. 31(2)C265-301. 1975
Page 32: azo61e00
FOREIGN COUNTRIES VISITED Country Date Left U.S. Date Returned to U.S. Purpose England, Norway, Sweden, Aug. 12, 1949 Oct. 27, 1949 To attend the 26th ISI meetings in Switzerland Denmerk, France, Belgium, Switzerland, I.etherlends Guatecala, El Salvador, Oct. 31, 1951 Nov. 16, 1951 Consultant for U.S. Government Costa Rica Guatemala Mar. 13, 1953 Mar. 29, 1953 Consultant for FIFD England, Belgium, Netherlands, Aug. 11, 1953 Nov. 5, 1953 To attend 28th ISI meetings in Italy Italy, Switzerland, Yugoslavia, Lebanon, Egypt, Union of South Africa anda, Southern U , g Rhodesia (Rhodesia) Venezuela, Brazil, Argentina, Paraguay, Peru, Guatemala, Puerto Rico June 17, 1955 Jhly 26, 1955 To attend 29th ISI meetings in Brazil (VO England, Sweden, Denmark, July 25, 1957 , Sep. 17, 1957 To attend 30th ISI meetings in Sweden Cermeqy, Netherlands, Belgium, Swit erl d It l F z an , a y, rance Belgium, Switzerland. England Aug. 31, 1958 Sep. 21, 1958 To attend 31st IS I meetings in Belgium Honduras Jan. 26, 196G Feb. 6, 196G _ Program Chairman Consultant for United Fruit Netherlands, Switzerland, May 11, 1960 July 4, 1960 To visit relative in Netherlands and attend 32nd Thailand, Japan ISI meetings in Japan. Program Chairman France, Switzerland Aug. 25, 1961 Sep. I2, 1961 To attend 33rd ISI meetings in France . England, Belgium, Switzerland Aug. lo 1962 1962 Oct. 2 To visit relative in Australia Consultant for , - Italy, Kenya, Uganda, Tanganyika, , , . ISI; Chairman, gducation Committee Southern Rhodesie, Australia "I 89006SE0
Page 33: azo61e00
P Country Date Left U.S. Date Returned to U.S. . Purpose Canada Japan, Thailand, Jordan, Egypt, Aug. 21, 1963 1964 July 21, Aug. 29, 1963 Sep. 29, 1965 To attend 34th ISI meetings in Ottawa, Canada Statisticel Education Advisor to Govcrnnent of Syria, Lebanon, Turkey,Creece, Itelp, Switzerland, Austria, _ - - Thailand, 7/25/-9/4: Crogrnm Specielist, Ford Foundation, University of Cairo, 9/5/64-9/21/65- vi a i gel d Yu oslc th ISI ti T tt d T Gcmeny, I'eth^.rlnnds, Belgium, Yugoslavia, England a g e, g + mee ng n o n en e 35 Fiji Islands, Australia, Singapore, Malqysis,Thailand Aug. 4, 1967 Oot. 1, 1.967 Attend VI International Biometric Society, Aug.-20-25; 36th ISI meetings in Sydney, Australia, Se 8 - p. 7. Aug. 2 `reden, Switzerland, Netherland, Ireland, Scotland, F.ngland, Wales Aug. 14, 1968 Oct. 4, 1968 Touring Greece, Thailand Dec. 5, 1968 Jan. 10, 1969 Conaultant, Ford Foundation's Bangkok Office England Sep. 3, 1969 Sep. 11, 1969 Attend 37th International Statistical Institute, London Italy, Greece, Turkey, Egypt, .. Dec. 1, 1969 Dec. 22, 1969 Attend the Inaugural Conference of the Scientific - Lebanon - the Institute of Statis_t_ic_a_1_ Computation Center end 16-19 1969) Studies end Research (Dec . , "~M Ireland, Wo3es, Scotland, England, France, Belgium, Germany, Austria, Sryitzerland, - July 22, 1970 Aug. 26, 1970 Touring;attend VII International Biometric Conferencf Ifannover, Germany, Aqg.--15-21,-1970 Liechtenstein, Netherlands . Thailand, Hong Kong, Taiwan, Jaaan - Nov. 1970 May 25, 1971 Work for Rockefeller Foundation; consulting and teaching at Kasetseit University - Scotland, Wales, England, France, Sep. 27, 1972 Oct. 30, 1972 Touring Portugal, Spain, Switzerland - - Soitzeriand, Austria, Greece, Iran, India, Theiland, Indonesia, hil J Aug. 16, 1973 Oct. 19, 1973 39th ISI meetings in Vienna, Austria ippines, apan P Ge,many, Polend, England Aug. 25, 1975 Sep. 19, 1975 40th International Statietical Institute, Warsaw, Poland - - -- - - ssoosseo
Page 34: azo61e00
30 Dr. KASTENBAUm. Dr. Gertrude M. Cox is professor emeritus of North Carolina State University, where in 1940 she became the first woman department head of that institution and the founder of the Institute of Statistics of the greater University of N orth Carolina. A building on the campus at Raleigh bears' her name. She is a member of the National Academy of Sciences. She has been elected a fellow of the American Association for the Advancement of Science,, the American Statistical Association, the Institute of Mathematical Statistics, and the American Public Health Association, She has served as a consultant to the Rockefeller Foundation, the Research Triangle Institute, the National Institutes of Health, the Bureau of the Budget, the National Research Council, and the De- partment of Health, Educationy and Welfare. In response to other consulting requests, she has traveled to many countries in the Far East, Near East, Central America, Africa, and Europe. She has written numerous scientific and statistical articles as well as an advanced statistical text entitled "Experimental De- signs." Dr. Cox's views of the FDA decision are best summarized in the following two paragraphs from her submitted statement: It is my considered judgment that the FDA has not correctly evaluated some of the statistical analysis information available when it reached its decisions on new labeling requirements in regard to oral contraceptive pills : and smoking. This probably is a good example where sound professionat statistical advice should have been secured and used before government policy decisions were made. There are several capable statisticians in both HEW and FDA, but were they used to help with the analysis and interpretation of the research results? There is little doubt that properly designed studies could provide much sounder informatiom regarding these questions which concern FDA and the public. Mr. FOUNTAIN'. Thank you very much for your testimony; Dr. Kastenbaum. I will ask each, of the nongovernmental witnesses this question, starting with you. According to your testimony today, each of the studies has defects which, you say, on an individual basis do not permit a, scientifically valid conclusion to justify FDA's decision to require the antismoking warning in OC labeling. Do you have an opznlon as to whether collectively the results of the studies considered would provide a valid basis to support this decision,,and if so, what is your oplnion ? Dr. KA6TENBAIIM. Congressman, based on the evidence that I have examined, I do not feel that collectiveliy-and certainLy not individ- uall,y-these scient'ific papers justify the decision that the FDA made. Mr. FOUNTAIN. If these statistical deficiencies of the studies that you describe are so serious, do you'-have any explanation or any thoughts as to why they were not considered by FDA and its advisory com- mittee when they made the decision to require a boxed warning? Dr. KASTENBAUM. I have had the. opportunity to read the transcript' of the proceedings, of the meeting of the Obstetrics and Gynecology Advisory, Committee in N ovember 1977. There were some comments in that transcript by an FDA statistician named Mr. Litt, which confirmed my conclusions about the validity of' the statistical data being considered by the advisory committee: I
Page 35: azo61e00
31 have reason to believe, therefore, that the advisory committee was advised about the statistical weaknesses of the studies under considera- tion. I do not know why they chose to ignore this advice. Mr. FOIINTAIN. Are you familiar with recent papers by Jick and associates and Petitti and Wingerd on the question of increased risk of circulatory disorders to OC users who smoke? . Dr. KASTENBAIIM. Yes ; as I indicated' earlier, I do scan the literature as part of my job, and I am aware that these papers appeared in the literature some time this year. As I recall, there were three papers by Dr. Jick, two of them in the. . Journal of the American Medical Association which had to do with a study of 107 women. One dealt with oral contraceptive use and nonfatal myocardial in- farction, and another with noncontraceptive estrogens and nonfatal myocardial inf'arction. As I recall Dr. Jick concluded that physicians be advised not to prescribe the pill to women predisposed to certain illlnesses: He also added "particularly if they smoke." These are the last words of the paper. But his primary advice is that the pill not be prescribed for women who are predisposed to diabetes and hypertension. This is not what the boxed warning says. It tells women not to smoke. To me, this is, tantamount to putting a warranty on the pill by suggesting, that women who take the pill are safe as long as they don't smoke. Dr. Jick is not saying this: He is saying that when women are predisposed to certain illnesses, t'hepilll should not be prescribedl The third paper by Jick had to do with stroke. As I recall, in the summary of that paper he said that stroke was only weakly associated with cizarette smoking. The Petitti paper was based on a studw of 11 women. It is incon- ceivable to me that these authors could have arrived at their statis- tical conclusions on the basis of such a small number of women. For instance, they have found 11 women who are characterized as having had' a rare type of hemorrhage called the subarachnoid hemorrhage. This physical condition is outs:3e my area of expertise. Nevertheless, I do know, based on what the authors indieate, that this is a congenital type of illness afflicting women and men: of differ- ent a;es. It' certainl,y is not a new disease. Women who do not smoke or who have never smoked, women who never took the pill have been stricken with this disease for a long time. Yet none of the 11 women fell into the category of nonsmokers and nonpill users. Somehow they were all doing either one or both of these things. The authors seem to imply that you need to do either one or }rt.h of thPse thinfrs ' in order to suffer a subarachnoid' hemorrhage. They indicate that 72.7 percent of these women both smoke and take the pill. This means that they observed 8 out of 11 women who did both, but they observed no women who did neither. So. I question the fact that i2.7 percentl. • of their cases represent' 72.7 percent' of such cases in general. When they talk of smoking characteristics of the control group, they are looking at everybody; that is smokers, nonsmokers, pill users and nonusers, but the small sample of 11 women did not contain any person who neither smoked nor used pills. Small changes could affect the conclusions of this paper drastically. • ~
Page 36: azo61e00
32 Mr. Fou:vTAI:v. I gather from what you have said, you have had occasion to statistically evaluate the data in those papers. Your opinion, then, is that the data do not lend support to FDA's decision to require an antismoking warning on OC labeling; is that right? Dr. KASTEN-sAU:K. Neither before the fact nor after the fact~. Mr. FOIINTAIN. Mr. Wydler ? Mr. WYDLER. I am int'erest'ed in the remedy that you are pursuing here today. I can understand why the industry could be upset with what they consider to be an overreaction to certain reports on the part of the Government and possibly more important to them, I presume, the precedent it might set for the future. However, are there not other ways you could have attacked' this? What do you~ expect the committee to do? I am not sure what you~ think we can do about this situat•ion, Dr. KASTENSAUrz. My interest at this point is purely in the scientificc aspect of the FDA decision. The FDA alleges that. it did a study of the scientific evidence and based its conclusion on this scientific evi- dence. They enumerated an& listed the scientific evidence that they relied upon. I feel that the FDA conclusions are not supported by the scientific evidence. If a regulatory agency of this Government operates in thiss manner and if there are committees of Congress that have oversight over such agencies, then I think that such information should be brought to the attention of these committees. It is certainly not within my realmi to recommend what course of action vou should take. Mr. WYDLER. You see the thicket that you are putting us in. We are going to try to sit in judgment on what is essentially-a scientific matter and' evaluate, as a committee, whether FDA might have been justified to do it this way, or whether it should have been written differently, or whet'lier it should have been different words, or whatever. It could be boxed in or not boxed in. We could go on with that. We could have an awful lot of hearings if we would examine every admin- istrative judgment. I d'o not know what we would gain. Let.'s say that we would come ont with a finding that v.ou overreacted and it was not' justified to make such a fuss about it as you have made. Where does that leave us? At the end of that road where are we really? I'm just curious about this. Wh,y didn't the industry try to obtain relief administratively? I don't know whether they did or not. It seems to me they should have tried that first. Failing that, maybe they could have taken some iudicial review of the matter. You could let the courts determine this. Generally, when somebody goes too far, the courts could step in. Dr.KASTENBAva1. I believe that it is notl necessarily the tobaccoin- dustry that has been harmed by this decision. I think that harm may have been done to a large segment'f of the American public; namely, women of child bearing age. In that sense, perhaps it is within the jurisdiction of' a conaressional committee rather than in the courts. Mr. WYnLER. As I understand it, the point you are making to the committee today is this. The warning is predicated on~ insufficient . ,
Page 37: azo61e00
33 proof. If I understand what ,you said, you read various art'icles and decided that there did not seem to be much proof. Then the question becomes : What is the responsibility of the Gov- ernment official where he has some hints2 How much does he require? How much warning does he give? These are judgment calls,,to a great extent. So, as a member of the subcommittee, I am trying to grasp what I am supposed to do about this situation in order to see that the American public is treated fairly and that the indlzstr,y is treated fairly, whether it is the tobacco industry or any other industry that might have a product on the market. How does the Congress set standards in this area ? That is really what we are talking about. Dr. KARTENBAIIM. I can appreciate your dilemma, Congressman, but I don't have an answer for you. I have my own personal views on this. but I do not think they would be appropriate. Mr. WYDLER. Did' ,you ask for administrative relief ? Was an ap- peal taken? Was an appeali made to the agency t'o d'esist from doing this?' Dr. KASTENBAUM. Based on my recollection of the chronology, we were faced with an accomplished fact. 11ir. WYnLEx. An order must have been issued requiring oral con- traceptive manufacturers to place this warning on their packages. I presume that was the way it worked. Is that riaht ?' Dr. KASTENBAUM. I am sorry. I am afraid that is out of my areaa of expertise. Mr. WxnLEB. All right. Then I will move away from that. I will ask future witnesses about that. If I understand your professional opinion correctly, it is that this warning as written and as published on the oral contraceptive boxess and containers is not justified by the scientific evidence available; iss that ri:rht? Dr. KASTENBAUM. This particular type of warning; yes. I am not arguing or necessarily agreeing with the FDA about. its desire to indicate that smoking, is contraindicated. Rather it is the way t'hee smokinfr warning is worded and the fact that it is highlighted by a box that worries me. It is being cailed~ to the attentiion of women who use, oral contraeept'ives. Most women will see this and' nothing else, and' they may feel reassured about the safety of the pill as long as thev do not smoke. I am questioning the honesty of this nroceduTe.. Mr. FOUNTAIN. Mr. Wvdler, in reading the statements last night. I came to the conclusion that one of the things that they are question- inp,, is that prior to January there was no such labeling. In January this warning was suddenly required to be put in a boxed format in the OC' labeling and they were not given an opportunity to com- ment• on it. Mr. WYVLEte. I don't know exactly what the administrative rules are. I don't know if they followed correct procedures or not'. That is a question of administrative law. I am sure the institute has a num- ber of lawyers available. I just do not know what they have done. I am not exactly sure where we are on that.
Page 38: azo61e00
34 I am questioning your forum here. It is interesting to hear this. As a general proposition, I can see where the Congress has to concern itself with what the agencies do. I am not against overruling them~ when I think they are wrong. I do not want you to get that idea. However, I do not know about this. If we are really the forum! of original! jurisdiction5 I am not sure. Maybe this should' be handled somewhere else to see if it was proper to begM with. I start with~ the presumption that some of' the language in the box is probably consistent with the warnings that Congress has already required. Cigarette manufacturers are required to put the Surgeon General's warning on all' of their products. I don't know what the words are exactly. I think the evid'ence, in my judgment, is pretty irrefutable that cigarette smoking is bad for people; it.causes a lot of complications. It is not the only thing that is bad for people. There are many other things that are bad for people. I understand all of those arguments as well, but' cigarette smoking generally is something you do nott do to improve your health, whatever the reason you may have to smoke. This warning seems to say in this particular case that it is superbad. From what you have said, frankly, I n-.oultl say that it is an over- reaction. It seems to me they have taken a few general articles and' made a mountain out of them. Maybe we will hear why later today. But I think up to this point you have made that. point. However, the point I am trying to make is this: I do not know exactly what this committee can do other than to say to the agency, "Do not do it any- more," or something like that. Or perhaps we could say, "Withdraw your order." I do not even~ know if we are in a position to do that. So, I am just questioning the way the whole thing has come about. I think you might do better somewhere else. That ism,y thought. In a court, at least, they say t~o the agency, "cease and desist." Theree you are pretty sure the agency will do what they are,told. In our casee we can tell them not to do it, and they may or may not pay any atten- tion to us. I have been around the Congress long enoughi to know thatt sometimes they listen to what we say and sometimes t'heyy do not. Thank you, Mr. Chairman. Mr. FoLNTnr.N. You make a good point. However, there is the point that public disclosure of all't.he facts willi enable the people to decide f'or themselves to some extent what the situation is. Mr. W-I-nLFx. The only trouble with that, Mr. Chairman, is this. I am afraid that if it is determined that the warning is not j;ustified, assuming that we were to make such a: determination, and if we said to the agency ini our report, "You, should not have put it out and you should not have it on there," and if they go on and leave it in effect' theni the public is still going to be in the, same position they are now. They are going to be reading that warning and thinking it is right. I do not know if we would accomplish the results you are after. That is the only point I am trying to make. Mr. FoLxTnix. I do not know if we ever accomplish much more.in these public hearings than to disclose the facts and let the public judge. You are quite right. These are not easy decisions to make. However, I think alli positions and points of view are entitled to be heard.
Page 39: azo61e00
r 35 If we should, of course, discover that in our opinion the Food and Drug Administration erred, then I think we should say so. If we are not in a position to do so, then I thsnk it would be a matter of public record as to just what the evidence does show. The public can then judge for themselves. Atthis time we have a vote on the floor of the House. The subcom- mittee will recess until we can return. The subcommittee stands in recess. [Recess taken.] Mr. FOIINTA'IN. Thesubcommittee will come to order. Our next witness is Dr. Jean D. Gibbons,,professor of statistics,,and chairman of the applied statistics program at the graduate schooli of the University of Alabama. Dr. Gibbons, would you please come forward? We will be glad to hear from you at this time. STATEMENT OF DR. JEAN DICKINSON GIBBONS, PROFESSOR OF STA- TISTICS, CHAIRMAN OF THEAPPLIED STATISTICS PROGRAM AT THE GRADUATE SCHOOL, THE IINIVERSITY OF ALABAMA Dr. GI$BOxs. Thank you, Mr. Chairman. My name is Jean Dickinson Gibbons. My current position is pro- fessor of statistics and chairman of the applledstatistics program at the graduate school of the University of Alabama. I received the bachelor's and master's degrees in mathematics at Duke University and the Ph. D degree in~statistics from~ Virginia Polytechnic Institute and State University. My previous faculty appointments were at the University of Pennsylvania and the University of Cincinnati. I was a senior Fulbright-Hays scholar at the Indian Statistical Institute in 1973. I have been associate editor of the American Statistician for 6 years, and have served as editorial collaborator on severall statistical journals, including the Journal of the American Statistical Associa- tion, the Annals of Statistics, and Biometrics, and am presently serving as a reviewer for grant proposals for the National Science Foundation. I am a member of several professionall societies and have served two terms on the board of directors of the American Statistical Associa- tion. I was elected a fellow of the American Statistical Association in 1972. My publications include three scholarly books in statistics and over 25 articles in refereed professional and learned journals in my field. Mv current curriculum vitae has been submitted to the subcommittee. Mr. ForNTAIX. Withotrt objQctlion; your curriculum vitae will be inserted intothe record at this point. - [Dr. Gibbons''curricnlum~ vitae follows:]
Page 40: azo61e00
36' Address: Box J, University of Alhbama University, Alabama3548fi Te1..205-348-6085or 205-34i-3750 SUPIDlARY.OF. INRORNASIOWCONCERNING JEAN DICKINSON GIBBONS Present Positdonc Board of, Visitors Research~.Professor of Statistics (1974- ) and: Chairman of Faculty Committee.for the Program in Applied Statistics (1975- ),and~ Professor of't".athematics.(QP76- ), qraduateSchooL, University of Alabama Personal Data:. Age 40; Female; Excellent health; Married to John S. Fielden Positions Held': 1971-1974: Chairman,. Department of~ Statisticsand Quantitative NeChod's, College of Commerce and'Graduate School of Business Administration, University of ALabama 1970-Present: Professor of Statistics, University ofAiabama 1973Summer:Research Associate, Stanford University. 1968-1970: Associate Professor, Department of Statistics and Operations Research, University of Pennsylvania, 1963-1968: Assistant Professor, University of Pennsylvania 1961-1963: Assistant Professor of Mathematics, University ofCincinnati'1962-1963:. Consultant in Biostatistics, College of Hedicine, University of Cincinnati 1958-1960:: Instructor of.Nathematics,.Kercer University Professional Recognition: Fellow.of American Statistical Association, Elected 197.2 Education: A.B..magnacum laude, 1958, Duke University, Nathematics M.A., 1959', Duke University, Mathematics Ph.D., 1963,. Virginia PolyteehnicInstitute, Scatiscics At2ended Columbia Universitg full time. 1960-1961 Academic Honors: PhLBeta Kappa (Junior year), A.B. maana cum laude Phi'KappaPhi (graduate scholasticc honorary). Pi Mu Epsilon (mathematics honorary)~ Chi Alpha Phi Chanter.of Nu.S{rm:n ?ho Beta 6amma Sigma (husinesshonorary). Biographical sketches in:American )Senn and',n tiomen. of Science DictienaryofInternational Biogranhy OutstandingEducatars of.America Personali~ties of the South The Ttorld sthc'sbfio of; Nomen Who's Whoo of American biomen who"s Nho.in the South and Southwest (statiscdcshonorarv). 1 .
Page 41: azo61e00
37 International' Professional Activities: Senior Fultiright-Ha,vs Sch'oLar, Indian Statistical Inst'itute, 1973. Delegate to~YATO,Stat'istics Conference, Greece, 1972 Delegate to~VS InternationalBiometricsCongress, Australia, 1967, Professional SocietyNemberships:: American Statistical Association (p.S.A.). Institute of, Mathematical Statistics (I.M.S.). Alabama Chapter of A.S.A. Current National Professional Activiti'es: Executive Cammitteeof, the A'.S.A. Section on,StatisticallEducation,. 1977-79 (nationallyelected:o€fice). Southern Regional Education Board Ccermittee.on Statistics, Universityof. Alabama Representative (1972- ).;',Treasurer, 1978-80 Associate Editor, Thedmerican.Statistician, 1972- A.S.A. Advisory Committee too the ULS.. Office of Management and Budget, 1976-80 Reviewer for grant proposals to National Science Foundation, 1976- EditorialiCollaboratorfor various stacistics journals, including (at various times) lournaliof~ theAmerican, Statistical Association4 Communicationsin Statistics, Biometrics,, Sankhp3 „ The Annals of Statistics,. Technometrics,.Journai ofEducational Research. Selected Past' Professional Activi~ties:'. Board of, Directors of A.S.A.., elected for 1975-77 Chairaan,A:S.A. Committee on Meetings, 1975-77Program,.Chairman,for the1976.annual meeting of the A'.S.A'. Advisory Board of National Cancdr Institute, DHEW, 1974-76Presidpnt,,Alabama State Chapter of the American Statistical Association, 1974-76 Chairaan, NationalComm2ttee on Women in.Statistics of A.S.A.,. 1972-1976 Lecturer forV.isit.ing.Lecturer Program.of, A.S.A. and 1973-1975 Represencative-ac-Large to~the CounciliofA'.S.A., elected for 1972-74 Constitution Committee, A'.S.A., 1i972-74. . Committee on Elections, A.S.A., 1972-71- Recent Publications Editor, The?.merican Statistician, 1970-73 Ca-Director, CBMSandN~SF Conferenee.onriultivariace Statistical. Analysis,.Summer 1973 Vice-President,. Alabama State.Chapter of the xmericanStatistical Association,. 197,2-74 Presidenc, PHi Beta Kappa Chapter of the University of Pennsylvania „ 1967-69
Page 42: azo61e00
38 PUBLIC.ITIONS OF JE.1M D. GIBBONS Books: 1. Nonparametrie Statistical'Inference, McGraw-Hi1L Book Qo., New York, 1971, pp. 306.Thisgradieate level reference and:textbookis included in the McOraw-Hi11 Series in Probability and Statis- tics, edited by, David Blackwell and Herbert Solomon, and also inthe International Student Edi'ticn Program~. 2. Nonparametrie Metkiods for Ouantitative Analvsis,. Ho1t,.Rinehart and:Winston, Inc., New York, 1976, pp. 463. This graduate level textbook, reference or handbook is included in the International Series in DecisibnProcesses, edited:by Ingram 01kin. 3. Selecting and'Orderina Pomulations:: A yew Statistical Meth- odology, eo-auchored with Professor Lngram 01kin, Stanferd University,,,and Professor Milton Sobel, University of California,.. SantaBarbara, John '7ilev.6 Sons, New York, 1977.,.oo: 569. This is the verv first book on this topic; it is a.volume in the "iley Series in.Prohability.and Mathematical Statistics. - 4~ Concentsof Nonoarametric Theorv, co-authored:with,ProfessorJohn~r tt. Pratt, HarvardUniversity;yanuscriot submitted to John 'iilev, for consideration. Refereed;Publications: 1. "On the Equiprobability of Two Rank Orders.," Abstract withH.A..David, Annals of Mathematical Statistics, 1963,.Vo1. 34, 35'7. 2. "Effect of'Yon-Normality on,the Power of the Sign Test," Journal of the American. Statistical Association,.1964!, Vol. 59, 142-1Z- 3. "On the Power of Two-Sample Rank Tests on theEquality of Two Distribution Functions," Journal of the Roval Statistical Societv, B, 1964, Vol. 26, 293-304. 4. "A Proposed Two-Sample RankTest: The Psi Test anditsProperties," Journal of the Rovali Statistd.cal Societv, 3, 1964,.Vo1. 26„ 305- 312. 5. "Small Sample Properties of Percentile Modified Rank Tests,"'Tecti- nical Renort No.60,. Department ofStati§tics, ne.JohnsHopkins University, November 1966,.with Joseph.L. Gastwirth~as second author. 6. "A Correlation Measure for Nominal. Data „"The American Scatisti- cian4 December 1967,. Vol. 21, 16, with KennethH.. Ives as first author... 7. "Correlation Coefficients betweenNonnarametric Tests for Location and Scale," Annals of the Institute of Statlstica'- `!athematics,. 1967, Vol. 19, 519-526. 8. "MutualiyExclusive Events, Independenca and Zero Correlati6n," The American StlatistYcian.,,December 1968, Vol. 22„ 31-32'.
Page 43: azo61e00
39 9. °Estimationofn the Number of'. Critical Sites in Limited'Genome. Expression During.Viral Infection of Bacteria," Biometrics,.1969', Vol. 25, 537-544, with~Samuell Litwin~as second author. 10. "Properties.ofthePercentile Modified Rank. Tests,"'Ahnals of theInscitute ofS.tatisti'ca1 Mathematics Supplement, 1970,.Voli..6~ 9.5-114',.withJoseph.L. Gastwirthas second author,, 11."'A Unified Approach~.to 63ypothesisTesting.," Estadistica, 1970~,. Voli.. 28... 12. °A..Distribution-Free Two-Sample Goodness-of-Fit Test for General Alternatives," British Journal of MAthematicaliand Statistical Psychology, 1972, Vol. 25, 95-106. 13L "On the Design of a Random Alarm," Journaliof Physics - E:Seientific Instnanents, 1972, .VoL. 5, 634-637 with Sw.uel Lit•.+in as second author. 14. "Comparisons.of Asymptotie.and; Exact Power.for Percentile Modified Rank Tests," Sankhya, B, 1973, .VVol. 35,. 15-24, 15. "A'Question ofEthics," The American Statistioian, April, 1973, Vol. 2Jy 72-76. 16'.. "Estimation of the Unknown Upper Limit of a Uniform Distribution,:"Sankhv-a„ B, 1974, Vol. 36.,: 29-40. 17... "Reporting,P-valuesas an Aid to Decision Making," competitive paper in Proceediagsof.Southwest_Conference Meeting of American Institute for Decision Sciences, 1974, Cincinnati, Ohio, 18-20. 18L "The Visiting Lecturer Program," Letter in The American Statis- tician, 1974, Vol.28~ 35. 19. "Simultaneous Estimation of the Unknown Upper and Lower Limits inn a Two-Parameter.Uniform Distribution," Sankhva, B, 1974, Vol. 36, 41 - 54, with..SamueliLitwin.assecond authoi. 20. "P-values: Interpretation and Methodology," The American Statis- tician„ 1975, .Vol. 29,.20-25y with John FI. ?ratt as second author. 21.. "The Status of Statistics in~.the Seventies," Letter in The Amer.ican. Statistician,.,1976'., Vol. 30, 150. 22. "Comparing.the Mean and the Median as Measures of, Centrality,:"' InternationaliStatistical Review,1977', Vol. 45, 63-70, with,Gordbn R. Stavig as first author. 23.. °Basetialil Competitions--Do They. P1ay.Enough~Games?.,"The Americaa Statistician,,1978, Vol. 32 gg-ry5, -ith Ingram 01kin and Milton Sobel. 24. "An Tntr,oductionto Ranking and Selection,":fie American Statistlcian,. 1978,, accepted with revision„ with :ngram01kin and Milton~Sobe1... .
Page 44: azo61e00
40 Other Publications: 1. Author of statistics section, pp.. 106-108, ofYontraditional Careers for Women by Sarah Splaver, Julian Messner, YewYork,,1973. 2'. Editor, Roster-of Women in~5tatistics, July. 1974, .'IThe American Statis- tical Associationj Washington, D.C. 3: "Nooparametric Statistical Methods,"by:7yles Hollander and! D.A. Wo1fe, John Wiley, New York, 1973, Review in T_echnometrics, 1974, Vol. 16, 477-478. Publications in Preparation: 1. Parameter Measures of Skewness, witH,William G. Nichols, submitted for publication. 2. Some Parametric Properties of Distributions of Yonnegative Random Variables, with William G. Yicholis. 3L Application of Subset Selection,to.the Scoring.of.Testsy with Ingram 01kin and Milton Sobel, 4'. orvanizations for Statistical Consulting at Colleges and,iiniversities, with R. J. Freund. Recent Colloquia Presentations: v.P.I., February1B76SREB Summer Research,Conferenee,.June1977N.C.State, October 1971' O.S.C., Februarv1978, Alabama Chapter of ASA,.'*ohile, June 1978 1
Page 45: azo61e00
41 Dr. GIBBOxs. In February of 1978, I was asked to review the reported statistical studies that formed the basis for the FDA's decision to include a boxed warning in the patient and physician labeling on oral contraceptive products. This boxed warning states that cigarette smok- ing increases the risk of serious cardiovascular side effects for users of oral contraceptives; this risk increases with age and heavy smoking and is quite marked in women over 35 years of age and therefore oral contraceptive users should not smoke. I have submitted for the record a copy of the letter dated February 23,,1978, from Dr. Marvin A. Kastenbaum. Mr. FOIINTAIN. A copy of this letter will be inserted' into the record at this point. [The letter referred to follows :] THE TOBACCO IN6TITUTE, Washington, D.C.,,February 23, 1978. Dr. JEAN D. GISSONS, Chair, Department of Applied Statistics, University of Ala'oama, University, Ala. DeAn JEAN:: I am writing to confirm our telephone conversation~ in which you agreed to examine the enclosed material before making a decision as to whether or not you will review the statistical evidence used by the FDA in reaching its decision to warn oral contraceptive users against smoking cigarettes. Enclosed are the following items : 1. Federal: Register, vol. 43, No. 21-Tuesday, January 31, 1978, pp. 4214-4234. 2. Reference 52„p. 42ii0 (Jain). 3. Reference 53,,p. 4230 (Beral) 4. References 11,,12,13, p. 4229 (Mann, et al.). 5. Reference 14, p. 4229 (Tietze). 6. Ory, H. W., J.A.M.A., Vol. 237,, pp. 2619-2622, 1977. 7. Lawson, D. H., et aL, British Medical Journal, vol. 2, pp: 729-730, 1977. 8. Editorial, The Lancet, pp. 747-748, October 8, 1977. 9. Burch, P. R. J., The Lancet, p. 879, October 22, 1977. 10. Detering, K., and Hartmann, E.. The Lancet, p. 1023, November 12, 1977. 11. Haack, D. G., and McKean, H. E., The Lancet. p. 1023, November 12, 1977. 12. Beral, V., The Lancet, pp. 1047-1057, November 13, 1976. 13. Jainw A. K., American Journal Gynecology and Obstetrics, vol. 126, No. 3, pp. 301-307, 1976. Items 2 and 3 are the principal sources of evidence in the FDA report; items 5, 6 and 13 support or supplement! this evidence; item 4 is really three references from which data were excerpted to carry out the statistical' analysis in 2, 5, 6 and 13. Items 7-12 are publications which tend to indicate that some,debate oni this subject was still raging in the medical literature at the time of the FDA decision. If you should decide to undertake this review, I will, of course be available to answer any questions you may have. However, it will be best if you look at the enclosed materiall with as much objectivity as is possible in doing your analysis. I would hope to receive ai written report from you on or hefore March 27, 1978. This report should be your objective appraisal of the statistical evidence relied upon by the FDA in making its decision concerning cigarette smoking by oral contraceptive users. At the time the report is submitted, you may also send a bill based on an hourly rate for your services, plus any additional charges;, such as stenographic,, reproduction, etc. that you have incurred, If you need any other background information please do not hesitate,to contact me. Best' wishes. SincerelS„ ;1SABVINA. KABTENBAUM.
Page 46: azo61e00
42 Dr. Gmaoxs. This letter requested that I analyze the studies relied upon by the FDA with complete objectivity. I agreed to undertake such an analysis. The FDA's decision to warn oral contraceptive users against smok- ing on the labels of the contraceptive was apparently based primarily on the papers by Jain (1977) and Beral (1977). Jain used' the data in Mann, Vessey, Thorogood, and Doll (1975) that consisted of 63 women under age 45 who had survived a myocardial infarction (MI patients) and a control group. Even, though the data are very confusing, it seems quite evident to me that the numbers of women~ observed' in each category are much too small to allow any reliable statistical con- clusions about the interrelation of smoking and' oral contraceptives on the incidence of' MI. Table IV of the Mann et al. paper shows that there were only 14 nonsmokers in the MI group while 74 of the controls were nonsmokers. Table VIII shows that there were only three nonsmokers in the MI group and 13 smokers who were using the oral contraceptive at the onset of the MI episode. These numbers are too unbalanced to draw reliable conclusions; further, the numbers are much too small: Jain (1976, page 302), mentions t'he limitat'ions of the small samplee sizes for the data of table X of Mann's paper (1975). Jain (1977) does concede that the mortality data used in his paper are based on small numbers and may be subject to large sampling errors. This caveat on the author's part certainly should not be ignored by anyone who uses~ his conclusion to argue a point of view. Further, his analysis is based'on the following two unwarranted assumptions: (1) The smoking characteristics are the same for both the fatal MI group and the nonfatal MI group; and ('2) the propor- tion of smokers among, older women~ is the same as the proportion among younger women. These two assumptions are equally unfounde& albeit convenient. Jain (1977, p. 51)i claims hemade these assumptions, and I quote, ". . . becausethe relevant data ... are not available." That is a non seauitur, to sa,y the least. If sufficient and relevant data are not avail'~- able, no proper statistical analysis of any type can be made and cer- tainly no assumption can be j,ustified: Beral (1977)~ used different data which are also confusing. All of the numbers are still very small, especially the number of deaths for nonsmokers. She points out (p. 730) that : These estimates are based on small numbers and are necessarily approximate: Without more data it is not possible to examine the interrelationships of age, smoking, and durationiof oral contraceptive use.... I certainly agree with the author on this point. Given that the data available are. so limited and are not reallv relevant in most cases, there seems little point, in discussing t;he statistica.ll methodolbgy used in the Jain and Berali studies. Further, it would be impossible to verify the, result's given by Jain without spending many hundreds of hours trying to figure out %chat data were used and where the data came from.
Page 47: azo61e00
43 Nevertheless, I must conclude that even if Jain's calculations of risk associated with smoking and oraL contraceptive use were verified arithmetically, the reliance on these studies as support for the boxed warning shows alack of understandyng of what can and cannot be shown using proper statistical procedtires: The warning as stated implies that there is a causaL effect of cigarette smoking on the inci- dence of cardiovascular disease. No statistical study can establish causalit'y. At best, it can establish a high probability of possible inter- relationships. Neither of these studies illustrates the proper use of statistics. If the FDA wants to obtain and provide usef'ul and accurate infor- mation about the relationship between smoking and the use of oral, contraceptives, it should run a controlled' experiment, or have some- one else run a controlle& experiment, to obtain sufficient and relevant data, which should include information about many factors not con- sidered in the two data sets on which• the FDA based its decision~ for example, length of time on the pill ; number of years smoked ; genetic, environmental, and psycholbgical' characteristics; among others. Thank you. Mr. FovxTnix. Thank you very much. Without objection, the materiat relating to the study references in your testimony will be inserted in the record at this point. (f Mat!erial accompanying Dr. Gibbons' statement follbws :]
Page 48: azo61e00
1 44 Bibliography of Papers Cited 8era1, Valerie (1977). Mortality among oral contraceptive users. The Lancet, October 8, 1977, 727-731. Jain, Anrudh K. (1976). Cigarette smoking, use of oral contracep- tives, and myocardial infarction. American Journal of Obstetrics and Gynecology, Nol. 126, No. 3, 301-30 . (1977Y. Mortality risk associated with the use of oral contraceptives. Studies in Famiiy Planning, The Population Council, Inc., 49-54. Mann, J. I,., M. P. Vessey, Margaret Thorogood, and Sir Richard Doll, (1975). Myocardial infarction in young women,with special reference to oral contraceptive practice. Brit'ish,Medica1 Journal, 2, 241-245. W
Page 49: azo61e00
45 TUESDAY, JANUARY 31, 1978 PART I! _ 55-727 0 - 79: -4
Page 50: azo61e00
46 1214 ' . RULES AND REGUIItTIONS .[~I10-03],~ ---_- ~ theJudt(ment of the prescribing physl- ed to preclude a repuirement of label- ne. ]t~.d re -... .. tlaa the declsfoa resardlmi the coa- 1.9 dlmtsd'to the.PatMnt.whleh willD"e'~' _ tent and extent of cautlonary.and dl-Promote eafe and effective uee of - WAr'ral t-lOODAND Dtpa ADMINISTtA- roc'tlve lnformatlom to be made avall- druEa. The CommlSiloner noteas that _ rlOlp DEtAlTMENt' OE NEA1M iDDCa. able to.tl]e patlent. -. -'h1s authority to issue patient labeliIDE TqN, AND WEVAnE '. --_ T[te. CommlaflOner dlaairees ~with' reQltlrementA -has beeG 7rel/.nlna-w - these contentlotla The Food and:DruL . upheld by the one coutt that has re-.spsawtta.p-ematNxw.uNwE -_ AdmhtLttntlom's lefal authorlty: forvfewed'~ the matter.PAarrwceaticnJ~[DeakeLNo..7aN44a7)' -- reau117nE patient labeling was ei- dfan[tracturers AYsociaNon v. FDA; -_ plAlned .m detail In the Preamble to . Clv. No. 77-491 (D. DeL,. October 5, • EAE7 ]Ip_NEW DtU07 ~ .-.~ the ProDOSednew format for.presrrlp- . 1977). (order demyin[ PreUminaty -in. .. N.. O.up G~wn,n.nt ~hn l a.wy o(~.e.1 ttom drug labelln[ publl!lted In ~ the . lunetlon).~ .. : - . Fzrrzaa'REn tsiu.of Aprll'7, 1975 (40 . 1 Nanulaeturer aad dispeaserIt-'. . - .:-FR15392): and (twu alsodtsclived )n• eb'il(ty.Several commeatJ mntended. AGEPICY: Food aad,Druf Admtnlstn- thee preamble to the final rude.for pa-~ that patient labelln` could have a sub-. ._ t1on., tlentlabellnf for.estro9en dVut prod-. stantlal adverse effect on the le9al111- ACSION: Final ruDe. _'- -- ~UC+ published iro the FIDCIaa. REx:rs- abihtyy of . manufacturers and df.iPens- _ - - .~-_m of July 22,. 1977 (E2.FR 37939): era One mmment suggested that the , SIIMMARY: Th9 ru1e. revLlea~ the re-` TheComWstioner will not repeat thatt partlal-exemptlon fromm strlct tort ll- yutiementa for patient labeling for _. full dlwaaalon here. but he believes-_ abDlty afforded drug manufacturen 'oral!contrs<epU4e drug producta'3-h1a . that several pointe ahnuld: be.empha-_ where the drug product I!properly. action Is taaen-to provfde.eonsumerse sized. Section 505 of the~aet provides - prepared and: accompanled; . by ade- , with expanded labellns Informatiom that a new, drug applicatiom (NDA)- puate directions and warnlnEs would reflecting recent reports aboutt the may be approved only If a new drug Is be substantfally-eroded and posstblyrisk of, blood clots,:other problems of ahown to bee safe and effective In use- eLtnll[csted by a patientlalxlltl{ proh- - the ~circulatory syatem, cancerr and e4-h under.the conCltlons set forth 1n lts.la- sloa.'I'hls outcome would result. It waS fecta~ on the unbarn ch5ld- sstociated-be11nE'• and sqctlon 201(p) of the act aryued, beolae 1t1a extemely dlfflcoilt' ' with the use'of oral.cantraceptives.(21 D$C. 321(p)) slmllarly.provides towrlteundentandable.warninysaod. 'Phls new labelin;.wlllbe provldedtiy., an exemption from the reqnlrementd)rections directed to the layman, thedispenaertoeachpatienttowhomh foranNDAonlylf,the:dru[isEeneral- which would be deemed.le9ally.adc- LhedruE'.UdLspensed-•• -- ly recognized as aafe. and effective Quate. The fact.that patient libelDhi EPFECITYEDATO;Apr313.3978:~..••- -~der the-condttlons-seaforch In the. may havebeea required and drxfted'- labeling. Moreover~ -both-sectlona . by FDA would not protect the.manu- FOR FORT13gRINFORM_ ATION. 302(a)') and 505(d)') off thee act prohibit ftctunr from an adverse luryy determiJ CONTACi: PfescrtPtton dtu9 lAbelih7 that Is false natlon. on the 13sue of adettVaacy. - - Phi71p I: Faaulq Bureau ofDruE. or misleadini manyparticular,,andAnother.comment contended that, a (8FD+30); Foodand Dcu2 Adminis- - sectlon 201(n) of the.act exPDcitl.yproo-y patlent labeling requirement will Cratton.:Deparimant ot 8ealth.,Edu~ vtdess that the failure of, the-latielln` expose ph'armaelstaa toleEal liability utlon, and Weliare, 5600F`shen ~'reveal material facts will render Predllcated on the failure to,dLspense Lanq Rockville, Md..20857, 301+-N3- that labeling mlaleadlng•.Acroord)aEly. labeling or on the dispensing of:aron{ eS22D-'.. - -- - - the ~treQultes CHeCommivioner to or outdated labelina.and urped that . ~ - meke a determination that'thet infor- - FDAA consider this possible copse- SIIPPLF_MENTARY LYFOR13ATtON: mation contained in the labeling forr a aaemce before taklns Itnal act)on. In a notice publlshed'fn the Finanuprescnptlon drug Is sufficienato The Commissioner does nott aEree Rtorsxta otDecember 7, 1976 (91 FR enaute the safe and effective use oi that the teipoeltlon of a repu)rement. 53630), the Food and Drua Allminls- thedruE byy consumers_ The CommLO for patient labeling w(ll neceasarlYyy tratlon (FDA)') proposed to revise re- afoaer has eoncluded: . that.. in order affect advenelyy the standard of civll Dulrements for patient labeling fon-- th.at', censumen may safely use oral tort llabllityy which 1s Imposad on drug oral contt'acepnve drug products. In- contraeeptl:'e drug products„'specifie manuIIacturets or dlspensen. Whetherr terested penons were Eiven until Feb- Snformatlon must be provlded tp them or not a corporation or individual is to ruary 7. 1977 to -bput writtea mm--- tllrectly.about'tliese dtwes. be held liable ms a pven situation will menaa.. More than 190 md6vlduala; The Primary purpose of that part of depend upon the facnf . surrounding physiCans•manufacturen,.andtrade section S03(b)(2) oU the act, which -themanufacture,sale•and~ use.ofth.e and professional or9animUens enm- exemptvs prescriptlon.n druEs from the - drug product,, and bn the nature of:the mented on the proposal. The following - reputrement that their labelht` con- - In)ury. It evi11 alsoo depend on the ap- dLscussion summarlsess and ~responds to.. tain adequate, directions for use and plinbie.State law, wtiich- the Qomntla--. the subatanNal Lssues raised by the warnings, ls too avoid self-d)aEnoels and siener notes.ean be adlusted byState. cumments.- -- --- - self- mlwr•*••tlen of dtuEs that re-Courts and LeEislaturro tn lirhrof fac~ 1. Sfatuto+y'au(AOritp.Severarcom-. Quirc professional supervision for safe ton presented by the.use ofpatlent la-ments contended: that FDA lack.s the- use.'I'he.tYpulrementthatcercalapre- bellha: Moreover, the Comm'.itsloner authorltyto reDuire patient labellili _ scriptlon druas bedlstsensed ' only .. bellevea thatt prvvldiraa patlents with for prescriptfon druis- The commenta when aecompanled by. printed Patlent - written information on the haaardss as- ar9ue that sections S0]:505, andInformanon does notcontradlet this aoclated wlth the.use otoral contra- 701(aloftheFederalFood.DruE•and purpose.,Rather,thepurpose-.otsuch eeptlvnwillas:llkelyresuitlnreduced. Cosmet(oAct(t)teact)(21IIi3.6.152,Informatfonferoral-cantnceptives.)s 'potentiall)abillty,due.t0improved;pa- 355.:371(a)), c(ted as authorfty for the _ simply to 4nformthe.patlent of, thead: tfent',compl~iance wltit physlcian d[tec- - patfent labeling reautrement, do not "vantaaea and tiSYs associated;w1th the nons and self-molutorin9 of adverse provide any such authority. The eom- -use.of these druEs, and thus to provide _ effects, and a correapondtna decrease menta urye.., moreover, that section ~:nformatlonthat will better ensure In drtiQ-indGced inlury. It may; as w.ell',. 507ab)(2) of the act exDressly prohibitas th'eirsafe and effective use after they reduce the tnddence of malpractice ec- FDA from reQwirtint patlent labeling have been pveacDbed by the phya4ciaD.. tlons ua result of L'reater patient for Dresariptlon druTS.'I'he comments NothlniIn the.legLSlative history of. awueness that certain riska fnevltaDly argue..thatthe IEeislatld'e bistoryy of sect(on503(b) oranynthersection.of accompanydru[therapy.,andthatnot . sectton 50] shows that Conaresa left to the act suE3ests that Conaress mtend- all advene reactions result fromdefl'- - EEDEtAI'tE6ISSEE, VOl tl„NO. ]1-TUESDAT, JANUAEY 31,:197a . ... 1
Page 51: azo61e00
47 -RUCES AND REGULATIONS 4215 denelesldi the diue or on thePUt of,elan at thetlme heor she prescribes result of recelving.the.labellhg.: More•the Dhysicfan. In any eventr whether the drug nther.than by the dispenser. over„It the parUcular labeling may alter matJutac• The commenta sugzest.that dispenser patient 1s going to recefve patlent la- tttrera" liability In a given . Instanee distribution of patlent labeling afterr belfng containing certain tnformation, . cannot be considered ass a dlspualUve purchase of, the drug may«sultIn the physiciaq when lnltlsRyy pres¢ib- faeterr by the. Com^+nioner In reach- little U any improvement In paUent Ing the drut, will be able to dLtleuss Ing a decision on,the proposaL The compliance with Dhyslcians". instruc- that'Information with the patient and Commissfonen.concludes that to assure ttons or patient undersfandingof thee anticipate potential questioni. the.ate and effective use of oral coo- benefite and risks of drug use. If the Flnall.y„u the Commtssioner noted tJaceptLve products It Is neeessary.that Dhysictan distributes tlie.labellrti,.say In the.flital!regudaUOn for patient W the patient bee provided dlreetlywith questions or.probletPs raised by the la• be1ing for estrogen.drug products, the certattt specific Infosmation on oralbehnp it liargued could be answered pertlhent sectiona of the actdo not caaU'acepUOe drug products.. durlnst.the In1t1a1consultatioRSl obvlat appear to authorfu regulation of.the _ With regard to dispenser liabillty- big the need to arrange a costly' fo1-- prescribing functionoU physicians~ to preditatedon the dlistrlbution oflne-lawup. One comment urged ln partfcu• the extent contemDlated:bythe com- wrong labelitng,, the Commissioner ta-la[ that requiring pharmaclst~diatribu- _ menta. The Commissioner rern`nlms,. conf:dent that phatmacistscan devise Uon7)1 detailed patient labeling would nevertheless,. that certainn usera may' distributton systems that.wtll.enwe decreaae the involvement of doctors in visR,to have access to patient labeling that the proper labelingis dlatrlbuWted criUcal discussions c••n.e.,:f„ae benefit at.the:tlme the prescribing decision Is with each, oxal contracepUvee drug andrlskandtherebyshifttherespon-made and therefore.. he.strongly en•. prod0ct.Pharmaclstw already~ Prooea slbilltles of physioians to phaemaclstJt., eourages the voluntary dtstril5ution of andfurnishconsldenbUeoralandwrit-- The Commistiener-acuknowledeesPatientlabelingInthe prescnbeYs ten information along witn-ttte drug thatInmost psee,. the druS pnoduett offioe Of course, when the physician products that theyy dispense. Indeed,. wlU'not be d)strtbuted by.the physl•disperies theoralicontraeepnve drug supplylntlnformation regarding drugs clan but.rather b'ythe retaII pharma•produet. he becomes the dtspenser and'along with drug prodlJcts'tt a recng,- efat.Inthesa cases,.ttie Commissioner beara respomibilityy for diLstributmg n1aed part~ of the praeUee bf Dharma• believes it' 1! apPropriatee and consls- the Patient labeling- with the drug oy. The Comnilsaloner does nMZ believe tent with the. purDOSes of patient Ia- productIn these casrs the paUena- that the revi3ed~patient labellEtgft re- bellhg for, . the dispetsser, rather thanwill have the opportunity to question quurement M31impose responaibilitlea the Preecrlber,,to distribute patientla- and consult with the.physiclanabout on pharmaeists signlflEZntly different -beling- Iadee4 he vlewa the misgivings the labeling at the time the Ihitial. pve- !rom thosee they are already perform- concerning Dharmacistdistrl6unon en- serltiing decision la made. The Com-- - ing withh high levelsof professional pres+ed'by these comments as reflect- mLtsioner hopes thatmanufacturers in competence-. -. -- - -- ing a misunderstanding of, the Intend-_- thelt promotional campaigns wtll 3. Who are ditpenaen} One comment'ed purpose of patient.labelingt stuDplynoadlspensmg physicians with = que'stloned whether the diipenserdis- Patient labeling serves prtmarilyy a+ the patient labeling pleces and other tributionrequfrement; 2L CFRanirJformatlunaliadJuncttothephysf• suppllesnecessaryto:carry.outthe.vol- M .~501(a)(t).can,.be.enforcedagalnst cian-patient encounter and is intended untarY'd4stributimnprogram. _ , the dthtg produet, dispensers who are to reinforce and augment oral infor•'S: Physittan's discretion in distrilnw not.pharmacisa. The comment noted matlon given bytlie physician to.the tinn of labelinp- One comment sug-that•the dispenser oforal~eontracep- pat)ent.at the time the drug Is Drt- gestedthat distritiution oflabelhtgto tt••3 rmy be a physician,:a nurse• a lay tcribed. The physician,-whoby' train•. the patient should not be mandatory . personor seml-profess6onal In a family mg and -experience is best.equipped to because in some situations It may not "- planniiagcilnfc.or student.health de- tailor diuussionof drug therapy'y tobe in the bestintereste of the patient partment, as well as a pharc++acist, and the needs of indlvidual patients;, hass to recewe . detalled; informati0n. In . stated that it is not clearwhether-.any the primary.responsibilityfor advising these situations. It.is argued..the phy- of these.Dersons, other UientheOhar-patients about such ihformaaon xs di- sician= should be allowed to exercise his macist4: can violate the misbranding rectlens for uae. uutlons - againstt best ludgment and request that the ia- sectionofttle s.et ' mistJse, and warrings about po.ssible belJng not be given to the pstlent. -Neitherr the reaelt.of this final rule adverse reactions. Patient labeling will The-Caaimissioner acknowledges nor the underlying act Is confined too not shift that responsibililtyy to the dis. . that there may be drugs for mhien the Dhatmae'sts. Thepertment sections of pensitg pharmacist Even,when physi- distribution of, patient labeling mDght the act-3011,303, 502(21 Lf,B.C..331, cians.relymainly on written drug In- properlyy be 'discretlonarywith the 333. 352)-:-apply., to "aoyperson,"fortnation to inform their patienti and.- physician_If patient labeiing require- which, in the viem of,the Commission- when patient labeling wlll, therefore• ments for such drugs are implement- - er,,clearly. ineltedes personss in the oa serve as a prtmaryy infnrmationaJ ed. theCammisaioner would consider cupatlonal groupss mentfoned' Inthe:n source to patienta, that labeling still providing DJSysiaans with an option to comment.. When a ptiarmaelst,physi- acknowledges the pFimaryy responsibil• direct the divDenser to withhold ttie-: efan• nurne, or other person dlipepses ityof the p,hysic)anand suggesta that labelmg.. Because orsli contracepUves the drug Droductto the patient, he or the patient make ddtlSions regarding - are ordinarily. taken--electfvedy by - she it the dEspenser.vritYtin the mean- uu oII.the drug IA consultation aith healthy women who haveavaiYabieto L.g of the act and regulation and bearss her physician. • - - them alternative methods of treaU therespanstbility forprovidingthepa- -The ComJnlasioner does i:ot expectmeht, and because of the reiativelyUent.with patent labeling- F'ailtve to that thed)spenser distribution re- high~incidence of serious illnesses as- diatrtbutethe labeling or distrlbution. 7ulYement will result Ina signfildantly . soefated with their use,, the. Commis-.f the wrong labeling would result.ln increased need for foilowupvisits to:2 sionv believes thatusere oU these the misbran.ding of thee drug- product enable thephysleianta- respond to,, drugs should, without exception: be and would sutilect the dispenser (aad . questions and problems: ralsedbyd the - furnished with written information . theproduct)e to the sanctions for Dus- revised patient labelinF A 1975 survey telling tnem,of the drug's benefJta and: b-anding. ' ." •"" "' of women whomeived: the original, risks. The Commissioner conUudes;, - 4. Aoa11aM14t$oJpatlenty tn/onna•. patient-labelhAg2oreralcantraceptives there:ore,. that a DroNsion allowtngtien on oral conlraccptivea Several i Indlcated that more th-85pereehc of:- dlx7etionary withJholding of the label- cotnments . reetmtmended that patibnt the drug's usens did not.Increase the mg IA nott apDropriate_ andno such labeling be dlstributed . Cyy the physi- freQuency'of . physician oontactsu a option has been provlded IEDEaAL Rtbifrte„ YOt ex,.NO. x1=NESDAY, JANVAlY 31„lerg 11
Page 52: azo61e00
Z 48 4216 RULES AND. REGULATIONS 6.LaDefinlrJorlndiee0ionoUterthan one dose at atlme.The comment contraception. One comment suggea4 stated, moreover, that in InstituUonal, ed tnatpharmaciats and physicians be settings. pharmaciitt,:physlcians,:and~ exempted from providing patlent la-auaes cidsely monitor a therapeutic bet:nl when oralcontraceptiven are courseand:ttiat apatientcan rely onpreseribed for approved Indications penlonai contact with these paofes- other than contiaceptlon. Alternative- siunals te assure safe and effective _ ly, Itwas.recommended that labeling drug ase. mak/Ag distribution of e;-, be amended to Include a d8xussionof plamtory.labeting less necessary: The noncontraceptive uaes..The comment comment recommended that.the pro- argued that failure to provide for non- posed distribution requirementss be re- contisceptive uses, either, byexemp•vtsed by.permitting health,cesemstl- tion orotherwise- would result in con- tuUOria to provide patient labeling to fusion or unwarranted concern on the patients before adm4nistration of the part.of the patient.receiving the drug firs[dose-or.ifaloag-term-care.faetll- fornoncoatti•acepUVe purposes ty. before fii•st administration and ibiLKyy without', causing significant dis- The Commvssaonmer acknowledges every.30 days thereafter.. ruptloru to presently utfliied'prmt)ng ~ that.certain oral contraceptive prod- As providedln the final rule reqWir- andpackagingprooessei uetd are prescribed~for approved Indi- mg patient labeling for estrogen drug 9- Lobelinp needr o/apeciafwer poD- -utlo, s othq than contraception (for.- producu (see 42 FR 37u36), the Com- utolions. One comment recommended . example, hypermenorrhet and endo- missloner agrees that hospitals and thatpatient IabelLUg bee made avail- metriosis): However, comparedtouse.otEer health-care utstitutions sliould able inSpanlsh and Portuguese.and for contraception, the useof.the.drug -have some flexibility In meetingg re- other languages where a need's dent- for otlier hidiratlona~is extremelyqulrements-regarding distribution.oP onstnted. Another com,mentlpointed'smallJ representing less than 2 percent. patient labeling. The comc.issioner to the needd of b1indusers oforal con--of total drug product consumption. . eonciudesthat It would be impractical traceptlves- suggesttng that.patlent la- Tbe Commissioner believes that, pa-, and'unnecessuytoreqWire patient Ia- bellitg should be made available In tlents receiving oral contraceptives.for belmg to be made available to the hos- braille or on tape.. ' honconti-aceptl0e ihdicatlons should _.pital9zedor institutlonalized: . patient The Commissioner does not bel0eve . receive the patient labeling. pieces every time a drug if admitustered. The that tie can presently justify, the re- sincemuch of the information. ioclud- f)nall reguiation, therefore, has been quired preparation and dstributlonof ingall the information.regarding the revised'by.adding.anew sentence to labeling meeting the.needa of special dangers of drug-use. 1s.equally pertl-...¢310.501(a)(t)which atates that, mn user populations. The practiral'diffl'- nent wtien.the drug puoduct Is usedd acutecare hospitals or lomg-termcare cultlesi m preparing comptete. and for other indications. Information ono- tactilties- the requirementss of faithful' translatlonsof labeling Into tained In the patient, labelittg that 9 310.501(a).are met /f patient labeling all the languages spoken In the United plainlydoesnotpertaintononcontra•. (both the summary and detailed pa- Statesare IiOtely to be significantceptlveuses. for example- Information tient labellpg)) is provided to the pa- Moreover- sueli, a requirement would regarding effecti.venesss of other means tient before first administration of the Impose significant' administranlve and of contraceptlon. obviously will not~ drug. and every30 days thereafter. logistlcalDurdens on manufacturers" apply, and can be disregarded byUie. This revision in the proposed regttla- and dispensen In preparutg, storing: patienL Even where there is a poten- tlon answers the ob/ectionrrisedby and distributing proper labeling with .. tial.for confusion or conce-on tbee thecomment-.but.avoids the :eme- each drug producr covered bya pa- pa'rtoQthehoncontraceptiVe.user,.lCwhat. compli3tedpeoeedured thattientlabelingregu5anon:.However,the can be successfulByy addressed by the would result,from having dlfferent.re- Commisseoner pointt ouL under21 physician ind'ucussions with the.pa- quirements for aeute-care.and long- CYR 301.15(e( for labelingdlstributed tlent at.tlie time the prescribing deci- term-caree fuliitles. -- solely in theCocnmonw'eatth of Puerto- sion is made.. The Comtaisioner g.:TVpe.size re¢uiiemenh- Tw.o.cem- Rico orIna territory where the pre• agrees;,nevertheless, that intheinter,...mentsn objected totheo proposed're- dominant language's one other than ests of fully informing noncontrscep- quirement~ that 9-pointt type size beEngiish, the predominant language tive uaen of oral contraceptives;: man- used in the detailed patient.labelingt may be substituted for Eng113h'. Al-ufacturers and; o.ther druglabelersThe comments.suggested thatths re- though in the rest of the United'should have an opporttlnity to add to quirementwoWd have a significant en- States allreqylred labeling must the labeling a discussion of the.other vuonmental and: inflationary impactt appear in Engllsh,.theregulations do approved uses of the drug. AbcrordSng- byy requiring replacement of.currently not preclude the distnbution ef labeP Iy, the final regulation wlll provide used priptingpressesand signifiCantly' Ing in a language.other than Ennglish- th'at, for those oral contraceptive drug Itlcrea.sing the demand for paper and or in a speciai fo=at or In braille productsaith approved new drug ap- energy. The.comments reeommended along withtne conventional English plications for indications In addition to that the minimum type slu require- language labeling. The CommWioner contraception, the labeling may Identt- ment be changedd to 6-. 7., or g-polht enrnwrages the preparation of labeling.- fythese other.approved indi'cationaif type, and suggested thatths change meeting.theneedsof spedaliuserpop- lt states as.well that the information would result' In substantial economie ulaoimns asdongasauch labeling fully. In thee patient labeling relative to eon- and envuotumental eavings with no and faithfullyy complles witnh the re- traindlcatlons, the dangers of oral con- lossof legibillty. - qusements of the regulation.traceptives, , and the. safe usee of the The CommLssioner belleves that an 10. Briefsummary reQUiremenL Sev- drug also applies when thee drug is objective standard specifying mini- eral commentsubiected to,the req.uire. used for these other indications mum requirements for the printing of inent'that the user receive, in additlon T.:Distributiouo/lobeiirtyin health~ detailed patient labeling s necessaryy to detailed patient labeling, a brief care iastitutio*u One comment noted to ensure its adequatelegibility.,8ow- summary containingcertam essentlal - that In health-care itutitutionss provid-ever- the Comm4ssioner spersuaded points of Information also contamed in ing.unitdose drug distribution to mpa- that specifying a partleularpoiat type the longer detailed patientt labeling. tleats, it Isl impracticai to provide pa- size. s no0- by itself, snffieienL Re The commeutssuggested that the use tlentlabeling whenever adrug is dlf- -notesthewidevariiationL-legibilltyof, of, two nLSert.t one a summaryof the pensed'beeause the drug Is dspensedprlnted material that Is possible with otlier, is redundant and imposes an - EEDEEAL aEO15TEe, VOL N, NO. f1-tuEEp1.T;,lAE1UA[Y. ]1, 1978 changes In tyyee style. liy;itnes: of the type, and spacing of.thetype on the line and between,the lines.~. Abtrorditu- ly, the final regWation has been re- vfaed to specify that the mlEsimum type size shall be ai least 1/16 inctt m helght The height pertains toQower- caae letten and'M1t is the lowereite "d" or itsequivalent th'at shall meet.the minimum standard- The body copy ahall be 1-pemtleading and noncon- densed type, and shall inot contain any light face type oc small capital letter: The Commlssipner_beLeres that this reqmifement willepauPe adequare leg-- . t
Page 53: azo61e00
49 -aULES AND gFGUTAnoNS . +217' unneeeasary'and tvoldable burden on latfon. storage, and dlstrlbution of oral fying the.pharmacist as an acussDie taanufacturers and dispensers An- mntixepUVe drug products and label- source of lnformatinn regard)ng oral other mmment sugyested, moreover. Ing. - contraceptive drug use. thatUse avallability to the user ot two The Cmnmissloner believes that it Is. The Comntissloner realizes the velu- pieces of patientlabeltng, in addition in the best interests nf the publieable mntribution that.can tiemade by to physician label)ng. could readily health that the revised Patient label- pharmacists As wetl aax other health confuse the patient', and not mntributa ing be provided to patleats. as soon.asprvfessionalsia responding to mnsum- to her understaeding and mmpl0ance possible.. To base the effective.date on ers' questions about the ecacents of withphysicianinstructlotss..- Utiet1mee when the drug prodtltct is -patient labeling.. However. the Com- R'ttlle it Is tiuethat the information packaged wouldlmake the provtvion of missioner believes that the.avai1abil:ty coaUined in the brief summary is alsothe labeling to consumers contingent and usefulnels of phanaacist munsCl- descrlbeC Inthe detailedpatient label- upon indivldual ~manufacturers' inven• ingcan be more appropriatel7snd ef. ing. the Commissioner does not agree tories„and.couldresult in significant flefently commtndcated byProfession-. ttiat requiieddistribution.of the brief delays.ia providing Patients withthe al organiutioru and consumer grcupss.:mmary in addition mdetailed pa- lihtlibg. The intent of _ thee effective as part of ~ their, general program to tteht labeling isunnecessary:. The brief date provision is to,preven.t any fur-educate cprtsumem aboutdrug use and: summary Is intended to perform a ther dllstributlonof a drug product tmt safety. Patient lsbelingisintended pri-function mmplementary to the func-contalning the revised detailed pat2entt marily as a vehieleto brutg important . tlon servedd by detailed patient label~- labelhsgg information . vtthout; st the speclfic Information about specific Ing. As noted In the preamble to the samee time, requiring a potential.lydnrgs to the attention of.ttie pa!;ent, proposed rule. the value.ofthe brief mstly and time mnsumlAgrecall sndratherthanasaveluc:etoeCUcte.thed summaryis:tvrofoldf: (1)'tne'summaryrepacitagln; oforal contraceptive consumer on general matters pertain- is short enough to be Included within products already, in the channels ofIng to drug wage..Nevertheless. whth the package dispensed to the patient di'stribut)oa-. ' - this regulation is primarily concerned. . aadwould be more llkelyy to be. avasl- Tht Commissioner aeknowledges with identifyuu certain speeific pointa- abletothe:patientthrouBhout~thelife that.an expeditedeffecttve date may of infora+aUonn thasthepatientlabel- of the package-, and (2)'trie summary require mme increased effort Dyrean- Ing must Include, itameanotptehlblt can.be.quickly and easidyread by the: ufxtuzen and dispensers for a shortthe manufacturei' Srom adding any .patient,-will:call her. attehtion mthe time. However, in exempting the brief other information that be deems - longer labeling piece, and will urge.her summaryy from- thee catchup, require- usefud which doeas not . mLSbracd the to read the mmplete labeling. The ments, thenecessity' of recalling Drod- drug.. The C n+oner would not Comrnissionerconcludes thatttiese ad- uets already in distribution channelss object to the Inedusion of a statement vantagess outweigh any possible addi- and repackaglag these productes with encouraging consuraers to directquea-~ tlonal ~ burden to manufacturers andd the nem brlef summary is obviated and tions mncerr.ing oral mntracepUve pa- dlspensern.and:thatttie bnef summary theburdenon manufacturers and dU-t tlentlabeling to their pharmacists as reqjire_mentshould be retAlned - pensers reduced to what the agency is -weil asosher.Cealth profesronalt- - - The: Commissioner does not antict, confident Is a manageable leveL in 13. fJi.tfri6'utioa o!~ dctniled patient' - . pate that the availability of three this respect, the. ComunLssioner notes tabetinp- One manufacturer objected Diecesoflabelingwillresult,in:patient that the.effective date provision is tog310.501(a)(gHli)(9010i.501(alaU(iw contusion.Tle.informatiommntailsedw identtcal,.asapracttrslmatter,tothat ss.Droposed), which wo•:ld requfrein thee brlef summary-ISconsistent contained )n the final regulation for thatdetailed patient labeling b'e.in- with the Information containedStt the patient labeling for estrogen drug cluded In or accompany each package detailed patient labeling and physkian Products, whiahvvat put Into effect. intended to be dispensed to the pa, Iabeling,-The brief summary, by Iden- Octotier18, 1977 without.the mnse-tlent. The mmment argued that.this tl9ying and highllghting some oUthe qvencescitedinthemminents requirement would subatantially ir mostimportant information also con. Although physicians who dispense crease the cost oforalI contraceptives tained; m the more detailed latieltng these drugs are considered to be dis- to the consumer as well as .o.tDeman- pieceo should - therefore.-Dromote pensen under the regulatioa theutacturer by requm.ng that tne.Da- rath.er than, deveasee patient under-Commisslomer has conoluded that it tienc be furnished detailed patlenC standing oftfie relevantmaterlaL - would be Impracticable to require the labeling not only when the cnginal 11. EiMdited el.reeeibe dara.'Pto forwarding..before the effective date: prescriptionis filled but with, eceryy mmments otiiected m the requirement of.separate.patientlabeling to physt- reffll: requirittg 12 detailed labeling ih 4310.501(aXT)~ <9310:50Q(a)(5): asdaeu.for those products already in pieasayears proposed) thatrevised detatled.patienttheir Possession. Acmrdingly,.the re- To provide.for the continuing avail- labelingbefurnished b'y.the manufae- quirement that oral mntraceptive' ability of detailed patient labeling for turerorlabeler.tothewttolesalerand drug products be dispensed arith.Da- the entire period that.a Patlent.uses retailer Ln . stuffPcient. .numbers -tv tient labeling will not be effective for oral mntiaeeptives, the Commissioner . permit any retaU purchaserto obtain- supplies inthe possession of physi- mndudess that the patient should:re- the-labeling with the.Droduet on- or clans on the effeative.date„ but arill- ceive.the labeling whenever the drug . after the effectivedate. The mm-apply only to'supplles received there- isdl5peased to her. The Dro~vtsion of menta recommended thatthlS "'catch- after. - - --'~ - detai)edpatlent.labeling with every up" Drovision be deleted'snd that.the 12. Pharmac'at.co,nnselin4.A.Drofes- new pre'.-~criDUOnor renewal wUll- effectlve date be based on the-date on sionat prgatLLtatSon remmmended ttiat assure,. moreover, that in the event the which the oral mntraceptive-'drug~ theproposedruleberev{sedmrequirelabeltirgisrevued the Patient w.ill re• t.roduetss are packaged One comment the Inclusion of a statement in the ceive the.mostcurrentversionof pa- complained:that as manufacturers do brief summary encouraging the pa- Uent labelingwitti each newDutchasea noa.know the exacti inventory of oral tient.to ask:herDharmacist anyques-In any.event„as oral contraeePtiveamntraceptives stqcked hqy each drug- tions about', oral conttaceptlvea and are frequently prescribedand reffaed Seore andwholbaleh theywlll be tkieirusea:. The mmmentsuggested for periods longer than 1 muntti, the- obliged to aigtllL¢anUY oversuDD)y tottiat pharmaeist counseling provides a Commjssioner does not anticnpate thatt ensure adequate cnvetage_ The. [ valuable health servieee by increashng the user will ortlimarilyy rmeive. 12 - mentt noted further than the expedt- understanding about and mmpliance piecess ofdetailed patient Iabeling a ed'effective date will have a maJar ad: with.a therapeuticregfinea.The.cpm- year as argued bythec+omment-Under verx.Impact.on dispenser msts of m1-ment noted'the Importance of identi-- this regulati0n only one detati.ed label- rEGEgAI eEG1aTEa, VOL. 43,: 040. 21-flRfDAY, .JANULRY.3l; 1979
Page 54: azo61e00
50 4219 . - RutESANO.RFGULATfoNf . Ing piece must be Included in or ao Ucular side effect has been oeB estab. cotn?anq. aaeh Dackage dispensed to llshed However. tn,thecase of toallg- the patieni Thus. UIa patlent reeetvea. -nant hepatic adenoma, only a few a DresCripUon for a g-month supplyy of , cases have been reported and nodetl- ort,l mntracepUVes.. the dispenser nise a^^c•rion. xith oral oomtracep- must Ineiude only one.detailed pauent " tire usee has been demonstrated. labeling D4ece in or accompan.ying the Therefore, th!s commentia re)ected entlre package dispensed to the pa-8owever, If evidence of such,an asaoci+Uenti. This requirementapplies, noG atlon becomes avallable..the Commu- Witlistanding the.fact.that the entiree elOnerwtllactpromutly~torepuirethe Dackage may be ttutdeupof a number Inelusion of the.ihtormation in the ]a- ol lndividiul prepackagedunitsd belhtg or take other action necessary Coatent ef~yatient (abetinm SecUon )h thepubllc interest:.- - 310.501(a) sets forth the information 18. One commenr:noted thabth'e riisk' to beincluded'in the brief summary, of myocardlal Ihfarctinn iss absent and detailed patient labeling. A from both the brief summary and de- number of,comments were received on tailed paUent labeling rcquirements.. these reqyirementt. The COm.nicuon... The cOmmentnrged areviSlOn to re- er'sresponses to the significant ccm- quiree a statement regardtng risk of ments on this part of the regulauom myocazdial infaretlon, since the.sbso- fOllom._.= lute risk oldeath due to.myorardlal . 14. One comment su~ggested that in intarctlon,. though not the relative ¢J10.501(a)(1) the second sentence be rlsk isgreater than the combined risk rev`sed to indicate that the required of death from aB otherkYlown sideef- Information must be put into lay ]an- fecta.- guage m,the actual patient labeling. Because.anassociauon between oral The comment stated thata literal contraeepttve use and myowrdlal !n- readih; of . the'pToposed requirement farcciom Is now establJshed, the. Com• would result In the reproducuon of all missioner agrees that this risk cate- medical terms asspeciLied m the rule. gory should be added. Section The Commissiomer aYrees thatthe310.501(a)(2)(Iv) and (3Nvin) . Sntento0 the regulatiom is to provide (;310.SO1(aH3HVil).asproposed) is re- iabeling. in language that is under-viaedaccordingly. - standable to the lay.Dub110.He thus 19. Onee comment obJected to thee agrees withh the comment and' Wnrding in. §310.501(a)(31(viii)'. 4310.501(a)(1).Is revised arcordingly-(4310.501(a)(3)tvillas proposedP: "The. - 15, One comment suggested that a ability.of.estrogen to tause malignant requirement be.incltrdedinboth thee tumorsinanimaLt,endometrlal'nncer brief sumrnary. and detailed labeling in womM and the evl.dence that se- piecesofiaformation concerning the Quentlel oralcontracepUvesmayin-cnntrsindication of oral contraceptive crease ttierisk of, emdometrlal catlcer use fur women with sickle cell traitor in women, must bee mentioned" The d5ease. - Lromment contends that,aa the medl+ :IOreteren¢eswerestubnuttedinsup- cal findings.on this.subiectare.equiv- Dort of this co.mmenL and the Com- ocal, the smtement is incurrect in sug- missioner is not aware of anyy data gesting an absolute relationship be- that warrantt rontraindlcating useof tween estrogen use and;endometrial oralcontracepUves in patients:ith cancer.. siekle.cell trait or disease. The cOm- The Commissioner believee that ment is therefore reJected, there is a well-established associatfon 10., One commentsuggested; that-between chronic estrogen use in posL 43L0.501(a)(2)(vui) (§310.501(a)(2XV(i) menopausal wome-and:an increased a5 proposed) mqulting a statement in risk of endometrial canaer. However, the brief summary that oralcontra•l theCommissioner agrees withtheh eeptives are ollmo.vatue in the preven- cotmmea[ that the statement'~ that es- tion or treatment of venereal disease trogen use causes endometnal leancer be revised to indlote thu oral contra- mayDe subject to misihterptxtauon. eeptive use increases susceptibility to He also believes the section should certain kinds ofvenereal disease. contain a clearer description of the ap. - No data have been submitted'tosup- propriate Inference . to,be made in the pot2 this statement, lt is not known 2t contextof this regulatlon from anlmai present whether oral: contraceptive data. To clarify. therefore. the sen- use increases susceptibility.to venereal tence that beglru'Theability.of estro- disease. This commentis therefore. gentocause4 4 0." sctlangedtoread al3nrejected- as followz. "The following shall be. . 11. One comme¢t recommended thatmenttoned1 (a) i Fstrogens have been the regulation be revised to zequire a ahowan to csuse cancer11s anlmals; statement Inthe Cetailed patient la- which showing Justifiesthe Inference beting. listing malignant~ hepatic ad- thatt estrogens may'y cause cancer in enoma among the serious side effectss humans; (b) there ia strong evidence. assaciated with oral eontraceptive use. that estrogen use- inereases che nsk of Fora1111sted serious side effects,.aa end6metrialcancer In.postmenopauual' association between oral contraceptivee women, (c)'.there Issome evidence that use and an Increased risk of the par- sequentlal oral contraceptivea (wbich are n0 longer marketed) may lncxease the rtih' of endometrlal canar ih womem;(d) .studia of, an assoc7atlom between oral coIDtiazptlves and breast oanter are largely negatlve except for a suggestion of increased ris1.1 study) In women with benign breast.dlsease, and: there is no evidence of an !h- eressed risk of uterine cancer in users of oral contraceptives other than se- - quentWs." - - - 20..One coatment suggested that a statement be required indicating tnatoralic0ntraeeptive use altlen the and enviromment,, of~~ the vagina, allowiug overgfowth of yeast, hemophilus, and _ttlchcmonas. . No references were submitted insup- port', of tbe comment The Contmis- . sloner is not aware of any eHdence thatoral contraceptive use is assoclaL _• ed with an overgrovrt)i of hemophllus or trichomonss. Physieian labelingdoen inmcate that'vaginal condidiasis haa.been reDorted in patients receivu)g oral contraeeptivq and isbeli:eved to be drug related:However, as the inci~dence oII' randidusu Is neither common nor serious, dh'e Commiisien- . er.conrdllde3 that It is GGt neee3dary. tG speciffcally require ihcluslon Of the dLiease m the iistmg of side.effecta in the pauent labeling:- ..21. A' commenturgedt that - S3do.501(a)(3)(tx) (¢310:501(a)(31(vW) as proposed) be revised to requlre mention of intertility: The Comment noted thatt for women who have not borne previuuss children a sig[lllicant difference in fertllityy remains. 2 years after they have.stopped using the ornl contracepuve: The Commissioner agrees ttut'de- layed~ return, of fertilYtyy shoultld be listed'in, ¢310.501(a)(3Hir) as one of the "Ot^er serious effects.:"Theregu- lation is revisedaecordmgly~. 22. One commemt',suggested thatsuit cide be Inelmded in the required listing of serious sideeffects. The comment stated tbatthe Rova1 College of Gen- eral Practitioners' stud} of oral con- traceptive ose'rs reported twitee the number of suicides m oral contnuep- tive users as in nonuseri . Alt)iough it is true thatthe.Britishstuudy reported an increased jneideneein suicide among usen of oral cOntra-. ceptives,. the same study reported noM evidence that severe depression (which would be the.type.most likely assoeiat, ed vntti suicide) ~ i, more common In orsl contraceptive usets than in non- users: Furthermore, thee reported tn- Creased!incidence of suicide could very well be due to a nummber of factonother than use ot thedrug Itself, and. absent otherirdormation: the Com-missioner• Oelieves the f:nding Is of little statistlcal shgnul ance. Thecom-ment.is therefore .reieated.' 23. One comment addressed the second sentence in ¢310.501(a)(3)(viii) ~ (§ 310.501(a)(3)(Vill.3S prODosed) Which _iEDEIAL aEnISTEa, VOL A . NO. f1-fOESDAY, JANUAa1, 71, lyra -
Page 55: azo61e00
51 requires the listlnk of ttirombophlebl- UL pulmonary emboWam, _.retlnal artery thrombasla, and streke as serl- atla side effects And which state-paren- thetically that the relation of these ill'- neaw to estrogen doee Is too be. men- RUUES AND REGULATIONS auentlUs, The commentt suggested that the werdlmR of this proposed re- vuldementt would: become outdlted If future references were published which also suggest an iacteaetd rlsk of breast cancer ln orala coatraceptive 4219 socfaUon Between Oral Contraeepttves and Myocardlal infarctioa'- Journal of the American Medical Aasodatmn. 237:2619-2622,.1977: Jaim. A. Ii. "Mor- tillty Risk Associated with the Use of - OralConttacepUves." Studies in Pamily Plannlag. 6:50: S4: 1977; Tierak V:.. "Mortallty Among Oral Contracep Uves Osers,':' l.anoet. Y727-731, 1977.) A revhieddnft of the patient iabelins .. baaed on the aew' informanon. and .. analyaes.was presented mthe FDA's Obstetrita and Gynecology Advisory CommitteeNovember 17,. 1977., The CommiU.eerecommended that the pa- be revtsed toretleot the tient Iabelingr variable ri+ksfOr smokers and non- smokere of:suffering serious nrdlovu- cular side effects. The Committee also ,. recommended'the inelusion in the !a- beling of a prominent boxed wsrndaa advising women,who use.oraUcontra-cept(ves not to smoke. A copy of the transcript of the advisory committee's dixuss/onof.oral contrueptlve label. InQ has been placed on file in the - o(flce of the Hearing Cerk F'DA The CommissioneraQees with the Committee that the role that amaklhe plays In Increasing thee risks of rerieua- eardiovascular side effects should be brouaht.to the atteptlon of all women who are presently usina or contem- UAced."-Phe comment argued'that the. usen. T'he:comment reeoaim•^ne+d the relation of estrogendaae to each of adoption of the following: "Tbere these Wnesses ii not known end that. ahould also be a statementcoaceralng - to clar137,: the words "where known" thesee studies of in sssociatlan be- shoUld be added to the.patmthetlcal__tbeen oralcontraceptivesand breast remark.. - . .- . - -c.ncer that arelargely negative In . The two stltdies upon which these contrast to atleh studies that suggest requirements are based (they are cited an iaereaaed risk in certain groups of In the Ohysician labeling text) suggest Women: the statemtnt'~shall also relate that thromboembollsm as a diseasesuehevidence that wouLd indicate In-ennty taestroaen.dose-related: In fact.. ereased risk of uterine cancer in users . the Vessey'study:(Tnmw. W.B. W.M. of oral contruepttves-other than ea:. P. Pessey. B. WesterhOlm, snd A. ESt- puentfals.:" gelund. "'Thromboembolic DLteue and'.'. The Commtaaldner re)ectsthis com- theSteroldal 'Content of Oral Contra-' ment.. To Incdude the wordity sugant. ceptivec A reportto the Comnuttee on ed by the comment "• 'I tacontrastthe. Safety of Drug.' British Medieal- to such studies that: sugiestan In- Journal, 2:203-209, 1970) ahows a aesaed risk lnn cerWn groups of, strong correlation between estrogen . women •-• •" would not be correct ass content ind: thrombophlebltls; pul- It -would Imply that there currentlyy monaryembollsm, aad strcke (eere= exists more than one study to thecon- beel thrombosial.~ Although retlnaltrsry: Also the recomtnended pbrue. artery thrombosis (a relatively rare.. -' '' the statementshallalso relate type of throm0oembolltm) is not spe• suchevldence that would indicate in- cifically, mentioned in the study; It creased risk of,uterine cancer in users _vrould be unreasonahie to assume that- of oral contranepnvee other than se- Iti unlike othertypes of thromboembo- tiuentlaJS"' would incorrectlyy suggest l1am, is not estrogen dose-related. To that there currentlqy exists such evl- plating the use of oral trontnceptives.m add'the term "mheie.known"' to the dence.. No such evidence currenUy 'Accordittdy, he has inUuded in thlt -~ paretithetlcal remark would not be in exists,: The Commissionerconcludes final rulee a. requirement for a boxed keeping with the seneral findings of that the current requirement seeu- wareinain both the briefsummary . the relevantscientifie literature. This -. rately reflects the present state of the and the detailed:pattent.latieling stat- cuament is therefore re)ected:- scientific literature.. Ing that smoking increases thee risks of -21. Several comments suggested'that28. One comment recommeaded tliat'serious adverse- effects- on the . heart . the proposal' turevised to require a the ditcusslon of the risks of seriousand bloodd vessela, and advising women statement in thepatientlabeling re- side effects associated with oral!eoa- who use oral contraceptlvestots to garding an avociatioct between oral tnceptlve use should speciflcallyy men- smoke.. He Ismaking two additional- contracept[ve useand certain vitamin tlon cigarette.=olnng ss. an Indepen- revfsIons: (1): ¢330.501(a1(SHvHl) deficlences. In particular. It was sug- dent factorwhlch signUlcantly In- gestedthat mention be made ofan as- creases the risk of myocardYalIaSart- sotlatfon, . between oral conttscepnve tianitl drug usen, Thecamment pre- use and viteminH-B deficlencyas weUseateda statisticalians]ysis ota Brit. u Serum feiite)evel depression,: tth retrospective study of oral contra- The. Commistioner re)eatt theae ceptive users to demoostrate that the commenta. The aasociatton between risk of myocardiat Infarcnon ~In women ora) contraceptfve use. and vitamin )3-B who Smoke and uae oralkontraceptlves deftciency'ia disputed and not weID de• istonstderably greater than the sum fined. Moreover, the ciirucal stantfl- of the rlskr for these woman who eance.of such a deneieney; if Itdoes smoke and donot, uaeoral contruep- (/ 310-501(a)(l)(VU1'as Praposed);is re- vised ta Provtde for a dlu(uslbn of therllatlonship between the occllnence of serious side effecttand aee, smok- 'Ing- and other Vndittons; and'(2) ¢ 310.501(a)(3Xx11): (4310.501fa)U3kxf) as proposed) IS revised'to require that ttie comyuisonof rlskof death from= various centraceptive methods de- - scrtbe the risk faced by both smokera-and:aoatmcken who use aral tcntra-' exiat, has notbeen well establtthed -tlvee and'the risks for those women - ceptives..Theae.revislons should pxo- While 1t.hae been shown that serum _-wlto anlyusey pntl rontntxptivea snd .ide woman wtth a clearer understand-~ folate levels may be depressed with do - tmt', atnoka. The comment urged Ini of ~~ the effects of smokir>s an the - oral contzaceptive nse, It', is not certain that the discussion of serious side ef- risks-of oral contraceptive use.. an un- _ ;hat this represents aa true vitamin de- feeta would be of most value to all ~ derstaading which in the Commisiion--~ flefeney_ because no cllnicalalgnlfh -drukusers-both smoken and nen--er's view is essentfalfor a proper.as- - ' tanoe hau been attributed to the db smokers-If this "synerg'LUc" Intet9o- sesament of the drug's salety. creased leveL ;- - - - : Uhn between smoking and oralcontra- - 27. Onpo(npg di.rtribation . of patfent - 25. Ohecomment addressed the ~ ceptive:use were expressly described. -taDelinp: The Commusloner is alea fourth sentence of 4310-501fax3XVflU,-; The Commissioner has carefully re- tevisipg .. . 4310.501(a)(6XW)- "whLcn reqyilres a statement that stud. . viewed the comment to determine the (S 310-501(a)c11[LLI) ass proposed) i to--Sea of an associatlon between "ora1 con- - need for revisions in this final rule. u-provfee that in thecase of oral contts- ~, traeepnves and bresst cancer arewelluln-theguldel)nePetientlabel- ceptivesInbulkpackagesintendedfor - largely negative except for a sugaes- ina text. AddlUonally he hss reviewed multiplee dispensin4- a suffietent-.t tion of increased rlSk (one study).inthreereceatly.pub4ishedstudies,.allof number of patientt labeAinipleceawomen with benign breast dtsease,. and'.. which suggest that smokiae, along (both the brief'snmmary and detailed .,. that there is no, evidence of an in- -with oral contraceptive ase markedly patient Iabehag) "shall beincluded In ceased risk o!'uterlne cancer m usersn inareases the risks of seriouss cardlo- orshall~ secompaay. eachbultpN-lt- of oral eontraaepUven otherr than se- vatcular side effects. (Ory, 8. W., "As- , age" to assure thatboth ,pietes ran be . fEpeaAl aaOlSteg, VOI. U,.NO..fI-nIESpAY,.lANUaaT]1,tera -
Page 56: azo61e00
52 4220 ' RULES AND REGULATIONS furnisbed with each package di.t- pensedd to every paUent: 'rherevislon is Intended in make elear that patlent labeltngmust phydtslly aa.rompanythe drug peoduct but need not be actu- all.y. placed Inside the Immed9ate bulk package.eontsiner.. .. . The . Cammissioner anticilr•tes thst', manu[acturera and labelere - will employ a reliable statistieal method todetermine .)tee sufficieney off tnee number of patlent7abelia[ piecn to be included in or with each bulk pacgage. He recaenizes. ttowever, that ffi somee cases additional patient labeling Dlken may, for a variety. of reasona, . be re- t(uired The Commissioner is thetefore. adding a sentence to ¢ 310.501(ax6)(W). to provide that the toanufacturer or labeler mayalso employa supplemen-tary. distribution system tosupply-ad- ditional Patient labeling to the dis Denser: That system may twt'; howev- er, actas a substitute for the repuitt* interest of the Dubllc in these.matters..e and itIs that type of mput Which 1+ contemplated by the Administrative Ptotedure Act. The Cocasuiislooer vealius the salue of public.participatlon in draftittY the best possible Datlentlabelln[ text:'= in fact, he.bae Invlted commenta from In- terested persons on the labelPug text and has carefullyy considered the nu- meroua comments that have been ttm ttived. Hovever, the Commiaaioner be- lievestlxat the mformstlon that the regulatloa requires be provided to uaet] of oral tontraeeptives, 1t1 partiml- 1az. information on thromboemb'uliee and thrombotid df.4ordets. (sncer, and adverse effecta on the exposed fetua. is of suehsl[alflCanri totneo public health that Le cartnot )ustlty the long-delays in communlcating this fnforma- tion to oral contraapti.ve usera that. Would result from notice and oommant consideration of the labeling text;tom- ment that patlent labeling be supplied', meata. -In or with each bulk Paeltsge. -:.:. ..- The Comadssionei.concludear there- 26: PatientlaLelinp ond xtf-medtta- fore.. that the present procedures tion, la'response to a comment on the shouidbe retained aad'that the py estro[en patient labeling propoaaL1 the tient labeling text should not be lnwr-ComIDL4atoner revised thee edtrC[en'pa- porated IntC the final Tme. " 'tient labeling ftnal rule' to require a The ob)ectloos to the ~Drocedureem- statement adv(sin[ the patient tAat ployed In sepsratelyy publishing the the drug had been prescribed for.the notice miscoasNUe.the le;a] purpose ir.dYV(dual alone and autioning thein_ ahd effect of the patlent',labellAgtext; dl'.tiduai'sgauut tyvmg the dtug to The CommiSsioner advlses that.,tbe anyone e11e. The Commiavionerhasr text of, the patient'Iabelmg Ib not a concluded that a similar mutionary substantive formal i rule: The labeling publtshed versions of the patient label- la[ text. the Commissioner haa fur- ttiet.tevised the patient labeling tea Published elsewhere m this Issue ofthe Ptaa,tt Ractaxn Ls the prectxlanguage of'ttie revised patientlabei- In[ text that will be considered tn meet thee repuiremente of the final rula As prevlously'y suted the Com-r missioner advisesthat; . this tert Is m= tended as a guidelme (IIl CP8 10.90)..- which....U followed.. w01 enable any-- person to comply with the require- - men6t of ¢310.50t(U. Those manutacturen and suppliers Who have deferred prepir{ng. patient labeling until the pwblication of, the final rule have until Apri1'3, 1976 to implement the final regulation. For . those.-mYiulaCturera and suppliers who elected: to use the December 7~. 1976 guideline text (as revised on May- 27,.1977), this earlier text will:contm- ue to meet the reQulrements of '310501(s) untll May 31.~L97A.After May 31, 1976- ttiiu earller version of - -the labeling textcan talonger be relied upon as meeting the reauirt.` meatsub§91D.601(u.* , 'Pherefore..under the.BEderal Pood:-. Drus. and Casmecle Act (se¢ 502. 505, 701(s): 52 Stat. 10504053 as amended,'1055 (21 D.S.C..352,.355;.371(a))) and under authority.delegated to the Com- n',ivloner (21 CFR 5.1)„ Part 310 iaamended byy revtting 9310.501(a)'.to read at follows: statement shouddbe included in oral Is, rather, a guidelme which, while ' 31oi5o1 Peepara(iuns forr watraseptiosrtrontraceptive patient labeling and has stating the ageney's vlews of how the labeyng y-W tO tbe yaeal- reviud the regulation accordingly. . - reduirementaforlabeling for these ' 29. A commentsug[ested thu-the products can be met, does not preclude. (a) ~ Oral I contfaeeptiws. (1)''Phe. procedural regulationgoverning the changes bssed on the best )udgmeat'~of Commiasloner of,bbod and Drup eon- aduption mm use oflabellam6 Without Indtvidual wmpames.aslonguthela, cludes thatthesafe and effectiveuae advance approval by FDA should be belm[ that is distributed still ton- of oaal contraceptive drueProduct+ ra spec(ficallyldeatifled. ~formstotheregulationsandappllda. 0~tlthat.patientsDefullyInformcd The Commissioner agren. To darSyy ble sections of the act. The procedures of thee benetits aad risln lhvolved m the. regulatory procedure by which employed are mtended'~t o~to effect.more the use of these drugs. Informationm holders of new drug applicationsfor timely Publicationnf approved label- ~Qy ~~a~ conoe W tng efteetfveness- atnindfestlon, arniu[a,- p:acau- ora) contraceptive divg products shall In[ reflecting the moet curreatmedl- ' uoru, and adverse reactions shall be (mPtemenct the patient LalxLLn[re- cal and seientl,lfa.lkarnm[ and are fa n,,,,;,hty, to each Datieat', rexivin[. ;31Db01(al (9) A' n t s l f i h ormancew Qu emen - proceduen con t FD (§ 310.501(t) (7).aspropaaed) is revised regulatlons (21 CFR 10.1 et seq.). 1f to state that supplements must be sub- the agency Were to take action again.st . m6tted under1314.9(d) of the regula- a product,, tt W.ould not rely selaly oa tiooi the guideguideline text but Would undkr- 30', Siotoa of patient laaelinp tls6. take to prove a violation bascd:on the Several c'otmments ob)ected to the pub- underlying rule Lnd'statnte. The Pub- llcanon of thee patient label1ng te:t In Uatted labeling could'bowever- serve a notice separate from the rule redulr- as evidenaeof sucha violatlon. . Ing the.labelmg. The comments.argue - Section 310.50~1(aK6)requires that' that this proaeduree removes changes. FDA make available aad: publiah, In in.patient labeling language from the the PmtAwr. Raptsru patlent labeling full and; open comment required for forr oral ~ contraceptives which is re- rtilenmtin[ underthe.Administratlve sponsive toall items - specified in Procedure Act. Choice of language and'. j 310.501(s) (2).aad (3). The suggested'edltorial style,it in araued- mayy bee text of patient labeling that met the substantlveWuea aad as such proper-- requirements of the proposedd rule Was lyy sublect tu pubUcation and comment published m the Fmsau Rxetszat of procedures.Onecomment contended Dece®ber 7, 1976 (41 FR 53630):and; . that the fact that a large number of revised' In theFmuu Rpctsznt of comments have, in the pasti been sub- May 27, 1977 (42 FR 27303): Se<ause mitted to FDA mresponae.to a notice of comments received on the proposed'of proposed rulematdne relating to pa- ru)e;,as well as comments and new In-tlent labeling (ndlntes the oa.idespread formation received on the previously oral centraceptives. This informaUon sha1D be [iventon the patient by the dispenserin the.form of abrlefsum- maq of certain essentlal i Ihiormation Included m each par.gate dispensed to each patient. and in a longer, detailed labeling piece in or actompe.nylhg each Dackagee dispensed to each pa- Uent. Paaent labeling for drug prcd- uctt dfspensed'I6 acute-caree hospitals orlong-tu-mcare facilities will be coa= sidered to havee been provided in aceor, dance with this section U provided m the patient'beforet admin(:+*atioa of the first oral contraceptive and e.verp 30 days thereaftery as long as the ther-apy tbntlAue.t(2)Thebrtief snmmaryshall~spectfl- tallyInUude the fotlowmr. . (i) A statement'that oral contrscep- tlvesare effective. Dut thatany failure totake them in aaeordance with the recommended dosage int3easess the cban[e of plegnanry.,- . rtCeaAt etP(sTFR,.YOt. 41, HO..21_2tltsoAr. lAN[lA[r 31, 177a . ,
Page 57: azo61e00
53 RU[ES ANO REGULA7IONS 4221 themanutfacturer, packer, relabeler or ehildbirth. espeefallyif s)ie intends to(6) Patient labeling for esch oral s tributor.. stributor. ' -breestfeed. the baby, pointing out that - contraceptive drug product shall tie d (lU A statement aft?ie speelfieltems' including the relative risk (where of ~ history to be told the physidann imown) faced by users eompared'to that.would lead the physician not to nonusers and ttierelationship of the prescrtbe oralieontea.eeptlves (Le.- the sfde effectJl to age, smokmg.,and otherr epntraindlctlons to use).~ - ._ condlUens. The side effects menNnned (1f1) A statement that oral contr~p- shail Include thrombophieblt)e. pul- tives should'be.taken only under themonary embolism, retiaall . arterycon(dmJed supervision of aphysiciana thrombos(s, strok'e (the relation of (iv). Al)sting ofthe acrioua sfde ef- these too estrogen doseshall be meJS- feetsororal cantraeeptives. sueh, as tioned), my0rsrdlLL.infaretion.berilgn thrombophlebit)s, pulmonary~ embo- hepatte adenam.s.. fnductfon of fetal LL'sm," myocardlal infaretlon. retinal ' abJSormalltleAand gallbladder dlseue. artery throntbasfs, stroke, benismhe- The follo.vutashall be mentlOnedJ (a) patlcadenomas;;lnductlon.of fetal ab- Pstrogens have been shown to cause normalitles,and;allbladderdLUsse., eancer in anlmaLl.:whlch showina jus- (O) A statement', inttieformof a tiflesthelnfereneethatestrogensmsy boxed warning that cigarette smoknllt nuse cancer 1n humans (b)', there lamcrea.ses.the, risks of serious sideef- strong evidence that estrogen usein- feefs on the beart and blood vesselsereases thee risk of endometriil cancer from oral contraceptive usea and adv)s: in postmenopausal women: (c):tnere h in[ women who use oral tontracep.- someevldence that sequential orsl Uvesnot to smoke. - - contraeeptlve (which are no lonyer (v) A statement of the most marketed may4mcresse thee rLk of en- eommon side effectaM sueh aa nausea dometrial'cancer in womenI (d) studfes and vomiting.. weight change, change of an associatlon between oral crontta- m menses, and breast tenderness cepUves and breast-cancer are largely lvil)A statement that the estrogen negative except for a suggestton of In- la'oral contraceptivess has been found creased risk (one study) in women to nuse breas' , mncer.aad otherr cats- with benign breastdisease. . and tlsere cers in certain animals and that these isno evidence of'anincreased risk of findings suggest that oralicontracep- uterine cancer Inusers of oral mntra- tlves may also cause cancerih humsns ceptivee other than sequentfals. but that studies to date. mwomen -(!x) iA statement af common aide eL taking currently marketed oral,eontis tecte. tiJcluding nausea and vomiting, , ceptivesheve ttot', confi'rmedthat oral weight change, darteningof theskin,.. contraceDtlves cause cancer m changes In menses, and a statement of b-(m+^. - . -other serioua side effects, Including(vili). A statement that'-af contra- worsened mi!'raine, and worsened'ceptives are of no value in the pnven- heart:or kidney disease due to fluld za tlon- treatmentof,venerealdlseaset tentlon., growth of uterlnefibroid (l:)'Astatementcallingattentlontutumora..depression: JaundOCe,.delayedtlie detailed patient labeling and a rec- return to fertility, blood pressure ele- ommendatioo that (t'~be carefully read vation. decreased aluaose tolerance. (3) The detai)ed paticnt labeling sn.i elevated blood llpida... ' shall be a separate printed leaflet in. (xD.A'statement'ofTeported slde ef- dependentof any~additlonal. mater,ali feetsnot def)nite1yrelated!to:orallcon- It shail specifically Include thefollow-traceptive use. ~- /ng;, U(1):A'statementcautloning the pa- (i) Name of the drug. - ' tlent to consult her phys4cian ~~ before -[LLlName and place of, business of - resuming the use of . the drug after (lil) AStatementthatoralcontracep- the hormones Ia thee drug are.known Uves are effecti0e but- cause. cer- to appear )m thee milk and: may de- taln serious sideeffecrse aeaae the flo- (IV):AStatement', that oral contracep- (2i11 AcomparLson of the risk of tives should;be.taken only.under the deathh from various oontracepUve contmusd supervision of a phYSleianM methoda (oral contraceptivess m smoa- iv) A'statement ofthe.effectiveness- en,oralcontraceptlvesmoonsmokers.oforal contracepti.ves, )neludmg the IIID.,eondom o[disphraam1 condom dlfereneess in effectivetKss amona dif- or diaphragm ~-with abortion in the ferent typess and the relationsliip be- event of pregnancy. no contraception tween - effectiveness and estrogen _ but abortion )n the event of pregnan- dosage. cy, and:no contraceptlonn or abortlon).1 (vi) A'summaryof the effectiveness -. (x111). A atatement of thespedflc of other methods of contraceptloa ltema of h)story too be tuldd the physi- (viU: A tioxed warning stating that clan which would'lesd the phyplefan cigarettesmokiisgincreaaestherisl[of not to prescribe Oral eontraceptivee aerlous side effects on.the heart and (Le.,thecontraindlcatioostouse)., blood vesselss from oral contraceptive (xiv) :Astatement'ofspeclficitems.of use and advuinY women ~ who useoral history thatmight. (muse the physician contraceptfveanottosmoke- to sluggestanother method (e.g., risk (vtii)A warnmgregardin[ the serl-, factors for myocardiall infaretYon. ous side effeets o4oralconttaeeptlves, family history of breast cancer.or past history, of fibrocysUC disase or abpor- mal Jaammogram, gallbiadderdlsease)', or would require the physician'sspe-cial attentfon(e.g.n tnigtaine, asthma,. epllepsy, heart or kidney dSxa.x, fl- brotda, history of depressfonl: (xv).A ststement that jaundfee, de- presslon, breastt lumpa: , and: the par- ticulu wara)ni slgnals of ttiromboem- boldc disease. thrombotic dlsease, and ruptured; hepattc adenoma- should be reported ta the physician. - . (xvi) A statement uf how.to take csal contraceptives properlyand what tu do m the event of bnaortwo missed perlods.. (xv1U YA statement: auttoniag the pa- Uent.thatthis, drug has been pre- tcribed for the particular individual onlyy and that thee dtluII maet notbe - given to others. (xvili) ~TDe date. Identifledas such, . of the mos0 recent revision of the la- belina prominently. DlacedimmediataIy'd after the last sectfon of sucn label- h[Q. _- (1) Fbr.tnose oral tdntracepttve drug' prodIDeta.w(thapproved new dtve ap-plfcationa for indl<ations In additlon taW contracepnon- both the brief sum-rv and'detailed patient libelinamay identlfy these other indlcations.. If the other.mdfrationsare Identif)ed the la- being must speclfimAly, include a statement thatthe Information in the patient labeling re4ative to coptralndlJ utions. thedangen ofors.l'centracep- - tivea-:and the safe use of the drugare also applieablewhen these drugs are used for these:other indintions. (5). The detaued patlent.labeling-shall be printed in accordance with the folYowing spec(fleanons: - - (I)~ThemmirJlmum letter siu (lower• case letter "o" or itsleQnivalent) '.shalls be not less than K- incn in height (LL) The body copy sh'allicantaJn 1[ pointleading and noncondensed'type, and sha1D not contain any.liaht face. -" type or small capital letters provided too the retailer by the manu- facturer, packer- relibeler, or dist-ibu- tor as follows . (1):TGe.brief summary patient label- Ing stiallibe included in each peekaae Intended to bee dispeoaed to the pa- - tlent. [hl Thee detailed patient labeling stiallbe lncluded in orahall'locompa- nyeach pack'ageintended to be dts- pensed m the patleni - - - (Lii) In the case of -al icontinceptivq.. drug produetsin bulk packages mtend- ed for multiple dlspensing: a suffib:ent . number of patient labeling pieees shall . be Included In or shall zccompany eacA bulk package to assure that bonh pieces can b'e furnLshed vntti ~ each package dispensed to everyy patient: - Each bulkk package shall be labeled with Instructions too the dispenser to include bothpatcenth labe4ing pieces tap[aAL eEnlsrEL-VOl 43. NO- 21i-rUEEDLT, JAJ1ULaY, 11, lera 41 '
Page 58: azo61e00
9 0 5 094 -i
Page 59: azo61e00
SG006SE0 I I I
Page 60: azo61e00
56' 4224 - NOTN3FS of' the :nnual number of deaths genital tract Including ehanges In.then amongoral conttaceptlve usen in cervtcal mit¢ua (which lnerease the dlf• terIDa of whether or not theuaerfldultyof sperm penetntton)'andtlie amokea eadometrium (wh1cE reduce tGell1te11- The Food and Drug Admfnlstntlon hood otlmplantatlon) may also mn- wlll regardd as m(sbeandedd and sublect trlbute to contraceptive etfecttveaea. tcreaulatory act)oa aayy onl'matta-ceptlve drus produet that Is sh4pped In roa, ra0ormoetx oa.u. Cowraaevmss Intenrtate commerce by manufactur- ' omT ets, repackera. relsbelers. Oe own-label ' The . DrtmarY -mert..nt..- tlupugh . d'utrlbuton,: on or after Aprll 3. 197B.whleh.(IEUertname of.drva):Dreveats wltGoutlabeltne w)Lch Is aubstantlally concepUon la not kaown, but progesto- the same in content as the physiclangen only eontraeeptivesare mown to labeling set forth in thU notice. Qnder -. alter the cervieul mueuag exert a Dto the proviafons of' } 314.&dl i(21CFR gestat.fonal i. effect on the endome- 314.8(d))„such labeling may'be put'' in •.trtum: . mterferb5g wlth Implantation, uae before aDproval ot a supplement to' and. In some patieat; suppress .ovu]a- a newdtug appllnqun. - ' t(oa. (Manmfacturer to Ibtol'ude Infor- Roldera of approved new drug appll- mattop nn absorptloa dtstrlbutloqmtlona for oral eontraceptfve dlugg ellinaNoym and phe^=S:_Itlnetfca If prodncta shall 'submlt supplementa on pertlnenCalso on dru; interactions or before Aprtl 3, 1978, to provlde for pertihant to humap use,) the revised physiclan labelhtf- The Commissloner advlua that the"leptatztotn Axn IIs.cs pat(ent.labeling text"et forth la this (Invert name of drug) Is Indlkste:d noti¢e complles wtth the patient label- tor the ing ftnall rule (¢310.501(a)) and can be - preventloa ofpsegnancyla relled:uponbyaoypersonto:meettlte. 'eOmen who elect ta uu oral~contra- CYptFve3 as a methmd'~o1 contraception. rule's requlremeata For thox: avanu- (~.,~~uren who havs: other ap- futurers and suppllers.who elected to proved ladirattons for oral'cpmtrscept ux the December 7..1976 gultlellae pa- tlves (EnoWd S' mg. . Orth'o-Novum 2 tfent laEelmg (as nvlied' ~~ on May27. ~ Ortho-Novum 10 mg) snould men- 19T7)- use of the December 7, , 19768 tioa thox Indlcatlons bere.l tezt will con[Inue to meet'the requlre- Oral mntraeeDtfvea are ~latilyeffee• -menm E-the tlnelitWe (g330.501(a))'. t1ve. The pregnaney nte1n womea until May3t. 1976. For ell~.manufac- ~~ ronveatlonal'cumb(natlonoral'~ turen and suppllerswhou have de- contraceptives (conta/.dng35 mcg or ferred dlstrlbuting revised Patient Is-more of ethinyl estrndlolior 50 mcg or bellns b'asedon the December 7. 1976 ro of mestraaml] Is generaLLyrepolrty guldelEne, use of the earlier text ~ ~ yv lav than oae pregaancy per 1110' cannot be relled upon as meeting the womaa-years, of ux.. Sllihty higher requlrementa otrthe rule.. - rates (somewhat more than 1 pregnan• The phYSidaa labeling for ortl eoa-cy.oer 100 woman-yean of use) are ro- traeeptlve drug products Is set forth aa ~ for some combinatlon ~ptroducts fol3ot¢s ~ ~- -- cuntalntng 33 -cB or.levof ethlnyl es- Ow.Coertsweerrrvclaer,t.uo. ttadlol, and rates.on-the order of 3 Descamzox pregnandes per 100 women.years are reported for the progestogen only oral (SO er sprrtrED aY NulDIACl'aaLl/) - eoatrYCept)vea.. . :.+.-. (Descrlptfon should Include the These ratea are derlved from seps. tollowimg:nformatlon) rate stvdies conducted bY dlfferentln- veat!gatom in sevetal pop>rlatlon L The proprietaryy name and the es- groopa and cannot'' be compared pre- tablisned name tf any. of the drug dsely. Furtltermore. pregnancy xrtes Droduct .. . tend to be lower as elfnleal studiea ate 2. The same qualitatlve.and/or qaan- contmued,.possibly due to selective.re- tttaUve Iegretllent. ulformatton as re- tepttonIn the.longer studles of those qu1redforlabelt. - patients who accept'the, treatmenc3. The pharmacologfeal orr therapeu- reglmen and donot dlscontlnue asa tte class of the drug product; result of advente reaetlons, pregmncy', 4. The chemical name and structural orother reasons.: -formula: ln clinical trlaLV with (Insertn'ame of, ~c•• P~w~ = drug) (tnsert number of) patients com- .. plete -- cycles aad a total of - ron COSatN42rox o_ CCNSawCaTrvm pRgmundes were reported. Thit repre- om= - - senti . a pregnancynte ot -- per - 100 weman-yean. (Maaufacturerto Combination ory'contraceptlvea act sdd'other.(nformatlon_related to the prlmar0y through the mechattism of pregnancy rate with his partlcular gonedotropln suppression due too the product~ if needed too provIde adequate estrogen1c and'' progestattonalaettvlty prEscrl6'utg Informatton to the physl=f of the myredienta. Although the prl- elaa). marr mechanism of actton IE mtiibl• Table 1 gives rangea ofpregnaney tion of ovulatton,~ alterations Inthetates reported In the literature (ReL 1) for other means of contneeptioa The effiracy of these meaas of contra- aptlon (exceptthe lIID) depends upon the degree of adherence too the method.: ~ . Taau l PRgsteurcnts Pat 100 Wosme-Ysaaa IDD, less than 1-0;'.- -. Dlaphragmwlth apermlddal prod- ueta (ereams or lellles). 2-2(k. Condom, 3a6( - Aerosol foatos. 2-29: . Jelllks and creams.: 4-3d: Perlodle abstttnence .(tyhthm) all types, leaathan 1-47: L Caleadar method. 14-t7; " - 2. Temperature method, 1-20: - 3. Temperature method-lntetocusa- only m post-ovWatory phase. less than 1=7; - 4. Mucus method. 1 -25; No contraceptloa. 60-6(L - . Dosa-Rawrm Rtsz or 'Pmtosaonoo- vsat Fsou Oau CoMxa.icaPrsom Two studiee have shown a posltlve- acsociatfon between the.dose of estto gens in orall conNaceptivea aad: thee risk of thromboem.bollsm (rets 2and. 3): For this reason. it ts prudent and In keepinY with good prlnclDlea efthera- peutlm to m(*f....ise exDosureto estro- gen., The oral contracept(ve product presc:(bed for anyglven patient should bee that product which coatalds the least amount of estrogen that fs compatible with an.acceptable preg- naacy nteand patient acceptance. It Is.recommended that new acceptors of oral conttaceptlVes be etarted on prep- aratlons containing 0'3 mg or lesa of estrogea. - -- ' CoHriun.munors _ Oral! contraceptfves should not be usedmwomen with any of the follow. . Ingcondltfons - - L Thrcmbophlebitis cr t)iromboem- bclle disordets.. 2. Apast history of, deep vein throm- bophl6bLt(s or thrumb'oembolle tlisorr ders.--. . .. 3. Ceretirsl vsseular or coronary artery dlsease." 4. Known or suapected carcinoma of the brea.t. 5. 8nown or suspected estrogen de- peadent neoplaala." 6. IIadlagnosed abnormal -genltat'. b1eedins•. T. %nown or. suspected pregnancy (see'aaralnr No. 5). - . W.uxares - - Clsvette smokma= mtreaw the rt'k of se. r(etu c.rmovascudar sfde effecta lrom ora) [OLaraceptRe Ure- ThJS risk ILcreEseE w1tE aEe and with heevy smoklna tIS a ds..lttle Ver w) aCdIl awte.ma'kea mver 33 rean et are Women sho me era( [onUaeept(.o /hould be stnOglr aCVtretl not W m,uk'a.. EEDEaAi IEGRtEE,.YOL 43, HO.21-rYESbAY, JANUARf 31, 1970 ~ ~
Page 61: azo61e00
57 Nonc.>s 422s' The use of eral contraceptives isas- however, were fauad to be a clearr ad-deeree'of'thromboembolic risk assacl- soeiated with increased risk of, severali dl¢SOnalrtskfactorl ated with progestogenonlyoral con- serlousconditlons including throm--. In terms ofreiatlre risk,.it has been traceptives have not been performed boembolism. stroke, myoeardlal Infare- estimated (Ref. 52) that oral contra- Cases of thromboembolic disease have. tlon, hepatic adepoma, gall bladder eepttve usen who do not smoke (smok- been reportedlin women using thesee dsease, hyperteps[on. Practlt/oners Lna iscons4dered a major predisposmg products.. apd they'ahoula nmt bee prc prescrtbing oral contraceptlves should condition to myotardial iafarction) are sumed:1o be free of excessrtsk.s be famSllar with the following ihfor- about twice as llkedyto.havea faral - - _ matlotrrelating to.theae nska - - myocardlel infarc:ion as nonuxn who Esmtaxtor F]tt2ss Montamr Faox 1. T7eromboentbelie Dieordert: and do not smeke. Oral contra<eptive usen . ClxemaxoarDisusrs OWr Vaacudnr Problems An increased wtlo are aLso smoken have aboat a 5- - rtsk of.ttuomboembolieapd thrombo- _ fold increased rlsit of fatal intaretton Alargeprospectfve study (Re(- 53) tic duease aa5o¢tated wlth the use of campared:to usen wha do not smoke, ' Carried out in theII.K.estlmated the oral contraceDtives Is well estsblHshed. butatiout a 10- to. 12-fold mcreased mortality rate: per 106.OU00 women per:71ree principal stvdies mGreat Brit- risk compared to.nonusers who do net syztem for ~en and nonusen ot orsl am (Refs.4[hrouehB) znd three in -~oke. Furthermore, the amount of eooltzacepqves accordi¢g to age, smok-. the Unlt,ed States (Rets. 7 through 10) =oking i3 alse an Important factor, In-. Los habits; anddmation ofuse: The have demonstlated an inereasedd risk determining the iu.portance of these pverall excess death rate annuallyy of fatal andnonfatal venour throm- relative r'sks, however, the baseline ntes for varlolYs age group3, as shown from circulatory diseasesSOr oni eon- boembolism andd strake, both hemorr- W qybje 3 must be: given seriouse con- traae.ptive usen was esti.mated to tie:20~ - _hagic andthrombotle These studies sideratloa The:importaaceof'other per 100.000 (agesL5-34-5/100:000;: estimatee that.usent or oral[ contracep, prMicposing conditions mentioned ages. 35-t4-33/100.000: ages45.-49- tives are 4 to 11 times more likely than above in determining rela[lve and ab- 140/ LD0.000); the risk being cnn<en. nonusen to develop tltese dseases soLute risks has nmt as yet been quanW trated in older women, In those a:ith a- without evident cause ('d'able 27. _ fled: it is quite likelyy that the same . long tlnrtion of use.. and :n ctparette synergistic action esists;,Wlt perhapesocoken.Itvrasnotpossibie,.however„ Caataaovascvtaa DtsoxnGts to a lessv eztentto. exam!ne the interrelatienvhlps of - In a collaborative. Ameriean study _ age„smvking, and duratfon ofuse, nor (Refs: 9 and 10) of ceretirovascular dts-T-a mcamDare the effectsofoontlnjous orden in women with and wtt[iouC pre-Fvttma[ed anoWJ mortallRf nte.perl00.000. venus i¢tErm1ttent use.,Al[hof1.¢:. ;he disposi,ng uuses, it vJas estimated that mev tram myourdW fnfarcqon by use stndy howed a 10-fold iacreaxm. ad ozal ~ caninr ptfve; .~mkiaa habua _ deatti du'e too circulatory disesses in thee risk of hemorrbagiostrokewu 2.D usea for 5 or moreyean: aliat these tlmes greater inusenn than nonusen _ i°d'a° (l° r'a")- -- --. -.- -- deaths occurred ln womep,95 or.olderr apd,the:risk of thromhotiastroke was 4 to 9.5 times greater invaen than in nonusers (Table 2). - - " __Tasu2-:. ' rm us: or nuosoorsuo- 0.4~LLN braea ra04 Until larger numben of women.under. 33 with continuous use for 5 or moreyeane are available, It Lss nmt possible too assesa the magttitbde oftheredative riii for this younger age group. The available data from a var',ety of sou:cr_s have been analyaed'(Ref. 14) to estimate the risk of'dcathastociat- edwith variows methods of contracep- t1on. TTe estimatea of risk of deatti dar each methodinclude the romhmed risk of the rontrsGept.vemetbod (e.g.,. t'[romboemboLic and th:omDOtic dise ease. m 2he raSe of oral cOntr3ceptives ) plus therisk attribu ta ble to pregnancy or abortion in th'e event nf method failure. This latter risk varles wi¢h the effectlvenesa of the contracephve method The fmditgs of ttilsanalysia are shown (-F7gure 1below (Ref. 14).. The 1tudA concluded that the mortal- Ltyassosiated with all methods of: birth mntrol Ls low and below[hat as, sociated w4th u'ildbirth-.with theex- ception of oral oontraceptlles in wcmen over 40who smoke..(The.r.tus pven for pill onlyismaken for each age group are for smokenal 3 elesl For, "heavy"' mokers (moretb'an 15 ciaarettes aday), thentesglven would bee aboutdouble: for "llght" smoken (less than 15 c:garettes a day), about 50 percenL). The lowest~ mortality is associated . with the . condom or diaphragmbackedup'bym early.abortion - . - - 5[rouf W.1 Wvrtlun . w ~~ ~.u.d .. ib19 ea women - . 4a-N anakSMLemb. n.em Nueu+us e.cnNVOr.em Aamokrn.- [0] i0EaY IS.Y .~ laA 5.1 ial ]L] SMfcelw u[oi~ 6'! : ~.e s'1 - - _ - rut NYmwten- lA 1.r :f i 7.4 _ . _.. . . . . . ~ok.,.~e - .- ~ w - r... . v: taev.w< uvemawmnour ans,. _ mnmaken_ ak. eS lii 1" Patmen uvmmEOempellkmmpbuWn.__ kJ .HyR woo- ta ef mon aprtts• eac Aar.. SLmmm[f! sssot. _____ 4'r.f pi,nm JYC A 8:. StuCl~ m IamJ) P..vWna, a50.. eemertasreavote f - 1~... . _ YSOOtCW.Wa~en0a YI]' . _ .. . MxocaansaLIxraSCnos . -. - .. -Rna or posx An: increased: risk of myotardial'Ln- faretion ,associated with the uae ot oral contraceptivahas been reported (Refs... 11. 12; and13), conf)rming a previo(uslysuspected association. These studies, conducted In the Unlted %ingdom.. found as expectedthat the - In an analysisof data derived from several i national i adverse reaction re- portingsystems (Ref. 2), Brit[shinves- tlgators concduded that ~ the risk of thromtioembolfsm including coronary thrombosisIs dllrect.ly related to the do5e of estroge-used. In onl i tvntra. greater the number of underlying risk meg or more of estrogen were asaooiat- tactors for coronary artery dlsesseedwith,ahlgher, rlskof~thromboembo- (dgarette smoktng,: hypertension,.- llam tdan,those.eontatnlne 50-80 meg hyperchal6aterolemia obesity. dilabe-- of, estzogen. Their analysis did suggest,, tea, historyoII preeclamptlc toxemfa): how.wer, that the quantityy of estrogen _hee higher the rnsit of developing myo- may not be the sole factor involved eirdial infarctlbn, regardless of wheth- Tlltss finding 7[as been tonfirmed Ln er the patient was an oral contracep- the United States (Ref. 7): Careful epl- tive user-or not Oral tontntceptlves, -demlological studies todetermfne.the fEpEaAl !e0(LrEe,:VOL C5, ND., 21-rpllpAr. JANDAa7 31, 1979 eeptlAea preparations containing 1D0 ~ t
Page 62: azo61e00
58' 4226 a,.. ,....,.......,,,,....... _... ~ ~... r..,......,..~.. ,a~.-... .« ..,.... ,.._..: _ . v a w a aa .~.u....~.. 2i a.........~..,. u.e~......,..~r . Q M...+r E];' -.w al. !J r "..f' l uo.... ~ I The risk of tluomboembollc and The physlcian andthe patient thrombotic disease aasociated with shoutd bee alert to the earliest manifea- oral icontraceptlves Increases withaQe tatfoos of throctboembolle and throm- after aparoximately age 30 and„ foe myocardial infuctlon, it further I¢- botle dLLSOrdeta (!;, 'LhmtOtiopblebltll, creased:by hypertenstoa, hyper- eho- pulmonary'embullsm..cextirovascular .. lesfetulemia. otiesi[y: diabetee.. or hLs- iasnf11c1ency.. coronarg' occlusion, re- tor4 of preeclamptic toxemia and espe- tinal thromtioais,. . aod mesenterlc cially by citirette amotins. . thtvmbnsis).: Should aq9 of these Based on the data currently' avaJl- ~~ or be suapeeted, the dtvL should able.. the followine ehart pvea a qtav - . estmlate of the risk of death Irom cir- be dLxnntinued Immedl'ately: culatnrydisorders svoclatedmith theuse of otal A four- to afz-[old tncreesedrLsk of oontracepttves Srmmm aut?aum Oren< Psau-_ poBU sur¢ery thromboemhCllic compW Cowornee.-Antdssecuzm Wsa IIsa catloca has tieen reported In otal con- or:oa.f.COxsaaensnn ttaceDtive unera (rels. 15 and' 16). If feasible,OraJ contnoeptivea showld'be Ma~ dLuontmued at least 4 weeks before H....ramoten C a. AL{fpC®ak.or n C e xoo®ece. ~ n...auw. a. c a c w ®w~mn. <o~necrw.c cx aA eeemeeo.. A-nr u.oeueN.NE.en trlea n.t 8-OM ~eoel.mG.nE Npi rfa C-Ua. Wne1.4d'.IN moAnV. rLY.. t>-ar.uou.ua.nwa.esa proloneed ImmobLllrstlon. 2 OcvlarLe.rfoaaThere have been repory of aeuroocuJnr lesionas such as optlc.neuritia or retinal thrombosis as- sorJated :with thee use of oral coIItra- ceptivee. DLimmtmue oral rnntcacep. - surQerp. of a type assoclated with an - . lncrea.sed risk of,thromtioemboliSm or ttve medl¢ation T then is unez- plained sudden or itadual, partiil or complete 10ssof vtcfoq o05l4 of prCD- to3iss or diplOpl'a:' papilledema: orr re- ttnal vascular lesions and tnstit>ote.sa- propriate dta¢nostic and therapeutic measures. 3. Cortinoma. Lotia-term continuous administi•atfon . of either natural or syntheuo estrogen In eertaln animal speciea Increases the frepvency.of car- emoau of the.breasti.cervlx, va¢1na.. aod llVer. Certain synthetic pro¢eatn- gens,none.currently contained in oral-' contraceptlves, have been noted to ln- creaae the Inddence of mammary nod- ules, benien and mzfipnant, in do6s. in humana- three [ase mntrol stud-les have reported aa Increased risk of endometzlal oarcinomaaseoeiated with thee prolonged use.of ezokenoua estlo- ¢en in post menopausal women (Refs. 17,.18; and 1s).:One publicatlon IReL F®F7.4L lFO1STER VOL 43..NO. 21-TU(SDAY,JANWAa7'31,.1771 \
Page 63: azo61e00
TM 59 20) reported on the first 21 cases sub- mltted by Physlcianss to a re8istry of cases of adenocarcltroma of the eado-metrium in women under 40 on oral'emtraceptivea Of the eaee found In women w(tGout.presdtsposlne risk faa~ tora foradeaoearclnoma of the enda ~ metyum (e.e-,- irreeular bleeding s.t the timeoral eoatca4eptiveswere fikstt elven: polycystic ovarlls). nearly all l ootUrred 16 women who had:uaed a ae~ NDi1CL3 taking oral ~ coatraceptivea The tela- Uonahlp of these druP to this type of maLtttancy'b not:khown at this t)me. - 8. . Use in or fmmediatNy Precedfn9 Preaancy, Birth DeJeets ia O1ltyrinp, and dfafibnancy in FentaL Othprinp ' . The use of femalese:e hormoa- both esttcserilc and proeesfatianala8enta-durin8 early pre8namcy' may seeiouly damage the offsprln8.. Itt haa been shown that fema1 Quendal oral conmeeptive- These utero too dl'eUiylrtllbertro4 a nonsterol- - reelmea. If the PaUent has not ad- Producta ara no longer marketcd No dil estroeen,. have an inereaxd'risk of hered to the . prescribed schedule,, the evidence has been reported su88eatitie developtne !n later)ife a form of,vad-poelbllltyof pre8nancy should be con- aa Increased: risk of endometrfali nal or retvicalcancer that')s ordinarllysidered at.the time of the firat mimed' ~~ cancer 1n usen of conventlomal tvmbl-exttemely rare (RefY. 31 and 32). This period (orr after 45 days from the last-t natlon or Proeesto8enculy oral con. risk has tieen estimated to be of the menstruslperSod if the Proeeacneen traeepU4es,. _ -- ~ order af 1 In 1,000 exposuresor less only oral conttsceptives atausedl, and Several studies (Refs. 8-and 21 (Refs. 33 and 47). Although there )a no further use of oral conttaceptlves through 24) have found ne Increase. mevidence at the presenttluae that oral ~ should be withheld untli presAancy breast caacer !a women taking oral i cuatrueptlves further etttieacee the has been ruled outi If pregnancy )s contrscepUvesorestro8ens.One:study' risk of developing this typeof.malls•oottflrmed..the patient should be ap- (Ref: 28). however„ while a)ao noting nancy, such patfente should be mont- prlsedof the potential risks to the ae uveraall'l incresaed risk of breast tored:wlthparUcularcarelftheyelecth tetus and theadvlabillty of conuinu- rmcerIn women treated with oral' to, use oral i contraceptives Instead: of atlon of the pregnancy should be'd)s-contraceptlvea, found an excesa risk mother methodi of runtrxepUon. F1tr- cussed in the light of these.risks.n thesub8roupl of oralcontraeeptlve- thermore- a)ti8h percenta8e of'such(Manufecturert4suyplyappropriate usern with doeumentedbem8n breast . exposedwomen(from30to909e).have InformatlomforusemendametriosLi)dtseax.. A' reducedotsurreeee of been found to have eptthellal chanees It.ila)so recommended that'woment benlen breast tumon in users of oral ot the va$na and cervia (Refa -34 contracepttves has been well-docu- through 38). Although these chanaes mented (Refl. 8,31. 25. 28, and 27): . are h)stolopeally benfen, It Is not In ^•m=••y, , ttiere is at Preaent no - kaowa whether th)scondlUOn Is a pre- eomffrmed evldlnce: lrom human stud- cunor of- vaginal malignancy. Male les of'anmcreased risk ot cancer aaso- children so e=posed may develop ab- eiated with oral contraeeptlves.:Close normalltiea of . the uroeenital tractt clinical' surveillance of all women (Refs..48,,49,and50):Althoueh.simllar thklfaY oral conaraceptlves )s, neverthe• data are not avallable with the use of lesa essential: In all cases of undife- other estrogens. 1t cannot be presumed: nose7 persistent or recurrent abnor- that they would not laduce similar mal vaginal bleedlhe, appropriate dis8- chan8es . - nostle. measurea should; . be taken to Ac lnareased riskk ofcoaeenital ' rule out malignancy. Women with a anomallea, includln8 heart defects and; strone family history of breast cancer llmti defecta- hea been reported'with or who have:breastnodu)es, fibrocys- the use of sex hormoaes- includdn¢ tic disease.or aEnormalimammoerams eralcontrsceptlves, m preenancyshould be moniared with particulhr -(Refa. 39 throueh,42, 51). One case care. Sf they elect to use oral. tontra, control study (Ref, 42) has estimated aa cepUvea)nsteadof other methods of 4.7-fold Increase In risk of hmb-redue- contraceptton... - Uon dlfects In If<famts expesed )n L. Ifepatic 71- Benign hepatic utere to sex hormones (oral icantrscep- adenomashavebeenfound:tabeassao- Uves, hormonal withdrawaltests for ciatedw)th the uae.of orall contracep- pre8nancy.or attempted treatment for tlvn ( Re"s. 28.. 29. , 36. , aad'46): One threatened abor[foaL_Somee of chex study(Re1. 48) showed that~oral con- eapoaures weres very short and In- ttaceptlve formulAtmns with hi8h hor- volvedd only a fewdaysw of treatment momal Potettcy were aasociated with a The data suggest that che risk of limb- higher ri+kthan lower potency formu-reducttondefecta in exposed fetwea is lltfons. Although benign. hepatic ad- '6omewhat lesa tltanone,ln 1;00011ve enemas may rupture and may cause blrtba. death through Intra-abdom(na4 hem-In the past(i, female: sez- hormonea orrhaae... This has been reported In have been uaed' during pre8aaney m-n shorLterm as well as loni•term usen an attempt to treat threatened or ba-of oral rontreceptlvea.. Two studies Ditual abortion. There IDs considerable zlater'akwf2hdurstlonoEuseoftheevldencethatessroeensareineffectlvecontracept:ve, the r1skbelna much for these ludidstloas.m and theue Is no greater after 4 or more yeara of oral evldence.from well comtrolled'studies contraceptive use (Refs: 30 and 46). ..that proenatoeenz are--effeetivefor While bepaUC adenoma is a rare these uses. - 4227 teptlvea (Ret.; 43). EbtEeyea with these anomallm are virt(tally: always aborted'd spontanirously. Whether there )s anoverall ln¢reeasse In spontaneous abor-- Uon of pre8nanciea conceived soonn after stopping oral contracepUvea L. ualmowa It is recommended that for aa,y pa-Uent who hasmiEsedtwo consecutlveperlods, preenancy should be ruled'out beforecontmulnee the contraceptive who discontinue oralmmtrattptlveswfth the datent.of becomine preeaant use an alternate form of contracepelon for a period of time befon attempting to conceive. Mfnyy cllnir/.ns rectsm- mend 33 months althoueh no precise mforaatton 13 available on whiah to basethisrecemmendatlun. . The administration ol proeestosen- only or proIIestoten-estroeen comblaa- Uons to Induce withdrawal bleeding should not be used as a test of pree-. aaney: 6. GaU~BladdsrDisea.ta Studles(Refs. 8; 23;,and 28) ~repoct an increaaed risk of surgically con- flrmed gall bladder diseaae. tn.usan of oval eontraerptlvesand'estro8ens.. In.n o0estudy: aa!lcreased ri51t appeared after 2 years ofuse aeddoubled after 4 or5 years of use. In one of the other studies, an mcreased risk was apparent between 6 and 12 montti.s of use:.. 7, Carbofuydr.ate and Lipid bfetaboG(c FJhctt- . -. . . A decreaae in glucose tolerance has been observed in a s68nificant pectent- aee of'patients On oral oontraceptives... For this reason,. prediabetlcanddla- betio pattents should be carefully ob- served while receiving oral contracep- Uves. An Increasee in triglycerides and total phosphollpids has been observed m pa- Uen(s receiving oral comtraceptlves (Ref. 44).: The clinical signif)cance: of this finding remains co be defined -- 8. Plevafed Blood Psesrura -. An mrzeaae m blood presvure has- been reported In patients receiving . lesion, it should be conslAered'Ih There Is some evidence that trl-oral coatraceptives (ReL 28). In some . women presentina abdominal!pain and ploldyand pcaaibly other types of po- ~women,, hypertension may omnr tenderness, abdominsl mass otshotlt lypleidy are increased amon9 abor-wlthin a few.moothsof begtnnina.oral i. A' few waes of hepatocellular urcl-` tuses from . women whobetomepre8• rantracepeive. use. In the. first yearof noma have beenreported in women aent. soen after ceasiae oral concrao- use, the prevalence of women w)thby- feDeAt lloR7ER, VOl 43,NO- 21-rUt5DAY, JANUA87i 21, 1978 III
Page 64: azo61e00
Z 60 4228 -_ NOTICES - pertenslon 1alow In uues and may be. 11. LrtoptrPrepnenep:.. - 7:Oral controeepttte usen may haven no hlgher ttian that of a comparable J:cCople as well ae mtrauterlne Pre=- dlsnu Wncea In nortmal' tryptoDhan group of nnnusets T2.e prevalence in nancy may oteur in comtraceptive 1sLL+ metabolism which may teault In a rela- usete lncreaw, however- wttlt longer urea. However, m Progestogen-only 11ve Pyrldozlte deficieney. The ehnfcal ezpoture, and Inthe fifth year of use oral contt•eeptlve failurea- the ratio of at...+n.,.nce of thla b yet to be deter- la two and a haltto three ttmea the re- eetopue mmtnutetine Pregaanelea Is mined ported prevalence in the fJrat year. higher than in women who ase not re- - g-. Serum foiate jerela may be de_ ~ Age isalsn aeroagly eamlated with r<lvfng eaal centraaptlve, alnte the presaed.tiy oral contraceptive therapy. the developmentt of hypertmtslpn m dzuga are moraeffectlve.ln pneventing Slaee the pre=snt woman la Predfa. oralcnntracepnveusers Womeawho IntrautertnethanectopldDreaaandaa, posedtothedevelopment~offolatede- prevtousty have had Eypertenalon 12. Bxaaf Feedtn0. flmeney Inanx withr*•.+++_ •~ duringDregnancy may be more likely Oral wntraoepqvea given in the ttpn, It isposaible that if a woman be- mdevelopelevation~.oIIblood, presure PAstpartumperlodmayln},erfere.with rnmes.pregnanLshortlyafterstoppings wben given o[fi contteeeptivea Hyper- lacratlon. There may be a deereaae m. oral centreceGtlves,M she may hsvee atension that.develops as a result of the quantity and auality.of the breast greyter cbaoce of developing folate de- taking oralrnnttxeptlves usually re- mDlt. Pttrthetmore, a amala fractlon of tlCencyy and cumpldcatlons attributed turas to normal after dfsrnmtmuing the . hormonal agents 1n oral oontra- to tAis deiltleney.. the drug. . __ -. ceDtltee has beenldentlfied' in the 9'. The pattiologi.it'should be aftised 9. ReadceAe. ~. milk of motherss receiving these drugs of oral contxacepttvs: therapy' when The onset orr exacerbatioa of m1- (Ref. 45).. The eflects.. If any. on,the relevant speclmens are aubmitted grame or development~ of headache of breaat fed child have not been deter- 10. Certaln endocrine and liter fune- " a nev.pattern whiehIa recurrent, per- m+oed. If feaaible, the use of oral con- tmn tests and blood components may sistent.orsevere.teeulrea.dlscon6mua- tr.cepti.vea ahouldl be.deferrsd untll be atlecced byeatrogen-coatainimgtioa oforalcontraceptivea andevatua- the Infant bas been weaned. . - oral contraceptives. tlooofLheouse. - a. ]:ncreased sulfobromuphthaldm10.~ Bleedinp Irreputaritfea retentlOm- . Breakthrough bleeding. spot.ting, . . . .. esasau, .. b. Inmeased:protbrombin and tac- andamenorrliea are frepuent reaseas for pstlenta diacuntlnuing eval cvnt- I.. A comilete medical aad i tots VII. VIII- I$ and C decreased an-. ~y' . ceptlvee Iabresktlirough bteedm[. at hlirtoy aaau1d be taken Drior to. Che tithrombtn 3; Increased norepineph- -mdu¢ed Plstelet aggregabill.ty-.. m e11 cases of Irregu1ar bleeding frum ~tfstlon.oforaJ contra¢eytives, Tbe ~~,.,e_.ed thyroid binding Uabuiin, . the vagtna,, nonfuinetloaal'. causess metxatmeat'~and periodie DhSSica1, ex- [TBG) leading to imcreased cixizilating ed ahould~d be persiatent borne m or mmd In recuareat abnor undisg- ~natlona ahomld inrlude sDedal ret-, ww thyroid hormoae, u measured'b no+w erenm to bined pMUurq breests, W protem-buuad iodlne (PBI):. T4 by mal bleeding from the vagiry aGe- dmmen aad petvte orgatu,, meluding rolumn, or TS by tadloimmuno auated(agnostfcmeasureaaremd!<at. ~D~tDla°usmearandrelevantlatio- a~Sy rataryLesei. AS a genezal[ule.l ors3, Free TIresin.uptake3 is decteueQ re- ed to nile out.pregnamcy or mallgnm, ppntruepti.vea ahould nut - bee pre- C~t1agthet elevated TBG„ free Ta ey.,If pathology has been exduded, rnncentntlonis.unaltered . time or a change to another tormula- ~b~ tor longer thaa. 1 yesi.witliout d- Deereased pregnanediul exccetloa " t:on may solve the problem..Changlny ~ntherpmysiesl examinatioaDaing ~ R~ucedreaponseto metyrapane to an orai conttaceptive with a higher perlormed ' test estrogeneonteot.g while potentlaAly 2- IInder the Influence otes[zogen- . useful~la.n;ni.nr_n.emeostzusllrregu-l progestOgenpreparatioo;,Dreexistingnvroasuaonroe.rJ$rastmlaritT,.should-tie done onlyif neces- uterine ]eioatyomata msy laveasem. . sixe.. See Patlknt Iabellng Primted DelOw. sary sineee this mayinaease the risk of 3. Patients with a history of psYCCic thromboembollcdESease.. depression should becsrefully ob- aeaoarmt,.axons Fb4luwtng paragraph tobe laserted . served and the drug dlscontmued If de- Reduced etficacy and increased mtl- tlves for progestogenanly, ora] i contrsttp• , pressioa reeura to a serious degree. Pa- deneee of breakthrouglt bleeding have' "-tienta becoming signifacantly de- been associated~ with eoneom)tant~ use An alteration in menstrual patterns pressed wtSJlee taking ordlcontt-a[ep- of rtfampin- A simiuar asSPCiation ha1 la likely.to occur m womenussag Pro- tivea should stop the medieatlon and _been suggested with barbittua[es, geatogenon1y.ottl comtraceptlves The use an alternate method of contracep- plae_.iylbutazoaeL. plienyminsodium,n amaunt and duration of,flow, cyde tion inan attempt to determine and ampicillin., _ - leng[h,, breakthrougl7 bleeding, spot.- whether the symptom isdtug related tlna and ameaorrhea.vrtll probably be 4.. Orat contraceptives may cause eaamat^xzr+xs:a ' Quite variable.. Bleeding Irregularitln some degree offluld reteotioa They See.Warnmgs section for informa- ocettr morefrepuently.wlth the uaeof shoWd be preserfbed'mtth cautlon, and tlon on the carclnogenie potencial or Progestogea-only oral contracepttves~ onlyy with pareful;monitormg. m pa- osaltantraceptlves_ than with the combinatlons and'the tikata vitti.coad(tions which might ba - dropout rate duee to such conditiena is. aggravated by fluid retentioa. such,as raecn,tncT hlghen, co.nvulsfve dllsorder~, migralAeayn- - - - Women,with a past.hkstoty.of o1180-. drome., atthma., or cardiac or renai In. Pregaaacy'.category'. %. See. Contra• meaorrbea or secondary amenorrhea sudflciency. , ind9ratiuns aadWarnings. or young women without reg[tlar 5. Patients with a Dast hittory, of ' yruasy c>corr~a - cycles mayy have a tcndencyy to,remafn lauttdlCeduring pregnaney have an in- . . anovulatory or to become amenorrhelP creased risk of recurrence . of Jaundice Sec Warnings. -. after dttcontinuatlon of oralicootra- while receiving oral conttaceptfve Apvrur.Rctcnoas ceptites Women with these Dreexist- therapy. Iflaundlice tlevelopsin any - ing prublema should be advised of thbpstlent receiving such druga, the medi~- An lnere&*d nsk of the tollowwg se- Dossibility, and encvuragedto aee r8tion should be dl.scontinued', rtous adversee reactiors has been asso- other, contraeeptivemethods.PoaUuse 8, Steroid hormones maybe poorly eiaeed witJf the use of,eral.contracep- anmvulatlon, possibiy prploaged amayy metabollt«d in patients. with impa/red ticea (seeWarynas);; a:ao omur in women,without previoua livertunction and'should be aGtnimis Tttrembephlebirtis. 'vregularitien. - tered with uution in such patients.. Pulmonary emboliim. ' 1IDfGl' aEGISiFI,. VOL 17„NO. 21-TUlSDAY, JANUAlT 11. 19yg l r"
Page 65: azo61e00
61 A 4 NOTICES 4229'~ Coronat9 thtomhasl&. - -" Flytherms aodosum. . Cerebral thrumbasia, . .. • "Hemorrliaglc eruption. . Cerebralhemorrhage.. _ Vaglaltis. - 8ypertenslon. " . " . Potyhyrla. Gall bladder dUeasa ... --: Jleprs ._ Benign heGatomat Ovtaaoa . . Congenitalanomalles..-. -Serlcua ill effects have not been re- There is evldenceof an association ported folloWiagacute lnaestlonofbetmeea the follovving conditions and large dosesi of orall eontraaptlvn by theuse oV oral trnnh•aneptives. al- youngchlldlen.Overdosaaemaynuse thouga, additlonal' conflt'mator7 stud- nauses- and trlthdr.wal bleeding nu7 le7 are needed:. . . occur ia femalks. - - Meseaterlu tbrombostt.: :. • . - - . retlaal DOOAa - ea~77oB . Neuroocular leslons . e g - . „ tbrembosis and opticnehritts The follovring adverse reactions have been reported Inpetlents receiving oral eontraceptlves and are believed to bee drug telated: - Nausea usuallyy the. ~most-eommon atlverse reactiun.- Vomfting, omirs In approzimately 10% or less of patients during tnetitat cycde.Gther reactfons, as a general rule, are seea much less frequently or onlyocC23100111y: Gastrointestlnal aymDtoms (suchat abdominal cramps and blmatalg): Breakthrough bleeding. Spottma. . .. Changeiilmenstrual3SoW.. Dyamenmrrhea - -Amenorrhea during and after tiea4 menL. ' Temporaty Infertillty after dkvcoa- tinuanceof treatment. _ . Edema. - Chloasmaor melisma wtiich mayPeralsL Breast c)ianges:': tenderness, en1arge- me¢t, and secretion., -- Chansee in weight (Increase or de- To achieve maximum coat_raceptfve elfectl0enesa- (insert name of tlrug) "must'tie takeneiactly.as dtrected and at interaall notexeeetlln; 24 houn. (Manufacturer to supp4yapproprlate lnformatton regarding eadometrimsls and hypermeaurrhea mhere appnn- ble_) (Manufacturer to suppl7 lnfurma- tton, on routine adminlStrstlon and _ specific lnstruetiona onhen^ln• prob- lema such as breakthrouah bleedHng., amenorrhea, ete.) i IIae oforalconL-aeeptlveefn the event of a missed meastnlal perioG - 1. If the patient has not adhered;to the prescribed dosage regimen• the possibility of pregnancy should be coa- aldered after the flist mLssedperiod (orafter 43 days from the last.men- strualiperiod It the.pr¢gestogen only oral contraceptives are used) and oral contraceptives should'be withheld until pregnancy has been ruled oult. - 2..11 the patient has adhered to the prescribed regimen and mlises two c•onsec'utlve.perlodi. Pregnancy should - be ruled out aefore continuing the crease): coatraceptlverealmen.- . Change in cervical erosion and eervf- a - How Smeclm " l seeretion.. - Posaible dlmhuution In . laatation , (Manufacnlrers to suppl7 Informa, mkien iiven immedlatel¢poitpartum.y t(on on avaLlable dosage forms, poten- Cholestatle )aund)ce: cy, color• and Dackaging.)--. - Mlgraine. - R~~~ Increase in si3eof uterinelelomyo-'- m8ta. L Tenulutbn ReyurLt" S<rles H. Raah'(al9er¢1ef'. - hum0er, .3- MiT.19Te; Ser1e.L..Nmmbef l. MerltaldePTesa[on. . I June 19'1{: Series B..Number 2; Januar7&edumed telerance to carbohydlatGw. 197l:',Sena H. Number 3;.1975:~ Series H- Vaginal candldi'asis. .. Numbere•.January.1979.(publLhed br.tAeChange.ln eornealcurvature (steep• Oeor9ePopOlislon IffiotmattOnOmvernt9Peoanm• 2']6edlralhe en1Dg).' . - - - W1-^ ^•L: - CenteL 200t~8 SL Nw_ Weshlertoo. D.C.). IntOlerenee m.OUntaet lensea. . 2 SOmaa W: R W. M. P. Veae7. B. Wes The following adverse reactions have ter¢olnt,end A Eoaelund, ~"ihromboemba- been reported fntlsets of Oral eo¢t[a- 11. dRease and the rterold.l mntemt o9'onel reptlves•~ and the esseciatioa has been toninoe9cl.vea A rcpoK m the Commtetea neither confirmed nor retuted: o¢ sa[et7 of Drv9s.•"B,it Ned ! 2f&7-0o9, . Premenstrual-like syndrome. 197& r. 3CO7Ler..P. D.l J. A. TonasrlL M- 3. '- ^ starac Changes m llbldo: . . Chores. Changes inappetlte. : . Cystitls-like Syndrome. Headache. _ _ Nervousness. . Diaiaesa. . . HirsutLSm. . Lnsa of sea)p hatr. _ Erythema multlfarme. . TGekman. P. S SartwelP• .l H. Rutled9e, . .. and di P. JamhA "Thromboas nltb lav<s- trorea oral m¢treceptlves:' /m J LPidemiol' 1021.197.203.1973..-' 4. RorY ColleR of General Preetltlo¢er; . 'Onl mntreeeptlon smd thromboembolllc.. dtseese "' J CaiJ Cen Prna-L 13:207.279. 1867.. 5. Inman,. W. H.. W. and ld P. Veasey. "7n- vetlastlon of deaths from 9Wmonary. mrtY ^ nary and'eerebkfl urvombesfs andembollSm . In s.omen of,ctiddbearlnt aaq"'BrIt.Med J 1:193-199.1999. & 4esse7. M.-P..a¢d R DoY-'Inraetlta- tlon of reLnOn b9sseeo uae of oral mntra- oepneei aed tiwmbeemhelle diteus A fun ther repart-" BrfE dfed J 2631-437, 1969. 7. gercwelt P. &- A T. 9fast 7. O. Artaaa. O. R...OreeM... ead H. L Smlth, Tbrum-... eClmbOltsm and eefl mntraee9tteee: an eDi-. demineoeleLL ease mnctof stud9•" AmJ LPf- demloi 60:363-3a0. 1069. a. Baeton Col)abor.dve Drug 90reeSLOet PrVeSm-"OralrOnlraee9uvea and eenouss tbromboembolle dlaee+a, surp0•1117 armed9allb:adder..dluraaandbreast. tamon--Lauet L1J00.1M9. 197& 9. Callabnraelve GeouG.for tbe Stud7 of. Blaokb m Young Women. "Onl mutxacey tldn andtlnmeased rtsk of cerebeal lseRemla or tbrombo6r-^ M Lnet ! dfed 2aa:a7127a,-. 1973. - I0. Conebnratlve.GrouP for the Study of . Strokee m Younk Womem "Oral mmtraeep- dan and ttroke m 7uung - assnelated'rht faefnrL'J.tArA 231:716-7SY• 1973., ' IL Mann..J. L•..andW. H. W; inman,. "Oral mutreRCttve and death frvm mYo- mrd7al mfarcdo¢."' Br11 . Mod J' 22~24a, . 1973. - 1z IdaM J. L, w: H. W. Inman, and:M. Thovo9ooa "Or+d ooatntsetive usa In older nmen . and faul myowrdlal tmarttaq•• . Brtr dfe0 J Tu6-447• 197e. 13. Mann. J. L. M. P. VesN7, 9f. Tboro- goo4 end R DoLL "MromrdW bitale[ba In rount oome0 with s9eClal referenmY tm oral mnireeeCnve.Practlee-" Brrt Med 122e1- 245• 1973: .. 1l Tlerz... C.,. "Ne. }htimates of Mortah tty AYo0lated with Pertfm7 CootroL' Famirr Pinnnix9 Penyeetloµ: 9:74-76.1977: - 15. Vesse7. I6.. P. R DuO. A S. FYlrbalrn, and ~O: 6luoer. "Pasiooentl.ee tnrombaem" holllat and tae uaee ot oral matneeptlv4": . B'+4t Med J r:127-126, 19701 16.Greene..O.,R.'. P. E Sanoeq "Oral mntraeep'n9e use In patie¢ts vlththrom-h boemboLLtm . folloolnasu[9err, . tiauma or ' 1WeRlda- Am: J' Pub MeaiA: 62:eaa"aa6. 1972. 11. Smith, D: C- R Prentlce. D. J.1Lomy sov and W: L Herrmana "Aitoelanon of exoaenoua estrOaen and vodumetrlal urEl- noma," N dn9i J:1red 29i:116{-11a7, 1975 1&.Z1e1 H. H- and W: D.. P1WdC .'1n• meued~ rtsk off endometrielnrdnoma " uaem-of mnlu9aaed estro9eCa" MLnyf I 3fed 293:116f-117m• 197& 19: Maek: T. N_ \L C. Plke,. B. If, Hender- anq A L Plerfer. V: R. Gerk3fla, !i Aitbur and S. E: Brovn;'•PStrogem and vudome- tr1s1 tancer tu e reWemeat oommamt7.": M. LrWI J dfed 2W:1297-1267• l97& 20: SllverbeR.. S. O. and E L Makovskl- '2Ldometrlal ratelnoma 1. young womea hkln( oral ntr.cept(ve. arentL' Obrfd G•Jnsmf~e6:503-S0a, 1976. 21... VeaseP. Ld. P., R DellL and', P. !& - Suttaa. "Oral' mORdneptlvd e¢d baeaatneoplasli a rettospectl0e etodr.' Br[t Mod J 7:119-724, 1972. - 21Vea5e7. M. P., R DoIL andS. Jo¢e; . "Oesl mntru:pcteeA aod brest canoer. Pro- grese repnnef an eoldem1o10poal-dy.' Lonnt 1:941+943• 1975.. 28.Boemn Collaborative Drug SurveU- Lnce Pxn9ra-. "9ur91Nl1 condlrmed 9a11b- Ldderd1aeese. W tbrnmboemaoiWn md breasttumers m rel.tion tnqoetmmo- ya ..1 estroren lheraPl.- MLn9t J Med 290.36-19: 1974 -24. Arthe4 P.O.•. P. L SaRVe1L and L P. Leslsom "The cILL eatreren;- and. me.-, brvaaL. EVldemlolorle uppcL-'. Caneer "29:1391i1394:1971i F6peU1L teW5T9e. VOL 41..610. 91-rlfeSnAV; J.U/uAt731- 14ra 35:-7270-79.-5
Page 66: azo61e00
62 4230 23. raeal. 1f".anad R. s P.ffmaaraer: "Oral cuotereeDUVee u related tn ttncer and Oam6n lesima of tlx.breaet.'J NeJJ Cancer f'a 33:767-773. 1971 26. Royel Colleae of Genetal Pntptleo• ety"Ona1I ContreoeDtt.ee tad Health,' 1nndon.P)tmaq 1974. 27' Orr, H., P. Cole, B.NarJtSahaq and R 80orer.. "Ora1 coptrEeel#tvea and xduced rifk' ofbenlElf bteaR.olseeaaa."'N Enpf J Msd 291 4 1 9-12 2,1976-. 18. Saam, J., P. aolta. J. J. Bookrtbin, and'E W. Slein "PeetlDle eaendatlmo between trcnl9. heaalOmae and oral eonarariDhve;" LancePt926-92$ 1973, 29: Maya E'I. W. M. Ohrlrtophetaon-M-_ bi SL.hr. .ed'' H. C- WiNllamE.--'BeDatlc chan6ea m y0unf eomen tnamtheg ronha- Duve E4eroidt Hepatle.bemerrbaae end Gftmary heDaOe tumue;° JAAfA. 235:130- 731197l. . . 7o..Edmondsen,:H..A" & HenOetaon. aCd B..eeatee, "f.lver<e4'adenomaa aswdated vttb uaeur oeal waOareDUve;": N Eaei JXm 2 914 70-17 7, 197i 31. Herbat,. & 1.,,H- LRfedler. andD.C Potkapaer„ ^Adeoa aremnwof the vapna:^ N FnYI J Med 28a:8R&d81. 197L 32: C'venmYd, P. J. J-.B.rlo...P. C 21.xy and W. Burnett, 'Taglnal cancer aftermatemal3reatmeot w1tS synthetlt e- - trefen;' N Enol J Nrd 585:390.]91" 1991. 73..l.bler. A- P" 8: L Nuiler, D. O. Decker, I. Elvebeckand I" T. B:url,ad, "CaMer end stLLtvtrol AloLLOV-uD of 17199 plewoa exDmed to.eepu[Rtnm utero aCd bem 1913-1959.' Yaso O-Pre 1617 93-7 9 9, 197>.. 34-. Hereet, A. I", R J. 8orm.n. andFt. E Scully. "Va6IDat aud rervkal abnnr)nWtfea atter zDOEUre tu adlb'esOCl m utexo,- Obslef Cyne..ro140Q267-298. 1973, 33.Herbat, A L, S-. J. Robboy, O- l. Hae• danaldL.and R- E ScuLly. "Thee eHeeb of localpfn6eaterene on 8[IIDeRxol-assoe4ted va9mal sdenuda" dm J' Obaldr Cvneeel 118:807-915:,197e. - 76. Herbet . A. L D. C. Po.4umer, .3. J, RobbeY. I: flledlaodel..and R L: Scudly. "Preoatal eapoture tn stflbertrol a urospec- t:vee wmparlsen of 'ezpoaed female offsprfna a'1Ih unexpwed oontrolL-' N'Enyl JNad282:731-339;1975.. - ' 37. StafL A.. R P. Yattlnaly, D. V. Poley„ W. Fetbenma "Clmleelidlzposls o9 vsp- na) adenoett' OD.tWe' Gyneeel 9]:116-12e. 197{: i8. Sherman. A. I:. M Ocldrsth,.. A- Hertm V. Vak'harh-a, P. BanoanL W:}t9eltaell P. C~oodmaq and 5. Browq °CeN)cal-ha6mal .deuoeli efter.m utero exposurt toarnthe4 le estroaeo;'-Obifel. Gyn[wl'4i531Vft 1971 79:. Ga), I. B. 83emm, end J. $tem 'HOr• ne Dte9nancrtesfa and.eonaenltal mat- formatlon." Natu.e 21e:8]: 19671 W$: P...A Cahen. end P. C. Pnuer. 'HOemone VeaUmene d~lrL^~a pxaa•ney anC tanteNwhe.rt delecta.' Lan«f' LY11.. 9T3. 41. Nera J. Jl end A H: Nor; -B1rth de- teeq and oral eontxeeptl.eq" Lanoer 1:9f 1- 94z 1973, 42. JaoertelL D. T.; ]+Id.P1Der, and D. NL OI'ebanL "Ot.l eonlrueDUVeeandwnEeol- ta1 llmbaeduetlon defecte,"'N Lnyt J Med 291:687-7Da. 19YL 43- Cax. D. H., "Chxamawme xtudll li releeted spootaneooe anornaneL Conaey non attlror.l contrar pt)vey'Casad Nra AaaorJ 1o3:3t]-]18~ 1970- ~4.. Wypn V.. J- W. 8. Daar, and~O. L .16i)lL"Seme effectr of otel Oen[rnceptlvea. MpTICES - oa Eetum-BD:d. end hDODf.otetn )eevla," LnncetY.720-723.196a.- 43. Lumaa IL R.. P. )C, KaptW. 8.. B1at- na6ar. aad. V. Ianaaa "Radloaet/91tym the breavt.m4lk of lactatmg vemen after oral WmmlaVatlon ot 3 H-nmethgnodel" Imer. JObeWQvneeof 98:u1-s13, 19a7, 46. Cenrer for Dleeeq Coptni. "In<rtaEed RLLt of Heyetocellular Adenoma IC WOmen nth Ln<-tarm uee of Ora.' Coutreeeptlve;" 4ro.D[dttr eud mormltfy )vpeu9 I6ePOrt. 25293-2K 197T. 47. Herbet, A. L P.Cole- T, CCltoa S J- Robbor. R. E SeWPr. "AafineidenC and R1ak ot DleCltyletIIbeatmi-xliiedClev Ce1P Adepararcmoma af the Va6fne snd ~CaWt,-' e m, J.: Obetet GyatcoJ: 12S43.SDI 1977. 46. Bmbe, tL ae- A3-Nayeeb. L n-tW-.OtIP. ai Nesetba' I,aL b7eeper,M- 5upek O. L Wted, "Fbne.-uD 3tudr of Bealeand Pemale Otfaprnlg of DISVeued67otbea A PxlSmtdary Repovt" Joar- N. Rep- . Med-; iSa9-32, 19TS.- - e9: OWi, W. H" O. F. B. 3chamacher. DL Blbbo. "9trucowal tad PuDCOUnal Abher- mailHea 1n [he~Sex Orguu of Mtle0efeDr-b16 ot 7aClhen Txated alth Dlethylstitber trol (DPS)," Juer.a/ReDre. Ned,.16:1i7- 163, tne, 50: HeoderEenB. F'8.. Senena 1d..Co.- 6]vve, J. HaDttata, J. AILr1eRD. TlownreoqW. HaR T. AEACk "171ueemtal RYaRAbupr- ma1R)ea mBeeeof Womeo Treased vleh DlethylitllbatrolJ' Ped:atriea56:303-507, 1976,. SLHelmnen, O.. P" D. Slona. R- R- Nomon, Eli.7iook, A 3baplro,"Cardbeaa- cu1.r Blrtti DefeW and AntenatN ~pmWoae toPemale SexBotn)oDn." N.En9L !.' Xe6, 296:57+10,1977. . Si Jela A. K-.. "Idortalitr R)sk Aafa:tated' a(th abe IIsee of Or.l Caatteeepnvey" Shd- 4i in iamAlr Plbnniay, 8'.50-34, 1977• 53. Hersp, V.: "mertallty Amon6 Oral Coo- traoeDnve Ueery".LaaceE MYf-731. 1977. The patient'~IabeLLUg for 0ra1COntra- ceptlvea drug ptroduct9 la set forih below: BauT Spev<.wz Parivrc P.csAOe I`JSaTr2 ClPrette rmokmgInereeee. the rfek cf u. rlbua adverx effecer nn the heart and b10ud vrsse1s from uraf eaon. Aptlve uee- Thu risk mereaw .tth eae apd'w1t11 heavy kfny (15 or owre enRxttes Der day) and 1E au)te marked Ip eomen over 75 yeaxof ee, Women be we ora wntraeepalvee shuNd nx ~o+te. ~ ~ ' Ola] contra¢eptlves taken as dilrecU ed are about 99%.etfeotfve in Prevenb lag pregnanOy: (The mini-pi11. howev+ er, Ia somewhatlese effectlve.)Forget- doctor'S eontmuous suPenlaoa be- eause they can be axscefated wtth sert- eue -s1de effectY whleh may be fataL Fortum1tely: theseoRur vet7y ittfre- Quent9y- The xrloua slde effecta are: L Blood Nota In the lega, lunga. , br1iL, beart o[r oLher or;aoi and hem- Orrtlage mto the beafn due to bulstfng of Erblood veaseL 2. Liver. t)lmOrs..vhich mayruDtt.re and nuae savere bleedmg. 3:. Btrth defetti 1f ~~ the Pt11 4s taken mh11e YOU are pregnan9. - . 6- Righ Dlood Dlt:nure- 5- CrallblAdder dlseaae. The symptome-assodated wlth theae serloua.side effects 6r6 d(+.a,.•ew'Sn the detalIed leafletglven you with Your suppiyofpills. Not3fy'your doct4r If you palee any.unusual physlcaldlL' turbanae while taking the pfll. The estrogen ln otsl tonttaceptlVea has been found to nusc.breaR raaer and.other eaneeraIn aertam a)llmala.. These flndlnp suggest thatorat o0o- 'ttaceptfvee mayaLtu,eause.efncer m ilumam- Hovxver, studiea to date 1n women taking eurfentV marketed orai eontraceptivee have nott confirmed that oral cOntraClptlVeB CSnSe caneer. tn hlmtata The detailed leaflet descrtbess more trompletelythe benefltsand rilksoQ Oral contraceptlvea. It 91toProvldea tn- formatfon, oa other formas of contra- ceptlon. Read It earefull9'. If you have any Ouestlons. consult yourdoctor. CatJtiott Oral eontrseeptlvesare of novalue m the preventlon~or treat- ment of venereal idisease. Derea.m.Pst'LxtrrS:,uv-oe0 ABAS LOO SHOInd KNOW ABOLR OAAL CUNSaAe@mr3 Ora1 a0atr3ceptivea ("thepHl") are the most effectlve say(ezcepc for StorillZBtlon) to Grevent pregLaOcY-Theyare atToconvenient and. for moat Women, free of Ser:oua oF un- pleasant aide effeeta. Oral cOnarattp- t14es mustalmays be taken.under the continwous supervf8loo of a DhyBiCRn. (If and oral eonaraeeptlve ts ap-Proved for indicattons oth'er than eoo- traceptlon (F]tovld 4mg.. Ortho- Novum 2' mg:. Ortho-Novum 10 mg), the.mAnufacturer may meation thooe tp ting tno take your p111s li)creasea the lndlattens In thelastParagraph't chance of pregnancy. " thls sectfon and state that the infor, ' -. Women who have or have had Olot- matienin thialeaflet under the head- ting dfserders, cancer of1he brea,rtor Inas "Who Should:NOC Bse.Or>,l'.Cotr sexorgaaa, onexPlalned va9inal EOeed- traceptlves.""The Dengex of Oral IAg, astiokea heartatfesk,.anglna pes- Contraeeptlbea,"' a;nd "How to6se tbrts, or mho9uaPectthey. may be Oral COptreceptlvea Safely" ls aiso aD- yregnant should not uree oral contra-phc8ble when theae drugs are usedlor cepti4es..- otherfnd)caaiona). . 'dort s:de effectsof the Dt11'arenot It is Ifnportant that'any.moman who aerlouB.Themostcommonsldeeffeets conf)ders usang an:oral eon6raceptlve are nausea, vomlting, bleeding be- understand the riska lf)voived. AI- twee nmenstroal periods,, weightgaffl," though the oralcoptraceptlvee have and breast tendernesS. Home.ver. ImportanV advantages Ovet other prolxr use of ore) oonRraceptives re-methods of, contraceptioqttley have. p)µltt9.t11at they be taken under your certain rui)ts thatno other mettlod: yEDElAI eEDnTEe,. Vot, N,H0. 71.-rnESDA7, JANDAa7' 31, 17y6 I
Page 67: azo61e00
63 a has Onl7 7oucan declde.whether the advaptages.aR worth these risks. Tt11s Ieatletm'.ll tell you~atiout the cmet Im- portent risks. It will explaia how you . tm help your doCAr Preac.7be the pitl as safely U possiblee by tetlhlghm about'younelf andbetng alert for the earlPest, signs of, t.-ouble. Andit wl"d teil pon tiow to use the p111properlY, so thatitwlll be as efleeuve aa pmsl-. ble There is more dltailed Informa- CiJG ;vailable Inthe leaflet prepared for doctots. Your pharmacLst',(m show you a mpy'. you may need'your dAo- tor's help in understaad:na partaof It. Weo Suovto Nor flss OaACColeus~rrvo A Ifyou, have apy of the following mnd:t)o s you should not usethe pill:' 1. Clotss m the lega or 1ungs.. 2'. Anglna pectoru - 3. Known or suspected caneer of the breast or sez organs. - 4. Unusual vaginal bleeding that has not yet been dlagnosed. ' _ 5. Known or suspected pregnancy. ' H.. If.you have had any, ot the follow- ing conditloas youshould not use.ttie, pt11: _ 1i Heart attack or stroke. 2. Clots in the legs or lungz ' C. ClaareMa smukimt maeaua tae nak of svlow advene elfeet. ao the . heart aod e)ooE ve.vela lroam ora) mntnceptiae use. Thds. rtlk inereaus w1tD age and with heavvyy smekmg(15 or owre efgaeetta per day) and )s eulte marked'm wumen over 35 yean of, ege: Wcmeowho use orai emtniteptlvnsitouidpat amakG D., Ib you have scanty or irregttlar periods or are a young woman without s regu5ar cyclk, you should use an- other method of mntza<eption, be- tause, if you use the pu1You may na.ve.dllttfcultybecoming pregnant or may fail to have menstrual i perioda afterdl.scontmuing the pllL Drcmtxp To IIsaOa.v. Co~eartvuIf you do not, have any ofthecundt- tions listed: abovee and are, thinkiag about', usitSg oral eontraceptives- tn tieiC you decide, you need mformation about'the advantages and nsks of oral mntraceptives and of other con(ratxp- tive methods ass well. 17Lss leaflet de- scribes the advantages and risks of oralicontreceptlvea.:F:zcept for aterll-. +za[fon, the IIID and abortl0n, whieh havetheir ownexc]usiverisks,n the only risks of other methods of contra- eeptfon arethose due topregnancy should the methmd;fsh. Your doctor can answer questions you may have with respecc tu Other methods of coo- traceptlon.: He can , . atso answer any ouestions you may have after reading th:sleaflet on oral mntraeeptives. 1. What Oral CoatraeeptivesAre and How They Work. Oral ContraCep- tives are of twoo typea.. The most mramon, often stmplycalled "the pill" NOTICF3 Is a mmblpadonof an-trogen and a progestogea the two ktude of female hormones The: amount of esirogea and progestogenn ean vary: but the amouat of estrogea)smod Important heau.se both tl1eeffecSlveness and someod.the dangenof oral mncracep- ttves are related'tnd the amount of, es- trpgen. This Innd oforal'mntraceptive works prlnlcyally by preventldg re- lease of an egg from the ovary. When the amount of estrogen Is 50 tnicmo- grams or more, and the pill Is taken as dlrected, oral mntneeptives are more than 99%% effective (Le., there would be less than pne pregaanay If 100 women usedthe pill for l.year). Pllls that mn- ta1-20 to 35tdtxograma of estrogen vary sltghtlyIh effectlveness, rangiag from 98% to more onan 99% effective (Manufacturermay lnsert pregnancy rate for thee manufactureYi produet found In clhthal trials,.if product la a mmbmation). r The secand'type of oralmntracep• tive, often called the "minl+'pi1P'. mn- tains only a progestogen. It.works inpart.by preventing relkase of, an egg from the. o.vary but also by keeping sperm frvm reaching the eggand',byg making the uterus (womb)lesa) recep- tive many: fertlllmd'egg that reaches It. The mini-pW is less: etlectlve than the mmb(natiorl orat mn4raceptSve, about 97?r,. effeetive.. (Ma.nufacturer may insert pregnancy rate for the manwfaettuer's product;found tn ,clhli- tsl~ttiaLt lfProduct.is a progestogea- oeily oral contraceptive.) In additlon, the progestogen-oaly pill has a ten- dency mmuae irregvlar bleeding WhiCh.may' be Quite inconvenient or ce.ssatlen. of bleediv.g entirely: The progeetogen-only pill LS used despite lts lower eflectlvenes in the hope that It will prove not to have some of the serloussideeffecu of the estrogen- mntaming Pill (see:befow) but it is not . yet, certain that the mini-pill does m, faet have fewer, sserious side effects.. Ttie discussion below, while basedmain)y on Information about the cpm- tiination pi1Ds,: shottldbe tonsldered Coa,PiS.aswedl to the mhol,.pi1L2: Other Nonsttrgieali Ways toPre- vent Pregnancy. A5this leaflet will ez. plaiil- oral mntr aceptives have: several serious risks. Other methods of,on. ttacepti©n have lesaes risks or none at all. They'are a14o less effective than oral mnttaceptives., tiutL usedd properly; may be effectlve enough for n:any women., The following table gives re-, ported pregnancy rates (thee number of womea outt of 100who would bemme pregnant 1A 1'year)~for these methods: .. Paeceraem Pa Im Wotmr. PnYrwa. Intrautlnhe devke IILT)). )esa thaa 1E; .. Dtapmarm rttd sprrmlctd.l Groductatereams.or lelIDesf. Coodom (t„b.Ver):3-re; - - Aerosol [Came, 2-29;. JehDee andereema-.4-0a; - . 4231 Periodic abstlnenee (thyUtm)) alll typealeat than 1-t7;.. 1- Caleodar method, 14ir, 1 Temperaatue! metho4 I-2p:.. f. remperatore metbod-ICtereaus.a only m Posc-ovulltery ydsae, less than 1-t:. 4. Muws method'1-28: Nm eoatnreppan, 6e2a: 'The flaures (except for the IQD) vary w1de19betause people differ In how.well they use each method 4eryy faithful ttsers ofthe various methods obtaln very good 1laflltd, eiCept for~ users oUthe calendarmetlimd:of peri- odic abstinence (thythm). Excettptfort tneIIID, effective use of thesemeth- odS requires somewhat moreeffort~ than simplyy tEldng a a:nglA pill everyy morn[ng, but It', ts an effort that maay couples undertake succeasDVlly. Your doctor can tell you a great deal more about these methods of mntraception. 3. The IDangersof Oral Coptracep-- tives a. Ci.cudatery dttorden. (dbaonnal btoud clottlnqa+rd rtroke dttetn.hem- orrhoOt ).: Blood Ttote. (in .vat7ouss blood vessels of the bods)'are tlie most mmmon of.the seri0ua side effecti of, oral mntraceptivea. A clot can result m a atroke.(1f.tlie.ciot ls intne hrain),, aheert attack (if thee clot Is in a blood vexl of the heart): or a pmmonary embolus (a elot which forms in the. legs or Pelvis,: then lireaks off and travels to the lungs).:Any of these Can be fataL C:otss also omurraretqin the blbod vpselS of the eye,.resulting In, blmdness orimpairment of visi0n Inthat . eye, There is evldencee t)eat the risk of,clottidtg mcrea.4Gtwith higher estrogen doses..It is therefore Impor- tant to keep.thedose of estrogen as loww v.s . possible, so long asthe oral -n- tracepuve used has an acceptable pregnancy rate.and dosen't rause un- a/septable changea :n the menatrud pattern.. Fluthermore.. cigarette smnk- mg by otal eontracepti,ve usera m- creases the rist of serious adverse.ef-fecLs on the heart and bioodvesselA This risk liureases with age and with heavy smoting (15 or more cigarettess per day)) and begsnsto becomee quite marked m women over 35years of.age. For thisreason,women who use oral mn•.raceptives should notsmote.. '1'he risk of abnormal ctottmg m- creases with age in both useraand nonusersof oral mntraceptivea, but the.inereased titk from the mnttacep- tlve apPears to be present at alll agei For oral t-o~ntraeeptive usen mgener- ai, It has been estimated that th - women ,between theages of 155 and'J4 the risk of death due to a circulatory dlsorderisabout 1 in 12.000 per yeu, whereas for oonusers the rate Is about 1 m50.000 per year. In tlie age group f 35 to ~ 44: theriskLs estimated to be about1 in.2~500. Per year for oralcon- traceptlve users and about 1 -10.000per year for nonusers-. - Even aithout, the: pilll the risk of having a heart attack Increases vdth FEDEGL gEGifTEg, VOL 43, N0. 21-rYESDAr, IAMtlAgt 21,.197g
Page 68: azo61e00
64 443s~' . NOTICES - age and Is aLto increased by such heart oral contraeepttvea have less risk of . sttatk risk factorsas hlghbiood pres. gettfng bendgn breautClseaze thann sure, hlgh etiolesterol- obesity. d1!abe- thosee who havee not used oral eontra• tes,. and1 etgarette smoking. Wlthout cepttvea Recently. strong evldence has an.y risk facten prtssepti, the use of emerged that estrogens iane compo- oral centraneptlves alone may'double -nent of oral-eontraeeptNes):. when the rtsk.of heartattack. However. the- glven for-perlods of more than one cpmb7natlon of clgarette sanoking.,es- year to women after the menopauae, pedkllYheavy smaking, and oral oon• psmease- therlait of rsncer of, the traotptive use greatly InQeases the Itterus(womb):'I'herealalsGaome.evi-. risk o1 heart attaclY Orsl contrecep: dence that a kind of orat eontraeeytlve tlve useta who smoke are about 5 time which ia no longerr marketed, the se- more likely to have a heart attacic~ 4venttal' Oral eontraoepttve, . may in- than usen who do, not smoke and crease.ttie risk of cancer of the uterua about 10ttmea more.ltkely to:have a There remains noevtdence,. however, . tYskth'annonusen of having gallblad- der d7seasetedutring surgery..Theln- treased risk may first appear wfthm 1' _ year of, uae and may double after4 or5 years of use•. " e. Orhsr sidt eAeets of orat trortfla. ceptiaea Some womeu usfttg otwl ~coa- tracep0lvesexperietue uppleasant sfde: _ effects that are nut, dangerous and are not likely to damage their health. . Some off theaemaym be temporary.,- Your breasts may feel tender, nausea and vomltlAg may otc¢r, you may gain or.lou weight..and your angles may 1welL A'spotty darkenteg of the akln- heart attack than nonusen who do that the oral coatraaptlves now avall• particularly of thefacx,as peaslble and not smoke. It has'been estimated that able Increase the risk of thfa e8neer. - may persist. You may notice unexpect- usen tietween theages oi 30aad 39 - Oral contracept(ves do cause, a1- edraglhal'bleeding.orchangesInyour rhosmoke have about a. 1 in 10.000 though rarely- a benign (aon-mallg- mensttual perlod Irregtllarblteding fa- ehance each year of havtng a fatal naat):tumor of the llver. These tumon frequently seen -tien 'uingthe mint• '- heart attack compared to about a l,ia do not apread- buttheyt may rupture pill or combidat.lon oral contraceptives 50.000 clunea in usete who do not Lad musel lnternal. bleedmg-, which contalning.less than 50 mterograma of smoke, and about a 1 in 100.000'may. be.fata!! A few eau.so! cancer of estrogen. ' chanee mponusen who do not smoke. In the agegrovp 40 to 44. the risk isabout Mn 1,700 peryear for users who smoke compared:to aboutl m 10.0000 for usetawho. do not smoke and to. ab'out I in 14D00 peryear for nonusera- who do not smoke. Heavy smokin7 (about 15 cigarettes or more a day) 'further mcreases the rlsk. If you do not smoke and have none of.tSe.other heart attack risk futorsdescribed'aboves you wili have a smaller rist than listed. lf you have several heart attack risk factots;. the risk may be consid@tably greatertLan llsted. In addition to b1Aod-tlottlhg d$or- ders,..lt has been estuaated that women taking oral contracepttves. are tvice as likely as.mmnusers to have a stroke.due torupture ofa E1ood vessel In the braihM b. Formot(dn of fumu+s.Studies havee found that'.when certain anlmalsl areglven thefemale sex hormonees- Loken, whlchda an mgredient'oII orali contn[ceptlves,.conthluouslys for long periods: caneers may develop in thee breast.cervlx:vagina, and hver: Thesefmdmgs suggeat.that o1H1 Con- traceptiveamaycause. cancerr in humana. However, studies to date in women taking currently marketed oral contraeeptivee have notconfirmed that Gral Contracept]veb muse cancer ln humans Several studiea have found no increase in breast cancerfn users.althpugh one studysuggested oral contraceptives mignt cause an Increase' m breast cancer Iu women who.already havee bemgn breasVdkease (e.e._ c9sta). Women tc/tb a strong fam[lyhistoryy of, breast~ cancer or who have breast nodules, Ilbrocystic dlsease; or abnor- mal mammograms or who were ex- posed toDES (dlethytsttlbestrol), an estrogen.durfngtheir mother's preg-nancy must be folloa'ed verg closely by their doctors 11 they. choose to use oral conttacepttvess litetead of' another method of contracept:on. Many stud~ les% haves'nown , that women tak3ng the:lfver have been reported m women ' More serious stde effeeta tuelude= using oral contraeepttves buttt, ls not yetknown whether the drug casued them. e. DanOera to a deaeloyinp child if Gi4f CpntraceptilteJ are:uaed in afr im- mediatefp preCedinpp prepnartcy. Oral coptraceptl.vesshould.not:be taken ~by pregaant women becausetbey maydamagey thee developing chltd Ah in- creased risk of btirth defecta including heartt defects and' limb defects., has been associated with thee use of . sex liormones, including oralj coatracep- tives., lit pregnancy..In additlon- the developing female chtld,whose mother has received DES (diilethylstllAestrol),~ an-trogen., dllring pregnanay has a risk of getttng cancer of the vagina or cervlxIn her~ teecu oryoung adi'llt- hood; Thls risk is estlmated! to be about lim 1.000 exposures or less. Ab- pormalitlee of~ the urinary and sex organshave been reportetl; In malee offspring so exposed It is possib1ethatother estaogens. sucli asthe es- trogens in oral contrueptlves.. could, have the same effect, m the ehild If the mother tak'esthem duzingpreg•nancy. If you stup taking oral contracep- ttvea to bemme prCgnant, 9opr doctor may recommend that you use another method of contraception for a short whtles The reason for this is that there. Is evldence.from studles In momen who have had "ml3earrtagaa" soon after stopping the pill, that the lost fetuses are more likely to beabnormal. Whether there fs an overallin(zease m "miscarriage" in women, who become. pregnant . soon afterr stopping the pill as compared with women who do not usethe pilllsnot known, but it~is pos- sib1ee that tbere may be. If. however,.. you~do become pregnant soonafter btopplhgoral conttaceptlves„ and do not have.a m* .a,tie, there is noevl- denee that.the baby has an 4.crea,ed risk ol,being abnormal..- d. Crrllbdaddef di.tea.te.° Women ~ho u'se.oral contraceptlves.have a greeter worsening of mSgratne, asthma- epl'1ep- sy, and kidney or heart disease be• ' cause of a tendency.for watez to be re.tamed In the body wtien.oral'contn-ceptivea aree used' Other slde effects are growth of preexisttny libtoid tumors of, the uterus mental depres- sion; and l.'verprobtems with )aundice (yeilowing of the akin). Yourdoctor may find that levels of sugarand fatty substances Inyou[ blood are elevated}; the long-term effects ofthese changess arenmt known Some womea develop high blood,pFessure whlle takfng orzl contrseeptives, which ordinarily re- tu¢ps to the.origlbal levelswhea,the oral contraceptive is ftopped. Other reactioeu- although aotproved to be raused byy oral co¢traeep- tlves, are. oceaslonally reported.: Theae. Include moree frequent urinatlon . and some discromfort when urlnating, ner- vousness: dlamess,. somelosof scalpP hau- an.lncreaetn body'y hair. an m- crease or decrease in sex dr(ve, appe- 'ate <hanges, mtaracts- and aneed :or a change.m cpntactlena presCZlptlnn or Inabllity to use contactlenses. After.you stop usEng oral contracep- tSves tRere may be a delay before you are able to bemme Cregnant'or before you resume havtugmelsstrual periodaThls 1s espeetally true of, women who - had irregular menstrual cycles Prior to the use ot oral~contraceptlves As dis- cussed prevlously, your- doctor may recommend that you wait > shortwhtle after stopping the pill befoeeyou try to becomee pregnant. Durtelgg this ttme. use another form of contra- ceptfon. You should eonsult your phsl-elan before resuming uae of oral con-traceptlves after childbltth, espetially tf you plan to nun:e your baby. Drugs . !n oral eGntraCEptlVp are known to ~. appear in the mllk- and the long-range effecton Infanta t5not knowp at.thls time. FLrthermore, oral'rnnttaceptlves may cause a decrease in yourmilk suDply.as well as Snthepualltyof the mllL FEpE"t EEGnTEE,.VOt..sa, NO. 31-tVeSpa1, JLNUAEr 31, 1171 I t
Page 69: azo61e00
65 • y 4. Comparison of the F.3kg of O[al Contraceptives and Other Cont.-aoepp tlve Methodz.lbe many atudies pn therLtkg and effecuveness of oral con- trecepttvea and other methods of con- t2-ption have been analy2ed to estf• matettie risk of death aasocfatedwith varleus methods of.[ontraceptioa. This risk ties two,partC (a) the tisit of thee method Itself (eg., the risk that oral contraceptives wfll~ auseddatb due to abnorma) elottfng), and (b).the risk of deathdueto pregnanc9or abortion ]n the event the method fatilti TSe reaults of this aaalysta are ahown NoTICES Inrthe.bar graph below. The.helght of the ban.ls the number of deatha per -100,000 women each ~ Year. Thete ate stx tete of bare, , each set referring to specific age group of women. WItWn eachset ofbata there is a sing)e bar for each of, tGe. d)fferentcontraceptive method4 For oral tootxarxptives," there are two bara-one.forsmokers and; the otherfor noasmokera. The analysis Li based on present knowledge and new tnformation,could. ofcouxx,f alter it 2'hee analYsis shows that the risk of death from all methods of bh-th controlIs.low.and below that, a;cocat• d233 3 ed with child bidtti,, ezcept for oral oontracepttvesin women over 40 who smoke. It showa that the lowest rtsk of deathla sssociated with the condom or dlaphcagm (tiadlttonNl comtraceptlon)) backed up bs early abortion fs ose of, fai)ure of the cond6m or diaphtam to preventptCgDanpY: BL4o, at attf age the.xlsk of death,(d+'we to unexpectedpregnaney) ~from the uae of,tradittona)contraceptton.even without abacknp of abortion. is genera]]p.the same se or leas.than that.from ux of oral contra- ceptivea.. - - ay..1.law.r1...rnw~rA`.e.rsr~.w,wal.Il.ay..,1i.m+a..w.leYll~wsi~.~.Y~wMewn> - .,rnr sr.,.. ! Q s a a x Et ~tis~ . aa - .w^..rr.. Qw..,..w 0 ~ye.e,wr- sn aa as M A F= a. +>s ROegI1 ggG6ige,VOL /I,NO. ]1-TU[SDA7, JAXUA4Y, 31, 1979
Page 70: azo61e00
711 66 4234 - NOTICES - HOW TO Dss Osa ComxACartvrl As Sudden seevere headache or vomlt- .1 Perlodlc Ehalnatlop. S.var .am F7rer.rrvaz As Posss• lag, dlazineae or falntiug.: dtsturbance. Your doctor wlll take a mmplers . n; Oercg Yov H.vs Ds®sa zo IIsa of vtslon or speeehor weaknessor medltal andfamSly history before pre- Tltmd .• . ... numbnesa m an arm or leg (lndlaat7pa scribing oral contraceptives. At that 1. WhattoTell yaur Doctor. . a Sudden poadible dtrake). •s ot tfine and about once a year thereatter. You can makee uae of the Pill ae D><t21y or mmDlete loa he v1llRnerauy easmine. 9aur Dlmod eafely as possible,, byy telling your ~on (Iadfmtlps a pas~dble dot inthe . eye). pressure, breasts, ab'domen,and Delhc~ . doctor If you havt atv of the follotw_ Brelat lumps (you shonlldaak your organa (fneludfng a Papsnimseu Lair. - doctor to show you how to exau^+n~ .smear. Le., test for rtncer): , - - a. CondStldns that mean you should your own baeasts):.. not use oral mntraeepttves: Severe paln dn the abdomen (1[Sd1daE- - Saaa.aT - _ , Clots 1n the legs or luags -' Ing a possible ruptured tumor of . the... Gnl mmtrseeDttan ax ttie.morte1. Clots in the lega or.lnngs in the pasL ]iver). A sttolte,.heart attacl4 or anginaSeveredepresslosl ' -- -- fectlve method, exceptdterilizatton, pectorle. - Ye9owingoftheskla~()aundUoe).fo+ preventing pregnancy. Other Known pr suspected rsncer of the28ow to.Take the Pill So.That lt Lmeohods,s wtiea uded mnselentiousiy, breast or sex organs. - - . Most E7fectlva. _ are.aiso veryy effective and Lave fewer -_ IInusual. vag1nal bleetling..ttiat haa (Manufuturer to.supplY lnforma-.-. .rlsl[s. The serious risha of'oral mntia-- . not yet been dtaaaosed. - tfen on dosoge and admfnisteatlOnand' ..reptlves areunmmmonand the °Dill^' ' 8nown or suspected Dnegnanry. , wltat to do it Datient bas forgotten to ]y a very mnvenlentmethod of ~ pre- b.: Condlttoms that you d8etor wiB take one.or twop4ls. Where appllca- ~ waattb watch closelq or which might ble, manufarturers shnuld dnpply ap- venting pregnancy. caux him to suggest another methodIf you have certain mndltlAn.s or of mntraceDtlom • --- _-- propFiate 1nfOrmatlon regarding _ Ux have.TStl thexe eond)tlons In the as[. Afamdly hlstory, of breast cancer. -for o[tierapproved mdtcations.:) - D Breast nodules,, flbroeystledliseax of.. At times there may be no.menstrval you should not'useoralicontraaptlves t11e breast, or an abnormal mammo- PeTlod after a c9ele of Dills. Therefore: tlersuse the risk latoo great. Thex g, Ifyoumise one menstrual Deriod but conditions are llsted mthe leaflet. 7f .. Dfabetes. havee taken the p1l1s exactlg' aa lrouyou do.not have these condittons,.and High blood predaun. toex x.pyosedto eontmue a.v ustul" decide tow ux the. ,.Dill," Dlease read High cholestereL ' --- -lnto thee next cycle. If you heve not ~ee leaflet carefullqso thaYyau ~ran Ctgarette smoklag. - - - - --- - taken the currectly and miss a men- . ~e the "DLLl^ most aafelyt aad effee• Migraine headaclies. -~ru~ Deriod or it-you are taklni Heart or kidney.disease.. mSnl-p111s and It Is 45 days or more ttvety.. ' _Epuepsy. , - from the start or Your last'menstxual: Ba+ed on Tis or ber assessment of Mental depresslon.. .. period you maytie pregnant and'yourmedfcal needs. Yowr doctor haaFYbrcid ttSmon of the uterus..- ahould s6op tsklhg oral tontraeeptlvea Gallb3adderdlsease.. - - until Your doctor determineswhether presmibed this drug for You. Donot. c. Once You are using oral ~mutraceD' or not.you are pregnant. Until you can gtve the.drug to anyone else- -- . tlves, you should be alert forsi`n9 oi a- get~ to yourddetpr, use another form (Sem.. 502. 505. 53 Stat 1030-1053,: ae serious ad9er5ee effect and'~ Call youof mntraceptlGn. lt twoco05CelltlVe aminded tY1 D.S-C..352: 35500and uoder d6ctorf1 they oCCur: _ menstrual' periods are mlssed-. You authCntr de1leated.tn the Cammtstlomer of . Sharp pam in the clSesti . toughSng . should stop taking pllla until it Is de- _Fwd sad Druas f 21 CPR 5.11... -blood... or sudden shortness of, breath - krmlhed whether you are pregnant, lf Dated January 18,1978. . (indlntiag possible clotslim the lungsl: you tle Secome pregnant while usfng Patn inthe.ealt cpossSble elotin-the Doaap Kc+ex>ar- leg). o>hl coptracepttves• you should discuss CommvsioneropFOOdnnd ' Croehing.chest pain or heavtness (:n- _the risks to the developiag Cntld crtth ' . Drups dinting possible heart attach):. y ouurdoctor. _ , tPR Doc'ts-23o1 ySed 1-1e-)t:10:45.m1 7E[IElAt REGIZrEe, VOL 43. NO. 11-TUeSCAr, JANUaly,l1,.i97e I
Page 71: azo61e00
67 w r Studies in FamilyPlanning Vofpnne 8..Number 3 CONTENTS OF THIS ISSUE Afi7Af=?YF " ASSOCIATED WITH THE USE OF ORAL CONTRACEPTIVES Anrudh K. Jain R.c.nt 3S4tUhs in et*sCBrtftaln and the 17nited States haw raisW serioua concem about the aafsy of oral, centracpdvas fon women aQed. 40 and oven This artlcle reanallrsea the datapubliahed in these studies and shows that a clear,undentanding of the ayner- piltie ef I.cts ofsmokinq and pill use is essential for.proper a.aesa• ment ot the safeyof orat'tontracepov.a. The present anal)sis in- dicates that (1) tite usa of oral dontfaaptivee" in the absence of smoking,.is eonaiderably sateethan no fertiliy'oontrol for all: A Comment by Sheldon J. Segaf MdrcA 1977 agea iAduding the.group aged '0-4::fA the use of oral contra- ceptivn among smoken aged 40 atW o.er Is substantially,more haaardous tnan no ferblityoontrol; although than ialitltie diHer- enca for light smokea: (3).we use of oral~ contvac.pbves.among heavy,snwkars in the group aged 9a39 may be more.hazatdoua th~an no fertility controt: and'(y).the use of oraf contracpb'+ea among heavy,srnokers in,the group aged SS-79may be mor. , harardous than any other mHhbd of fertility regulabon. 54, RESPONSE REL;ABILITY IN A LONGITUDINAL SURVEY IN THAILAND 55 John Knodll and Sauvaluck Piampiti Tfia two rounds of ttre Natbnal'Lbngitudina: Study,ih Thailand provide a usefuli opportunity to explore response reliability in alarge-scale social and demographic survey,in a davNoping coun- try. The results indtcate that.nonrandom reliability at the imdiviQ ual levell ranged.fram auite high (for several straightforward. faa- ua1 questionsy do quite low.(tor most atrotudinat qwastions)i Theesm was considerable distributional stabiliy, howaver, even for manyy of the variables with low individual-ievel reliability. in terms of ita . response reliabiliy, the Thai study eompares reasonably,wNlwith several leading US fertility suurveys.Howeva;.in both,countrin response reliability at ttieindividwal.lavN tbr attitudlnal qusatiGns is distressingly low. This dear.y,shbuld be a matter of major con. cesn for sociMscien0ista using survey reaulrta THEEUIRYONG EXPERIMENT: 97 A KOREAN INNOVATION INHOUSEriOLD CONTRACEPTIVE DISTRIBUTION Chai. Bin.Park...Lee-Jay'Cho„and James A. Palmore Although contraceptive practice has increasad'substaneatiy in, Koruas aresuitot'a strong nationelRamily,planning program. surveys indicate the eaistence of an additional unmet demand. A number of obstactas tend lo limit availabitiy'of family'planning services and supplies• The effects of totall availability, are bstng tested in a thnee-year experirnent in housenold contraceptive diis-tribution. for which an exploratory,study was:carned out in 1975 in Euiiryong 9un. Thrae distnbution systems were tned in three areas af,.Euiryong• each enlisting local v-oman to canvassand dis- CURRENTi PUBLICATIONS tributepitlb and eondoms (and nfv those desiring)VOs oratMll- [atlon to physioians). Results of the preliminary,tesi showed that household diStribution wes autturallyaccptabll and adminie- trativery feasible andd tnatit did increase contracpcive:use. Cosr-dfectiveness wasconsideredto be withintna range of thsna• tionar program. The mostsuccesaful af mathree aistribudansys- tems is being applied in,the mainatudy. begun late int976m C1eiu provincs.. . a ...,.,.~....« --- 49. A PUBUCATION OF TrI-MPOPULAT"ION COUNCIL 77, ISSN11079-3685
Page 72: azo61e00
68 Mortality Risk Associated with the Use of Oral Contraceptives AHRUDH K,JAIa In a recent analysis, Tietze. Bongaarts. and Schearer (1976) have reported that the risk of deathlrom use of oral con- traceptives after age 40 is higher thaa the correspontling risk of death from pregnanry.and chtldbirth,when no fertility con- trol method is used. Studies in Graat Britain have shown th'at, the incidence of fatal and nonfatal myocardial infarctionamong ptL users is signiRcantly, higher than among nonusers, amd that the excess risk of myecardialinfarction attributable to pill use-increases with age (Mann, Vessey, Thorogood, and DaB~ 1975; Mann and lnman, 1973i'.Mann, Thorogoodi Walters, and Powell,.1975; Mann, Intrtan.and Twrogood, 1976). On the basis of the finding of 73etze, Bongaarts,, and Scheaier, along with the results of theBritish studies, the United States Food and Drug Adotinistration (FDA) has pro- posed a revision of physicianand patient lab'slidg for.oral contraceptives. The proposed revision 'or physicians states thax"the. use of oral contraceptives ion women in this agee group (40 and over) is not'recommended.'-The proposed re- vision for patients states: "Although it is your decision, it is recommended that if you are over 40 yearsol~d you do nott use.the pill because of an increased risk of heart attaek's from . the pil!"' (USFDA. 1976). By reanalyzing the dau pubiished~ by Mann, Vessey, Thooogood. and Doll (1975), ]aiet (1976)'.}ias shownthat:in comparison to nonusenwho do not smoke the relative risk of nonfatal myocarddal infarction associated with pill luse.among nonsmokers aged 30-44 is not statistically significant, and the relative riskk ofmonfatal myocacdial infaretion associated with pill use in a society depends upon the proportion oB.4omenwho smoke cigarettes. Similarly, the esdmated'ruk.of death frompill use is also stronglyaffeUed byy the smoking habits of pill users, as demonstrated ian this article. It is shown that, if, the results of the British studies are assumed'ro be valid, the risk of death fromm pill use for women aged40 and over who do nm.r smoke cigarettes is substantially, less than the risk of death from pregnartcy and childbirth when no fertiity control Imethod is used..it is smokkrs over~40.for whom pill use carries a greater risk than pregnancy and childbirth. There are some indica- tions, moreover, that even for women aged 30-39 wh'o are heavysm~oken the risk of death from piB use is higher than the corresponding risk of death . from pregnancy and child- bvth when:notertility control method is used. This is in con- trast to the risk associated with pillusefor nonsmo'k'ers in this age.group, whichls is much lower than the risk of pregnancy and childbirth. The Problem In theiranalysts. Tietze,. Bongaarts, and Schearer com- p,ared the risk of death associated with'various fertility regula- tion methods.and the risk of'death from pregnancy and child- b'irth when nofertilityo regulation methodd is used. The estimated risk of death for each fertility regulation method includedthe risk af the method i tself and the risk of death due to pregnancy or abortion in the event of method failure. The method-related mortality estimates-as well as those for matemal~mortality, used by Tletze. Bongaarts, and Schearer were based on various studies conducted'priotariUyin England and Wales and in the United States. For mortal3tyrelated to female.steriGzation..lhe estimate employed varied by age be- tween10 and 30 deaths per 100.000.procedures. These esti- mues did not include mortality associated with h'ystereetomy, which varied between 150 and 230 deaths per 100,000 pro. cedures. Hysterectomy waa eachuded'because "it is clearly a more hazardouss procedure than sterilization" (retae, Bon- gaarts, and Sehaarer, 1976, p.:g). As an estimate of mortality from abortions, the authors used 3.3 deaths per 100,000 procedures, an estimate based on all legal ifirst-trimester abortions without concurrent sterilizations that.occurred in England and Wales during 1968-7L About619,000 abortions were performed in England and Wales in that period (Tetze and Mursteut,. 1973):about 70 percent were first-ttimesterr abortions without con^•urrentstenllza- tions;.7t percent weree fi~rst-nimester abortions with con- current . sterilizations; 19 percenrwere abortions performed after the twelfth week of pregnancy without concurrent steril- izations; and about 4 percent were late abortions with con- current sterilizations..Themorta)ity estimates for.these four groups were 3.3. 47.0,.16.3, and 6110 per 100,000 procedures, respectively.' The overall mortality estirnate for all abortions in thesee four groups was I 1 per 100,000.procedures, Clearly, had Tietze, Bongaarts, and Schearer.used this figure as an estimate of mortality from abortions instead of. 3.3 per 100,000. they might have reached a different conclusion about the aafety, of,abortion procedures. It is evident that additional risk factors, suchasconcvrrent sterilization orlonger.gestation period. influence the mortality associatedwith abortions, and undoubtedly the same princi- ple applies when considering therisk:of anymettical proce- dure or treatrnent. In using 3.3 insteadof I I as th'e.mortatity rate for.abortion,.Tietze, Bongaarta, and Schearer chose toezciude the influence.of.additional risk facton:.fn selecting mortality estimates for oral contracep[ives, however, th'e authors did not exclude the excess risk of moeaaliiy.attributa- ble to other factors, especially,cigarette smoking, known to io- crease the mortality, risk among pill users. Since it Sas been shown that theePect oforalkontracep- tivesf and, smokingg an the risk of myocardial infarction is synergistic (btann. Vessey. Thorogood, and Do61- 1975; lain, 1976);.the mortality estimates for piBl eseamong non- smokers should be considered for• purposes of comparing the risk of.death from various fertility control methods. T6etza Bongaaru. and Schearer mayhave faced a problem. however, because the British studies did!not provide estimates of eaass mortality'attributable to pill use in the absence of smoking. 50 i
Page 73: azo61e00
69 . r The excess mortality attributable to pill'use among non- smokers is estimated heteandli.compared:with the risk of death associated'with other.fertility regulation methods: Method of Estimation The mortalieyy rate among smokers and nonsmokers can be estitrnted from the combined mortality rate, the proportion of smokers, and thee relative riskk of death associated with • smoking. The combined mortafiry r¢te (J) is the weighted average of mortality rates among smokers and nonsmokers- the weights being the proportion of smokers (P) and noo- sniokers ()!- P)'. The relative risk of death (R) associated with smoking is the ratio of tha mortality ate among smokers (K) . and'the mortaliay, rate among nonsmokers (1). Thus, l'- PK +( G- P) l' ' andRrKllor,X- Rlso thatJ'>• PRI+(I -P)! and7-Jl.(1-P+PR), . The values of I and K.can be estimated'from the values ofJ, P. and R for users of oral contraceptives: The mortality rates among heavy smokers and light smokers can also be esti- mated by using the relative risks associated with heavy and light smoking and' the estimated vaWse of l;.e the. mortality rate among nonsmokers. The basic formulas alsoo apply to nonusers of: oral contraceptives, to users: and nonusers classi- fied by age, and to the mortality rate from myocardial infatc- tion andffom pulmonaryy embolism or cerebral thrombosu.. The estimates of J:..A, and R used in estimating the values of/. and'Kare shown in Table 1., AssUMPTtONs Two assumptions were made in arriving aethe sstimates of! 1I and K for.various subgroups.. 1. The estimates of the relative risks associated with smok- ing and the proportion of smokers observed inthe British studies of nonfatall incidences are assumed to hold for the fatal incidences. This is done because the data ab'out smoking habits for fatal iocidenees are not availablt, and it istsot possi- 6le to collect these data retrospectively with any accuracy sinee such information does lrotappear, in medical or any other records. L. Itis ftather assumed thanth'e estimates oftlie.relative risks associated with smoking and the proportion of smokers are independent of age and are equal ho those observed forthe total group of women idicluded in the relevant British studies of nonfatal incidences ofthromboemb'olism and myoeard8ol! infarction. This is done because the m:evant data for nonfatal incidences for broad age groups are not available from, tlie pubGshed:reports"r . Results The estir7tates of annual mortalieyy rates per 100,000 women from myocudialinfazction and thromboembolisum sie shown in Table. 2 for users and ~ nonusen of oral contra¢eptives• smokers and nonsmokers, and various age groups..lrcan be seen that the annual'motzaGlry.rates from myocardial infaretion among those pill users who do not smoke are lowerth'an the corresponding mottaliltyrates among tAose who smoke ciga- rettes. For ezample, the annual.mortality, rate among women aged 40-44 who use oral contraccptives but do not smoke is 10.7 per 100,000• while therate among pill users who smoke is about 62 per 100:000.~ Forwomen who neith'eruseoralcontruceptivesnor, smoke cigarettes, the rate is 7.4per. 100,000. Excess Mortality Since smoking and use of oral contraceptives are.both'risk faetors, the ~ estimates of annual mortality rates are highler TABLE 1 Estimated.annuald mnRality rat. per1o0,000.women Irom, mryocardiat Intamtton end thromboemaoilam, proportion of.smoken, and'ratal7ve riakassociated with amoking, by age and:useof oral contraceptives Rdanvr i:k auociaredwirh srnnkfn; Cosue cf dfark ase of ural' Morratity rure per 100.000 PropurWnn of smck.rs AllsmolsrsHrasry LaM corurareptivo. and o:e (in yearsl wumen (1). (P) (R) smuLers• s~no4ese Myocarnial inf4reti0n, Users 30-39 54 .43 5.78. 733 2.E6 40-46 32.8 -43 5.78 7,33 268 Nonusen 30-39 1.9 SO 2.15 423 0.77 40-u 11.7 ' .50 2.15 4.23 0.77 Thrombeamholihm Users 20-34 35-44 1,5 • 3.9 57 57 t,t3 1,13 3.14. 3.14'. 0,53 0.53 Nonusers 20-3a 35-at. 02 0.5 50 50 1:49 1:49 1.49 t<9 1.a9 149 . • Heary smokeA:: tt teaar Is cipnnes p.r eay. • Litht smaken: f- lhaa IS'C+aarettea pa day. suuaeel: Ntann and Innun.l197J1. 1lann. Inman..n6 7TOroeaed 119761: and Tisuc. eansaana. uW Schearer 11976) rw mometiry rxc fram m1^nydial in! .'en; Inman alMVesscy 119651 fnt monWity f-hrnmtwemlwl:ism: Nann. Vessey: 1ToruWae, and D-Al (1977)and `/ea.eyand Do1t1196a1 iorCroWrtion ulimoksn: Jdin (1976)for rtlauve nsk af myocardiaLinParetion assac.av.dwtln smokiat: aMFYedenluen an,l Ra•<nhVll (1970) forrttlau.e nsk' u( lhremlroembolnm mseciued. .ii6 snokine. 51 •
Page 74: azo61e00
70 TABLE 2 EsOmatad annuat mortality rata par.100,000 wom.n rrom myoeardial Intare0on and th~rombo.mboltsm• by uae ot oral contr.capbrea• amokinq habits,and apa On y}anl . Myoeardial info.erion Thrombormbol'um Womrn aard 30-39 Wemrn aC<d:0-l women ojed 10'-3a Womrn aeed3S :2 SmnFina hablu Utr.a Nonsaenr Users No.wtra Uian Nonuern Urui Nenu~rn AO.smokars Heavy, l.yht Wnnamoa.rs 102 1A0 4.7. 1.0 26' 5.1 0.9 12' 62.0 78J 28.6' 10.7 15.9 313 5.7' 7.4: 1.6 4 a 0.7 1.4 02 02 02 02 4.1 l.t 4 1 A • 3.6 0.6 0.6 0.6 0.4 8moken and rwnsmolhrs 5.4 12 328 1t7 1S 0.2 39 03 • Eiuntauld nua tor smnken and noaemob+s .ww 0.24 .ad 0:14 e:speri.elY. karn apqar dee wne 6eeauw ot nundiag. among all nonusen (including smokers and nonsmokers) than among those whoo neither use oral contraceptives nor smoke- cigarettes. Thus, the value of estimated excess mortalityat- tributablt to pill use will depend upon whether the excess is estimated from the mortality level of all nonusers or only from . nonusen who do not smoke. For.purposa ofcomparison„ excess annual mortality attributable to pill use,.shown in Table:3,.isestimated here from the mortality Ibvels of all nonusers as well ae frnm thelevels of aonusers:who are non- smokers. Tietze, Bongaatts, and Schearer estimated the excess annual mortalityy attributable to pill use by subtneting the annual mortality rate among nonusers from the annual morral- ityraGe for all users. This method. however, ovemuimates the annual mortality attr3utable to pill use because the ntes among users as well as among nonusers arcdiUPerentiallym affected hytftie mortality ntes attrtbutableto.smoking. The excess mortality attributable to pill use alone can be estimated accurateiy, only, bysubtractiegthe annual mortalityy among those who.neieher smokc.nor use oral contnceptives from th'ee annual mortality among those pill usen who do nae.smoke. Thus, the excessannuad mortality rates.from myocardia4 in- farction attributable to pill use aloneare.estimated to be 0.6 and 3.3 per100,000 womea in the groups aged 30-39and , 40-4<, respectively. Tietze, Bongaaru,.and', Scheater.esti-mated the corresponding excess tobe3.S and'2111, respec- tiveiy. It.mayt be argued that.the excess annual mortality among pill users whoalsa smoke should be estimated from the mor- ta6ity level ofnonusers who smoke eigarettes, attributing the excess.th'us estimated topill use among.smoken..However• this procedure will overestimate the exaeas:risk attributable to pill useamong smokers, becausethe e(-tect.of the two risk factorss for myocardial infascaion hass been shown to be syn- ergistic. Instead• excess annual mortality, amongg pill'usens who smoke shouldbe es6mated~ frorh the mortality level of'f thoaeu whoo neither smoka nor use oral contraceptives• at- tnbuting the excess thus ,timated to the presence of both the risk:faetors_ Thus, the excess annual mortaGty from myo- eardial infarctidnn attributable to pill use and smoking is estimated to be 9.0 and 54.6 per 100•000 women in the groups aged 30-39 and 40-44• respectively: The excess mortality of 54.6 per 100.000women inthrgroup aged: 40-G4IinclJtdes about 3:3 deaths attributable toptll use alone, 8.3 deaths at- tributable to smoking alone, and about~42:8deaths atttib'uta- ble tottiejointelfectof pill use and smoking. Risk of Death from Pill'Usa What is the etfectof.the above-mentioned differences in estimates of excess mortalityon the annual number of deaths associated: withh pill use? How dbes the an.nualrisk:of death from pilll use compare with the use of other fertifity resulalion methods, and how does it compare with theriskof deathfrom pregnancy and childbirth when no fertility control method is ueed? To study these questions, thee annual number of dea[.hs a¢tnbutableto pill use alone and those attributable to pill use and.smokingare estimated by summing tlieeaeeu mornlityy from myocardial infarction..throm~boembolism. and moruliltyassociated w3hshildbirth im tlieevenrof pill failure. The esti- matesofexcess mortality'attributable to pill use are taken TABLE 3~ Eadmatod escaas annual mortality per,100,000 woman trcm mryoaardial Infarcdon and thremboembotlsm• by' use of wat contracapnvea• amoking habits, and age Cn yaarsl Fscu1 mn.rafu+yfrorn myocardiel iWmion in rompevifon wahc Eccr.camonaliryJ m. rlvombaembo!'urn tn eompanron ~.rh:. Allnonaaeer. No.tamo6rr.nnu.urr Allnonn,rm NoRSmo4emm-era RilkJactors. 30-i9 10-f4 10-J9. tp>JS 20-3< 3S-1J 20.3a' 31~? Pin iuse ann smokin0 8.3 50.3 90 5A.6 1.41 3.6 1,4 3.7 Heavy 11':1 670 11a 71 3 42. 10.9 42 1AA Liqno 28 1Q9 3.5 21.2 0.5 1;a. 0.5 1.5 Rn use cnty -0.11 -i.0 0.6 3.3 1.2 3.] 12' 32 AllOdlusers 35. 21.1 4,2 254 13' 3.4'. 13 35.
Page 75: azo61e00
71 TAttLE 4 Annuet number of deatM aasedated with cuntrOt of tertWryft per 100.000 trenatama woman, byy memod and aga Ape.JAUp (ie. ytevt) MetlrodofJsMnryreerrel' 15-19 20-24 . 21-29 30-31 3J-39 40-J4. Orat cpwacsptbn (w/mout a bonion bacld101 1;4 . 1.5 1b 53 7.5 249 Nonsmolura 13. 1.4 1.4 22 4.4 69 Bmokera 1:5 1A 1.6 10.8 17.Z 58.9.- H.avy 4.3 4.4 4.4: 16:4. 23.4' g2.7 liqm 0.g 0.7 0.7 4.4. 16 231 IUO (wi0rout abort)en b.rJup) 1.1 . 12 1.2 1.4 1.6 1.4 Tnsdhkavl epqraeaptidn (without abonmon Oackup) 1.1 1.4 19 17 4.7 4.0 Tnsdi6olul contracapuon (Jritlt aoortialn baeaup) 0.3 0.4 . 0.4 0.8 1.4 0.6 AbOmon only 23 2.5 2S 52 9.9 6.6 NoINtflity.COnMd' 5.5 52 7.1. 14.0 193 21.9 feu.ca:Torae, geeprra, aed Sra.anr (197a)..+eeNfardara ee.mes.n rd ee.rauara wn: Ti.ra. aatparta aM sea.m alwawd aN coneacrpe.v. uN fUDt tebe 9e.w- rlacu.e.M eMl+ daul nrthoe+. ruea a cnndwn ad di.eAnpe. re b. 9o ryee.er decrl!n in pr..eming pnsmxy: Tb. eocemea monaOYy rue w ussmd bW e. 1..1 pFr 100.000 Y.e 6inaa, Abatiom ~.r .re atWmac tal. IqLL paforte.d eunne the ant Mnwr.r ar.itaeat eaeeorrent swi6nea... from Table 3and'the estimates ofmottaGty aasociated with childbirth in.the event of pill faslureare taken from Tiette, Boagaarts, and Schever. For purposes of comparison, the corresponding estimates for other feruliry control methods and no fertility control are taken from Table 4 of the same source. The.reaults, sh'owts in Table.4 atiove, indicate th'atth'e an- nual risk ofdeath from pill use amongwomen aged 13-29 is lower than the corresponding risk from pregnancyy and'thild- binh in the absence of fertility control; thisis true for non- smoken as well as for all smokers-heavy and light combined.t Because the disrinotion between heavy and light smokers can be subjeet to some unknown degree oterror in reporting the number of cigarettes smoked per day, the figures for these two categories should be interpreted with caution..But the figures indirate a Wkelihood ahac pill~ use among heavy smokers aged 15-29 may be more hazardous than any other method of fer- dlity regulation. though still lesss hazardous than usiolg no ~method to controlferuGty. The annuall riskk of death from ptl3.l use among nonsmokers aged :30. and ~ over is sutistantially, lower than the annual risk of deatE, .fromchildbuth.when.no fertl7iryeontrollmethod uuseds as it iis for younger women. For ezample, the annual i risk of deatlifrom pill use alone among women aged 40 and~ over is about one-third'ahe risk of death front pregnancy and childbirth when no fertility cotitrol'method is used (7 versus. 23), although it is of~the same order of magnitude as expected with the use of legal 5nt-trimester abortion without con- current stersliaaYton. For pill users between the ages of 30~and 39,.the risk of death among those who are heavy smokers is sbghtlyy higher than the correspondiitg risk fromchildbirth when no fertiliry control'method is used:..However, for pill usenaged:40 arld over, the risk of death among all smokers (heavy andLght combined) fs substantially higher than the corresppnding risk . 11. these e >. heavy seroken an NCU.r/a rnroW. IS ormora eiµnnn p.r Jey. litht fmCken iha.e ~hoamok4 rtr<r thm I S titaRnea per , day. from chrldbirtA when no fertility control'.method is used (39versus 22). The corresponding risk among hesvy" smokers Ia ebout four times as great as that associated with the absence of~ fertility control (83 versus 22)..Al(hough ,the risk of,deaN amon;light smokers is of the same order of magnitude as ex- pected wish th'e absence of ferulity control (23 versus 22). it is substantia)11y greater than the risk associated with anyy other method of fertility reguladon. :' Discussion It is unlikeiy that the relative risks of myocardial infarction or thtomboemboiism associated: .vith smoking or the propor- tion ,of smokers will vary, substantial0yy with age, especinllyid the group aged30-44:,However, tberelative risks associated with smoking or the proportion of smokers may increase slighdyy with age. In either case, it can be shown that annual mortality rues from asyocardial infarction among nonsnsoken will ~be Ibwer than those estimacedln this analysis on th'eu- sumption of age independence for wnmen in the group aged 40-44 and higher than those estimated for women in the grnup aged 30-39. Siitailirly. the annual mortality rates fromthromtioemb'oliSm among nonsmokers wfllbe lower than those estimated for women in thee group aged 35-44; and higher than thede estiaeased1orr womenin the group aged 20-34. Thus, annual mortality rates estimated in ahis analysis for those women (users and nonusers of oral contraceptives) in the group aged a0-44.who do notsmoke can be taken as conservative estimates. This means that even.if the relative risks associated wieh smoking or, the proportion of smokers atn foundd to increase with age, the concltitsi~on of this analysis in regard to thee safety of pill use.ammng nonsmokers after age 40 is uniikel¢ to be.affected- l t is essential to point out that the mortalityy estimates used in this paper are based on small numbers and ma,v,tie subject to la rge sampliing errors. Thesee estimates are also subject to up,w.ard, anddownward: biases, which may not cancel.eachotheoouc For exam.ple, it is possihle thatin, the preseno 33
Page 76: azo61e00
72 analysis the exceasmortality attributable to pill use and', smoking is overestimated because some of the women may be exposed to other risk factorsaueh.ash,ypertensions obesiry, and diabetes. If so,.givi'ng up smoking bythesewomen may not reduce theirexcessr risk to the level estimated for pil users who are nonsmoken. Some of the pill usen.who do not smoke may.also be exposed to these.addi'tional~riskfactors; therefore, the.excessmortality attributable to pill use alone is likely to be even smaller than the estiinatesused in this analysis. ©n ahe oth'er, h'and, these estimates may be subject too some downward bias as well, because pill use.truy'be as- stxiated with other causes of death not included in the present analysis.7he duration of pt7luse may also in®uence the ex- cess mortality attributable to the pill. The net eRect of these various factors cannot be estimated without additional! re- search, preferabiy in differentsettings, The choice of a particular method of fertt7iryreguladon de- pends upon many considerations. The safety issue, although a very importantfactor, is only one of the.consideratiotn im- volved'in decision-making. In assessing the relative safety of oral contraceptives, whether tlie.mortality, risk associated with pill use.is compared with the risks associated withothkcmeth- odsoris com~pared'wieh the risk associated with th'e:absencre of fertility coatrM will, of course, dlpend upon the availability of other altlernatives. The present analysis does not change the conclusions of Tietze,.Bongaarts,and:Scbearer in regard to the safety. 06 the 3 UD, traditional methods. and early abortion . without concuaent stenliution. AII'. these methods used alone or in combination are less hazardous than no fertility controt method. Pill use among nonsmokers is also less hazardous for all ages than no Settililtyeontrol method. References Fredenkeen, H, and R. Ravenhoit 1970. "Tbromboembolism. oral contraeepdves, and cigareu.s." Public Nralth Rrporq 85:197. Inman. W. H. W.,and M...P. Vessey. 1968.'9hrestiqatimn ofdeuhs ftown pulmonary, coronary end cerebnl th'rombosis and embolism in women of childbearina a8e." BriM.rh Mrdbaf Jeurnef 2i 193-199: Jain. AenWit 1:..1976. "Ciprene smoRing,. usa of onl tontraceprivea; aad'myoadial infaretion."Amr.:ran Joarnof.of Obsrrerics artd' GynecoroaJ 126(3):30t-307. Mann, 1. I" and W..H: W..Ihman. .1975. "Onl icontraceptives and death from myocardial infarcuon.' B•irisA Mcdicat~.Jo.vnd2:?a5- 248. Mana, J. 1. 1v..H: W. Insnan. andM: Therogood. 1976. "Onb con, traceptive use in oider women and faol myocardial infan:tipn." Bnrirh Mrdicd Journd 2:4a5-er7. Mano, L L M.; ThoreBocd. W. E Walten,.and C. Powell. 1975. -Ord~contracepavesand mypcardial infarction irt youn8 womea.". Bririrk Medicaf Jovrnd 3:631. Maae, l. 1., M. P. Vessey, M. Thotosood. and R. Detl.. 1975, "Myo- cardiYl int.nction inyoungwomen with speciat refnence to oral'w conoaceptive practice," 8nnah Medical Journad 2°2a1-2a5. Tietaw Christopher. and, Marjprie Cooper Muoteia.. 1975. "to- dueed aEortion: 1975 9etbook." Reponr on Pnpu(anunAFa.,rity Pfanntnt, No. 14 (2nd, Bdikicn), December. New. York: The Populuion Councul., Tietze, . Christopher, . John BmKaarn, and Bruce 3chever. 1976. "Moctality associatedwim the control of fertiliry.- Family. Ptan- ncmp Prrrpeedvef 8:6-I4.UnitedSutea Food.and Drvt Admiitistntion, 1976. "Oral conua- ceptire drua producu:Notice and'proposal'of revisedphysioiaa and patient labelliinj "Frdrraf RePir+rr a 1(236):536J6-53642. Vessey, M..P., and'R Doll. 1969. "Investi8ation of relationship be. tween useaf oral contneeprives.and thiomboembolk disease:A further report." BrfrerA' Medical lovmaJ 2:6J 1-657:. Alot1T Trii ,1I)THOR Anrudh K lain is an associate in the Population CouncII's International Progruns. Acs:xowcy:ooasarvTs This analysis was supported inn part by. National~.lhstitutesof, Health Grant No. HD-0567I..Tne author wishes to acknowledge the comments of S. Bruce Schearer. . Sheldon 1. Segali Irving. Sivin, and Christopher Tietze on an.earlier version of.this paper. MortnGry Risk with Oral Contraceptives: A' Cottmatot Thls artiete by Mrudh Jaindemonstrates tnat in assessing the excess risk'.attnbutabte to the use of oral conRroceptives, it is esseAnal to con3iderlyparqistic effacts of other risk faeton such as amoking. obesity, Irypertension-and perhaps others not readily Idennfiad.7Taanalysis shows that piLLusars who are nonsmokers ar. at much lower nsk el heart disease mortaliry chan woman who use pills and smoka cigarettes. Put in omer terms, average excess mortality, from pill usa li depenm verymuch on the eatentef ciqarette smoking by woman in the society underconsidera- ' ti011o Most ot the dara avaitabte on mortality associated.wtth the use ot oral Contraceptlves come Irom ease-cnntrot studies or Ionqituoinat, prospecd:ve studfes earried out in the United Kingdom - or in the United States. Ciqarette smoking by,women is mon comrmon im thesa.countnes than in most developing countries. To evaleaee the nsk-banelit ratio invotved in theuse of orali conara- captives,. policymakers muat determine the risk mat applies under the cundit/ons of'use In nheirr countries and not e.tnpolatedrrectty,from, reports of studies carned out tlsewhere.Indeee,: mee same.principle applies when evaluabng Ifiahealth bene/its: one of which is a reduction in the merta-iry,ra[e associated with pregnancy and childbirth. In countries wnere maternal mortality is hiph, :he haalmbenefits ofeontrolliny fertility will be even qreat.rthan in countnes where deaths from preynancy and Chilidbirtl are relatlvely rare. Both elements of the n3k-tfenetit ratio' must be assessed within the appropria[e national context Shel~don,J. Segat'Diirctor..Cenrrr for B7omrdieaf Rlsenrch The PoPulation Council 3t
Page 77: azo61e00
73 The Lancet • Saturday 8 October 1977, MORTALPI'Y AMONG ORAL-CONTRACEPTIVH USERS Royal College of General Practitioaen' Oral Coavaowsiba Ssudy' Srrrsmary In a large prospective study carried out in the United Kingdom, the death-rate fromdiseases of tlse.circulatory system in women who had used oral rnmncaptives was 6ve times that of eon- trols who had never used th'em:; arsd th'e death-rate in thoae w'h'oo had:taRen the pill continuously for Syears or more wasten.times that of tliecontrols. Thee excess deaths in oraltomracep'.ive.asers were due to a wide range of vas- cular conditions. The total morrtality,rate in women who had ever used the.pill was increased:b'y.40"k„and this was due roanincrease in deathsGam circulatory din- cases of 1 per 5000 ever-usen.per year. The excesa.was substantially greater than thedeath-rate from complica- tinns of pregnancyin the controls, and was double the deatfi-rate from.accidents. The exceesmortality-rate in- creased with age, cigarette smoking, and duration of oral!contraceptive use. fotroductioa Tste Oral'Contneeption Study of the Royal College of Cxneral Practitioners is aa continuing long-term pro- spective rtudyy of.approximately 46000 women of child- bearing age in t1ieUniied'Kingdom. Itbegan in 1968 and is designed to eval.uate.th'e eRects of oral contracep- tives on ,.hcalth~ Women were recruited by 1400 general practitionas whohas•e recorded all new episodds of.ill- ness reponed'bythe study population. A comprehensive account of the illnessesin those using oral contracep; tives cominuously ("taked')a those who stopped using them during the coune of the study, ("es-takers"), and those.who had neverused them ("controls•').was pub- lished in May, 1974.' This paper discusses the deaths recorded during the follow-up period until )une,.1976, whicls covers some 200 000women-yean of observation. Metb.ods The nudy, design, mahods of dno colleaion•. analyticai techniques, and'ahe potemialsourcet ofbi.a hsve alread,v becn discussed in detaill.' In summary, 23000 current, taken md 23 000 eomroli who - tnatehed by age and marital seuus to the takers,m vare reauited over a}4•nwnth pcriod. Controls wholater berame onl<vntnceptivc usen were included in the • 14inriptl auh- Dr. t•a-u Rcut., drpanmeu, ef inediut uaw.. tia and.piden.dopy.liomlon Schuetof Nieiderud Trnpieal MedF rinr. wCl. Dirator: or Curroaa R. Kar, RCG.P.:Resurrh Unit, M1nch.uar. •'taker" ccategory from:t the tinx of change. Es-taken wha rnumed oral. rontraceptive usewerethereafter raclSdad from the present analyses. At 6-nwnthtn intervals the genesal praeti• tioMrs report on the occurrcnee efillness, pregnancy, or death aqd on the details of ora1contracepuve use in she aud7 popu-lation. For a0l deaths the `eneraa pnctitionen repon the cause o( death in the same fonmat as is required for tlie death eeni6- cata. The underlying c.use of Lnth p•as eeded by,CR.K, using the gth nrvivonof the Imcrnational CJassification ~of'Diseasea (I:CD.)and checked by V:B.The dlaths in the takars, n-takkn,: and mmrols are related to the respective cumulative takndar-months of observation in eaah. group ud ecpressed as mte per 1001100wonsen-.c.rs. 1ncrrtain analyw thc taken and ex-sakere are Emuped together s "everusas". Monality-ntn are standardued by the indirect metbod,'1 using the toull populition rata as staa- dard. 17 women ,rho died (rotn conditionswhlch had been diagnosed'before their recruitnums have been eacluded from these analyse. ResultY Table t compares the monality-ntes from various auses.in eva-ttsen.of oral contraceptives with th'ose in controls. The number of dnth's,.mona1ity-ratea,.and women-ycars ofnbservation are shown in table la when all periods of pregnancy,and related deaths are excluded, and'in tabie sb when they.are included. When pregnancy is exditded, the total women-years of obsetvation is simi- lar in the ever-users and the controli.:The controls have doublethepregnanry-rate of the ea-takers, thus, when pregnancy is included the periodiofotiservationare greater in the controls than the ever-users..Irrespecti4e of the inclusion orf exclusion of pregnanry,. however, th'ereis a.40;o increase in total monabtyrate among the ever-usen. This is because the increase in the wameo- yean ofobservation in the wnsrols wasat:companied by only 2additional' deaths from complieationsof preg- nancy...(Tlsere is no program to ssandirdise the rates for -age,M parity, smoking,, and socal, class; but standardisa- tion for these factors when pregnancy was excluded 1tad no appreciable dfect on the comparisons between the two groups.) Since the indusion of pregnancy does na alter the comparisons of death-ratn inever-users and controls, and for consistency wiih earlier reports,' all periods of pregnaney, and'associated dnshsare cxcluded'ftom the subsequent analyses. Table ta shows that the standlydised:monaliiy-rate from the circulatory diseases in ever-usm is 4.7 times that of controls, fromnon• rheumatic hean disease and hypertension it is 4',0 nimes that of, controls, and fromm cerebrovaseular disease it fa 4.7 times the control raoe. The exceu total moQality in cver-usets:can be accnunoed for by the excess deaths from circulatory diseases. tOtt6Ti,e Lnm 4d, 19i7 ID
Page 78: azo61e00
. 72g Table n details th'ccauses of death by oraPcontraceN tiva use_The divervty of dreulatorydiseases among thce oral<ontracepti.e ever-usea is notable, although th'ee number of dlaths from any singledisease is small. Tab'le ttt shows the standudised!moreality~ntes from cenain diseases of the circulatory systern by.comracep- tive use at the time of d6ath. Themonality-rate from all circulatory diseases in the takers is 4•9 times that of the controls and in the er takers is 4-3 times that of the eon- trols. A miew, of the records of the 8 ex-takcrs who died from circulatoryy diseases revealed: that 2 had' disLron- linuedonl<ontraceptiveuse when they were found to be hypertensive,,and later died of.malignant h'ypenen- sion. The other 6a-ukers stopped using oral contruep- tivet forr non-medical reasom. None of the 4 wbo died'of subarachnoid: hanorrhage bad been repanned'to be hypenensive,. Table !v compares the suadardised', monafity-ratet from circulatory and'totah diseases in.ever-usen of oral contnmpeivesS and controls according to the women's cigarette consumption at.t recruitment into the study. The ratio of the awrtali[y-rate in ever-users so controls is 4-7to 1 for non.smokers, and 44 to I for smokers. 9eauseof the small number of deaths in the non- smokers, only the ezcess deaths in smokers is statistically significant. Table v.comparesthe monal0ty.rues from circulatory and total diseases in ever-users of onl contraceptives and controlaby the women's age at the timx of dnth. The raees increase.with age, but at all ages they are higher in ever-users than in .controls.. When these dau were re-analysed, classifying women, . by their age at emry into the study, a simular increase in eachh age-spc dfic monalrty.ratewas noted among the ever-usesr, standardisationfot smoking, parity,.and social.clau madee no important di/Icrence to the ntea- At, present 74 TM[t.ANCLT,OCTOi[agr1977 there is no program to scandardise the data in table v. for these.faators,.b'ut it is unlikely, that the procedure would materially alter the comparisons between ever- usersandcontrols.•' . TAaLa II'--C.WSa Or DLATN t1 ORALKDnTRAetRlvt USa Na e( deuM I.CD.-de tSe: ('-onueh 000-090 Infa+iem (I repiaeniL I Lep.dris) . 0 140-209 Alaliae.nu ds.a' 1. 20 2111 Atnnularytn'r -0 3No AlulUpk dese.ii 1 .00. Alaliananihpenn.iee 7 410 Acuu nmocanlial infamien 7 411 Atw< rnryorard'ul imu/fiOtee7 t 421 ~30 Subal einorr6aae 9 131 C<nln.ll haemrhate 1 132 Ceiebnl ~h'rwnbw's 0 M4;2 MlMeteric-anery ,h'rombesi I 2 .50 PuMqn.r!7.mmolr I 560 Abdomnsl heni! (Oastaperwn); 0 363 371. i tDcaau.ecnlrtw Grsnnu, cf G.a 0 512 t]tronie tteporir. 630.67a Camdioriar uf pepnn 7, , 796 ehitdltiM .ndrhe pupyvivm nl-Hd6nrd'qYb I I Fa00-9N9' Accidenu ' 6 F.990-979. Saiede 2 E960-999 (la.kide 0 E9t0 roiaprrina--rca.n rrnNq.n ~ t AO nrun 56 43 'fGxn deMhs is e.er urm: Ilaree inua:nr, 1~rmum; I p.nCaal; . 2lune; I rwm.sil.e riuue; 1 melannma; 3 arean; I ama; I ki6,q; I IUnin; I kuk'enu.. G„m dsarhs:in canudr:. I csspT.tu,; I na m.ch; I Iirtrin-:nr, I rm m; I peri~mneum; 2 tuna; 6 tneM; 3n.ary; l.uln:1 Hadakin*I dlsa.e, 2 kukaai., 7AalrI--MORTAt.I7T.ILtTa raa I00 000 wpMa)FTtAai ntOM VAaIOUS CAUSES BT ORAL<OIRaAC7JrlVa USa (.), odndin[ Drepwnq.(n.nd.rdued fw Se ar mu7. 9uny. .ndL dus, and MmYfeL), (brinchdieawrp 7,(vnnutlardfrdrne}. . MenaOry,rue(na.afdnrh.) Rniddrna i I.GD. cede Gus E+enwm Cauuels n e.<rrne. lammro6 (p)110-2D9 I -sp r..e.n t7a (tq 21 •1 (20) 0d 390-47a Aadns.uaoJUro . ' ldro.yry- 5.5(5) . 6-r .00"29. Nawrh'dn4lKrwndseaM. .ndl,9penemiaa 2.5(2) 1 -0} 130--t3a Cc.brwsculudawse 113 (10). 2-t (d) 4-7t E800-999: , Assidenu -d aidese t7,1 ;(16). 10-0 (9) 1.7 Ea00-9<9; E910, Aeudevo. 9-3 (9) 7-6(D) 1.3 tl930-999 Suidde 5.3(5) 2-2(2) 2-4 ... ONer r.n+s 3-o(z) 7-1(9) . 0-4 .,, TMd 636()6) 16-0(13) 14 tcamnw,w.n afnb..ru.n:w 91161) 91 521 I l61630-67t Comp6cuioa. JD^ra*~*4~ <A,7db.rrA,endpnrrprt:... 0~0 ta(~). .. .. Tad sa-0(d6) 40•9.(!1) I 1• It'..r ~.n eJ,.a;erw.ina 96 62/ 110065 -.<0•m.tr<o-0s.. . Fkcawe ihs naa (ar eeh nuse ue nandavdised repsr.uly rhms u!rc amall diherepaeoe l,- nhe rvm al the indi.idual nia snd ihe -iaH' rerC. ~ t
Page 79: azo61e00
75 TML 1.61+CtT, Oc70a1R 9, 1917 729 f:ALLL W-MORTALITY-RATL FLR 100 OOO wOMLM•YUaS FROM YARIOUS DISEASES Or TNL c1aeU1LATORY. SYSTEM aY' CONTfAGL7T1Yi OSL AT~TML Of DEATM (S7AMDAaDI3LD F0t AGl AT EMTItY, 9ARITY, SOr1AL CLAal, AMD SMOaSMO) . A1wlity-rau 14sia ef ra (ee•a(dlstM)'- tlwafcaueds LCff. code j Gvu T.kas E.-tsk;en Gnud1 Takm E.raken 41a1~29 A....Rea..e:i Ann-I:Ye.M.dl2perts.eia 11.1 M 7-6(1) . 1.6(2) 4•T 3-0 400, Mµipuea R.pmemiat 0.0 7•E (Z) DO 0-0 110. Akuum,rorardiYtidfataiaR . t.l'16t~ 7•3(f) 2-3(7) 3•7' l-l 430-431 G*rbnsars.l. /:r++. , ..12 4(¢), I 1s-7 (4) 2.{ (n' I I 44', S•7 410 Suborac6noidhatrhaae 9.3(.1). 15•9(I) 0•0. 411+d32 CershrY unomba.is and 4martNie 2i111. 0•d I 2.2 (I) 0•9. 0-0 A40-sf{. OrAmeurdnd:n.w 2-S(7) 29.(1) 0•0 .. .. 79WSt ADsix.4twy liruw )6l(16)i 23-9.((J. S.f(J) 1•9t 4.7t •r<0-0S. tf<OOI.. Table vi shows the mortalitynte from cimtlatoryy diseases by duration of oral<ontracepuive usa 77u analysis is confined to women who h'ad been using oral oontraceptives continuously, up to the time of death (tak- en). lCSth inereasing duration of use there is a striking increase in the ratio of the age-Mandirdised mortality- rate of ever-users to that of conirols. Those who had' taken . the pill for Syean or more experienced a nte which was 9,7 times that of controls. AI(the deaths in womenn with a duntion of use of. S yprs or longeroccurred at age.3S or older. As yet theroareonly small TA1Lt N-MORTALITY•aA7L FLa 100 ()0O waAUM-YEARi FaOM DItLA3E3 Of.GRGULATOLY 7YSTLM 39O-15g) AND TDSAL GLATMS aY SMOCING MAaIT AT pTRY AND ORAL-COMTIACEfT15'a USa (1TANDAaDISED rOa AGa AT tNTRY. SOI:.W. GL1lSS. Alm rAR1Tr) Gas- t Martdn«rau (na o(denRs) n d'Me n eru-asus .,RO vL a rntp I Cavar af d.uh E.a..am C- u eM u mnvob C.•MVlawc9 Toul IS•t (n 49 t (S01 TD(2) . 36•3~.(2d) 4:7 114 s..Ra, Grcala,.7 Twd 76.t ()6)i 9•9(n 57•9120t 4-4t 1.3 Sr<001. TARLL Y-MORTALITY-RATE )LR.100 0OO.w'OMLftYEAat TROM DnsLAStl os CIRCULATORY SYSTEM (Le.D. 39Q-45g) AND TOTAL. DEATMS, aY AGa ATDLATS/ (oM37ANDAaDIS1D) Murtaliq.ue J (_( 6eab) A{e a- .<diMe Eva-usm Gmro(. 15-24 Gneulawry 7.f ({)l D.0 TMaI 1 14.9 (n. 11u (n 23-34 GrcvMta7 r t (4) . Aj4 p). Tnot 37•S (17) 24.4 (11) 35-M Grtulatoq 12.6 (17) 9,6 (1) T.m at-t (z+) 77.4 (lJ) 45-49 ' Grculauq 140.9(7) 0-0 Tad 1.15(1D 1l9•5 (D 's<d•OS. Raln oL nu ~n e.er•.uaer. ho mntsdh TAat2 ~MORTALITT-RATE rrR 100 DOO wOMEM•YEAas IROM DISEAS[S OT tlacuLAtoeY TYSTEM(LeD.390-151)aY DURATIOM or OONTSNUpus oRALCUMTtACEITIYL USa flvrnion e(. aanrsaqRi.. - (-t 0 t-s9 6D• RSonaG,..nuc.a~ afdtatlnt V!ae (Tt timr af dtath) IS-34 3s-4;' 7.7,(S) ] {.f (n' D. 0 7~9 (n 13.0(4) 113•t (9) 'Tau fer IiRw uuMti r[0•Ot. Agepaedardnd f-l (lt u.f (7) S0.f (71 nuTnbers of women under ihe age.of 35: who have uken the pill continuoual}foc this length of time, lo thleir risk cannot be assessed sepanseljt'. The relationship of monaltlty, to the type and dose of astrogen and progcstagen was eaamined•' Because of.the large number of p,reparnions,.the secular trends in astrogen conlent, of oral eontraeeptives, and the smallnuml5er of deaths, no dear pattern emerged. It is also very un(ikely that diagnostic biases have occurred in the broad l•C•D. groupings.usedfor thele analyses. ifthe eseea deaths from circvlatoryy disrase among the oralkontra(epsive users werea diagnostic artdaa, then.it would be espeetedthat, the dealls-rate from all other causes would'be correspondingly reduced in ner-uters•.But the diHerenee indlath-nae buween ereswscrs andaantrola from other conditions is too small to explain the excess of drnvlatorYdisease in ever- users. Furthernmre, ttao•ehirds of t_he eireulaeorydeaths wereanributed to either myocardial infarction or subar- arJinoid hrmorrh'age..neither ofwh'ich was suspectedas a complication of, onl-contraceptive use att the time when most deaths ocevrred- Noris there any,indiation that . a differensial' losss to follow-up txtween the ever- users and controls could eaplaio.these findings. IXRten a woman changes her address and transfers from, . one general practitioner to another, she is automatica4ly withdrawn from the suudy. This is.responsible:fos the large maiority of the losses to foltbw-upI and theyy are unlikdy . to be related to the patient't mosMdityy or LoD- tneeptive usa• An important consideration in interpreting the dilter- cnca.in disease.panuns between ever-usen and controls
Page 80: azo61e00
76 730 is that oral-tontraeeptive users are a selected population. Theyan sel8-sdeard, as they choose to take.the pill, and they aremedicaltlyselecteda as they must obtain a prescaiption from a doctor. They tend'to have had fewer seriouspastiUatcsn,t•sto use less medication ,' but to smoke mome than non-usera.tx Adjustment can be made for the majnr differences byy excluding those with pre- exinent disease and naodardising for risk.faetors stxh as age and smoking habit. Small'.di8erences between ever-usen and controls may still remain after these adjustments, but they could not explain morulity diA'er- entiils of the size observed for circulatory.diseasa..On, . the other handa theymay,explain small differences such as the excess of accdents, suicide, ind'homidda and Abe defieieney of cancer and oth'udiseaxs in the ever- tak'en. These mayy be chance findingsnr they may reflect psychological and othQdilferentxs between ever-users andeontrola.. Conclusions The diBerences in death-nre from diseases of the ar- odatorysystem between ever-utln of oral comraeep- irvas and rnntroli is 20~per 100000 wromenper year, which represents an increase of.I death per 5000 ever- users per year..This accounts for the excess total ~morta- lity in ever-usen (table +). It is substantially largathan previous estimatea which considered onlyy the risks ofth'somboambolismf and myoeardial' infaretion,ibut' is similar to that based on observations of recent trends in monalitry among young vrornen! This rate is more than double the monallty-nte from xll accidentr in the studyy population. It is also important to note that the in- aeased mortality associated with oral-coatraccptiveuse is much greater than th'e excess mortality (148per 100000 womenn per year) associated-withd the la.-ger number of pregaanaes in the controls. Although the death-rate from dreulstorydiuases was increased byl death per 5000 ever-users pcr year, this risk was concentrated in otder. womat, in,those with a long duntion of oral conanceptive use,.and in.cigarette smokers..It can be estimated from table v that the eacen annual deatb-nte was about I per 20000 ever-usert aged 15-34, but it increased to I per 3000 ever-users at age 35-44 and.to 1 per 700 at age 45-49. Similarly, thee excess annoal'.nte waa about I per 8000 womea who Siad used oral contraceptives contiitsously far lessthan5 years, bus I per 2000 for continuous usen off more than five yeans duntion (table vs): Also, the c:cess annual deach-nte was I per 10 000 ever-users wtiao do not smoke, but I per 3000,evertuxss who also smoke (table sv) Tnese estiinates are.based on smail numbers and are necessarilyy approximate. 1Ythout more ditait is not.possible to examine the ihterrelationships.of age, smaking, and'duntion of oral<ontraceprive use, nor to mmpare the effect of continuous oralkontnceptive use with interasiUeat use. The excess mortality among oral+tamnaeptive users is of sulfiqent.siie to.warnnt a careful reassessment off pill'usage by older wcmen unless tliere are strong sodal~ personal,, or other reasons to use th'em. It~. must' be strrzsed that the size of the risks desaibedh'ere are ap- proaimace:. Moreover, they would'not apply in other countries where the prevalence of eirculatory.difeaseis di8erent. Continued observation of theseand other Ttta wtccr, ocyoaastg,1477 groups of women wip be requiredd to permit.a more dnai1ed'evaluation of the risks and bene9us o4oraF contnceptive use. Discussion Women who had'used orall coneraceptives had a morta- lity-nte from circulatory diseascwhich was 4.7 times that of women wh'ohad never used them. While these and comparable prospective data from the (3nford/Family planning Association study' supponth'e condu- siotts of earlier case-eontrol studias'a they alsaindicate . that the vascular diseases associated with onb-contrn «ptive use arc more varied than has previously, been recognised. The dnths,.all in oral<ontnceptive users, from subarachnoid hzmorrhage, malignant hyperten- sion, cardiomyopathy, and mesenteric-anery thrombosis in this studyand'from congenital land rheumatic hean--dsease in the Oxford study illustrate the divenity: These have occurred despite a lower prevalence of vas- evlar and other circulatory.diseases in users than in noo- usen bdore recruitment into the smudies.'a These hnd-ings.closely fit Bersl'spredictions, based on ananalysis of mortality trends in young womenn from twentycne countries! She suggested that the range of vaseulacdis- eases aRraed~by.onl~nrraceptive.use andth'esize of the «skk invo4ved' were substantially greaterthan thee combined risks of thromboemb'olism and'myoeardia6 in- farction- 7be present findings suggest that the riik k of.eircula- tory disease increases withthe:dunuan.oforalconen- ceptive use and may persist after the pill is discontinued. The increased mortality of the e=-takrrs couldnot, be explained by a transfer of seriously ill women from the "taker" tothle "u-taker^ category~ The ratio of the mor- tality-nte from circulatory disease in ex-cakets to con- trols is n7: 3.7 to 1~even afierueluding the 2 deaths from malignant hypertension, where ora4-contnceptive use was stopped becausc of the onset of.hypenension. This suggests that oral.contraceptiva indutce changes in the circulatory system wiiichare not immediately rever sible. Since onSyy small i numberss of deaths are involved: and'since the relationship of durationn ofonE contncep- tive use to the risk of'dearh fromm circulatory discues could be assessed only for women who had used onl . contraceptives eontinuouslri it is premature to draww firm conclusions about the LGag.term and'residuat vascu- lar effects of oral contraceptives; but clearly they reqpire further study as additional dita become a.ailable. Thec nsio: of the death-rate from arculitory diseasa in everusers to that in controls was similar for.sm~oken and non-smokera (tablee iv). This suggests that the rela• tive.inwease in mortality associated ~with onikontneep- tive use is independent of smokirg habit. In contrast, the ratio of the death-rate frorn circulatory diseases in ever- users to ihat in comrols increased from,2-0 at age 25-34 to 4-5 at.age 35-44~(tabld v), The incrase in monality: ratio with age in part reRects the lenger average duration of braliontraeeptive use by older women, buc the numtier.otdeathris as.present too small to perttsita detailed anaiysis of the imerrelationslup between age,, durationq of onl-comrsceptireuse and'riskof death ftom circulatory disease. - 11 i,is very,unlikelyahat'any form ofbias could explain these SnSino . The over-reponingof symptoms b7. .
Page 81: azo61e00
77 TNa uxrsr, ocromaR 8,1977 'a'amen using tx'al contraceptives waseonsidered to be the major source of biu in interpreting the morbidity findings in this population;' but, this is eliminated by can8ning the analysis to mortality data. General praeti- tionm should fmd:our about all deaths,regardless of whcre they oaw w•hen,as part of the routine adminis- «ationof the National Health Service,the decased's name is removed fromtheir practice list. Although the deaths would rlotbe missed, the monality-rate in this population is lower than thatin the general p,opulation.' This is because, as in the Oxford study,' the study popu- lation tended to be healthier and of higher sodal class than the general population, and also because 17 deaths from causes diagaosed before remaittnent were excluded from the snalyas The cause of death used in these enalysa is that recorded by the general practitioner. The validity of their recording of diagnoses for hospital admissions has alreadybeendemonstrated.' To uphold the important principle of confidentiality, w•hich is basic to the conduct of the study, copies of the death ceni8catahave not so far been collletted. If the gmeral ipraeailiomen agree to divulge the patients' names, it is hoped to aanine the mnsiseenry of their reoordiing of the cause of dath with that recorded on each woman'sdeathceni6tate. Eaalsr ination of the source documents suggests no ioconsir senrybetween pill users and controls inth'etttalmer of recording the causes of death. We thaY rhe saBD {awd praah'niaen wM an amsAoviq a8. the dau fa this rorvev. The nudyis rupporsd by s msior psnt from he Medial Resesrnt Cnundl. The coua of the p:lot tnah and surrem wpplammtanespatditure haw: heen ran by he Sriemi6c Fwnda• tion 8wrd af t6e Roval'folleae d.lxneralPnrtiuaxn The Berrdarstefuuvacltno.uedpe be receip d(unda (or meareh imo arat rnn- tnrepliaa froe! Orpnmt~bnraroria I~d,OnlwPharmaneatisst Carporniaa S-heriaa Ueminh I46,6. D. Sesell and Ca(.ad, S>rtni P6amaeewab lad, and Jahn mmh aad Beahatul Raquews far repran rhauW braddnes,edte CR. K.,R.CG-P. As.nsbm,r Rerearab Unii, a garlow htoer Rea4xLneluer M20 OTR. uanaacss t: Kapl Cdlexe Sr GnY P'naws,na pel (irnepf,v ue xia. Rn,~ Ir6- IIH. 1. t'ewe. f4 P. ,tMAtre,. K. }hoati IL /ww. lett; it, tll. x. r.\•w,rt.Al P.b.wd.y.ItM.p, lfa. ~. xl„r. y l.lanr a': iG c. ~y, ms.y Nf- S. qenJlarin f...w~ fs. tY, S,/r+( S,Me ia Yw,aa wwntal-Aw ri A.n 1119A71,118. • 14nL v: (i,rp.lns. i lart. t. tt-0uu.r e.wry's sut.el P.~_ if esp..e r.r.l.. Ins. xx S,ua,er rab. e. aaA., n. L,x4Ons..r. h,e.xet Irf., t,]aa. e. \'sa~.K P.DWq a.Ine: R.Sw. iwe.a.lflf.IM,N. Adderldtsm Since completing this report, we have been collecting death eerufiates fora comparison of coding.So far 97deathcanigeata havebeen analysed out of 173daths among study wornan in England and V('ala (a greater number than has yet been analysed). The coding was identical to ours in 68 (70',). In 28 there were minor dif- ferenca which would not have affected our analyses. in I case a control who haddied ofcercbral hannorrhage had been included in our analyses in that dtegory, but obesity had been coded as the underlying cause of death on the cenilkate. A full account of the comparison of the ooding will be pub9ished elsewhese. 35-727 0 - 79 - G 731 AtORTAI.tTY AIiiONG WOMEN PARTICIPATIDIG IN THE OAFORD/PAAtILY PLANNING ASSOCIATION CONTRACEPTIVE STUDY M. P. VEssLY K. MCPxal3ON BMDGeT JONNtON Departn,eu of Sitcia! and CmnmueityMediarne, Unioerrity if 0:for4 8 Ko6tr Road. O+Jord OXt JQN Sornrrrary 43daths are known to have occvrred! among the 17032partieipants in the OKford/Fantily Planning Association rnntraceptiveaudy up to the end of Apri1,1977- 9 deaths from cart5oo- vascular causes have been observed among the women in the oral.ontraceptive entry group (49 681 woman-years of observation) while no such deaths have been observed among the women who entered the study while using a diaphragm or an intrauterine device (39 146woman- yean of observation). These findings are consistent with the results presented in the accompanying report from, the Royal College of General Practitionen Oral Gontra-eeption Study. latsv3uetioa IN an analynis of deaths in the Royal College ofGeneral Praaitioners Oral Contraception Study,Beral! and her collagua'found the monality-rue Gom dis-easa of the circulatory system to be nearly five times higher in takenand a-taken of "ihe pill" than in con- trol subjects. This result prompted ustoeaatmne the mortality data in the OKford/Family Planning Aasao-ation contraceptive study. Methods A denailed descripiion of ihemeshods used in our prospec- dve study has been8i- ehawhae.' 1. brid the invati8stiw (*hieh began in 1968)if in pregras an 17 family planningelimcs in diRerent parts of Englard and Scodand and I7032 , wamen are under,ob-tion. At the tinse of recruioment, all i these women weremasried lChite Biitiih subiecu, aged 23-39years, who voluntarily agreed to panicipats in theprolees. 36owce using oral contratspuves, 2S~.wete using a dia- pheagm, and19. wertusing an intrautenne devide (l.u.n),. During fallow-up, information is collened about pregnaneias. contraceptive pranices, eervini knwrt, hospieal ref<rrali and, of course, deaths. The collection ofthese data at each clinie is eootdinatedby by. research assistan[. So for, data obtained during 19 681 wornall.veand obav-vatqa in the oralarootracepti.e enuy group, 22937 in the diaphragm entry gmupandl6 209 in the t.u.n. entry group, are availabte feranalysis. Follow-up is being mai~ntained with an annuaL lapse ratefer "relevant" reasons (withdrawal cf co- opention on loss of conlaa) ef 1ae than 0•5T:. We are cno6denrthat the asanainmrnt ddeaths among nudypart(dpanu is t.omplex since in this respect our roucine follow-uprnethods are supplaneniedbythe "labdting" scMme oprrared; by the National HedthSavic< Centnl Registries in EnglLld and Savdnnd.' Subjeeo Serne characreristits of the subierts at the titme of recruia- mcna to the invenigatiun are summarised brie8!y in tabk I. In genenl, the wonsenin theoral-emsraeeptive group compraae favounbl*with the a'omen in the otherewogroups in regard to thecdistnbution of cardio•ascular-disease riskfaaon.Oae i
Page 82: azo61e00
78'. 732 TAaLa t-fOMa OMAR.6Crrala71C3 Or sualaCT3 M TN! TNRa{ CmtTR.tCapTna GaOuPs AT TIMa Or aaLaYRMaNr TO STUDY . attl6ee eri ;a.wem Uaranaini2 Pvamale apd 25-19yan Pnav,u6e ie med rlw /arn• Pocmule amkin! 15 a mn dlarnua/day Meaa Quadn'. iMezf Pacwaa6e3:_N Liway dl H7penenoim Peesdluq,ie- Stn,6a RtxVnutic fear Rhwttr beandnn.e CenleniaW annd+.aas venew J.m,bmxld'na >f 39 17 2-23 0.91 112-17 I 0•76 0•o•03or 0.12 0.17 3S 4 7 2.26 0.67 16•I6 0.04 0•66 0.26 0.31 4:30 LYA 35 34 12 2.33 030 16-07 0.30 1-06 0.32 0.16 7•P6 •RaaipevGenmlYdaai6aaiwr - } ao4M (g1%dWs 1®)' SStmd4d(d.byiidhmmnJad for e0a ud.Peny Sa Va.q.nVs notable esaepti0n w this sutement, howe.er, +rJsses to unolo- iag babitr. . R,esnlta 43 deaths mt known to have ot.eurred among the study pani6ipants up to th'a end of April 1977. In 0rw of the findings in the Royal College ofGeoeral Praeti- tioners study,.it is of interest that al1.9 deaths from cardiovasenlarcauses (including 1 d eathfrom eongenital . hean-0isease)', involve women in the oralkvntraeeptire entr,v group.It should be noted that 2af the women who died (1 from mitral-valve disease and 1frmn an accident) . had previously been °d:leted^fromthe investigation because of failore.to cooperate; we woukl, af course, have been unaware of these 2'deatti.bus fop the assist- anee of the Naiiona!'Health~Serviee Central Registries. An analysuef the deaths grouped according to mapr disease categories is prnented in tabk it. The rates given have been standardised by the indirecemethodfor age, parity, soeiall class, and smoking habits nsingg the proaed- uren described by Veaey tt aLt 77u 2-0eaths in "deleted" women have tieenonutted from.this analysis, while the I death from mngenital1 baart-disease appears under 'btherr causes". Ih view of thCsmall number of dnths, data for the diaphragm and l.u.a groups have b'eea 7Tla taactT, ocToan 1,1977 combined. For comparative pucposes "expected" Oeatb- ' rates, based on the age-speeific monality figuresfor England and Wales for 1973, are a6soo givain the table.' The datain,table lt tmderline the dilTereaoe in.ear- diovascular mortality b'etween the womenin, the oral- contraceptive entry.group.an&those in the other two gtoups; with only 7 of the 9 cardiovascular deaths taken into aewunrt in this analysis, the di&rcnceis natiiti- olly sigaificant. However, the low.oavrdd nte of murta- Ilty among partidpanus in our study should also be noted~ this findutg will be considered again in the diseua- fi0n. Tabk.ltl.summarisn some of the clinieall features of the illnessn of the 9 women who died from cardiovascu- 1as disease,.together with information about, risk fac- tors, inclYldingpi11 use. It.seems reasonable to suggest that oral contraceptives could have played an stiologieall or aggravating role in every case. DiacuasPoa The data included in this analysis are few and, consid- esedalone, not.very impressive. Viewed in eenjirnction with the findings in the Royal ~College of General Praai- tionen studyv and in . the earlier case-control studies,s•6 however, the evidence is now eondusivee that the use of oral contracep,[ivn is assoeiated:wieh an increase in the risk of ardiovaseular death. Furthermore, as suggested by. Beral,' this inaeasein risk may involve disorders other than.vetmsu thrombo:mbolism,.stroke, and mya- ardiai . infarction. Oral contraceptives have, for ea- ample, been shown to have substantial effects on cardio- vaseular dynamiawh'ieh moght prove deleterious inn women with, say, rheumatic or mngenidal hean<ia- c,,e,.-u While the only valid:monal!ity mmparisonsfor our dau are those made arirhin thestudy,tlie lowovenW death rates experienced by the panicipanu in compari- son with those generally prevailing in Englandd and Wales require some comment. First, it mutt be remen7• bered that few women with chronic disease, sither physi- cal !or mental, were admitted to the studyy since those ree- ruieed had to be well enough to be sexually aaive and to attend a familyy planning clinic. Secondly, it seema Idkdy thauth'e.women in the study have a nwree positive attitude towards health than.average_ this is suggested' TAaLa tt--6TAl/pAa0lsan WOaTALt7T-aATla Pea 109 000 w06{A1FYa6af aYCAVaa AMD CmITaAQrrtVr OaOnr AT alRiY . . Mnlnddmnvacqim ia we m adun:en u O ~ •b (nR n~,"aca) tLue afdath ' Ord x~. ~ da,Ma,e Lo-Z3f. AOlnecpt.una 214(10), 21-3(R)' 34.3 190-45! AOeieo,lnae7d'ue,den• l3l(7) 0-0(O) 77-7 300-429. Nmrhevw,ktxandiw.e 9.a(3) 0.0(0) 7-0. U0~3a Genb.waaevtvdiuue .. I 34(2) . 0•0(0p " 6:1i 1oP-999 Arridoan and ioknn 3.4(2) 6.3(7) . I16•7 O,hvw.s 11•1(5). 6-t(3) 1 7l,0 A11oua 4b7(74) ~10(17. 91.7. • aQlrnKi6-bal ReW p anddiaphHp,l - LY.a. aM,p Ygai6ba, q 7 pCfmu.ln'& !•M,Dea efdo,M ar. p.en ie pvm,bnr<. ~
Page 83: azo61e00
6II06SE0 I s 13 I u s s a ~ : 0 g g at $°f ~ gyq Y ~ l11W ~ ~it[ ~i U9F i H ~ i s H-U? Itt 11 . S 9 °I ~g~ R : ~£ ~ {Sr4R 3p aa 9 ~~~ t ~ G ~iG ~ x 0 ., o E~ C ~r ! ~~C p ~ "•'TGRiR fiR ~C IL ~ ~ ~JR1 "r I Y . e ~ n , ~ ~agt ~. ~ ~ a I £ 1 k 1 >d IL ~~ ? 1111 I IaZU
Page 84: azo61e00
I atsnsn Mwlcu fnucnsu 3 Mnr 1975 ' 9orW ~HWd, Ory,nu[xm. Wrra:,rc•m,r n. ' rhu:..,r Nrurr'Uir.uu Nqrrcrn: Keamne/ Vnf,Y (~n f:urmly.~.rlu[,.~n/,ann, 19SI • Pika. N. G,.ana .\Imnmr, K, 11., Ilnri.AJnu.W Sf.l•r.v<wrsv. amf Sunu! Ma.lI10 •a, • Micm O S l.uvm . 1010, 6; 15.'• Gvldbovn, U, and Us.da4e.1. H.. L..usAJevnal a! P.rmmrnr ..d Suad 19t4, _a, 1! 6: , sa Myv,~ 1_ 1., and 1'fioroctW f il.. JnnaA Hen•r)nxrwel: tn ~reu. n Kar: G K. 5 N:\: d K'rlsaeC. S!.1969, :, 1221, w Vtyry: M, P., rr f, l ' ne /!£p-.+,ulor}, :, 9'/2, ~t. 119. '•Srota S d Wfrn•\, Lr : 4rl 16-~. ° Sluaun},~&,~,i1i re.J1 J s d u., 73; Sdp;•I. Ma 1JM1, reey, N:, and slaura, 0'.,Currdarrnn, 19ox; 34 ST., 80 215 tr Ol,rrr, \s. F., l11lnn,n ~tf:dirn levrnel; 19 ~0: :. _!0. • Vhrn: .N~. 1~.. 7nnr-f, 1 Y iJ a'5J. • lat r-5uc,ory sa,mm~r,nnr(,in I lran Uucav Kesvurc.~, GirrWan ,.19:0, '• Mann, 1- 1d I n, W. I I W1!.' irh. H dirn/ Jev.nal, .9"lS,~L ::5. s~pune, .1., fuwr7Y.l.Lmu,y >m.rrre r. Luylv,Wund Lurel. Landdn,IL.\L.i.O., I9nJ, . rr Uenc, nonci cvmmvmnnun, 2" a7nman.W HW_,«:al. .ti Id-i.ilJ f 1 07,701.. a` Inmut, W H W.,and N ry„ l P.,F.n -' llrv,rn /~.19ha,; 193. 's Vrsrcv,.tl. t'„and Dwl Ik- NN f7. •+w/,.'cM :, 199. •• Iteaun,,,nt, l. L, .r.l.:, U,:1f nst ,rAe'W Id fI .rh t).resezonen; i9-0, 13 a91, Oral Contracepiives and Death from Myocardial u-ifarction J. L. MANN, W. H: W. INMdN' BnaY M.dul:7wn+rd. 1975. 1, =i5-1;a Summary We Iavstir,nsed 219.deaeha from myoolydisl infarction ia wamen undcr the age o(S0.Theirhisrorin were eom- pared \vith those of itrin; age.matcbed <ontrob selec[ed from Ne same genera/ procticea. T}le frequency of use of orai eonsraeeptivca during the.monsh before death wu significan[•ygreater in the group, with infarction th'an during the.corresponding month in the.coatro4 groupand tivaavenge du'ration of use was Ionger..Vo infor- taatibn oneignre[se amoY.ihg waa availAble but the proportion,of womea being trea[ed'for Fiypertension or diabetn was greareramong thosc who diod'~ch'an aong the controls. Thisdid n atmr [he overall concluaion rharth'e risk.of fatal myoenrdial infarctiun waa creaur n che womcn usiag. eral <owtracep[ivea,.particalaely, in rhe oldor age graups. lnsruduetion I- and Vessey's :eport.to th'e C-mettee on Sa(:ry oCDrugs oo.dezchs from ?ulmonary,embohsm and coronarv ~nd cerebnl hhrotnbosis ia women ui chiWbeacingage •.vas pwillnhetl in 19681 1 Ia the eases of ptulrnonaryembaiism and! cerebr:d throrssbosis a stron; rclationship wv found with thc use of oral caatracr,t.tives when these dinordem accurred in che absence u( predlsposingeonditiotss.:\fore of [he women tyho.di'cd fiom coronary', ehrornbosis in die abseace of'.prediiposing c.ondthrons had a4so bem using unl ~convaeeptiv 3 th'on would have bc¢sespeeZed from the eaperienee of [he contculgroup, but fon ctis. aondition the di17 rence was noDqute scgnshc9n[ and a. d.risiw usoeiatinn wascomsldcrcd not proved•I Luerarudin°' were: tw[.cooc!osive andwe Ih'nughr it desirabld oo undr.nke.a fusrther Invesugstion of anthsdrvm.[uyenrdi'uls int.rcuon'in 1973. Ssllcxion of C.isesTranicnipm of LI ideath. cersi5eses rc!arrng :o women umdcr the a9e o( 50 ycan whe dred in crnsland antl Vpalu 'unng t973.and i.riv_h Url.eni,v'.of Oefurd, O][I Ia1H J~ 1. MANN. nt,0„ rn.p•, lsctarcr sn Snrul and Cummumn ItceL,ne cama,rme om SaLerr.aridf.d;:;nea. I.LnGen rC.,..l rr m. 11..W: 1?L,VIN. .x...ca , nr.r.c.n„ rnnnna.,4ct,cas 0[hrn had bcmm coded to. -brtc 410 sccordins to :he etghth reviiion ofthe In[emahanalGaaufiotioo of Disesses (myocardiai i,r(ire••~:onand tytwaym rma) w ,ned (rcm dse R<Fstnr Geneni, A tatal of725w ecewed:.qilr!death's us. menunderrnesge ~n 40 yeaes, er•ery secGad daath in aho i0-14yerr aqe gsoop, ar,d evrry nbth deanh in rhe.45i9-yearsgc;roup "•ere stleaed (br:chettudy, :inng ;otal of777 nsn.i/ab/el).~ Seiemnn: ,.u tmde mme<u[wdt.ss bo[cha of certlfet•ues were recei.ed. i.uu [-,V.wees .J!)mrA' Crm(frore F:rreiwd„Sereend fw .SrWY, lnee,rurar.l,a.dl.riuded,n Prrrmr.r.alyru ~ae r'wV [rPrr~: <~e a0.N ~ aS~a. Toul ue•eiderm'~rn,:,d .rn.ca . a: I ,xesa ,ra;wa. mN..r„ar° r. li', +'6i I I In' sslan. ~ s:sss ; u(i91 ''si4roel ~„mor,..r CBcrn wene m.de [o. m[ervicw: the genml ~nmri~onen who bad ared1 (uc the paue.'ui. -n 15 casea esthrr tLe regDtered with a coaor otb¢w osp,tal;eo n oeai, wuMd tould Viot idcnnfy.tum, andnFr~ mcGCal rrord: 03 nioe wnrS nen~hadbecn lost:a an - vadaole,..' •..' tngated~ i funher 34' deaths ¢ bmusa ~hc grn n1 pnc- uuld aoo be ntcm.ew.<d. nves:Thus SR21"~) of iS'e 2T7 ata[,y enuidctsot be srmdicd, Tha -emvning =19dcaJcr were ;'vesz:ga[rC by thea . . mc•d,ol~ heidl a(ficers•: v. cae whdcn..r urtS- kr 66 casest~.. - urJudcd.1n 37 o cvidea o thb aiagnos . o( mY'~nd~a: aa Nnugh[ ro be inadequa e Dea[hs .r•er.e:nd.vdN',rv chs 5 al'9nal~ssonly.what the d,agcsus •. ssedoy nac.'opsy Rrsdngsor . cory.o(sypiiol~cb'ntpasn[oge:herw,rhefeesoerdw- gnplu¢ n:~mane : rruuoa as ddr•ncd by.N'c Wurl•iHcaiN' Jrp .' 1u23'nses(1(15";,) ev a opsy a([K.:hc dearh ~certtfimtt had hec f.c:ed o 0enlco[hcr e . ruCgcsccd [hat dnth nns actrsb omeo~nma tn ~Saiaens cascse. A Lu eteludc•d hewse rhe!r•rong sex; age, -v.yr of Jeaoh had barn.g-.ve nA the ccr_siq[e• The rc vng:153 dosr[s provrde c bavs tcn ~ISia rcpor[. In 104 o es the diagnosu was suos[an- tiatN at newpsy: Proc<dure rnr4•e,ght m . mrre •nemaloshcres roo ian :n shc i:eldwor.k.e7ur:ng [hc ~n snesuon o _ ath ~c o( sLCm~ <onllcld .tl a aucsruonnurc a-u'.dy: ~ aid oi thc penurv prrnuoncr andd anY.o ncY doc:an h;,a a.aCnc rn panen[ JnnnvI iw • nal lnns. .i S,ncc :t retnr dn Ind '~ s• btt ~n reeur,-,ed tn thc 'ocad Ne psn hr t.:d s _. ne ~.. ~<y y [r`S. ervrv;wqblll at,azcls m l ~- . n[ad by aesi";ullapsu . nulr. ~-natufg-Cl.n,ct:aorCS;g necropsy *cpons; .:nd cuuer acorC ~uPpimd 16r, .oroner. M
Page 85: azo61e00
81 • In addiuen :o proridinS:dcrailsof Ute fataln h Scncral' p•srsiuonac pra In ^tanon. ab'uucnnrainuther~s cac h mhis cict. 1:a:all'.v s..uui.:n, co¢zexo:: mu:cicl :ur ii,.c-Y<oatt aSC :+up stJ..'~a tali st c i.titM iorrac.{ c+.e bvrn~ 1'hk' yr.i cnI prutinuner locatcJ iu hu s lh. `.u'h uvuL'.luve becn ueeul''.icJ 'by^, t: c!See. r tien['z~hc:•zJ:starJnx1hicb - vallq sUil pc.1p[N'L'}' .e:H.lfn Gnq l [~wan ben ~i her orw'ar IlrnJ'v. \SAvingan[ :Ji :rsdhad,lsfenYtumdnt pwit,ion th'e dmr.or sclc<sed:thc fesrrn scd ufcssc an tmG. ro a:pru(•rsm (ur.scratzdimaricsl' sCtw. Irrc:pec[sic nf:..v[sa-er he hasl :Ji dsa n mrary' Intorma[tpu abou[ thcie patiwiu thay. were aecepted as contcoll.: Si~mdar qu•nnanna:rt wercwmpmed for ach trni! 1[ snon b<~omo appaeena howevrr, Jut :ne prnceJure v vuWne consldenbk demands on Ne 1-1 rrnaunencn`tidse, ma wu laeer. hc9 sn th [ mntre~s ••~er,.;.•GU<stcd oniv' :br rhe panrnn 'saE mYocsrdiai infaretwn whb were vnd¢r :0' ynn ofagt and one fde tnese n- i0..la'he raaes af 31.oi c•.e 51 pamrrtu nrith.h ioiarction under ;0 ;he doc,mr was aete :o hnd [inc topru.ideintormauon for. only une wenui: GunrreL parienn were socsht for all.219'eaxs invnWSatN hut';.he daupres.nlsd h'cr.e.:nclude otJy rM Cautmle Ae osaa (153):ia w•h'ichthc1:51e:Wais bad'xen ^ . mn- e;tt5ly, substant'uted• Dit`c-:nca. bcnetcn the charaac:izta' c(the ptir.ts wrthh infuRion and theeautrols we.i. reatcl b..ihec ~' tnt or b',y Fsh'eth ax probatitllN tesr w:._a [So nu ., tmalL~ 3'.~cn co vLerin;the d:x'.cnaccs in aA aSc ecu -nles {Isand I`ri aummary x+ ea4ues wcrc alc_ c_d b'vthc .rrchad. a: ~;:cninen.' Ind:.idual mateC'ss~ was no n thc st-tiari~l c.~luuuo of nhe :esui[sh'ee~e~;u\satneapanmu'm-nin ir.ms of infurrmnaa not a.~ilamlc. Thc atmber ai ',amcpu tp ~..nom U[istpp'Gcd peea in eeti of ;te <ab!. ex rnuW. Ratuln The onl centaaepsi.c pnccce ef :.Ye pauenu with iafaratdn and ch< -ts a shown .c-Ie:IL Ih .vea tz; •;ruup Ih< propotcidn. e6'ev as gno r inrhe cuiesn slmr!n.'irehon. Ccrrent uao - ;:eSned 2, ma•:nttSe tnpntII.bccre:! _rhrinthe.nse..of:he p.nenn withinfar.'unl ce us. durin-rr tl:e same .~lendar period ;oe s"s x~d pa=e~s..'•"-c :-v7orr:cn ~d cf_ usas (sedcn wSo wM Nese p-~= se - ~c but nre ~at .cvr:ene uaan) were na apprecatilydi.Se-ns:benvr_s the tw.nSruuµ•s. 2Se ddu-atioo ot use by at'r:rat wesa is shavn w table 1II. Is a11 ag<=mtsps - sprmsla.qea/c,.LIOOnxa[. 3.\++Y ;97i rhe (nesenn'. with inf rcuun had been :uing [he prefanrions a nge :ue lon r~~han Ihe tol 6vom n l~us enJ •• ::cnv::cant an [he 5:: I:rel<w aIl arGn'ruups w . reJ tllc e-.nJasd:scr:beJby:\rma ., d.T.Sc_' ,rt:.WJd~nn..aralr~<J w nc Iheeesteo;a~s~eonocn[ci ihe\onl uscd• cut the ulsc uf hnsh-.bsc o,. traGCn ;.npanaons b<itlrc~lY~O.wunuc:careco.rtm ongtne.pat , ~ , Tresd hyp<r.ensionanJ Jlabetn wrre srgn: w1YV- .~ thc p:ulcnz: with i. iarer.on :han in thr conrrols 'ratie I~'Ce~Biood presscre and b~•uJgiucv:< Idcds.•ve:c noe acvl~oic {dr :+mzoeiIh< pat'ients andiwnuels aul the p resenec'oi'.hypcrtcns~on.fnd ~:i'sres muld rhereiure be dcr.ncd only.on lha ~hasis nf anyiorm oi .-ea[m-_~t foo thesc'mnliY;ons..Ou,r ngurn re thcre re likkly roundcreacirnat< rhu.prn-lcnae of hya<rsmton.arsd dubtto in b'eJs;roups. Tteased abesity,.thyroidJiscasq rcrssiidisawe, and pre-cc!arnpnc o ord twr to - repprud appreaiably mon efaen in the pauenn wr ~th mv ul idfarciun. There wce no siSmnnnr difderenm ban.ecn [he n•+o grouSS t[ snY. aSe •.'ith :egard [n soeiai Uaas snJ'>ansY, . _ . e prevaience. af prenuture mmopause ....9 Vthe ?ac- wdhWiia r_gc g:o.p. (Ei:.St par :cas. ci al1 tha pa::c-r. vith iofarc~on and'l°;,6 of the conrrol laanmts w<re Pasttmc: nFpcsal.l, We asa [:ipd ro d<[ennine whnhar the u -• and'.m.oc'au=iil irtiareuon could be~e<plLeeJ'~y r'. 'msbenve=n thvse prepantions.and ch- uther rsk ia . srudicdJOnly.f nfthc31e nta.wr:hnnnarcnon:.r:dc 19 conussls [urrenUY usinY aral i wnsneepticn we slsu ben.- vearM'ROC hypencnsivnondiabeces. Futehrrmore,. when`patiecn had reedveJ'any, :u oi.tr - forc[her hypeaensrun .:. dibbe[es were rxsltsdeJ :rom tha nro ;rouxs,26:°aoithe :-^v pstirnn \+ith"rr.iatcnon.and 11% ,af'the controis wem cw~re.. (P<OOll.illvssos.cushae:onl covacepuves e - -: svhieh is usdepessdcnt.ai hy~per aad'Siaa sC-'nc1'~. 11, cos[!d1not idcutigx[aule 7essi[ I<cpnfaundir•; earctnf a~a:cxt: mwkidp since:e+. prcr.ae dan muld ha abcatn<d Sy'[ne Te[cod oi ;nquiry,u d, whtcn nappcd sh- of im[enogaoin8relaui.ay:. DiscuasiouT:ieaerenrLn suaqes[thac[here is aa ineze.zed isk of deatS iiom mYOCafdial IinfarcCon - womcn u'si[g JGal'COCL cep[.ces. Tlie 6rz[ repoa [p tfte Commz[[ce on SaLe[y.of Drvgs cpsce: r:rd deasSs:rom myacardialzniarecea ia[he 204~yexr age g-oua >'asna -Ord f. yrm. r',mamr. nJ,Mysraw'.al'l.ierutew (M.L) mtl CatrrMPnrlara. P.rurtqrs <nr Giw. i,a Pmnrelum o,al caev.vpa.e P..rm ~ <w Aee Grs.n <1 m.n,m tYam/', 4YM ~ W9 Tau4 I AU. hu.n,. : G•.uW. ~ AlL. Pautmf ~'. .. ` ~ rr~mnr.m .. I=x (w~%l s L:~m : Ir~q. ' io~ iru. vfra+n.w n,.< .. es., . n asss. Twl .. n Oca m -.(topa) 4tats.>e.v .. .. .., ..,I . [ s ~ Oen.~temntmu...l n'~51:a_,iGP<a9a NS- r:p vpu6am.~. ~ us~n~. ~ 11 nes~. i I[m no!s. I ~ si/lco-al. I x y -11t9i V<p- P~401(FUM'a.vr[ Katl I :I'.1 ~r. GnvN1 I AlL 'lium0 GnuvM1 ' M.1. Aumn '~ Cc,weU N vsa ' arnn t ux ~ Usn . Pn ~ a e~A~ I (u . ~ i n a6i1n ~i to rlo:n ilr!n a assn I a<.-n i e[u-m'', a tnar »:mn Tr uoo-m: a/lepol' i 1s l[ao9N l Ias Um-af e9 (ta6op I.'. I x~ e ts -r.rd'Oran'...lureefOr.zC•.nrsteM.r,ls.bfy«ereufwn4.m.lGnrN.Parinvrlvlusweew.aa.umaPrrcrnvsrearrG' m,.i. ~~a.ati:r.r T.a O••r.u._ f N. . Ap Ciwa d mu.,aw tY.val Terah 4..a Nw. .. .. ..I ss aom i s sez I i I I I s•.a~n i us:el lii.~,a. a ovns;:+.-rr 1uxn. tu~ll i c_esla~mm.r.,.,. .. .. i tta-m i no-xl ~nn~o. 1 ttoas~ i ao.os a;~in Isal i s . ~ s'. ~ t 1 n ~ : uooor i xx ;~w ~ . Tuu( I a[[aaYl• ~ tiWa. •(t~MMI' i I[lea'ull xtIJTU1 I- - I M(ItYDa I l9'(IpUW •Pmslere.Y,w tw. Wr anr NamL - ,
Page 86: azo61e00
82 C:U)ISIt'.+Ia,ILaa)cuNV'at.. nMmr.l,ra •-T., ...: ll~........... Cir. ,. !`~ nrArr : GT,y..d> ~ " 10. ,5) )1'. . el uVJU/ w tm l) ~. s! fq n. ,"11' (72b [ ~ SJ tv) o) ui (Ta~ n Ia. (%si ~ ~~r "' ) m'r) : ~ l(SV). ~ )~ .la)~'r•• ) IsaC ~ l a-n I 9(i]) i a. a"u~ ) .. ~.. 1.Ull)'. ~~ n%: ! ! Isara) I 3 1-: ' ll VI ~'~, l'~. ll9)', ~'~ x~ 11 ]I'. 1 (l.r) ~ ) (l'a) ' i(t G) „Tm.l„' , L1(lo[IVI~ ~~tlYUOeS)(IOPU)~ p00%~ :7 a9UU1 . 51(IDO% ue(IVOC) ~r9a(I6n01l \u,tbrn ' ) !' r 3 I s j e P, f ryGl I P,eaos ~! I , P•PODS I r'.1Sra;', l.<P001 P M t. . . '~ af 1111) tIS100%.~'. a1 (9P7)~ S6(IC0~01 •' (ey)) ~)C (M-)) I lal (4l'S) I 19S (M~1) (]0 ~. ~j.. i ) ~~. (]~3)' II 1:-1) 1 5 f/~9f )~ (561 '~ (ZIl 1: (I3). ' S. {1~01~ 1 (Oi) s) uwm !Y(IOOV;., Sa(IOP%i Se({OOi adll(V0f 51 .1 J001. • 1))(ID)01. i )DC([UO% ~ l-a0s. I P-a-0x ' P-Pal ~ P-baox •PJ.ri.eJbYr.ucrrea~ whieh oen:r:ed ia 19662 Since then rhe u r oforaticonra- eepeivcsh~- era~..eJ ahn+os[ Ili.. !-~dd17"hi•h>s~r+ation,e togetherssath{he.Ixtrthar thc'ddadp.llicnts.iui this sstrdy appa'arcd la hbvcbccn .usir.,onll ca torh1n;yr han the con:rol pari.•nts :•-ho ncrc cJrrent u!crs; nra.y hap:o.explsinn tha more P-ltire )i ,dimps of this ,ruSy. In 198b, .iut'cox{ra- ceprive, haJ ib.cn in n idesprnii u:c, for oni} d, ccars. \o~siSnificant.Jic'r,cncc ~n nce u,u ui hir,li-•Ies;uoesrmgen preRleacianslr.,s fiv_r•d b'e,een,patrznn u. . n rion.and eon(ro!s ,°`u ho.l heen'using cni conva«pti+~c.Saf re 1970. Thia is, oerhart t ,crising in +'ICU . u4 Ihe pmitii-: corrclation ticrscen~u.•Im~rnh . and d); risk'oi'mv,xxcdhl In,rotiun foun.d m ~• of v.,pccud aulYCr,c rc ctions.to oulco•IC[a- plcs p{J .(1 C. S- - tDrusbt fOl,w uo n a ! ........ .I I c' t os d as. and oth; tt cn I' .- f t' ot' othxrstud7 I'. th )" J Is.smadl numFcrs \\ ' res hr ili p: b' 1 cd in 1945 and 1969 -U m d, F s f ca ez- of thrombocmMhvm •. ,71p <rcne in an txl.ln5 onl conan- ccpoivcs ont '; I U y ~n, 11rc pu ul t ti' +nr.l• . h[ h' I cnced th;creatrunshaws ooser+aJ In ,hus.stud 'mcude .ha f.z (hst 53 (21i ,' ', of tS , ths hk d f (- Idd b n cst' gat dca le r. e IiF t-ti r s or (r tf r. II r callxto Ir s~ I' o c n tb s o.W h bc' -' d f 1u a on b r 1 P t~cncs s:re selc d '1' 1 pr,rcpce I es ot o. -c. . g doc:ors. Anuthcrr p•IOIc soulr e isthat th re musc hn+c bccn sume du'a[hr ~.luu.to myu:JeJi.ll imt:rrcrion inaho rrieranc ses'and agc groopsof rrhich '.: . re not nuu)icd because of incorrxtd.cath « rtifi<atwn or c.LT.irv..tnri+'_ J.ilScuhiiv: InJzcd. we arc -xte~ of cwo.sucn J,ath's-.nc u'hich" d ir ..onnn u rngam nraf cuntraacpo.. •,.hiahwssrn~til~icJr.- as anad+'crserNction too th:<Comm.alcc on Saelyuf S4Ji.i'man, nnd anuthvein a p;mtent with Ioncstari ..,cJx~n-s; ('Phcsc ay r ascs + c no[, of course, In.d.ua.d m mrr anukcsis.J Only:~bc of Ihc il dAadis inw•omcn us -i m'a po. r. trmcor'J-uh .cr•4 hons.cr, Jz:1 Iha C mi 1 n SaOClyot .\t .cdr<mcs, and thisindinm {i~^cthcr •.•,lir dr¢f.ar tha[ Jdallis a+ere incdud d only,+hcn cJi_ . +,n6rrncd~J'Cnccrup•y or suh+tan- tiatcd bvuncqyl+ n~a Jomiaul.tnd IaN'•nlory'IinJ,n_s~m:,kc the possihliihy' Ihsr I . r tnltla4. mf:ircn~,n. u~as sim; I)° drJgnoscd morc oftcn in.<,rni.~~~nni.eptls'c'u.crw.<xrrccnclr.unl9::cly. I[ is; furth'crm.ac; mr.rc.l,rv xI ar :hc frz„ ui urd ...... a.epnvc uso w s thc ,. r p '.rucnt> wdlh u. rntion.+shm~c c.e of dcarhau scunnnncd nu mssy, ) In Iho,c h i •smns tlue diagnusis dc^cn.f!,J. ~,n a•In1a~Ii:In.II/;Jb,•mrory ;imhngs.. Of.grcatimpmrtnn¢r is the fact that the control.p?p..ulArimn,' shoulJ bo rcpr,3 nt.nire oRhc cen¢rai pop,ul]tli•a uf+YOmcn oi cli'lldi:caring>ge +:ith rcgneJ lo. thvle oral contracepnve praccice.,. C:ospar. 'uIH a rceer.s n aad.s y; ho,-acer, suggestsj Ih.c th'a-currcnt crateonrracepeive use in+c our eoncrol p~pu:nrioni under 10 yeaca.uf :Ige ++as;reo[cr.than mi3hchr:e beenJ expectedJ' - and this, IL ;rucy v:'ou!d hacc tondld to reddoc the ~ aswcntion found ben•.zen 1 Yoc.~rdial infarction and orol coru:rn¢cpri+c usr. A' passibld csplanation mighr be an "orcr- nutching"pnc n n if acr:acu pncticioncrshad tcnded'to bc ciel.ler, reglolar oraleoneraccptice praeribars cr nmt.. A.n assJelationhenrcen my'owrdwl in6aretion and onll conra- ~ cep(•us "h )hJ el Jtl ' ' 'g or Izn( p3' th o,. [ofrr'- sh cthcp c Sh3of ' 1 'nlq ra f 7 t d h e he cont 1 pupml - s hcred fro h a pr ts..ovdd t d to inctude more users' than thk g ~ p.p- n of tl eaa e gc group ~I Th p ,1 -dopt d h c r n1L u red icaore ,ire 1131 ' non f' h p. nc ts 'th. if re -un thrn ofh.arnrrol .)n Aatfo-m rconcer ng '.he dnadp, r ntsI o oghtt f hopt L case not family' pi ning I" ord c opsy rcp rdco t rec da. I 111![ s; tharemrcmccunad.rcd rhar mfo:nation o oralieomtra- e pc-- u mor I IAeI to bd tir . d Chcs t' ts h for tl c:ld t 1 Of h il7- ts f nd co Fa e h'ecn . .ug orl nr ) ca hc 39 had this nfo nuebn rtto rJcd hg ncrnl p annon ds v<n h pro -d J tt c pim y 'oC fomation, for both the pancnts ,v{th t vo arJ'ro4 mtaraion and thecontrols 1. Ah cssiyal:on of'.thlis typecan provide little evider.ce of the mrehrm.rn b, +.'hicti onl contnaepeives givc rise :o a risk of Imyps-.rrdiA .inlaraio.n. Thc facr :hat the patients with in6co- tian:ud bccn u Irtqthenn longer o eroge tlnn.thecontroL patian(s nw7-su_^cst rh'.m their usc conrribuncs ou dde-anherogcnic pruc.•sszs. 11 mu,c bz nmrcd,.hu.'+~cver, that current usc of oral contraccpuu.cs ws' mnrc Imporr nt than vse at somc timc in thc I posr:_\torco.eq 9J"„ ofnce nccropsy rcports o women who~ hafl becn o I cvnanccplr•es tch r ~ I'death men- r,unaJalhrumlus nthe<oonary,anaricr,.•'hc a only55 of ilse rcparts on patients x^ho Is.sd noe bccn us,nq the prePan- ti,n~ m,.ntioned th.is (indlnc.,This diafc:cnce is s,gn Seant at' Ihc 5, level. '1'ric hntcr- tndingshould bc interpre[ed"xlth grc:n_'.uwiotn, sinee the(cchniqucs for.e.Yalnining rhesearrerics r ca4 ,nd cuutJ clcarl+ . nur tle controllcdin any .vay.in muJ' o( Ihlskind. I''hcsc nso obscrvatimns; howc cr wciglll in. uvucr uf the IYyruthesu Ihata thrombouc ocndcney nncer than, im.rcl.cdatli..u-ac:r.pl5nns, at Icast to someea[ent, tlicImcrca.cd'd,k -unbulcd4w (he use of ahese drug). „
Page 87: azo61e00
83 :4g 'Rs¢ tL ta dcrivcd [ m dslss study and fnfbrmaciun un thc .•tmsxureof the fcnu4c pupubtion or Lncland and' NValcs dmvcd fromtiie R~;isrrlr (icncral'. «pust.ima.u it, poa•itile to estimaceIldu.nsart::lisp Irom m7vcusiul'infi•cvon iti ss-omca who usronl'c.•n[ncePrires and in thsncwho do no[.In the 30-39.)'cae ate `roup dsk ycarly d-h C :c iq. u'UmCn nut using these prtp;taatians.is ati:nared to be 1 9 pcr I00000 eomparcil with 51 per 1P+OOLYt'im -moes s-ho aee.usinethem. In. the 4s7-a4-ccaaagc5roup the ~urly d~-~[ISrata arc 11~7and 54-7 per I00 COO.respectively• It secros, tbm,.that the tisk of dath fiam myocttlial inLsraion n•u in(reased ahout 2~8 dmala nsrrentusers of oral ennc::iceptives aged i0-39.ycus assdabout 4-Triaus in wnmen aCed :O-1i'.years: Ia absolute cernw be attr.bucab'.le murrali(y.was comidenbly lower.in +•vmea a;ed130-i9y-n tiwn in those aged +'0'-•i4 yean, there beireg. an excess of 3•5 deaths 9. !00 000 users yearly'.inthe fomner group and 43 per 100 000 usmyarlyin the laan gruup,.Thesc exiauta of risk are simular to the in¢eased riik of non-fon4 infarcticn est'imaed by, Mmfn rt alP' but soiLLittced toNe iheL•rpceted -+irh nution„as a num'ocr of assumptions h- oeceesar.lr had to be ttudt in theiraicuWnion and she maegin of e:rror is.li'r-.r:y, tn befairly wide• lse d nn•essSdess, <onsider themto Se.helpful ia Providi:~ racde essisaate of risk of.dcsh frosn.myoorsliSl iatuaion.:n women sslrrendy using onl. c patncepti vca. ~fbere ]~u arctoo fra ro iu.rii.saYine m+~rn tor.asea ated35-e9 Trro, m~!wm ~nc .aw monu4q rsrr in 1973 (is wa and mwwm embleed .•as ]9-4'p¢ :0i 00P. . Mai: ten=- c° oi Labaur J..6L BEAZLEY, I. BANOVIC, NI:S..FELD Eervd,'Mdiral7ns.er, t9•n. z, z4e-25o Sumtnury In 160 sromcn large but variable amounta of.lsrtnvenous oiyror.ss were neerded w iaduce labourr wirhin a t•eason- ah4c time inseml'to S csn cervi-l dilasadon. 2herenfter 7'mU ofoxpsodarmin would sna:h.taia pregress. Siisee Iarser maintenanec dosu may nuae obsrertnc problenu it is roeammeaded thaca maiatenanceregimea should be med once labour has progressed to this scage. ENRLS1r MFPlGAL lOL•ANAL 3 fL1y I9i5 We arc gratef'ul re the m..linl. otfittn on the cnmm tfwno tarricJ uurUe inacrviun:, to ibe my ,.'onors wharrac. s_ th_iv roPm.'idd.us srith muonutwn, and to i,r Duil anJP 1'rol.snr •tl. 1'. Vesuy.n,r and erK - t: Dr. .sL :. Adclet:in an1 Ssr. J~ G. G.r.arJ. ofeSe O:iiee~.+r^~,+uuunn C.-u.u:.s anS 5¢r:.rys. l:~nd/y aman:;eJ ~Cr us to rceer•e.lhe cecti6wrrs .\1i.s A: ReoaalJ,~.\trs.:•LSpcllman. and .11r.. G. ~tl•ad'~ prv.'.d.al in.~leabl< secrc evl hdp, and \ln. %L Tauroqcol aed.\Ir- PGo W:::~ aa.l Ivs s•_.T of ih'e Garavd 14s_9rta1 i i•.^~[h 2.tat.4enty, . he:Qcd a-nh theana:vsrs atlbu resulrs.. J:•\L.ws.auppond byan.. LCIi Pcllo+sh,p andby, Nc aeaG. Johts Adama ,\leconal Cdl-FJp dunnS the pcnod:of th:s asud7: We arc gr-tc:ul to Sii E-.e Sac:••:a and th'e C snsrtec on 5afcsy of :t•,kdieiocs &r permi.ven to repprt these dau. Requesu for ra:peims sh'ould be addressed to: Dr. t7;1i W. Innan,. Comminceen Salcep ai :\Icdicnn, Fiaabur7Square House, 33:77A Finsbury'. Squore. Leodon EC2a.' 1PP. Refereneea ' tnna, m_ 5(. 6..aad Vna<.: 1L P.. addrASrrd~mliJ.x ~ 1966:3, 19T. ' Vc.acr, .'.I. P., aedDoll• ,., T~::u Sr.,Gia! Ja~.u~nd: l•rG9: ` 63 1.. • lnsue, R': !1. \V.,.r v., -r.:::n. a::1:.c: Jum-:, I i70, :; 203. ' Werld H- Orranna_:w. il'iu.nMGro:,een ilianr Juemr - U':H.U.. IteS:enu' G::i<erCn Er.repc.,(Yippwaam, 197I. a ~L1er~r..u.f~0: S..J:r-•.r-rr. iv.J, _a, 15.. •.Lmrtat;e: PSt+rvnca: u:.trlira( Rrxmri,. ). 363, Os:ced: -ala¢val Saeaa.ur; 1971:. Vrsr-..LP:,aadL,nu,.-:H.~,7~• - foearanr,wc,,,nawrn a( tAr Enn r, t9'i, aJ; ` r Vnu.~, SL P.~and~ Da'w it. S:va.:.ltr'._( )mm•cl,1969. 2. 199:. ,•9'one.:•L. t'e.-r.wrL'smaeue:~ritxi. 19"rl. 's Al.m.). L,.e d: inurir.ar:iral ja,v,u7: 1075. ]..U. IialI attessrion hu been p~id to the maiateaaae of'.abotu a1erssseh icqieeen• T+e.hare stvdied rhe maderap:ocess oi laocetsiimnilcion to see wheth'or a stage is rcr'd bey®d u•hicb oaytocn tequiretamss ue grotly redsrcesL Patieass and JSetL'oda Labour was [ndn¢¢d 'm63 rmsluPames and.77;rasdp,4rns -a bet..eee 1bc and :94'dan of cesnzria T.:e indi-as for.iodaetion are shonn- in nble 1: :1L Ne !e:u.--f preseancd E.¢ePr for vemen adaddd tienme or possible cPbbly:h oisytvporrion ®aesvdve paaeass - trared.'. I,.trqdnetion I'ffiuetion-0<tivery inter.aLs much twer 12heurs are rb loagec atuptatile. Smail qsunritin.oi onytotin (Ssatoanon).uafusod iaeavmnurl), forpmlongcd period indecc latiour in many sasmea who are near teraL T.a easure inducvon in all Patients airhia a rnsonable nsne,.ho.vever, meth'nda stach n axaocin sissazon' and an tacma-r•ag rata of infusion hare b'een dc+eieped, a.i.er+a. Droarws.s ar'.asu.rr:u ..d cr.am4op. u•.rPm tussuX J. ]I. elC.\T.t:EY; xo.. r.a.eGa. M1vfemr L S.{\i9t'a:. .,.a., it_a sL S. }._'11s, _.rssi, Rr~,vwar ruu t-,,.Pe,ew Jn. t.t.me. a Laas Hvtw• enww.~.•duCUtf+•+eupnrreo~l. Rdane.., en ~~ ! ro0 r... I ft--. i--- N ~ I 3 a , o Iw~a.rewv .•zrto® rmfsn:ve •.a.>nesedi~s,i~'sety s~e :e,.. ansn,otetmy in alll ;an s..\t :Le nme ef ~aranqtom.+ Se ll.iyp srae.. Mas esnmatNa.:.aniietp rn me aav ode IL T.W..e rrw^_i -seed for nscessang osyswn s rdr+n in table 1JL .1fz,= str~~ . s. ~ a
Page 88: azo61e00
84 PAPERS AND ORIGINALS Oral contraceptive use in older women and fatal myocardial infarction J LMANN, WH W IN.4fAN, MARGARET THOROGOOD B-irA Med'uf!T_el. 1976i 2. MS-ae] Summary A previous study of women who had died from myo- cardial infaretion. and of a control group ofwomen match'ed'd with them for age suggested  fivefold in- .crease in the risk of death from myocardial iafaretion among tssers" of otal contraceptives aged 40-44 yean <ompared with womennot using such prepantions Only asmall proportion of womra in she infarction and control groups had used oral coarraceptlves, how r, so the margin of error was wide. We therefore investigated a further 54 women in this age group who died! from myoeardia] infarctionand e mpared their oral contra- ceptive hiseories with those of age-masohed, Ilving contro4s. Combination of the findings from she present investigation wirh the previous results have enabled a revised estimate of a threefold increase in risk to be made. Although this risk estimate is similar to that pro- ouslyy shown for ayoungen agce group, the total' mor- nlityy sttributable to complicadons associated with the use of aral'concnceptives remained considerably greaterr among women over the age of 40. Introduction Previously' we showed an ineseased risk of developing fatal myoardial infaraion among wumen using onl mntnceptives. The dava suggested!saappmamaeely.fivefold inerease in the risk of death amoogusers.of the prepar.tions aged 40-44'.yeus O.p.rtm.nr of Sadal d Commataip Medieiey Uoiveeriqy nf Oerord.0*fcrd OXI3QN J I.NA.YN, nr„ nm, lemuee MARGARET.THOROGOOD, ssC, naeareb enirunt. CnmmJtte. en S.fery.of Ncdld,s~ ,Loodae ECiA IPP. w H ~ IN.~SAN:.+ao, wew. prindpal meeid ee5aer cnmpared wieh - nearly threefold intaease in rhe risk to younger women. The confidence limits of the.rifk euimaces,.based an a sampk of deaths in sheage group.4044years, were wide,, however, sinoe few.vomen of this.age were.usiog oral contn- ceptives. We therefore studied those dnths in this age group net investigaeed providusly to provide a moee reliable esdmac o6eisk among older women. Patienta and methods A detailed aeroune of'the method. ueed for seleeting wamm who h.c died from myoordisl Iiofarnion and mntrol patimo, tueethee with the In.estig.uuon ptaedure, was given.previously1 In brief,nwexipo were- obuined Cran the Reg-unar. Gener.al of all death tatifioees of women undlr thee age of 50 who . b.d died:in ~Engiand .M Wain during 1973„aad which had beeneodad ro rubrie 410 in tbo eiehth revision of the lnrernutimmf Clnmfirnaoe oJDt}raus (myoevdial infamim and syoonymous terms). All de.ths among eomen aeed under.40, e.ery semnd death.in the 40-44-yearage group, and e.eryy durdd dearh in tbe 45-49~year age gtoup were- sderted forr the initial iuvetigation.The 92 deaths in the age grouP 40;44yera oet in+estigaeed in that study fo- the basis of thia repore (D.n aee mt girm hceforwemen aged 45-49 years, rioce em few ioGrction snd oponolp>.omn in this gronp, h'sd been usimg onl oon- mrtptivea to mable any defntice wexlusioos to be made.) Of tb'e 92deaths,.24 mWd not be iorestigrted beouse the relevant ntediol recmds muld one be found (7) or beeause general pneinoneess wld oor be rneed (l1) oe were onable ro help (6).'.'[he temaning 60 wrre in.eseisated by madicai'Seldlasfrersemployed by she Com- mince on Safery of Mldicines_ lofontution on drug use waa obtained ehlMy from genera1 pncritionen, whoo were alao asked to pee.ide omtrol infomnt'no by selecting at rmdomfrore their 51es a woman who truerhed each favl eae for age and maritai srsm. L)mda wero induded in the5nal amiysia od9'whm the diagnmis of fid myo- ardial inNrrnoa wo subsonoaeed byy nec.opsy: 5ndimP or' by. a hutery nf ehmt pain togerherwith electrenrdiognphfe or enrymatie mu8rm.tion u defused by, the World: Health Orgaois.don.' !n 14 asn endence for the d"vgnesis was mnadercd to be imddquate: Tbe findinp fer the remaimng 54 dratht md contrnll are given Edow.. In four ose rhe sener.l wsctitioaer did not prmride mond din Togive mure.reliable risk enimates she rrsulo were mmbimed'with iofoernutim fro.a th'e raelier srudy' on the 52 death6 in thta age ssvop (40-44 ysata): Tbe sigrtifieaneeof dil6erenem in chsncterinio between the pasieao .ith infansion and the mntmla waa asae.ed by a autcbed
Page 89: azo61e00
85 4i6. sltsrlsst Miotru.7omtwst. 21 Ausasr 1976 Taau 1--o..F w.esprfw w•o;a..r"e .r..w sd.4-(lye.ra .Lsdidyrsw a y.e.e,ef :drmar (Mf) d rwMa o•al eaeuaaenir. - Cvrtn. wen (.xA elender Gwnanerod r manMw.~U ~Mfsn GuJ a.,.s rame w,).. , a~a~~ . ~GUM~ ~ p •, A TnW N« anere. . Cww.a•ae vlr~mnaWaww.a.ew~ .. m Tiamr YnatlWri' fsi.and aater NoIS)afWriens.inlMf~ Ne(X).afamwla rNn.('.).fw.i.'.a.+.L.Mrl NnlWdernreY M (04» i 10 (I4fln a Il~q laW)Y I •YIaaYl. f(ln<) !(t0al}IO(aDO! . . ~~ i.. Ta (T)M. ta (1Ta) la1N)}ral]Mi). Y (N-r): I I T(N)1 lii • lap,lallfTli ~ s4uaaY)', ! sa(1:OnM ~. lano00') ~IOtuWO). ~ a ~ I 2 I a II )~br•'V"am r I rf•4M;T<OOS pain merhod,•'andmnfidmee limio of:the relasive risk orimates established by using tbe methrdof Miertinen.' Standardisad relative-risk. estimates and summary r' hloa .ere elcu)atcd by u esan maximum-likdlhood smethod.•' R<sults Table I shows Creonl enntracepdve praedm of.the54patieno with: . imfaraion and the conr.mis., The ditlfermae in rrms uae between the evogruups wasIbs than that repurted previwuly and by nelfis na rtatisciully signffione For the mmbined series of 106m, aged 40-44 years lio had died Gom mryoardial infarerion iu England hnd Walea ib. )973 the use of oral cvnsnceptives in tha mnnrh'~ before deatL wa signifi(andy gruter than in the mntrol scr (P <0.05). ; PLe eseimaee of riskk for mntncrpuve usen aompared with' that for non-taas was 2.8 teI(95°r;: rnnfdence limte: 1.2 ro 7~.2): Evm witE his urger nlum)xr tbe mnfidence.lin» are.sti11 wide b'muse theuse oGaril.emtracepdvea in rhGs age group is uncommoa mhm theee dfuare mmidered inmnjunRioo.whh rhe Regiswar. 6eperalY statistics the yearly deatA nte from myoord4sl in( ion amnng womenaged 40-64 ycus wbo were nm wing these preparations uesrimatrdrm be 12/IOOOOO cempared witL ]2}tOD000 among ee who w.ere using m=-:Sintx aboue 95 000 nvwnea aged 40-44 . ynn in Englmd and Wales were estimated to have used unl mnen, . eepvlves during 1973, romx20 de.du frorn myonrdiai infarctiou in this age group could probYbly,h'av< bem.aeaibutedto Otal.mntn- cptive- Infara' wn and eontrol patients were mnidered~ro.be tiypertmsiv<or di.bnie if'theyhad recsi.ed [reatmmtfor eixher of tAeae.mn-didorrr. In tha combined seriea 23y;. of ehe iM retioo W rimts andJ:. of the.rnnnob h,d'bem tmted for.bypenmsion, and 10?`% oL rheformer and noneof the Ilner fordiabem..l3e risk.ofmyonrdial infarrsionassocfued witk the u oforai cnnrrauptives was noc appretiably'ahered after .Ibwing for the eReR of there two6etoes,'~ she risk esrimate. tieing reduced from 2-g to Z7... Disousafou These findings.again show tfuc.the risk.offftai myonvdiat infaraion is inapsed;among wottxn using ural comneeptives Zd suggest that this onnot be eaplained Dyy an association tween th'e:use of:the prepussions and the two riak.6crun for myoordW Infarttion (hypenension and d'utietes)that rv.c:were ablc to inalude in this iovesuigation. Eycombilning the resulrsof ihisand our previous study,we estinvte the risk of faoalimyocardial infaraimnamongwomen aged 40-44 yean using oral <ontraceptives te be about three times aa great as among .vomen in the sam< age group no2 using the p?eparations. A nearly fivefold iucrease in riskhad been apparent in this.age group whetl-the estilnates were Wsed on a srnaller sampie of patients:' The wide:margin of sampl4ng esTOn was strassed in our<arlier repor: and theinvescigation of apprecubly mere deaths pronded tlieoppurtrlnidy.to caliudate reliatlle risk esrimaeea. The p;resenn risk eatimarefor this oldengroup of enl contnoeptive usen is virtually identical ivnth, that culculared previuusly in the yoongesr age group (30-39 years)..721e aaributable monaliry, however, if muahgreater in the older gcoup (20 deaehs per100 000 users yearly inth's 40-44- year age group and 3~5 deaths per 100 000 uscrs yearly.in women aged 30-39 years): Thiresulisafaarudyof non-fan4 myo(aedial infaredon• suggest that the present revise:d:estirtute may, be too kna. In that investigation the riskk of myoardDal infarctiesis was found to be 5-7 times greater in women aged'4d-44yeats who were using oral mnanceptiva tlunin women who had:never used such prepantions• T)sis ra6culasion was based on 40 women with myocardia iinfaretionl of whom 11 were using ural con¢a- cepuves: Aitemprs ba.ve been made to draw up a balance sheer.of advamages and disadvantages of oral contraceptives.. In.one such smdy' it was cunduded that the mornliry assoriated with the use of onal contraceptives wu of th'e same magnitude u the mord1iry from cnmpiintions ofunplanned'peegnaaQes whm Iess:e95dalt <ontraceptive methods had been lued-'I3e calduL- tions were made as a tidne when myonrdial i¢farc[ion was not . recognised as a complication of.ora6 tnneracepriWe use. Saveral i ,erious ad•erse reactions to oni mntracrpnives have baenestablishsd} but thmmboembolisml and myolzrdial infarrtinn" probably aamunt for mosr fatal adven< reaaions. It tlierefore . xenn timely to attempra comparison of deaths resulting from the use of oral contncepalves with dnths from the mmplin- tidns of unwanted pregnanoes that might be expected amoog users of lea effeaive methods of'mntnception-for erample, tE<diaphngm. 1l)e mortaliryrstirnates givra in tablo II for the compllosiorls cf oral kontnceptive use are hased on the results of ahis and an eulier study.' The age:breakdown.wu chosen so tha mm- panbl.e statiieia from.the difterenostudies muld'be shown. It sbould be remembered th'ar the ddu for pulmonary,and cerebral thmm~boembolism were mllectW~at atitne when onal~'rnntra- ceptivescontaining rnore than 50 µg of oesrrogea were widely .vct 1-Y.a•fy .vmh'qq. Jrwn nyua.dial i~arctien, pdnwary rw/ rnekd'rMe,.hnMdii.•, vd Tq~•mv.y~ d.liw,y, n.d'W p-pmuu,. bwptvnwd prermrnti') n•wV w..e. .n tA. ta.ora.rwvr vriw wdrperauDriwr w didyA.or.t I ly^rlr ~ry (o« loU -) a.pouGd~enam.(rr.n):.II 10.N, )Lw r e,.., r r m i~. u,.., , f.3 Mjd+.dWinfarn:an. ltal h~v.T~ d~4 rrtabNA,mmbo.mlWi,m rl I ) ~T.anda-o-l os T«at I a3 i laL v..,..rwM..e. . Pr.s.unr,. Aeli+er: uW onaNe~wa . I I ~1 ~ f9 Yacns mnruliry atrribvu0k w eN tanWU a,e okni,.N w R"a , rul.re nt< or lo•:'. .vinnt aaN
Page 90: azo61e00
86 tssed. The figures for myotardial infaretion refer to the period wh:ea -1 conmcepeives cvntaining less than 50 ug of oestrogen were more oftenn in use. The present mortaliry fiosn pulmonary and cermbral thromboembolism attributable to oral contn- ctptives may therefore besligh9ly lower than th'ac given in the table.. We assumed that oral mnenaeptives are 99°;:. effecaiveas a taethod of contraception and that ttie:failurente among users of the diaphngmis l0°;.a year for women aged 20-34 years and 5°o a year for women aged 35-44 yean. There is an excess of 141deaths per 100000'in rhe younger age.group of womea. using oral contraceptives and of 9-5 per 100 000 among women aged 35-44'.yeam. The excess namb'er.of deaths:acsributablo to compliations of oral contnceptir'e use wouddh'are been greaeerifthe 40-44.year age gro.up had been considered, but Ielevanr data for pultnonary.andcerebralthromboembolism' were nocavailablG. Tliese- simple calculations did aot include several other aspects that may' be rekvaeu. Surgery for gall-bladderdisease- stuue common among usen. of:onlf comnceprivesr'•-may. be asooiatM with a signifionc morrtaliry,,as may cerebral liaemor- rhage"'and other,Jess cornmon adverse reactions.1eFurther- more, widespread use of lcgah abortion of unwanted pregnancies arssong uu rs ofth'e.diaphragm might reduce the number of, dao.hs in this group. These observations, and the.fact that a pessimistic failure nce for the diaphragmwas assurned,'s suggest thac rhe estimare of the excess deaNa amungg oral lronrn- cepuive usera may have been a conservative nne.. In England and Waksth'e iua+scd risk of:death from chrombocmboliim co usexaof oral contraceptives aged:20-34 yeasamaystill be less. than half that of death from rcad-tnlHe.acnideats, but for those in the 35-44-year age group the increased tisk.maybe doublr that from ~road-tnffic acoldrnts.!' Whether this is an aceeptablee risk for so effective a method of, mntneeption remains to bee decided: The risk escimates for death from rhromboembolism in -table II were made before the ihtroduaionof oral mntnccptivess contaihing.less than50ug ofoestrogen. Si~nce a dose-response reWtiomhip has been ahownbetweeu higher oesrrogen doses and deaths from , . cbromboembolism," probabhy oral cootn- cepeives.cnnraining smaller amoun7s of oestrogenmay be asso- aued wsch aa appredably.reduccdrisk The.presuibingof alreznative methods of coacnceprion for wotnen u riikof mypnrdia6 infaretion for orher re soos-for example, hypar- cholesterolaemia,, heavy, cigarette: smoking, hypertension, and diabctes-ld further reduce the number of deaths associated with onl cnneneeptive use.' We thank' th'e medital oQlcers nn the eommittee's snff who carried outrthe interviews; the many doaors who gave their t'u'ne mpro.;de us witAinfonsmtion; Sir Richard Doll!and Professor M. PVessey.for.. adrice; Dr A M Atleisrrin.md Mt.f GGerru7, of the OlRce ofPopWation Cemuses and Svrveys, who arrangedfor us .ipn of the death certifiates; and Mrs A Rrnauld,m t Mrs M! Spellnw, and'~Mrs A Read fon seuesuia4 help. Refereaces "Mane, I lL rnd Inmen,:W H W, B.in.A'MrJiaaf]w,nW; 1975, 1:245. ' Wond'.Holxh Orpnvauon, ipe,.b'rq Grena aw lra/ar,rue Nnrr.O- RTirurr. CepmMgm„WHO'Retional Orffise for Ewmpe,.1971. ' Pile. M C, .nd Mdrtow, R H, B.,n,A Jnwf'aJ P•nvnriw aM Sesid, . MnGm.e, 1970, 24, 4z . ' Mieronec, 0 S, Ammean. Ja.nN ~LEyidewid~pr, 1974, laa, 515. ' Mietdnsn; O 5, Bio..mo,..19]0, 21, 75 . ' Mann, 1 1, rr W; Brin'~i.1 M.dienl Je"rnuf, 1979; 1.391. ' Pma, D M; md 5'.Ier, G I M. Bnru6 M.~ul BWI.n,,, 1970; IE, 2d'. ' faeux, W H W, and Vmser, M P, B~uiA. Mrd,'rd7wmnd; :96a, 2. 193. , ' tbrmn CuUetrond.e Drua 5or.eillance.Prntrsm. La,wt, .1973, 1; 1399.. 1e Sw0e7, P D, er d, Ane.iru Jnvrnd 4 EP~uoldaY, l9'/5,: 10'{197. 'r CsWbo,nti.e Gnwp fw nc~ Stud7 00 Sueke u Yowg Womee. Javnafef rAe A.enrew Mdr'ml Auxration, 1975, 9l, 718. 'r SaWl, D M. NIVEnrfwd Jn..wf.oJ M:&ri,v, 1976,: e94' E19. " Vare7, M GP, ued WiaDris, P. Canrrau9rien, 1974, 9, 15.. sr Resiuuv enenl,: S<auiriml R.n._ nJ &qIn.d ad Wufn Jm 197D, Pus i, . Innday HMSO, 1975. "(nmen, W H m, m ef, BnnlhMe6la1 Jaw,ul11970, . S, ]p3. Comparison of propranolol, metoprolol, and acebutolol on insulin-induced hypoglycaemia RAY.r10ND J, NEWMi4N, - Bnrir.l Meb'eaf Jwe„oll 1976. ; M7iN Summary Metoprolol and acebutolol,, tvroy snpposedly. wrdio- seleccive beta-adlenergi'c r ceptor blocking agents, were tested In liealthyvoltsnteers.against.propranolol; a non- selecrive drug, for theireffect on blood glucose levels during iusulin-induced hlypuglye.emia..lhere w s no significant difference b'etweea propraaolol' and meto- proloi, whfch both po9entia9ed che initial hypogly<semic action of the insulin and delayedthe return to n mo- glyeaemia. Aee6utolol, e en though potentiatiag the initi:l Isypogly<lemia, did not possess a significanr delaying effect, A sirnilar trial shouid be andernken in Se /amn'1 Hoepiral,i Leede 9 RAYMOND I NEWMAH: aae,,u,hm.ee phlt•'ai.v.(oor devwnsrmme, Depvm,enu of Aouom7. f-Rds.lldiemsir9 S<hool of Metlineel diabetics to determine with certainty the safety of such diugs in diabetes mellicru. Introductfoa One of th'e body's most important resppnses totiypoglynemia is the release of ad-enalihe from rhe adrenaf ineduL'a. This hor- mooe stimulates glycogenoi7nis, and the subseq9teoorelease of glucose tands to reswre rhe blood glucose Ievd'eowardi:oorrnal.s The beu-adsenergic tecepror blocking agents available are noo-seleRive intheir action and are thougbtm inhibit glyeo- genolysis. These dtwgs would theseforecbe expected to possess ahypoglycaemic elfeet, and iadeed hypoglyca mia precipitatedd byproPraaoiolhas beenreported.s-' After such.reports irwas suggesred that drugs like propr+nolol sh'culd: be med with extreme aution in ihsulln-treaced diiabetia and fan patients prone to hypoglycaemial'-s The evidence pub'lished in the pasr dende on the etTect'of beta-adrencrgic blockade on blood glumse levels bas nor, however, been unanimous in suppoeting this recotnmendation. - % r
Page 91: azo61e00
. ' TGIT.SIr.SCg.p1CALJOUILYµjMAY1M 87 PAPERS AND ORIGINALS Myocardial infarction in Young Women with Special Reference to 0ral Contraceptive Practice J. L:MANN;~eM. P. VESSEY, MARGARET THOROGOOD, SinRICHARD DOLL 8r:ruA MedieaM1JwrnaA, 1975, 3, 241-245 Sumrnary SLCty-tF.rceromcn discharged' from hospital witlti a diagnosis of myocardissl infarccion and 189 control patients.w e studied. All were under 45 years of age at thee time of adrniisfoa. Current oraJcontraeeptive use,, heavy cigarette smoking, treated hypertension and diabetes, pre-ectamptie toscaemia, and obesity .vere LIl reported by, and' type I] hyyorlipoproteinaemia .vas found more often ia, patients with myocardial lnfarctiea than their cuntarols. The relntionship between myorardial infa retioa and onl'eoneraceptives could:not be ezplafned iis terms of an associatlonbetween the use of these prepantions and the otherfaetors. Thce combined e6eec of the risk factors wuclenrly slaerg7stic. Iatroduetfon Oral tontraeeptives aeeapt to produce arterial as weil as veaous. throm6osis:l • Evidence aboua their.role in mryotandial iafuction ii, asowever, conllicring.' Myotardial imfarction is uncommon in young womcn,` and little relevant imformationmay'be expected &omany.of thrl%rge prospcaive studi'esof this disease nowin progress.' • We aherefore undertook a reuospeaive studyof women under 45yesn of age who s•.uvived a tnyonrdiali infarction as th'is mcthod: oft¢ed the possibilityof'obtaini¢g information direttfrom enough patients to enable a dtfinite condusinn. m be reach'ed. Severai auhet factorsare known to rause the.diseu4 or aresuspected of doing so, and we snudJed them zlio to see whethec oral ~cootraeeptivess acceJ independently or synergisticadly: Patients and Methods Adaxried:women under 45 ynn ofage who had been wted fmr myotardial infanetion during 196g-72 were idenoided in she dischargee reconds of hospitals in two of the 15 hospinl regions of Englacd and Wies.. A/l hospsals whica treued gcnenl mediolmd s•_-ya! r- icnn and S.d.amre than 300 beds rere inciudad iotbe study previded shht a satisf.ecory, die,gnosuo inden had been naiuaieed Altogether.84..paderus were idennifiied from.,he.recerds of 24'bos- pitals. Padenu were indudei only' when the diagnosis easiafied the vir.eria of the Worid Hcalrh Organinrion.' Si.rteen of,Le pauenu tddied ia h'oapital and'fin•e.had died.subsequanrli. Only limiied enaauon cvuld be nbtaioed about these 21 pauents md'tbty' were= therdore ezduded(rom the srudy. The ages of,he remaidwg.63 padenu ranged from 25 ,o 44 years, v.ith. a nuaa oU;01 yean-. Fifty eclisaified accarding :o the Wonld'Health Orga,uravon as having had •"definire m5-dief infareiion,°and 13 as h'avirigg bSd •'probablec myoardiall iofarcrion."' \mehad. beenaddnit,ted on spre.ious ocnsion fon defu+ite oe suspecred infarcdorL Thsee other pssienn.were taka asmnvols for euh parient w•ieh irdfaec.i"n. Tliey were selceted ar rarsdom troo women wbo had been diaebl..ged atrv rrratment for «rrarnaaure m<di W or surgical eondiuorn or .fter «rnin dcccive surgiol Ipruced,ures,, and matchcd a~s patient aith infucrioo ie ,espeet of ovcinl sutus, fiveyear.age group, and year of admission. The condihioos from m•hieh rbe mnerol palieots suHered arc luted io ubldli r..sta s-M.dicd mdSvniref Ca.dlriov in.0ennoL Gcmy. Ns; af Gc,de AaurmaAibl kvedluus .. 1,s . 1e ~ ~ t in[cctieu peoeuti. .. .. . s umvmlrr or o=rord I. 1. l.ttV V, w:n.. r,aw• Lcaurrr inSedal rmdC.immuni:rl l•nedlNnr M. P. V FSSl1', ,n,..,:, ,n ~n. • Pr~rcnor rf Sencl ved. Connniunny' Mcditine MAR'GARiTTHORO0Of7D, Research A'>.iscant , SiRRICHAiiU DOLL• I-, r.s.4r.; r...x, iteg,ur Pre.rnsor af ~tcdioue I
Page 92: azo61e00
88 :.A2 Permisaion vns obeained frorlL huspilal consulunua asW general pncutionen to idte.rvicse the patimsa iP-th'eir h'emn. All thc patiients with'h infaaclio¢ and 174 of tlse 189.controlIpaticnte (A21 traced- Those:who hadmoved:to.imcccs,ihle areas or ,vho.tefu.cd ro s,ee Ihe mmviewer wece.sent a pm,[al.~unaionuire.: In afew mu ees th . ge 1 pn r tA ugh that ta paa' [ I Id nor be t r.ie .ed l4h h, oc rred and Isoh n a p t nt cfused tobeuite <d or lo. rnpl tpost 1 Q4est nna-dlegcnerale practit oner g.ve the d.ni.red Iry rmorion The. umbcrs of pa6;ma. snveu.ipted by eauh of these mecbods xe shown~ in Iable 11. I11trfISFI MEDIGL JOWRNAL 3 MAY 1975 Tsau sr~:tncuac.•./in:. Xaiu. N, MyouNid'Jn/a.etiaw ed Gnrof. Petlrrx. V kkd E.. xndkn . 11 d. ls- 3a3 e I Nw(~)d1N-(laar M-LPaurnw I tonrola. 13 axwa LrretmrMr ..11 9.lIS:)) I (3s1 TaeYIssllooCl i IsT IlOOOP ' Tess Iar nnwr umd asnma cn smdi.a auprin: Y(I) .]0<31 P<Dm1- Tut-a u Mnbd N. Darv CaRrrrivn. Jrew IpAa)a Ssrin N.6'T M)xaNr'af JN .enow (MJ.) m,d ) ]i cmnvl Par:mte Inler.ia...:,h p.uant Pesrd v~wnn.,r. mnPlcwa - heycm~ea w~vcoumrr~Ltvd. \a.(;aar N4(;)af ALI-P.uenu Omvd, ss (se l) 4 (0Y1 6 (Ri) l3~'.(TfU)i 23 ' (13-f) , 1T (eY) Taul I 61(ImL)I_ _I)r (IIpO) The ; and queationn.im s e. designed, toobuin iotormarion on the paienn' medidal,l obstetdq,social,.family, and nccpuve h'iscorin before th'e relevanaadmission to hospital. Physied esaminadon waf noc eaaried ou6, but fasrint bleod samplo c ebtainedd for lipid'analysis whenn<r pouible In all ufn [h'e sample was obtained mme than six'. monrh'o a6ter the inSrcTioo had occvared The sutisrical IsigniFiemee of the. «sulo wu rn[ed, when appr.op- riate, by [he method of Pike and 4forrows for the anaiysis of indivi- dually', matched ose-eentrol studics. When the number er'exppsed patients was Iess.than fitro Fisher's- tesr waa.used isuread. Statdsrdi:ed'~relari.e-riih 6rinlatea and acmmary Zt val5ea N-erem cahWlred by W esac lnozrisum-likeli'hoaC method. • ponding mntrol ptimts, but the numtierss were small and Ih'e diMerence muld ca.il¢.have bem due tochanee.(8ggoof,he patimsss with infafction and 76°0 of the eouiro)s had used these peepansions for ovcr 12 momhs). • Gsarett<. unokidgg way reporred more often by.the parienm with infarction than by the controls, whichwas due airisost mtirely ,m Me indufion of a largcex.cesa of rnodente and heasy sntoken ((able IV), In cumparison.wiuh nm-smokets, th'e relenve risk increased fvan 1.2 to 1 in:wommsmnk'mg tewerthan 15 cigareeres a day.to 4.1iro 1 smoking 15 to.24 a day' and 113 to I'ih womensmoking 2Sorm morea day,. Tables V and;VIgive rhe nwnb'en.of patients who hadbem treated fon cettain medipl condie,ons which meght i-some way hhve predispored to the developmenr of the disease. .\tomof thee patimtswiN infarction.than the mmrols had becn.treated forhypertmsion,r diahetn,, pce-eclatnpsia, aadohesiry: Blood pTessure. and" bldoA glucose w<re not mcasuved in iodiridbal puients, and oua figures ar.e likely ~to underestimate.. the prevalmee of hypee:msion and'diabetes in both grvups, Foureeen of . the patients witb infarczion (233 e)!g+ve. adnr hisnory of seeking advice aeeause of obe:isry compared with 17 (9'9. o) of the cnncrol ~palients. a'r'eighc and:heighe at rhe time of,adtnission, knawn fnr SD of she patients w'.ith intanction md 119 ofthe cvntroli,.. u< d'te akulate QuetrJei s.imdea af obesiry (weight:hsighe-; teefr'); The mnn vatoes of this iadex did'not differ si`mfimnsly, Results . - -Ioformation wv sought from eaeh palienr ah'ont all thee itiems ineluded: T"as'e v-Pcnprrimct N,Mya.wdivJ J/ n;in and Caetre/Ram.nu T,rvted' li 7 HJ'Pr"cn`°°rawD/airm im this at buc the s w d ao l / A ponse u y re as wayacomp em ea geoen ,-pracsisioner, for esampie, might bave known whether his patient had'been t[nted for h7Ma¢nsufn without knowing the ocNpation- of h<r husband or wh'erhershe smoked. The resulu for the ditdcrent iiema,, therefore,.relase to diP.crens'.tosals..IX+itih the exception.of the bloud lipid anall'ses alI th'e :erali:nried bxnvccn g-t°o aod95^e; of'the panenu .vithh infirccial and tierween 77% and 99°0 of the mntroll patients. The proporrion of~panenu:who had wed ool mnenceprives during Ne~ mmnsh beforee admossion wu signi.icntlp higher amoagdte yatimts,vish infarcvon than among cb'c rvnrrols (P <0001) (nble 111), a was the proponion of those whohado used urai cunrnc<ptivn at aaY ume (P <DOl).'fheaa was noo appreeiable diflerence berwem Ne two groups, however, in the.proportioln who had used orul oontn- Nw..e.,mmr Reau W arux .hmOr rar. bs nl. ncew.. a~,a,nir+n'.ro~ mor: ~ w.,, ) fe.rs. . . .. Inlnminrvt mw,m -. .. Nu.(::) af ' (::)ar I MLPUnmc ~Cannab N3ywru.rv. .9. (a1-r)', 16S (%:s) . 2 (3./) l l . 1 (G'al 1z (sll' J • I 1 (I~z1 ~• T (11-ry 3. (hS1 J. Ton: aU(Im-01 111(10001 DietoNYwr. .- ..-• ~~w31 ((IH71. I 1TIO,IOOe). o~v„al>a.a+v.emru,ea;n :.I i iinilS? Iolt mptives oniy ac solne time in ihe.past (103 ;o:compared with 120°;,). - n,W The.relative risk of admission, for myoardial Infarctionin women - I an(IOOU) i ITl(loOY). who hadd been using ora1 oentneeptivn in rhee p?evious moneh • u.nv,(enrce.rv„ud1 :.>,!(U-tzzs:e<oml., mmpred'~witA th.r in who had neveeused'shem untutnatcd }ul. rn•rnu .: conr.vla': P: - pcua r.,,a. P(eoaa'usrtawomen n from~ thesefigurn to be ai to II .Ydne of'the.nia wometu w'ho had'h'ad a prcvious myoardial'infaaaion had ever.ued onl contnttp- rivn,.b'us.two or. more of,she other riskfactors mmtiooed below were psosmt,in all.of thcm.,The 17 patimts wiN ~[aion who.had'o ObsuY a.d P~~IwnP~ 7t3'are~dial InJvrurow. and Co+nal. Pat,rnn cilF. bemusing or+l contnceptivn during the month befoee admission . ~a (•:) er w: (•:1 or (rableIlI) had bem using them Ionger on a.erage shass the mrres- ` xt.l. r,nons I [•o„i.vla TAau n-O.•i t:en:.or:Pt'me P..niw oJ Afra<erd:e) lnJarrnon. snd GntreJ P'vrunu NMCLm,~u.e+.dset . - n tRal'. rll~ m,II. aUn'ren~~n F euf.rr~f~ub.w.I . .. I 13 ~(_1~n,'~ I 3i~ (13~• Tn..l 60 (Imnl ; I (tnuo/ I Mu.t~]or. I Nur(~7ef P.r-rr).nronrLraen,:e Pfaasu ~ ('mrr„Is Ndrve-crh.vmcm.,c,,,~ w ,6sar Iss (-oaf. I ( 1s1: ~ ~i 1/ (DOO/i i j If (II!It 3 U A d•ne --N bi! .J m6wn.. la (.L... I l.n Uorcrum Iv M IX aa1. nue V 'J. ly ma . lu ..,i~ wuna,. . -. ~ 1 (I91: li (Yel. ]'eW' I Sa.(I'm~o) 1 I(d(-li . Taul 1 60(10007'. . •M1I-'t.nalicn r'.ili• sa( PP<00t2 . ~......••„,~,.J,.. -9]6.<OO-.. li ~ii ii (u+) ~ Ia qoa) I so pl al ( ~ `I () Oa 1] r]! 61 a I1l s1 170 (loYq .
Page 93: azo61e00
89 3 > bA)T)SN,MFDIGL IOVRNAL 3MAY 1975 bnwew he iv.ro groupa (346and 240 respecri.vely),i but aigniBonWy' re of the paticnn svioh inGcesion (8:16?'°):than the controls (6; 5%) had'an index grcater than 3-0 (P <005); The inereased'freQuenry of pre-eelampsia (S0-0v6' compared with 11Q°;Jwas nou dua to an mereased ~ntunber of pregnancies (see table V'Il).'. No apprcciabli diRermce wes remrdrd:bcrw¢en the groups in the freqrsenq of prtviova renal on tltiyroid disnsn Iatormatien on previous. payehisivue h(story waa . dilTiouli to mterpret and will i be reported sepamtely. .. Table VII shows the distrlbution of the patienn by social class, eoun[ry,Of origin,.parity, and tnmopaulsal state at thee tirtle of ad- mission. No sn ally signifima di8ermees w e found between rhe two greups:with respect to any'.of these fatmrs. The psoienn with Infanerinn,.however;,rmded ro,belang toshe higher soeideconemfe groups:and to havc h'ad fewn pregnanaes. The proportiun of wvanen whowerrpostmenapsusal on admission was dosely simildr in both gmurya. An arufi¢ia6 mrrwpause.under 40 yeus.of age bad b'een inducM id only oue paoiene„ a smerol. My°rardial Sn/vuCwn and Cmr al PO- aP T)ry- asrd'~ mnpwwl Sau °J. II Nc (~.)'d M-1.: Pats,u ]] (al NI !)S) ]2 3 (e5) Pdo.ed'nrdi:ortW .. . { (T-S). T°ul (. 53 (100U)~. C_ dQwm SI (la l) aa N Sp 31 (IT9) Il (e'n ISe (IOpU) Lleinelt:nadom.. (ssn. 13]14(as x l) Comm°nrulNsamein .. R'.)'. a (1"tl F:,r°ym,r°wvua M,..avnu+rs i'fln. uF) Tmal~ St(10nMISa(IOn07 NJpipuow~ .. .. .. Pp~ a (100). lt~ (}S) 1 mereM•^'is ., lel (aPl) M wn 3 Vrepunca 10~ (IaT)', SI Tud ' a0tlaco>~, )6a~a00a). awv..ra snrr srm monusw„a . .. vIsast ua (vo-a). Pmwwprvw. .. I a Ilaxl I.' Is (e9). T°W I Se (]OOd) 16t ~(IOtrO)', Blbod; lipids we samined in 44 (770"<)of the patients with infarctidn. and 84 r(48?e) of'1 th'e contro!s.. The rnudts, . which ar, rcpomed in derail cisewh'ere,~s showed thae the mean levcla of both m" choldssetol assd serum tciglyceride were subscar.tially hi;her in the patients. wioh infaretion., TypF IIa hyperliboprot maetnia (ehoiesrerol b'9.mmo1!I (267mg~100 tnl) or more)• or IIb.(tholesrerol 69mmnl,l ornsoreandre;glyceride 2-3 mmuL,l (207'mg1100.mq'.or sesote).waspresene fn 18(q1°o) of thepatienu wish infaraidno hurin .. noee of,sbe rontrols. Ih eoncrsr, hyperiipoproteinvemia of type IVornured with'almoss equa4 frequetsey io both groups (68°o and8!3 ; reapecti:ely),. With the fihdl'usg of so many associatiuns it is neeessarytoculTsider whether any of thent,are seeondary, arisirsg bemuse the risk factors are thmssdOn interrelaced. Foe esnmple, m ~Bnrain oral Lorsuacepcive " sooatedd withc gar tu smokng. s and cigaretce ssnoking is assnciatedwith -yon dul ' famion It.is therefnre n nsary to egami ne Ihe data sep~ranely for different smokSng aalegeries. Tltis is done in t bl VfII Thdimmn5rm th heavy, moken tend to u¢ on1 cunrncept. s. ofren tAan ligh . mokers and show thae the rclanonship b-em oral ~centracep -ve uu and myoerdial inf.rctloo u presenr in ca¢hh ategory: The nurnben of abservations in eaeE otegory.are, ho,.ever, small and''the di(ferenees between peuienn with infaraion and. connoli ne signifiont only: wh'en all th'e moki,ng categories:.re taMidered rogeth'er (P <0-Ol)..The relation- shipappcansmorc pronounccd in..smokers rMn in noa-smokcrs. TableIX gi~ves esrimotes of the rclativeriskassoeiated with ehcuse of oral co acepci.ves after sllovdngfoe shc cffen:ofYour other faaors associmcd withl myonrdiai inffrct,on. Of the lour factors on)y •Lipid SbnormaliYies wrre elani6ed as rrrommrnCrJ b..ehe Rerld Health organivul°n O mwlsfikaion of mt Fndncx,nn clandioo;nn)i' The upper Iim,v, of nomm~l fJr Avrn <Iwpsntul and ~r cnJe x aL tw ce the undara. dc.uulor. Iwve he m v°I ipo~uio n, °nd he 6pepror° n panrm~ oiaam wnrmrW+a <aammcd hnec sacum, e)d lon,aroPn«ensl 743' auvttr-0rdCem npt'.' Pr°criraa( Afyeceedidl~f+JerrrunWC°vrd. P°urui .a D,$rrmr Sm°tny C°rq°riri iar Na.. (•;') N Pasimn Urrne SmWeG ~a,ltlm Or0 fiwrm°f ip.eAe ~ Tael onses °r Ey.+°de 'AtlPauan IG°nud.~ ASI'Paimu I Convda N°ne .. .. ~I 3 fls-.a ~ a aoJ) I la (1c00): ~:0 Oco~Ul i I13 I'ni)~ af10001~ 'H(I00L)'. ! °r mpe 11 ~)) : (I1~5) 11000 ~o): I la (IJ[1~9/ ~, Ted ST~ ~ i IN 'stS a-&nmarrd Rrf°rinr Rir.1 n~. MYw.d.'d 'lnJprrtidn, in Pcrimn carrnn%iy'Ur(.e Ord Cauexryr,vr P.rPv¢rrnv.oJro S+..ddrtie«im Jn Puu+aL C° leurrdirr[ van.,Arrs V.rwde snn<ereiud Rebti:e Ri.k F~,im.u I Y Sie°ia~°~ L..er. GauemrmWna . Nrprner,um 32 •l aaa Io-TS PeP91 o-oon P ~ Prredamvueroinmi~ OOnflr.. .. re .e lo-u 1]Ya eo-01 P P<oUDU no wxe..risnie .i,i,w,a°.ws/r .. 3r. 59) P<oo] agarette mkmg has. nynlatenal Cfa `.e. tnf'risk,rcdotmg th uotr 45 to 1 to.3.2 to I%Vh n eig smoking isallowed fo th . he ds f cn show esfen '4 Diatinra is nor m idered since thce wete only f v im~n M n ro hare thsa disease and nonewere using ora' enrwxeptives. Hlood 18pid le.els wen known for only'4a (70°e) of ihe:p?uenus with InF nction, .nd the above dculauitms do nor take a¢counrofany possible associanon with blood cholestrrnl.l !t is diT)culr to seewhy high veluea shouldd he associated with'h he use of'ora) mnea¢epiives exxps' in so far aa he cansncepuves muy th'emseives. ... ans inerease in b'lond cholesrerol.`r Indeed,.patients who are known eo have hyperchbleslerolaenua ar.e 1'kcly o be ad+ised t use a:other fnrm of contraceptioa...Th t tiis may h ve bnn ao gg -d by theobservauon that.only dsree of[lie 18 patiems wth mfarction kssow.n ro have type 11 hypcrchoksrerolaenia we:e usin; nral contn- eepuves cumpaned with eight.of he 26 patiinn with infaresion io whomrhe bloodlevels were considered to be nosmal according to our Crirecia.rs' prrFtNCf1oNOF FARONS It would nced wery,langcnusnbers imdeed to es;ablish the Q~_nriutiie efte¢ of eaeh mmbinanon of riskfaaan. 1. an asner..pt eo disen"angle the selasionahips, we showic nble Xthe propoaronsof parients. know-n ta ha.e h'ad vari ous numbers of risk faeton, suEdi,iidir.g those whohadonVy one facror acco:dingtoin narure.,Icformonan onthen rau:a s-RfJ Facsem is Afy°rurdi°IleJ.rafanu,d Cswrnd P¢naw Na .)°t I Vw(:)~e(1'~ Au. ~ eno I - N°nLL(.r,: . .,I . .. . On 4! ror 1 Ilf (aaf) 11 i (le~l) TYPr Ir yDrr 9.Pro¢inumia ..,( e'.. (6T) i I eDd'p~~ W-na (IS ot m°re CWr1.ll 0()l.T)' I xa (i39) Cunent N-r 1(lR) I Onr.~in( . .. I 3(50} II le (31 0> Ta~nak'.~I a I f1)9.(Sll) Thr« °r mnu rvk r.cwra 1s (xsoJ ](1-]) T¢W I aY(IOO-0) I~I t»(tooU) . presenceof one or moor: factun.wv aot ohuainedfor a few paeic•nta, and lipid'annl)ses were carricd eut an only, i0"'k of the pa,imn:uah inf]raien, soIhu thee number who mcre rxposad to one riskL^ccar a afl is likely tob'e cvrn smauer wn. ~PPF°rs:r -n iFi nblc.It ihoulW he -!7rd'ulso he h)Tertensien and ili.ihetes u ord.doniv.whcn thc pati,ass hadbecn eated fortheseccoad:rions before he inssrnion.occunred. Thc dsFssoim]acs derihed from the data in table X, howcvcq ssronglry soggest that he comb'm<J esfea of he faezors is synergislie.: In companson wnh paeicnrs.not khowa
Page 94: azo61e00
90 44 N . s have any, risk factors, the rdaeiv< risk increased from 4-2 to1 in with one fetor.tn10'5 to 1 in women .with two factors and &4 m 1 in women with three or more fieters.j )iseusaion la[ FALTDaS 's'evious studies of, myacardial infaraion have,, wieh one xtrption, imcludcd few.obserntions on womenunder 45 years d age,°''• O4ivet "• aloae has repocted pn sa appreciable sumber of ose=.. His series cnnaisted: of81 patients who ttended the Royal IIDfirmary, £dinburghi during 1953-70 and sofpartienlar value:benuse all the patients were seen and nvestigated by hioa Unforruauelya. coatroi sariea waa not nvesti'gated in the sameway; andlcomparison, could be made udy with data from other surveys, some of which were carried ,ue independently.in di.Terent.areas and at, diferent times. Jespise:this, O6vernshowed that,an abnomvldy high,proporuoa if patienu (79°;) were exposed to ona or other ofthe three sujor risk factors known to be associated with the developmrnt sf myonrdial infarction in a'tcn and olderwomen'•; 48"0 of »s patienubad hypercholesterolaemii (serum , cholesterol 77 nmol/I (2700 mgll00"m1) or asore), 39;',bad hypertension diasrolic pressure 100 mm Hg or mom), and i3^e% smoked 20 :igarettez oamoee a day. The imposvvnce.ofthese factors is :onfirmed :n our study- Ona or more of them -as present in a nodi8ed formin 80°;' ofnhe 44 patients for.whom blood lipid evels were known compared with 22% of the 94 rontml mients. Oliver's dataa are less conclusive in regard to four other -aetors thatwere less prevalent-prematur<menopause, obesity, iiabetcs meWitus, and:use of oral'cootraceptives. The last twofaesnrs were found to be dearlyy related too thedise.se in our study, and in a concomitant study by. Mann and Inman.'• Obesityalso emerged as a risk factor ih.n aua srudy,,b'utnot prcmarurer menopaux. OtLSL~ cormtA^SrrtvEa The association between:myocardialinfarction and oral contra- ceptives sha.un by, these two. studies is ualikely' to be due to bias. There u~ano seleeaion in ehe choice of caaes(otherthass the necessity for survival), which svu determimed by'srncr diagnostic c:iceria audthe arm of hospitalization. The method of se1lctingshe controla enn:red th'aat the puienrs in both groups.\s'ere comparable for age,.marital status, and year and hospital of admission, and they were.foumd to.be comp~bie in country of origin and social class. Some patients. admined to hospital fon elective surgi¢al procedures are advised to stop using oral conoraceptives before adnuuioD,, but.none of our control patients had stopped using them in the tAree months before admission, Mbreover, . the frequency off onl contra- eept.i.ve useoy.the con¢ol paaients was.the sameuthat found in a surveyof contrxepeive.pnctice amongwvtnen in England and \ValCs ia1970." •' lo b'othstudiesth'c proponion of women aged. 30:-39 years who reported using onl contnceptives was 13-8 ;e. Whethar.c5e ~ssociatiunis.causal orrefteets the association of onl oomraccpoive use witH some ouher facmr,is more diffsluLc to decide. ?levertheden, the signifiont«tarive-dsk esrimam, evrn tvhen zIDowimg (or tAe othcr associated vari6bl-s, and;the rclationsh.ip.beewecn the frequency of casesreponted to the Commincc on Safety of Mcdicines and the dose uf oesuogeni" all argucin favourofa causal'.rcEatiunah'ip...Thefinding.thac. women on oral contnecptivcs who developed m,oeardeal }TTcfar, thar lilood lipidf rere evmincd in unlJ.ai (JN^.) of ihemnirol pancnu.is unLkay io ha.c indarnad dieuczn- loany apprceubles.typc II h.ccrlipopsvwimemia n mally uncpmman.m .ounS n~In a. fuun~ m ana-cf ,hc mnuul ~Daoicnu trom ~sumhlood ~~m~rk..°rc abuincd. aRtTlStt lHEDIC.IL /OUFNAL 3 MAY 1975 iofarction tended to have been using them.longertlian eentrolpatients wishother, unrelated diseues'• may go some way'to explaining why.cariier case-control studies based on,(cwcrcascs failed to shosvaclear rclitidnahip.'ssa If,.however,.as these findirtgs suggest, the riskihcreases with,prolonged use we might h'ave espeaed:to hnd'an appreciable though somcwhat smaller risk in women who had used oral contraceptives . in the pastt bur had seopped using thembeforeth'e infaraion occurred. The numbers are too snmW to exclude sucha possibiLry,,but neither this study noe that of'.Mann and Inman providess any evidcnce to support it: The strong suggeuion that she combined cffec[ uf~risk factots is synergistic may.h'ave important pnaiml impliationsforthe use of ora1 contzaceptives. The appreciable increase in relative risk in.women with more th-one riskfaaor fbr myonrdial in£arction suggests, that other . snaehods of conaneeption should be copsidered in sueh ~ases.. tYCtDE`XE OF MYOGADtA1 IhTARCrIOIP . We ers make a rough estimate of the incidence of m)ronrdial' infaraion(u determinedby admission~m.tlospital) both,in the general population and in users ofroral contraceptives from the data cvilcaed ia the Nor[h-Weu Metropol.itan Rcgion, where the proportion ofhospitals induded in th'e.study was greater thanim the Oldord Region-. The hospitals participating inahe study.contained aboun5: °~: of the femaleaeute.genera1 madical beds inn the Norrli-WesrMetropolisan. Region. An estimate for the Regionof tliea.unbcr of morried women agcd30+44 years admlc ed to h'ospitall withh non-fatal myonrdial in£arrtion dusing 19K8-i2can tii<refore 6e made by multiplying rhe abservcd aumhcr of patients by. 1-75, onn the assump{ion that th'e.b'eds covaed'in the study served 57 ;ol of theRegion's.popu'ation and that the admissions to thesebedi were rrpreseatativc of the whole Regiaa.Similaely; an estimate on be made.of the car- responding number oLpatiants admitted in the Region n<ho were cvrrently takingonl contnceptives. If the data acquiredfrom the controls are a umed'tobe represeutative ofthe general rrtamed populationis tlie Region,,a estimace.can tie-made of the numh~er of womemtaking.oral eemrzceptires aean•v,point in time. On thit tlasis the yearlyy hospiall admission rate for non- faaallmyocudialinfaresioa is 2 ~l'per.1001)D0in married woman aged 30-39years whodo noruse oral conuaacptives and:5-6per 100000'in married scemen aho do. In the 40-44-ycer age groupthe mves are 9I9.and 56•9 per 100 000 respectively. The inareased risk.ofhospital ad'miuionfor m ocardinl InGrcrion in wmnaa currendy using oral contnceptives (2?. times in the 30'-39-year age group and 57 tiracs in the 40-44ycar age group) ~ is siinilirn to the incrnsed'risk of death frorDthe condition estimated by Mann and Inman.r• We are mna grateful ro the mnauhann and general pnaitionecswho a0owed us.to studypa«enn unddr their n and1 proa-idedinformanon, tothe medical reconds oihecrs ofthc hospitals and theii srag who gave wflling help in many w.ayf,.,and:mth'e patiants whb gave up their time:ra amwer.ourqu<stions. The \ScrVpineFnun.dation kindly' loaned us a mocorar,.wittiout which! so rtwny paeienn cuuldinet have been inrerviewcd, and 6nancidl suppors was received'from th'e D:H.S.S. Gale Mtadiprovided in.ahmbld eecretarial help and!Dn. A. JJ Hon ur.Sasc much advi and cncou~ngem f. ~4. was supponced by an I:C.I. Fellowship amd :by, the Cecil iJoh'vs Adams Memorial Fellowship duving the pexied of this setsdy: References 'Ve.,.y„\!: P.: and Do1Ll g., l:.itii4.IAedi.W 7uv.nul, 1069, 1: 651. =CnIL.M~rati.'c (:mvpfon S~~ud•' ni S,r- ,n 1'uunf J,:,Fea a Vc,se., ~ll P., Chm:~A rrh:n:,r,.,. c,u/ 4~*:cwncr: 1e73, 11, (,S. ' R::nenr Grncnl, .1'rearr,.allA.rr._ af.l:'.rtvui uRJ It"ul:r, l'urt 1, 19^6.. l:ondnn. H..11.5.0:, 1972. ''f: rlLr. l.. A., and U.,u,tcr, i.. l., L-rr, J912, 1865. • Kanncl, ~Y. C., rr~l.,.iv,rWr.eJ hnrrnuLtr:J::,~.; i9:1, 13, t.
Page 95: azo61e00
91 wor.la Haltn:Ore,m,rarinn, IFwliq.Grn.ponhaAa.~ric.lf:arr Diuare /treLr:rr.,W:ki.J:, R<yianal OMrc[or Europc, Copenhag.~n, 19i1. '' Pike, 11. 4, and .\Mrrow, K. H., anri,A')uurnal rf Prrrrnri9r md Seddl At~nai :1llc r 0 5L -a 19 0. f, :S. ' Gu1JWur . L.,and -M..a+lit,J. H'„ Grv,J1n.nW e/Prre..nav aM Sarial ,tlrJrc,na, 19:+, /Y; IIG: 'r'h)ann. 11 L, mJ'I'EOrog,rd, hL, R.iti,A lka.,7+••^'v' Ib. prn.. ti e: (:< G Sn 1. A:, and I:iahard. 5', L 19E9, 1, I3.1 Y i\L P. rl ' IA r.uan,m,al ]nvrn 1 rf EpL.vslarv. 197^_, I,;119... $ 4eL5 dA.1 V,Li•wr,19]I,1P%. Beasnw , C., d<rn et:e,r. srn,,.finnoi.a. +v77, Su9p4 Nd. 5+9, . 'r htdka6, , R., Hick<f, N., and Maune, g.,lwraadu,ue. 1967, 36, 577. Btiri,a:.tfrdiiclldi 1970, 3. 110.. 'r pli" rv" " Oh.cr, M. I'„ Uruui..41;JiYa! Jarrwl, 19YA; +, 133. 'r ImrnSar,cryCommiubn (on Hran iDiuau Rrauurm,Cirnlm:w., 1979,D ~Smn; ) I d 1 n, W H W, Brin,A s6rd,r.llnn^'d, 19'r'., z, ^-e7.. tu ,q ht I. !. I•ron iy s ro. u Basl,odnna Ipa/r,. lsnam, H-m _S O , 1. , J. - Cnec, `.1~, p.r-etul co.nmuniulion. 3973, ra/l :!; 19i0, 3,'pl.. ~ Inmm,. W I,'t'.,.r ol.4 LA..tlrG "Inman,W H W dV ht.P„R,n A dka!] I1968;;193. A 'r ee.um M j~~ d D 11 R; D ~ r Asad- ! 1 d, 19eg anos, 1970, t'eft. DdGri. Ar IYOrld N aGa O• +7 89Y. Oral Contraceptives and Death from Myocardial Infarction J:,1. MANN, W. H. W. INMAN BridrA Medird Junrulr 1977, 1, 215-1ig Summary' Wr.invesdgated 219 deaths frommyocardial inf3ret]on wommunder th'e.age of9p-Thcii histories were.eorn- pared with those of living.ags-msshed:comtrols selected from the same general practiees. The frequcnqy of use of ota] contracepaives during the month before death was signifloansly greater in the group with infare:tionthan during the corresponding month in the control:group and the average duration of' ux was tonger. No iafor- mation on cigarette smoking was available but the proportion of women being treated'for hypertension or diabetes was greater among those who died than among the eontrola. 7his did not alter the overalt, concluslon that the risk of fatn] myoc-.ardial infarction was grester in the women using oral contncepcives, particularly'in the olifer age groups. Introduction Irlman and Vessey's report to the Committee on, Safety of Drugs on deaths from pulmonasy em6olism and coromry astd cerebrai'rhrumhosu in womea of clsildlbearing age was pubLislud inI96g:' In rhe nses of puLmemry' embolism andcercb'ral'thrombosis a stttsng relationsh'ip was found'nvith the usc of oeal I contraceptives when these disorders occserred in ahe absencc ef predisposing conditions. More of the women who died from:m coronary thrombosis in the absence of predisposing conditiom bad also besa usiogoral crontraceptivea th'an wouldhave been' <rp<aed from the caperience of tlie conerol '.gmup; but for this condition the difference was not quite.siipsificanr and a dcfmiieaasocntion was- considered not ptroved. Later studiea' r wereoot conClusive and wc tAoughtt i¢t desinble mundcrtake a further investigation of deaths from myonrdial) infuction in 1973. Selection ofCaaea Transcript's of a9 death cerrifiotn rdatierg lo wamea under the age of 50 pesrs wh'o,died in En:jland and Weks duriog 1973'and''whidh Uni•crs+ry of O:ferd,.O%) 3QH', ). 1. MANN, ars:,. rw.a., tcmurcr m Soaal md CommunrcY Mld:d.neC9mruloue os Safcty'. of AgedlelGe., t.ondun ECU IPP W. H: W. INMAN,:r.e.er., M-r.rx, Pr4,aipai Med,el'Ofgccr bad been coded to rubric 410 arrerding to the eighth revision of the Interrutior.al Clasaifiation of D'ueases (myocrdial inGraion and syponyrnow terms)~were obtained'from, the Regisrrar Genenl Atcnl'of 726 were received. Alli desths im -umen under the age of.a0. yean,. r+ery secorsd dealh irv the 40-dd-y... ge group, and every.fifth deauh in the a549ryrar age grouID were sdlcted far the srudyl giving a eoul of 27,ases (table 1).i Stlee[ion was.madd cnmecvnivelyas bstches of cernfirateswerereaiveds u t-NioM.n IDcarA Cni/Imr Rern's!ed; Salerred Idr Seudy,. Lrrs:ianred,and frlel.dlJ fnPruanr Analywa Taral Aarnovrtsorsla ta-+e we. .a i00 `o. in.nuprM (77 . }9' bt .>inrL,deA,ermiry.•.. ..I Sllal)HI36). I .a(311. • V,m,Een o( iannvl enrimv ~'e ~.rn in. e.rmrwe~ ns n7 a)9 IS7 (1%)'. Efforts were made to interview'rhe geoenl pnn(tionrn who had edred foe the pdenes. a In. 15 a ei tier Nee wom a r not rrgisiered witli aruydocto[ or thae er, hospital,:ot loeal oeiL muld noe IdmtiEy him,:and the medical nrnrds of en had hem Josr and no other dam source was availahle. A further 39' death'a wee noc.in estitated1tienuse the gcnenl pra - coudd not beinoecviewed. Thm 58 (Y3efthe 277,de.oW could~not be srudied. Tbe rrmaining219 dnths were Iovenigated by'y ths.comrni«ee.'a medical field offieers, rs esuh of:wh:cb afwrta'er 66osuwerr eat:uded, In. 37 eses cvidence for the diagnosis of myoerdial hifarcuun wu thought to he inadequate. Dcaths were.inehadod in thefwal .unlysis only when the diagnosis wassubstaotiaeed by necopsy finEingf oe a Nstory of typical chest pain cogcusvr with. elecuoardio- graphle oe emymade wnfirm+tiun aa dcfimed by the World Health Organiaa on.'. In 23 ases (InSqo) a ruv.npsy earried out after the dnuh cert;fimte had ticen complnted or (tess often) ottier eridenae suggeued rhat dnta was attnbueable te a dilfereac ouse. A further ezduded bcousethe wrong sor, age, or ync of daathh had bcen given onthe certifieare. The rcnuinong 153 dearha pre.ide ch'e basis for this reporu ln 104 nse. the diagnosis wn suhsnn. tuted ae ne'erapsy. Proceddsre Forry_~itht members of the wmm(au's sutf of xncd;aal blficen rook pan in the field wunL Dmriing the inveniiariun- of rach dnrh urc of . rhamcomplet<d a 9nestionnaiue as fully.rs pnssibk wiahthe aid ofrhe gcneca3 pnaiiioner:an.l any other doc.un who had anendedihe patient duringg hertetmmal illiness,.3'ince the gencnl pnaiCioncr'st eerds hzd muallq'been rrtumed to the loal erocvtive munal after th¢petidot'i death hr - aaked ro e c rhem berorc heingw.ed. These major s0urm of indormation were oftrn supple- mcntcd by; h'ospiia1 o notn,, famdy planningctinie remcds,. nccropsy rcporn,: and mun reeords suppdiedby.acoroner. , s
Page 96: azo61e00
r 92 New Estimates of Mortality Associated with Fertility Control Btr r-hristopFter rarze Foutneen.monghs ago, Family P(tirming Penprctiaespublished an article in which estimates were made, by age, of . the levels of . mmtality in dkveloped: countries associated with the use of vari- ousmethods of fertility co.ntrol, iockud- ing risks associated witA th'emethodsthemaelveaand those arising from un- planned pregnancies. as compared with the risk of death due topnegnancy and childbirth when no fertility control method isused.s'Cn aes appeod&c to that anicle, the authors.eapressed the hope th'at'-new' and' .. more precise meas- n'of mortality would pennit a revi- sinn oFthe estilnaes.~ Since such new data have.bceame avaiGblh,.it isappro- priate to update thefindings- The data not availible.a year ago c'on- eern: evised estimaees of, mortality as- aociuced with legal abonion.durin~g thee fust 12 weeks of Cregnancy, e new estimates of',.pillire)ated mortality mong,moken and norumnkers, and revired estimates of IUD-rdated mor. ta)ity. In addition,.maternal mortality ratios in the Uni~ted Statea during the three-year period 1972-L974'have.been4 substituted for th'e.19TJratioa usedin the oarlier computations. As was noted in t)ie.earlier article on mortality risks, it must be emphasized thu'beeause the assumptions as to cr.n- traccp0ive dfediveness.. and mnrtality'y used'in the cnmpurtations nm dcrived ea- clusivelyy from data originating in the United Statea and thle.United Kingdom, tA.ia,oh.r TLw s a Se.ior Fdlw.isA TM Pep. ybt imi Cnuneil .nd di.<.v Jir.l.w,-e,,,esevah ao- .Ue. of d,u, m9an,uuen. Thi..n41d updaeer tM 6ndinp nf'-1dws.lity.Arrnit.md -uh'. Dmtml of FrnnnY . le r .,,d hi. Guneit knlta.mua.. )ohn B.,,a.ersr..d Id dv.la.w,..riFelwuary 1976 oumh.r nf P.nyeo- TM .w„pmx m,dei .., ~.hk'h na nnnellly. m.mua w ba..d ir di.mt.d m tb.t anser. 74 Recent data on mortality related to pregnancy and to use of various contraceptive methods show that up to age 30 the risk to life among noncontraceptors from pregnancy and childbirth is far in excess of that experienced by users of any method. After age 30, the mortality risk experienced by pill us- ers who smoke rises dramatically, but among nonsmokers the risk remains relatively low-and is lower than the risk of' death among noncontraceptors even after age 40. Safest at all ages is condom or diaphragm backed up,by abortion. the ntodel is appl.icable o.nly, to women . living in developed countries. Extension . of'the model to rnuntries with higher matemal nnortalityy and lees aelequate n.Yliud facilities and serviws cannot bemadcon the basis of availatrledati Table 1 summariseathe revised'esti- mates of rnortollty.associated with yreg-nancy.and childbirth.induced abortion, and the use of onl contracepti-a and'IUDs. The estimates for pregnancy and1 childbirth are slighdyhigherfor.each, age groupeacept 30-34 than those pub-lished in the earher anialc(ice ref. 1, Ta-ble 2);.for abortinn, the mortality esti-ntates a e comi~derably lower; and' fon IUDs, they.are elightly lo.'.er for young-cr ages and higher for older ages. The es- tisrotes. fnr.pidl use are not cmnparable with thep.evious report,, s ce they are Table t i etonality associaad w.itn pregnanry and cnudblrtn; ley.l atsonion.ua, ol urai, . rantucepo.yes (by' amnting atatua), and'NDa, by age Aqe pmip P.19- ta9Y Ps nmat l , w,cy d.r6 ww• aEC,-. "an" Nen ust : 5rop f 1S-t9. 11.1 13 13 11 pg t0+2.'. ,p.o 13 tz ,. 0.8 zsn,zs,a tz 11 ,.o atiJa'. aa.9 SS59.a.o I t' 4 ,.e t.e as to.a ze ta 1 4 .pL T1A t e e.e 51.1 . 1.. 'Rano Geaou.o]o s.. t.uy IeovuCbq rrxwan4 U.619T2-19T... rRaw Cw iE6:opo ftnsynsw' hrreore. U.s. 19r7-19Ta . tNaie p~ tpp.p9p uran paryear. eFSlenatae elY'l K: Jr,.li.e•N4o.oa.6i here dieaggregated for smokers and nom that had', been used!as the basis of the smokers. e:arlienarti¢le. Annual mortalityntes per As ntntcd alin.e,.muin. nf mnternal Im,ISnOn.ersof'.nmis(mainlycumbined umrtality, pcr 1i1QWD Jive binhs frnm prngostin-estrogen formulationd;are de- complid'ations, of prercgnency, and child- rived from estimates of ezmss.mortalitys birth (exclsding di<aths attributed to frompul.monaryembolisnt orcerebral abortinn) sre based'on eaperience in the United States during 19 i 2-L974.s Ralios of maternal mortality' penL00,00p abor- tions are based nn all legal fint-trimester sLsnrtionr without.rnncunem sterilii'a.- ti'on ideneified by the U.S. Center for Disease Control (CDC) during 1972-1974- replaeingth.e British data 'Thi rviar rre den.vd fM,,, wmb.... nf 1.6J rhe.. n1.M.d ,e ih. C.ntrr..ldr Du- . Camd I I9TS1' a.d TL. nb„ Ommach.r Insreute (19T3 ud 19T+).; reawddna to E., w'tioslorL n of., -Aboaro. Ntt,ir .nd S-+aes - 'h. U„nd Stu.+. I9TL19]5._ Fa.~.Jy Plmw~wa. PFfCSnno. 6:56.. 19T6. Table 1: and erimu.d diien6uslee ef sber-mms by I-.od ef'eesuuon. wm..l..se. ud.d,eh.rrn.v,emt nr..il'vaian.vpr.fennd` , Fa,eily Plannins Paspeeti+es V
Page 97: azo61e00
01 f th'romboaifcamong women 2U34 and SS-44 years of age,.and itom n,yocardial infarctions among women 30-39 and 40-44 years of age, based on retrnspec- tive case control studies in, England land Walks. Uiing these data and others as th'ee basis fbr h'is nleulations, AnrudYt ]ni n of Tke.Popplatioo Council has recently es. tio,ated ratess of eacessmortalityas- sociuted'witk the use of the pill, sepa- rately for smoken and nonsmokers! Jain found thateaeea.mortality.ddm to uralcontr.ceptives is ma.ginally, higher for emokers than for nonsmokers am.ongwomen under 30: yeaes . of'age and is . ahout twice as high among those aged 3l'1 and over,.Howard Oty of'the CDC h'asmade similir computations for myocar- dial infaection.only. (but including pre- di;sposi ng conditions oth'enthan amoking suchash'igh blood'pressure and high cholesterol leveli); and 'arrived at simi lan results.' ' The rates ahown in Table 1 are suliject too h:iaen in npposncdirectinn. widoh mayy nrnsay'1 not u-ePuut., The upwurd biar relates to the fact that some of the deaths under.investigation omsrred ata time when pills with in estrogen uantentof more than 50 micrugrams were.nsore widely, used than in remnt years. Fur- themsms,.it is g!enarally acklnowVeshged th'at duringz that . perfod1y potential0y. sus- ceptible w.ommnn were less rigorously~ screened than ahey are.todiy. The down- axd b.ias derives from the limitation ~ofth'e mortalityy estimates associated with pill use to pulmonaryembolism;,cera bbal Ithrombosis and myocardialinfarc- lion.,A recent report suggests that the use of orulf may be associated with addl.ition- a1 nrdiovaacui,r di_.eases.sThese a not incl6ded i thisrsport because the additional dsks haveuul yet.been snb- stantiated'by otlier studies, and benuse deaih otes sper.ificc forage and smoking status are not available. Pill~use has also been assocfitedwith ash increase in the occurrencee of sometimess fatal (but c txmely rare) benign liver tumors uwdl i as gall:bladder dise:ue. While it ir rea- sanable to assume tharsonm eseess mur- talPtyfrom these conditions may be at+ tribulable to pill u cidence data-ar<' not svailable withSV)iich to make olcv lations There is little information available onn method-related mortaiityy with use oP. IUDs. hfost of the known denths have buers as.oeintedd with epti. alwrtii,ns„ generally in the second trivneater, or with . rupturtdl ectnpi@ pregnavoies: TLc rat.Y Vtilume0, Nun,l.cr T. Mart-hPApril 110T7 35-727 0 - 79 - 7 93 Taole 2. Innuai manEer.'ol t,lhewelated, melhod+Nated,~ and total deaths assaelatad with control:ot /ertl0ty per to0.001i nonatenle women..by re9imen of contml and iqe ot wotnan npm.e a wi+a. •9• 7- ane eum,. tsta 2o-24 2r:29 7a-3t 3" .O.a. tb.•onuet' e.m-rwted auentee erey . tANwo.Maee sa t3 e..t t.a. Ia 1.6 109 ,.i7' 20.3 0 ' 2tA 122 ei0 erlyrnanenqY.n Birefral/IM ueuroe.aaro v.t 12. 0.2 ,_a os 1_2 a.. ,9 ae . 79. as 5.6 Tqatdeaeq 1Z tA 74 iT ' .S 7d vtn er.yn,.o.ers Bi/T+Wtee 0.1 n.3 0.2 0.4 a.a 03~ asemoe.Ma.d L. ,m_4 M . ~. M_4 TatLdealM tYp. m,/y, e-ee : . t3 at os t.a 03~ to.e ~ 04 ta. ~. ~ 0.6~. so ~ 05 swroa.w.d 0.8 o-e t A, , D~ 1_4 1 4'. TdatGeey 0.9 1L tz tN, lo " _ 1.a~ Tr.ettrww,,,wnoeewy esiawNN•d 1.1 . 1.8 Re~ !A~ 42~. TrWnbna/ mameas: etw eCamen MMed.reWed 0.2 11.2 0.3 0j, 0.3 ~ 03~ .aiww„ in Tablc 1 am bnsed pri,narilynn: • 17deaths ftom septic abortimn- ss- sociated with the useof IUDs-identitied by the CDC during d972-1974s • in estimated'sia m.illio,rcwuman,years of lUD nse d'uringg that peried, derived fn.- tlsc Natinual'. Surveyy of Farnily Crnv.th in 1973;saand • a further, estimate that two-th'irds off all unwanted pregnancdeswitlh an IUDn the uterus w atemrinated by legalabortims in the ft sttrimesterand-there- for notezposed to the risk.of septic abnrtina in the seeond triin<ster. Since 13 of,the 17women who died of . septic abnrtion diuring. 1972-1974 .. had .:ed the D:ilklm Shield, a tytse of IUD no longer marketed in die United Statea;, morrd9ty 6mn septic aliurth~rn shnuid1 have declined substantially srrce th'attiine. Ectopi¢ pregnancies prohably c.,n- trinbueeless than-0.2 deaths per 101),000wnmen.per year to the meth'od-related, monn{ity nf I U D usen. Tahle 2'ar,d. Figure llshow the num-hersof binh-rclated death.,, method+ related deatha, and- where applicable:. the total of these two, u.ociatedwitfl, e by fe,,und women of the several I methods of fertilityy control and withfailure tno use any, method off control, based vn the revised m~ortality,ratios and! mtn show in Talsle 1. The outstandingfindingg ar . • Arn<nig wnmen urxlcr 30 ycars of'agc tire mr:d iri.k to ltfe asenci,tnl with-h . nf the fonr. fw utcthoiie of fertility cuncrnl (pol, IUD, diaphvagm orcondom sr,furst trimester abatrtion) uredalone is aluout equal, and is very'low'(1=2'per 100-(00M women per year), signifie+uly. IoWer than the bi rth-reLated risk of death without fertil ity cnntrolJ • Ileynrnd~age.30the risk to.life.(which is ahnoet entvely, methnd-related) in- creases rapidly fou pUf users mho smoRe, until,.after age 40,• it ismuch higher than the risk emperieneed by, women us- ing neither contraception nor abortion, as shown in the tmp.line.. For ell oaher nethods-- ti.e risk remalus eonstanl or (in the case of nonsmoking pi11 usen or thnse.u.sing tmditimnal methods without adwrtion,Isackup)' iocteases moderately, bue rcm,ins far beiow the levelof',nor- talilyassociated with eomplications of pregnnn y.and childbirth without fertil'u- ty cvotroL .. e At sll sges, 6,e lowest level of mortal- ity, by', far, is achieved'by'the use of tra- ditional meth'rxlswiths recourse to enrly abortion in case offailuse. . In smmmary; this update demonstntes low. levels nfimonalityy associated with all major reversible methtsds of fertility, control-thc pdI,.IUD, condom or dia- phngm,.and aitorti-npaced with the risk of deoth from pregnancy'andy childbirth, when . no fertility' control method'is used, with the eaception of • 1,.1...4 Lw -o ....e rt.aa la.;p- .,tar. d+e wnrt~n nrk e.a.ed. th4 u,ona ..IJ, 1 „/ 1 feniliiy amud by. cc iE 6.<4d~ kk ~u,Mv a,d A: K 1sin, t9T7 mt 0) i 75
Page 98: azo61e00
z , 94 •NW£stinwuean(P.forsofiiyAhocfatedmitAFersififyCentmf Fgure 1. Annual number of deattn assotaaled with conorol of YsAiBtyanA tw conlrol per, 100,000 nonstenle:.omen, hy•e9lman of txnerofl and a9e ot wornan Ar....l0aaar .. so 511 54 u sa H -4 .Z .0 y Je a. 32 x © EI za a m ta 16 3 10 a 15-19. 3031 A" Herp.wrdm.rd' o NpmMpda ILGn40ntYMy. 3SlY]l4'SA ® P9mNy/ npevnptws pill use aftdr aBa 40'by Women who smoke. The updatee also confinns the vety, low moeality'comb4eed with100 percent <Rectiveness Ihat'en be nb- Uined b'yuse of the condem and dia- phragm, when the:e n+ethoda ara backed up byearly induced phortion of'al1 prcg, s that.result Irorm contrxepuive failuee. IM(wamosa 1. G Tlene. J. &+nyru ud Bi Srheuer,-Merta4 ity Asa.a+atM ~h ihe C.mncl.o( F.nllity, : Fmnay Plnnn-9 Pmp.c- lA, 1976. 2.. Vv',.,.1 Cmte. IM Hnhh: Stuln~..l)111!RV. (NCHS).:-5umman Rrpnn: Frwl Alvnalil, 5ta1'u- lit -Manthly YaIS/NWinA.yo2VO1.2T,Ne,6, 0 PJpVyI snw.ws ED IUOsdYy~ SiryrplnrcxPA- 19'~. TaMe l; V„1. P], Na. 11, Snp plen.ent t'U.', I975. Td.l.6:: udVd. Si, Nu 11. Su, PI•"'e"s..1976 T.We 6. l Censrlc. DP'saa.eG.i.nl, n..prrhli.h.d t6r.. I. - II. W. h„wn and N. P. V.n.ay,-Inws.tlatirx. u/0emh.. fiwrr Cor..ury vd Cardthl Thn,n,4.in ..l E,nlK,hk,n in w,arn.d'CAild• Bc.rineAee.- B.ai.A Nedkd lwnrel 2197. 1968. S:1•1.MannmdW.H.W.hmrn.~ OnlGmn <p rw +nd De.uh (ro-n. Mwranlial Infii<uaw.: H.v- bA Al.docd Jn.^`^4 i2,S. 1995, v,dhl. Minn.lw, H. W. Imn.n and M. Th-d-On1 Cw,tnarp•. s Ua in.0hk. Wn,rretuatdFnaI MrneardulIr fantinn; BwtwhAlyd4all^•nn1i2:Aa3;197d. b. A.. l' l+i•. "M-tality Misk:Ai.,s.ated L.nh Iha U!e .4 Or.l Cmm'acapi.n: Stndir- i. Fa.rily 9-.Jrll 19T11 `, "Cear ne Smnk,ne. Ufn n1 O.aLGrrcncepli.a,.nd M-di.l lnlartipa 0 \ I I . Tra6tiaW 'wnra. ® Trpi4prW mnra- aWd^o+} propdi ana alrapo. A....:d., J.w,.f p6OS.eark, o,dG.,e,.ldev. 116.7111.1976. T. H. W. Ory.-A BneE R- ntth.A..edaiew Bn ~n 0.1 . Centrw.qi.. Us..nd'De.elePrn.M of Mn++rdul In/ua....; wpe. prs~n.ud a tM .n- srul of Ihr An,n:r.n PobIK Hs.hh~ As- ursatrr„ Miasni Bcsh..0.L l7-zl. I97ti 11.. V: 6enl.. -C.rdin...n,lv-D4m. Munality Tr.,.l. andOnlLMn.Rpt~.. Us. m Ywna Wo... cn- Tir lar,rer, 2,104t 19 i 6. 9. w. C,tn. H. W. Or.: R.. W. Rocbm.nd C W. Ty. I- .T,. L.lautenn. D.riee .nd peadu IromSpnma s AMniort: Naa Endyid Ja.nnda( AI,d41.., :95•I lSS.19]fi, IO.: NQT I5, -Cnnt-ii,. Utiliation amont Cur- mtlyAl{med w,wnaw I~J Te.n nl Ate Unit.d AIunWP Vii./ Sierut/pAapert, Vo1.26,.Na. 7. SuPd- 19T6, To41. 4. Familr. P~lannine Persp: +i...
Page 99: azo61e00
95 Special Communications Association Between Oral Contraceptives and Myocardial Infarction A Review Howard W. Ory.,MD~~ e Anatysia et the assoeiation between oral'contrae.ptive.use and the dt wlopment of myocardial (niaretfon In women lesa than 50 years of . age sh'ovrrthat dqareMe.smoking Is the most'Imp,onant factor In Increasing the likelihood of myocardial Ihfarctlon. Thla aEtect Is Independent of oral'eontm- eeptlveuse but oral contraceptive use also appeans to be a risk faetor how- er, their use In tAe absence of other pr.disposing factors appears to have y.a smell affeet on inoreasing the risk of dyhng from myocardial Infarctlon. Z . s small Inoease.Ib of the.same onder.of magnitude as the lncreased risk of death from thmmtwemboita disease. Oral contraceptive users more ttian30 years of age who have other factors that:Inaease the Iikellhood of myo. cardial Infsrction appear to have a substantiallyy higher d.ath rata (JAMA 237:2619-2622, 1977) REPORTS' published by Mann and aolleagues`-' reldndledinteeat in the possible association tietw..rn oral rnn- trieeptive use and devdopment of myonedial infarction..This communi- cation is a.review of.the6ndings of Mana et al and the findings in earlier major stndies that touch on this asso- ciation.`•Ovetsll, the reanalysis sug- gestlthat women mon than. 30 years of age who have eonditions that ptr dispcae.to myocardial infaretion msy be advenelyy aHeeted: byy oral oontra- eeptive use. Women (even tbose more than 40 years of age)', who use oral contraceptives but have no other pre- dispadng conditions appear to have only a dightly,higher risk of having a myontdial'infarction than nonusen First Reports In1968,.British investigatots'+ re- ported' ans association between oraleonttaceptivense and various tEmm- boembolie diseases. The report on am, vivors of thesealnesses' induded only Fimn" er Eptlrnbbeic alWiee aeanCR Fus.. M F%MM Ev.luseai Oir6/eeti (:Mer kk OiF YN Co11Ver. AaYRL F~W O~lon V.r annual mwlkq a11M MilL Can.Mpib Heel1 Ameuaoll Nilani eYCq OR 1a. 197a. NeMM mew.u to aunw cr EpiO.mcbqr. Caetr b nNeeee Cenrtet' 1000 CWme N6 M- mee, W 30333 (Of Ory} ... 17 women with coronary thromboses. With so few esses, no reanalysis is possible. The other reporN contains data on 205 women who died of mxo narythmmbosia; these data are re- viewed'in detaS Tbe investigators divided thesa.205 patients who diediato.those with and thoaa without~ risk factors that pte- diapose to eoronary.artery disease. Theyy speeifieally mention hyperten- don and diabetes as such risk faetora. 13ey, also looked for 'other conditions whirh might have contributed to the terminal illness." However, the inves- tigatora did not have information on smoking hab(ts. This is unfortnnate, beause a number of studies,,iadud- ing tbae of Mann et ai,- atrongly implicate smoking as a eause of myo- eardisi infacctiaa. However, Inmann and Veaeey presented su)ifeient data to allow os to make a detailed'ree=, amioation of the84 women who are stated Whave no.predisposing risk faUonformyoca=dial infarttion (Table 1),. The confidence intervals aredeulated with usn of:themeth'odf of 3tiettinen-" All x+.alues an ealdt- lated: hy the method of Mantel and HaenaaaL" In thesen data, the age- panty. standardized risk ratio of 1.8 is thebest.polnt.estimatee of the excess risk of myocardial infatetion for oral JAM1 bLne 13, .19T7-voll 237. No.. 24 contraceptive users free ofmajor risk faetors. Inman and Vessey, dalso looked at.tha data on 116 women who died with eonditions predisposing to myo- eitdial iafarction. Of the 1100 for whom they had'eomplete information, only five were using ord.aontraeep tives, whereas the number expected from the contcol'e:perienee isI26, Thu difference is statistically signifi- nnt (P-OS). This finding is at vari- ana with the results of Mann et all'-' One possible explanation of this in- verse association is either that physi- cians were reluctant to prescsibe oral contraceptives for women with seri- ous obtanic dixaaes or that the women themselves made less demaodfor oral contraceptives. In fact, then is considerable evideneethat fewer women with chronic illnesses use oral contraceptives than erouldotherwised be expected to based on their demo- graphic chancteristiea•-"° In 1968, Oliver' published a aue series of all ?3 women 41 years or younger admitted to the intensive can.unit.of the Royal Infirmary at Edinburgh with myoeardial'infara tioo. He noted that 11 of these women used oral coattaaeptiver from aa.in- direet estimate, only two would have been e:peetad to use them from their age-parltydiatabution. Ten of the 111 nott taking oral contrseeptives and nine.of the U whohadtdten oral eontrsaptives had factors that pta diryo.e tnmyoradial infaectioo. In a sabsequentt artide published in 1973 on all women under the age of 45 with myocardial infaeetionss admitted to the same institution b'etween1970 and 1M Radford aad Olivee showed that 27% bad been taking oral o,n- teaeeptives, while indirectly, they estimate only 2% to 17% would have been expected to be taking them. Ev- ery patient in thia series had at least Oral eonaae.ptlv.a.Ory. 2s1a-e'lo:. e
Page 100: azo61e00
96 onep:edispming factor for coronaryarteryy disease; the average patient bad 2.6 risk factors, not including oral contnceptivee use- Their datashow that myocardial infarction in women under the age- of 45 years oemrr predominantly in,.women with one or more risk factors for coronaryy artery 111lease.. Also in.1970, Inman and associates' 7eviewed the reports of the Eiuropean Drug Safety. Committee on thtom- boembolism in oral oonttaeeptive users. While it is diifleuit to actuall!y. • quantitate, they observed a itatisti- cally significant (P-.01)) trend of in- ceuing-a risk of myocardial infare- tionwith increasing ,amounts ofmestranol.taken. The risk~of infaro- tion is somewhat greater in users of. 100pg prepantions; thann in users of preparations with less mestranol The risk is greatest for women ua:ng aSOµg mestranoL The same pattern is presentwhen comparing two dosages of ethinyl estradiol. While the authcrs acknowledge th'eprob'lems of these data, the dose-rdated increase in risk suggests a causal sasot:iation between oral routtaceptiveuse and coronary thrombosis. Up to this point• the studies of the possible association of oral contraceP tive me and coronary thrombosis had all been 19ae series or astcontroL stadiei.In.19T4, the Royal College of Generai Pnctitionea' reported on a prospective cohort study of approai- - mately 73,000 oral contraceptive users and an equal number ofnonusers Five users of oral contraceptives bad had myoeatdial!infateCions, compared with only'one nonuser. Corrected for age differences, this gave a risk of havingthe disease for ttsere 52times greater than th'erisk for nonnsets. Becalme of the small numben,. no breakdo.on by risk-factor status was possible. ReCent Repor4 Agaiitst this backgrotmd,. Mann and his colleagues reported the re- slilts of twostudies.in 1875'•' and one in 1976.' One 19T5.study itSclvded63 survivors of myorardial infarction (all of the survivorsidentifted in.two of the 15 hospital regions of England in 1968.to 1972). The other mmmtmica- tion reported women under the age of 500 years who died of myonrdial'n in- faretion in England and Wales in 1973 This latter, study included: only 2620 JAMI. Jtiufe 13. 1979-Vot 277. No. Tabla 1.--0iatribufion ol Oral Conlracepdvr.(OC) Use-t Aae• T 0.3 q.. - 04 4T . Panly. Mlj'. Cont.al MICanba/ ~Mtt. Controf~ N~.'. No. 01 Momen 13 632 2 Se 53 337, la 70 WomeneunentryualoeOCa.% 53.8 . 123 Oo 3" 1l.1 12 te:e 22.9 'Oan bonl IN,ten and Veayer,l tAer and parity-ahndanRed rlaa' ratlo equala 1:a; .a!% conadence 1ntarnl• 1.0 tei1. =MI OMitalea mymardiat wareeue. Table 2:-Dlsfn6ution of Oral Contraceptive (OC) Use' •a,9e. 7rt <u ea-u -~---~ ~-r-~ MI CeMrol Mf t:onhol No..ofwumen, 477e 106 107 Wonae Ourtenttr.urne OCL 7L 4.7 724 17.0 a.e RIeF. ta0o of w!n cun.nt oC - wen (sSx ennndeeeo idlerval)' isrlla.t) _ __ iat12-7~ 'Olu trom uann at .U.a LN.I indlcates myeurolY Itdarrflon.two controls• aged 40 to 44 years, who relationship between oral cmotraeep- were current vral contraceptive users tive use and development of myocar- Benuse.of.the inaccuracies inherent dial infsretion as one of cause andef- in making risk estimates from suchfect, theinvestigaton applied the small nomben, Mann et, ai gathered relative risk estimates to the yearly' information on more deaths and com- death rate from myoeaedial! infaro- bined the two series in the 1976 re- tion for women not using these prep- port' Only the combined series is tiis- atations and estimated death rates cusaed here, for women usingthem. Combining In the combined report on deaths tliese.rates with census data, theyy fromm myocardial infarction. 3% such estimated-that for women 30'to 39 deaths were identified from deathyeaa of agenot'using oral eontfacep- aertificates. Ultimately. 204 of these tives, the death tate from myoardial deaths were the basis for their report. infiretion is L9 per 100.000.nonusen, The others were excluded ntegorical- compared with 28 times that (5.4 per lybecause'aflogisUaproblems,.incot• 100,000).for women taking oral con- rect death certificates, and various ttaceptives_. Thia.means that.there is other reasoms. Ecaept in the cases of an excess of 8a deaths (5.4'-L9) per 26 women with no physician, the.rea- 100,p00.users in this age groupIn the sons for exclusion weree unlikely to be 40, to 44-year age group, . the death related to oral mntraeeptive 1me. In- ratess are 12 and 32 (;-12x -2g) perr formation about the deaths was ob- 100,000 for nonusen and users, re, tained mainlyfromy the general prao- spectivdy. This implies an excess of titioner, who used a. systematieZO deattis.per 100,o00for.users.in this approach provided by the iuvestiea-age group. to7a to select two living women, age- In the studyon survivors, there was and marital-status matrLed;.as con- no selection process in aaeepting ttols for eachh ease. cases for analysis.' AB 63 patients The data that bear most an this dis- aged 25 to 44' years who had been enasion, areabstractedia Table 2 treatbdforandsurvived amyoeardial'. These data are not theresult'of dif- infarction fmm 19688 to 1972 in two feresces between the rsw and eon- hospital' regions in England and trols with regard. to social class,:. Wales were included. The investiga- parity, prematnre menopause, or by-to7s obtained detailed information on pertension and diabetes under treat- risk factors for myocardial imfarctionm menG However, the investigators h'ad. For each ose, three controls wensa- no information on cigarette.smoking, . lected at random from women who lipid abnormalities, or untreated dia- had been disoharged'' with certain bete or hypertension. Accepting the acute medical mr surgical aonditiona 2a Orat Conbacep4ves-Ory
Page 101: azo61e00
M 0 w 97 Table J.-Distnbution ot Cunrent Oral Contracepnw (OC) Use By, CurrentCigarette Smoklrlg Statua'f N-a/ vromen . OC usen. % Nwes smakeo I~ p M rol 16 14.4 Risk ralie uf MI in curt.n4OC useN (955 nenridanea inlenalt. 2.0(O.S-0.61 -nau trom Yann at.6s 1NI1 indfGlaa mreeardia/1Mare0en. 1.442_o-14.71 Table C-Exeess Risk of Oeatn From Myoeardiat iIntarcnon (Mq by,Age, Oral ContraceptNe (QC) iUse„end Fredisposing Conditlon. All Wnlt Condaana WiNOUI CondWbna Aqe.7I .~-~ .--~ r--~ ao-af ao.fl 7Fif{O.Ia 7F3l. ---• IOJa NGGI wom- in Uni7.d suttl(x1o+qrt 11.52e 6.162 ],sa 1JMlf,o7o ..01a r..dyno..d.a1Tal.emMiV/ 617. 1,107 .e.aaa 123 22t Esu'mauo MI d.aln rate/10" 5.4 1a.0 ta3a7.e 13 0.1 R.wfr- nak of dlam Lar OC us.nt 2.a 1.1 20 20 EatimatuG death ratet lof. OCuaen 40.0 134.1 . 3.0 101 Eatimstw deau nr. aotib- . utabie lo OO uset 25.7 66.2t.s 5.1 . 'Rale eauau yeany dntns dl•idad by number ot women tn. Vnitad Staln thnn 1.OOLL7/late louab INaove nsk oP dealn ler OO us.n timu.slimat.d MI Ceatn raua- tRale puala .stimated d.am Ma tor 00 wara minus tlelnelad MI d.atn ntt and arete matched for marital status, five-year age group,.and year of~ad- misaion- pndings The teostt striking finding in the study of Mann and crolleagues' has ooth'ingtn do with oralcontracep- tives; it is the strong assodation be- tween myoeatdial infarction and cigarette smoking. In comparison with nonsmokers and :ez-smoken, the relatise risk of myocardial infarction increased from L3:1 In women smok- ingfewer than.15 cigarettes.per dayy to4.5:1 in women smoking 15 to 24 cigarettes per dly and 1191i in women smoking 25 or more cigarettes per day. The z' value, based on a test for lineartrend, is significant (P- Sx10"). A atrrentttser of onl contracep- tives.is CS (95%confidence interval, 21 to.9.5).times more likely to have a myocardial infarvtion than is a woman who is not: currentlyusing oral contraceptives- Adjtitstingg for the effect of cigarette smoking, the esti- mate of the ratio of the risk of mya nrdlal infarction for oral contracep- tive bsers and:for nonusero goes from 4:SL to 32L Adjihsting for allother risk factors predisposing too myoear-dial infarction on which they collected JAMA. June t'S, 1977-ve1.297:.No:.2a data (hypertettsion, diabetes, P'Y- edampsia, and ob'esity).made no fur- ther change in this risk estimate.. Instead'of standardiang for theef- fect'of smoking, we can look at, oral mntrariptive practices of patients with myocardiat infuetioos and,con- trols by smoking status (Table3),~ Among nonsmokers, oral contracep- tive asers have 20 (9Sb confideneein- terval, OS to 8.5) . times the risk of having a myaeardial infa7etion as nonuaera. Beause the confidence in- terval includes L0, chance variation is a possible explaaationof this finding. ' Among smokers, if a woman is an oral i mntraceptive user, sDehas 5.4 (95% mnfidenoe.interral, 20 to 14'1) times the risk of having a myoratdiil infatetion than if she isa nonuser. This resulU is highly s7atisticalPy. sig- nifinni. (P-.001). Since themn8- denceitVtervals overlap, we cannot~ say for certain that oral inntsaceptive ttaera who also.smoke have a greater risk of infarction than nonsmoking users; however, the data suggest tharthis is the wse- The data reviewed from the report by Mann etal on the survivors of in- fa7etion lead too the following conclu- sions for women Ikss than 45 years of. age:(1) smoking is the strongest risk factor for myocardial infarction; (2). Cur.mt SmokN MI Conirot 7a at 76 10.3 31.7 79 for nonsmokees,, oral contnceptiveuse isassot5ated ith a two fold,, but not statisticallyy significant, increased: risk of myocardial infarction; and (3) i for smokers,, oral contraceptive use appears to be a substantial!additional' risk factor for the development of myocardial infa7etion.. Tb better understand to whom these results appl-v,,considerone more of the findings. Ia only one of the 60 survivon of myacardial infarction thatwere studied' is oral i mntracep- tive use the single risk factor' The implication ia clear tbat,tbe risk esti- matesfor myocardial infaretion apply only to oral contraceptive users who have at least one additional risk taca. tor. That this is the lase is not surpris-. ing. bfann and:colteagues note that Oliver' reported on all 81 women with myoeardial'infarction undertheage of 45 yeaa vho nme under his . care from 1953 to 1912. Seventy-nine per oent of thesee women had atleutone of the three major risk factors (hyper-choleaterolemia, hypertension, smok- ing one pack of eigarettes or more per dsy)known to be associaied with the developmentof myocardial infarction in men and older women.'• Mann and colleagbes go on to note that in their own study.'of myon7dial infarction, oae or more of these faMon were present in 80£% of the patients for whombiood lipid levels were known, compared with,.2Y.5 of controls. Fur- ther, in the 1968 report tiy.Inman and Vessey,:69.(3i%) ofthe 205 deaths from coronary thrombosis oceurredin women with no predisposing: factors.('1'hie percentage would'undoubtedDyd b'e lower if smoking had been taken into aacoant) . The data of Olivet' Mann and collesgues;' and Inman and Veacey sfionglyimply that apprmd- mately 80% of ibfarctions occurring in.n women [ess than 45 years of age omtr in the presence of either hyper- cholestetvl6mia, bqpertension,, or cigarette smoking. Aceording to.the data of Mann et al,' about 66% of the wnt7vL popu- lation have none of.these risk facton.Since the controls should approximate th'ee general popnlation,, we can roughly estimate that 70.°a of women of reprodnctive.age are free of crondi- tions that predispose.to myoesrdiai infarction. At'Shis point, we mn make the foi- lowing statemente about women less Oral Contrac.peves-Ory. 2621
Page 102: azo61e00
98. than 45 yeara ofage:(3) Apprmo-. mately 80% of the women who had'myoaedial'infatrtions bad one or more conditions prediapaing them to ' this disease. (2) Approximately 10%of% the general . population of women, of this age have no conditions pre- dispcaingto 'myoardial infarction. (3), Women with predispoaing condi- tiona who use oral!conuaceptives are 28I times more likely to have a myo- ard'lal infarccion than are similar women who do not use oral contracep- tives. (4) Little information is avail- able with which to estimate whether oral contraceptive aseinrieases risk of myoardial infandion inn women without risk factors. The data from' Tables I and 3suggest, that oral eon- ttaeeptive usem are twice as likely mhaveo myocardial infaretions as non- users The anfidlna interealsfor both of, these estimates either includlor touch L0;.th'us, no sure association is the best interpretation of each of th'esesetaof data takenn separately. However, because they both point in the same.dilection and show risk ra-- tio estimate'aof thesame magnitmie, an oral ooatraceptiveuser free of other conditions ptadisposing to a myorardial.l infarction may be..rivice as likelj^y to have an.infarction as a nonoaer. Extrapolation of Data With this informatim tEeUnited States ansus," and vital'statisties data,,• th'eeccess rish.of myoardial infarction by age, oral contraceptive use, and by presence or absence of other predisposing conditions was es- timated (Table 4): Fach age group for women with, . predisposing conditions contains 30% of allUS women.and 80%of women who died of myocardial infarction. Sina80%ofthe women in these age groups were not aurent users of oral contraceptives in M0,'T' the death rates given are a good esti- mate of thedeath rates from myocar- dial infarction in nonusers of oral an- ttaceptives. hfultiply'ing these rates by the risk of myoardial infarction in oral contraceptive users gives an esti- mate of the death rate from this cause in usen..Subtracting.the deathrate for nonusers from dhat, for users gives an estimate of the nteof dnthat-tributable to oral contraceptive use- In the absence of otherpredispa- ing risk faeton,.cral contraceptives appear to cause only a small number of excess desthsfrom myoaedialin- farrtioa In fad, these excesses an similar to the SSelaest deaths per 100,000uxrs per year that Inman. and Vessey' calculated as the number of deaths from pulmonary embolism for oral!contsaaptive usen aged 35to.44'yean. The situation is fardiffesent for women whoo take oral contraceptives and have anyy of the three major risk factors prediaposing to myocardiai in- faretioa. Above the age of 30~yaara,. the risk of deathh fiommyocardlal in- farction attributable to ornlimntra- aptivense.appears to be at an unao aptably high.rate. Though based on the best estimates available, the numbenin Tab1e 4 are speeulative: 1Vhat is not speculative is the mndauion. If one.man.ipulates the g01Z0 and 70/30 spLih, one sees that, astheybeame more extreme, the risk of death attributable . to orall con- traceptives decreuesfor those free of risk factors and increases for those with them. On t}ie.other hand, if one varies the ratios inn the other diree- tion, the attributable risks do not vary enough to substantially alter the condusions. For example, using In- man and Vessey's study' showing that 34%~ of .. women who had myoar- dial infatetions bad: . no risk factors and derreasingthe proportion of women free of risk factors in,. th'e 40- to 44-year age group from.. 70% to b0%. the.yearly death rate from myoar- dial infaretion attributable to oral contraceptives would be 1Zper 100,000 users (compared with.5.1 un- der the assumptions in Table 4)_ At the same time,.for women with pre- disposingconditions,.the death rate attributable to oral contraeeptiveuse decreases to 42ti. This does not alter the conclusions drawn,,however. ' Conclusion . A.erall, this review suggests that women free of.risk factors who aremore than 30 years of age should, be informed that using ora( contracep- tives may slightly increase their risk of myocrrdlal infarction, as theya=e now.ihformed aboutth'e risk of stroke and:venous thromboembolism. How- ever, thesewomen~ should not be warned.against.the usa of oral con- traceptives. On the other hand, women 30 to 39 years of age whoh'aveanyof the three predisposing condi- tions should be warned againstusing 2622 .W1.t June t3, 1977-Vo1237;,NO..24 oral oontraceptives..If they are 40 years of age or older and have.any ofotherf predisposing factors, they should not.take oral'contraaptives- If a woman wlioo is more than 30 years of age is a smoker, and that is ber only risk factor for myoaedial infaretion, she could be counseled sligh'tlyy differently. If she stops smok- ing, her risk of myocardial infarction would probably ntttrn to that~ of a nonsmoker and she would again be a good candidate for oral contraceptive use. However, it takes, as of yet, an anspedfied:time for h'errisk to recurn to its presmoking level. Until more data is available, each physician pre- scribing oral antraceptives.nill have to make a ase-by-ase determina- tion. NNennees L Nann !LIpnu. wHtN: pat nnveeep- tl.u and dnLb f'awe m.arWf.i'inlartuoa B. L"ed J ~•iL33S lYi +_. Y>ipp JL vorrr..17P. Thavroad X.e ak ajaeuaul ieLraien in..amg - s,ib..pe- o.l, nf- to onlaveueeqtire.praoua B. xd J 2341-:15,1Y3. Y Yann JL Ineme wRW, Iyme9aed 1L Oral ooavrnpdn ur ip oM.r.amen.apd (a1a1 m7s e.rit+l!inf.neuen. Br Ifd J_61b-L+7, 1M l.lAmu 7P7ttt. V.nerliP: Inresi'aauoa a(dplif rraa- pWmanuF: weonu.. .nd artbnl thnmbnsir aM mbuli.m ia avmn of ta0d- bearioe .ae- Br .Y.d J °_199•199'. 19aa _ S..t'.aey,3iP,ball: B: lmvsrirrtion af nir cpe bnaeen o.e af oral'l mnvu.plin. and tlummbormbNie diSeaac A fcMer mpaet Br,Ved J.2$51-677. 1%9. . a. 01i:.r 1iF: Or.l mmr.ceptl.n andmyoor- dial iinfarruon. Br Jlei J 2:1'Fa3,1ee9.. 7. ItadrondDJ: O1i.er )1F: ON annvanptirer and m7ocarolsl inraeadan- B...Ve{ J 3:43-134, ts1 a lapue WIIW. v.aer XP, wenerbelm e. a.t Thnmbnemboiie diaear .d~b'd neevid.l mpunp Yf oral 1aaCU.npeives .{ nporem.uw Commince on Sadety.o( Druga Br 71rd J t'AL 22. lro: s. Ro-.al. Cd4r. af G.nenl . Pnaieanus. Ora! Caa+enpeinr a.d Needtk .4 PMiwi.arr' Brpert: IN.r York. Pimun 31.6.0 Pubiiabina Ca LUL 1974.- ta Ni.viaw.o5: Simp4 i-af cimuian - or titY nYa. A.. J6pid..iat 1m03ts:19:4.~ 1L Yaacd N, g.eaaml. w: 5wtsiol a.pw ot the aeal7rtr of daM feom nunp tin nudin ef d'uene.l b'e1N Candr 1+a/ 2271474s, 193s. 12. Ysy YP: Don IL Petn IL ee .L• Cb.rao- tunstie of .omen usinr df6enns m.uhpda of mavsaeptine:5an. q:l.. .~mnurr 6ndiege fmm.. pwpeeti.e .udy. 1.1'J Epid,wid 1:113M . u'rt 11 Gbb.r G, DpD $ Faiah.i in A3.ra! P.p de ularation and lhe pse of oeal mnnsnpd.es A nea.u.emrManwn anribunb4 totbe di.. eWE BrJ.As.at Spe bfed 25:1141K. 197L 14. Inuraaae.Y Canmiuiap ton HeartDia• --- - Cindatiw 42'.4i9a. 19Yn 15.Cnueu ej.%VWutiea lM C..r.vl Pepr: 1atiaw ClarpetmaHq Ffne! 8rpoee-; P'abl4stiop 1-HL 1S 3- af ehb Censu. 19Z 16 Vled 5faebelo.nJG. L-na.A 3tatu, IfT6- Nweaiilaput8.: \'au-l Cevler (or fl.auh 5latisnm..-o1 1: 1Tl' 17. wesmnCF:73enmderniaelaeofl5nn. era.pa.. peacrtita Fo.. Ra.. Prnp.a 4S1Z 15'ft Ora1~.Contraceasves--Ory
Page 103: azo61e00
PE' $~~ do g. $: n .~G4 ~~~pg~ . tU1tI aSEc~~ ~ a too v P.a ~ C p o. ~ p 0' ~ p• °~ 0 0 pR~ iN9! d p~'r Op & G' Sn r ~ ~ ~g s xr ~ ita ~ O P. T$K Fo~l6~~~~p$ .C C.0 tl~ E~ p@6 ~.6n~ ~a! 5 ~a.p~'J.~'aB~Op~g~ B~,66g~'~pe p L g4 °~ n na 0 ~ to U N ~ 1 asi a ~ o pE~E" li N N q i ~.a~. . ,. ~ ~~e~e >~ ~k?~~~~~7~~~~~d~'~ a d ~3 . 4W&. O ~ r tAa'°r' ~i a ~a~ to, ~ ~ ~~ ~ ~ E ~ 1?. "jo2 &~q~ «a y a 4.j ~n .'~'~~~ j~ ~.-"G~ ~ H~$ ~ ~Gtr °~ ~ ~ ~ Ra ~ .~. @ .8 xRp. dE& 0.~ YoB9p~ ~ 6.ad l S °Fy~ ~i ~ ~d. i 'l~ - ~L6'Pw A'~' 6' ~o sp na A It I ~p 9 p1~ d 4 1t~L ~ S IG~ E C. ~~S 8 ~ _ i{ E, p. g~. m -~1 3 j a 6 ~E 7~~ ~ .o~'A 5°"~yI Q~n ~E~a„ ~ n~G~ a1 ~ R ?s~p r r ~ < t ~ u:3 i ~ ? ~ D~~ da ~a fi~~~~3 6? 0~~~~ ~a9a~as~~ o $~ ~~ P . a ~ 6C.IQVSEO m 4 dx trw' 9 ~ I I I
Page 104: azo61e00
r 03590140 e, ~ , i 0 0 ~ i ~ I ~ , I ; 0 v f .: n I l.~ilt [] v .9
Page 105: azo61e00
101 747 0 THELANCET Mortality Associatedwith the Pill Tstu weelt's Lancer contains reports from two major British prospective nudid oa the iaAuencs of oral oactracrative medication on the health of usera.-DrEatut. and Dr IGY report on 101,dcaths in the ~6 000 womee taking part in the Royal Co1- lege of General Praedtioners ()LC.G.P.) uudy, and Professor VzaseY and his colleagues have anaiysed 43 deaths in a series of 17 000 women attending Family Planning Association dicict. Both groups of workers found an inaased mortality &om dia- aaes of the circulatory system in oral-convacep- .tive users, confirmiog the results of retrospective studies. The adverse efecsa of age and smoking wex noted apin in the R.GG.P. study and'the risk of death was also related to the duration of on4contraecpvve use. Some important findings emerge from these two papers. Retrospective nse/control'studia have not shown an inereased risk at myoeardial infacaion" or strokd in exrssas-i:e:, in women who had used oral'contraceptives in the-past but were not current users at the omet of their llness. The mor tality risk from drenlatorrd'ucax in the RGG.P. study, however, was inereased: about fourfold not only in women who used oral!mntraeeptive duiing the month of death, but also in ez-usos, Vasszz and his eollaguesSound 9 deaths from ardiovas- cular eausa in the oralt{ontraeeptive entry group aod nooe ia-those-womm who had entered the study using physical methods of contraception.Of the 9 women who died, however, 5 had aopped taking the pil1- before the terminal ilUsea Meta- bolie changes occurring during oral contraceptive LH..S 1.LC.rhLL /.71•.7+1A..t1A L4. r1J..I1TLi 3aL L M.~ I:L t.r.l. x.l. a:r.r. r.L L C.a~wee Giw f. W A./f.! fuM, : h+i l.ra X.. lyi 7.Yi ITTLSY.an. mediotion geaaally re;ress 2-3 months after the pll' is discontinued.' The present findings imply that certain, as yet undefined, circulatory disorders may persist for longsr periods. Furthermore, while prcrimu rencapective analyses have shown an in- aeased risk of death in current users from wefl, dchned circulatory disordets such as myoc=rdialin- famio0.r pulmonary embolism,° and cercbrovascu- lar disease,l the present studia sugYess that oral contraceptive medication may predispose to a wider spearttm of circulatory disorders. In particular, 9 deaths from subanchnoid baanorrhage occurred 'm the R.CG.P. oraltontraccptive "ever-usd' group whereas none were observed in the controlsh lllthough the numbers of deaths frmrc specific eon- ditions in both studies were usually too small to achieve statistical significance, the estimated over all escess mortality from drculatory, diseases in ever-usea (20.3 deaths per 100 000 wonuo-yean in th: RC.G.P. study),is higher than previously cited' figurc This estuaate dbes agree with B[tut.'s earlier analysi3' based on mortality-rua fmm certain circulatory disorders in young women from twenty-one couatries. The number of deaths analysed in the R.GG.P. study is small eompared'with the numbers farmiisg the basis of published retrospeccive case/control uudes, and'the estimates of acas-mortaliry rates . ia oral-contraceptive users have wide margins of error;.sonse indication of these would have bem hdpfuL Furthermore, some of the earlier deaths 'st the RGGP. series may have beea associated with the use of pseparations captaiaing more than 50 µg of oxtrogcn, and these are known tv cany an ixr eeased risk.' The absolute number of daths, how- ever, is relatively small and such risks are ditScult to esplain to the prospective user. In 1973. the mor ' tality-rate for deaths from mad traffie acaideats in England and Waia was 4•6 per 100 000 women per year for women aged 25-34 and-35-44 yearss This figura is similar to she excess mortaaity from circulatory diseasa in the R.C.G.P. study for womea aged 25-34 years (4-4 deaths per 100 000 woman years) but considerably smaller than the rate fornider wanien aged:33-44,(33•o deaths per 100 000 woman years). The highes ezass-mortal- ity rate in older women confirim a«ttmpecsive analysis of deaths from myocardial infan:tioa in whiFh ilte ezceu-mortality rates in oral-contraetp* tive users aged 30-39'and 40-44 years `.erersclc- oned at 3• 3 and 20 deaths per 100 000 women per year, respectively.s The RGG.P. uudy a;ain cnphaaises the impor- ...ly.J.e. N. cY..[.l~..IfrW.lmr. )01. t, ~w.µl.Fwr,.Kr.a..ri).tM1.'itfs. • Nnd.4.l.w.w.l,T,.aaaT.. ta-l. IL !- rw.f. M r..wnYdi~ a. r+,i7. A F. ~). 11AaAL L a.p_ Qw....f. ]u...d 11....~ .! [wd~ r l.1. Ir W 1~ 111LtIJJ. ]/-TlMln. f. M.ry j 1.1..r. ~. M. ~-fl~rqr.KM.K7.IfH. ti ~L
Page 106: azo61e00
I 749 tattca of additional risk factors for ardiovascular disease. There is evidence that these faaors, such s0 cigarette amokins, hyper[ipidania, diabetes, bypettension, obesity, combine with oral<onua- eeptive medintion in a synerginic rather than additi.e manner in certain dreumstancea.l' It would seem wise to consider other forms of contra- eaption for women with more than one risk factor and also for those in the older age-;toups. A study by VasssY and' Dot.Lt' of oral'eontracaptirea, in- trauterine devices (LC.n.), and the diaphragm suL- =eua that the ccas-monaGty rate for oral eontra- eepti.e users in the United Kingdom may be two to ten fold greater than that for women using the Ltt.nJdiaphragm, depending on the failure-rates assumed for the various methods. Entirely different ontdusions, howeva, may be reached in countries with higher matcrnal-mortality rates and where the population may also be Itrs susceptible to vascular diaease. Further eaperimce &om the prospective studies will allow drmv estimates of the risks of orai-:emtracepsive mediation. On p. 757 we publish two assessments from some of those closest to the fatts who have so far had a dsanee to examine these new data. Their conclu- sions are that renewed caution, moreor less, should preKal Those who havn e~rea mase dilfiealt judfi ments to foem am the doeton who must decide whether or not to prescribe the pill for a particular patient and also the women to whom they must explain the tis2s L3ttlc in the week's banrer is going to dismay the many women who regard oral cnntraoepti.es as a blessing which carries a minute risk of prattature death. The doaors whose task it is to interpret this ne+vs wlli have to say that the danger now appean to be ptate than it onee seemni, but it is still, in the absence of other risk faeton, very small indeed for younger women (and the su;gestcd age for reconsideration of conttneLp tive metbod ia 30-35j. Aplastia Ana:mfa: Seed or Soil? • To the cliniaan, aplastic anaaaia is peripheral- blood rytopeaia with chronic fatty atrophy of red marrow. To the esperimental haastologist, it is _ the clin3¢al ezpre>uon of reduced output ftom the - ltamtopoietie stenreell pool into the more mature, ditferentiated lampattments of the marrow.l "13ere is, as yet, no technique of counting the ancestral, piu:ipoteat edls in hutma marrow, and we must Il W..*Jl!-t#qLhaR_Jw.L 1174IMM. LN..wI,'11.X.Y.LLAwj.x..r.lns.Jt.-1ltF1r. 102 itta t..rcxr, ocroant a.1977 rely on assays of their closely related dacea, dants--thoseapable of forming granulocytic and' erytbroid colonies in culture. Most cases of aplastic anacrnia show reduced numbets of tucS colonies.=- Iraving aside irradiation and cytotosic drng; e:o(;cttous toxins can be idcnti6ed in some SO^o of cases of acquired~ aplisia, though th'e manner in which they interfere with stcnrcell replication and ditferentiatioaremains ann c.v=ma.. Some agents known to cause aplastie anzmia, such as chloram- phenieol, benzcnG and theh'epatitis virus, am cp- able of producing eh'romosomai damage in cultured marrow oe11t; a moreover, the persistence of mor- phological abnormalities in the marrow of patients with partly recovered hypoplasia,t and thefaa that a small numliarlater prucr.d to pamrytmal!noc• turnal hasnaglobinuria' or acute leukaaia,' maybe cited as evidence of permanently damaged stem cells. Ncvertheleas, there arc other possibiiities: for eumpUe. Booos ind Booasl• have postulated a dislocation of stemtdl kinetia in chronic apiuia, so that didermtiation occuri before the parent stetn-csll pool is adequately replenished. Again, agents causing aplasia might do so by producing changa in the humonl or cdlular microatviron- mettt which render it inimical to stem-ce7 growth. We have little knowledge about regulation of the size of the various stemtell wmpartmcnts. No bumoral Gaor is known which tsgulates the di•ti ision of pluripotent edlss; indeed, recent work has ehallenged the cottcpt that the hormone erythro- poietin conttuls the miiatioo-rene of the "eom; mitted" crythroid siem txtl,ll though it certainly iaAttenees maturation . of its prcBoY. and ImEn cases af red-cell apiasia am due to interference with its aaion.'r The cellular or "ttromal+" eomponeat of the miaoenvironment is also poorly understood, but there exists an animal'modd which iliustntes its importance. The Wll9` mouse and the SlSi' mouse both have eongenital'aneznia. The former has defeetive stem cds while the lauer has abnor- mal' marrow stroma; transplantation of SllSI4, or normal, moase marrow will!aue the anamsia of the W/CC` mouse, but neither normal nor W/+JP mar- rolw wifl cure that of the Sal51O mouse. However, implantation of W/W' spleen timte w-ii1 improve the anania, beovse the SUSie stct cellsran thea grow within the matrix it provides."-1• f:xosrs L G.w+wyr. Lsa+..s. La/.d ifn. a.n7. L saar.N.ry.. n.Mw.e-c.Wa.RJ..j.x.- ltn:]s.abL .. H..r4 e-a.A. L lY-.4 M. Hn4 s.;a.wi. L i:L 4Y M. }.y; 1.11.a{': LiCe4nayx.IY.L LM li. Yf.. ~ aLNL0.i.A...Y, t } xwwf.., IMf.Ji:at: 7. r...,. C n, n,.r.. a Y. $..r. r. 1C. Nt...1L A. t+r1..t +L l ,.. j.x....~ Is,..n. tn. .. -Lt.*swp.rlr.da.lf.e.u>x !. w.....M.las....aw.~.x- r-el.r.+.lw.-n.laeL le Il.,p.n a_nq+saalu.nrt,4lt. I l.lr.a K K Cvq r....Lf....t Iflf.la, fa IL IMb. C).. j.x.-w.lf)l.11:-wLI, Ml . ILrus-~ L+L.s....wsLTi}~Law.4La.L- a.L/IrLIMLY7w. Iw SY....s LAa j.s- IH4ttf."Is. I I . t
Page 107: azo61e00
/ p IIiIHUIHU!:!11 o d g" e o° ~' ° x s:+~a~rg L'~EsB~~~g d~~2 ~~` ~4 ~g~~s , 3t ~"g V~ 9~5 u ~ ~sx~~ a~ ~St3a_s~~~ ~~ ~ E Ysa ~Yff"v~ ~~. d o~5 EB~~~'=.11-~YlESI aY~.a< ~ ~ 5 ~ ~ lg 7~i.=s" ~v d ~ ggX~8 ~ e_ w o:ar ~.a E C~ `~p;3 ~ 8 p s e=~a D o t7 ~ ~ 2~ 1.9 l ~7S'~E i~t~~~~~„ig~~ 5a. ~~ 8g5E.5Y~~,~P~a ~~ S~, " ~ ~'6~° 5t 1 ~3~~~d1~ ~cn ~~~~~~ ~~€o~s~ ~~- ~ 9 03590143 y
Page 108: azo61e00
03590144 ;~~~k ~ ~ ~ I ; ,WK I N S Z4 T. ~ a ~~ i ~aes~~a& .. ~ A ~,~4 ;.a.~~; _. ~~ ~ ~ gd -y,~ -"^9~; e~ r ~y,~~•.~I~,;~~.AM $~op ;`~~`dr^ .~I.f! ~.9 a~ ~ t• fwn i FA tl:,, • ~ J{ h ~ 3 !1_ o 1 . . u2! ~ ] w ~IL#'a KH!$`dei`~~avK~3~BJ `.•9 I _ ~'/.ll^~~e.~,~t~ Ji~~iE~~ v'92l:9~ 3z''~ gJ s'"~~5~ z'.:\Fl\'lSl(f~t~~ '~ '~ $$$ N d o~N S,~~, ~ o~'7G.~o%_~.. z u;~ ° a 00601 ~ $ " ~ ~ ~ A _ ~~3 .l ~~_~l j s _7Ti1'Ri {c : a. '~'`!WItTO! 11, i • ~ ~I u. i'.`K.~ C`rdr ~ 4.~
Page 109: azo61e00
105 t l 1024 Sn..-Your fssnt of Oa. 8 mnuissed two epidwlologieab nsdfe rcmding a higher morsaln.y-nte fremheart anddmr- latsets diserdertamong oral-traeeptive uaen than.asnong ssmt-sssert Higher soonahty-rnn were obser.ad amorsg older wotnrn,;heary smoken, and/or those who:had taken tba.pi8! for ownthan Svcan. The mssclusion was thatwamen ave 3Ssh'ould eonsider ahcnative fonm of~contracepcion, - sug- gesrion supponed by, the presidmts of the Roy.l!Collegc of Obmevidaas and Gynseo(ogisum andtbe Royal College: of G<neral Pnaitioness and, editorially, by 77Ye l:nscer and the Brviea McdirnlJavranf. This seerm lo us asrhjenive inunprnnion of the finding.. The data, apetia0y.thwe:frttm.ttie IIC.G.P., are dilBcuh. to undlnsand It is eot ckar-what is.hiddtn in tha sundatdised'.alsro b'ned on 100000 women-vean. Data feam . Cewwateenm faBowed up forw a long timc result in the same ssan- dardisednte for1D0000'womea-yean as the data of many wonsen foBowedupfor a shonn periud.The raaes Ihereforee canrqt be traced back to the origiial data, and the,v aBow only aaiooorssplete rnrieval of the basiomformation. Epidaoiabgieal studies, whether prospeclive or rnroepee:, ti.e, do not permie uumleasu about causality. They nn only point todidaences bnweeo ttie obwved'group.:, By sdF-selecliot4 which is how thedeosion to use ae oral crontraacptive mssss be regarded, the eva-usm an a speaal postp wDo diHerfrom the oe.er-wen,and awt iten in their methods of rnnts+ecption..This is mentioned in the R.C.G.P. papn. In both studiasllie ubin ahow that evv-usenand never-usaa are diMenntiued tiy.the frequenry.of nrdnomas, amdenu, suiade. asdagarette mnsumpcioo.. Thener-usen comprtsc uscn aad ca-usen. Fdr.these sub jeets the risks fnr all circulation diwrden show the same magai- tude-namely; 26.8 and 23.9, respectively, compared.wish J'S dnths among nwer-usen,.in 100000women-ynn (table m in the R.C.G:P. actide). IA your ediuorial vou referto evidesKe that the ary-.o.er dfen of the pi8 aflv discontinualiou. lnts a few weeks only, and n most:3 months. Itu reasonable, therefmre, to assuose, Ihat the eaeela monallYyeeen in CiYiera is dtK ln ididerent pnognosis of circulatory 6sorders in the ever-ussr group cam- parrdwith cnmrolsThis di8'nenl prognosis i. seen allo in oehemndiiians. Even if the pill ieausn bypenensidn, this calsnot be sha only esplanation. Eavseesnill hadahigher monality-nte from circulation disorden than nevcr-Iners cven when those with' hypertcnsion vrere e.el5drd. 71sis finding ean be eaplained by an a Driori dildcrena in progaosis anangevec-usen. We arc still of the opinion thst the pfnmptlon of hosmanal eontraeeptives msut be carefullr mnsidered inn every,case aad muu be adiuated to nn indi'vidual oeeds: - AfuC,nl'tnlan„re, - KUtE'f DERan:o Wskammmrer&aaer!wSi.ustia. EbaAFti HAaTalAlnt SrAUxa AG, DI gehb af; iw, Gasur Sta,-The posaible synergistit eRen, of ssooking andorab ontneepri.re:use in premmopausal women is worrnng: The ntensive pmspeaive study by the Royal College:of Generai Pnnroionen prm-id[r a hne opporcunity in invntigale the rNationship of age, smoking,, andpilti usage. The care with whicA the studv was sn up in terms otage-matehing uun and non-users is ro b'e commended. tL'e are mncerned, however„with ,some aspeas of the data handling„ andwouldd iuggest that the daua be sausinisedfunher:. No attcmpvwas madeto asuss levcl or duration of slsn6ng (table lv); If synergistn is real the length of time the ~twoagenu have to intencs would be npected to be rclaled'di'r- ectly lu the inareascdn4 of circulator,v diaorders. Alio,.tbe Ievds of smoking wculd tend to incrrne risk in the same direc- tion: Other vudie (e.g., Jain's') liave implied such a dose-.ee- L.Je,u,AL.1a.l~oeut4Gme..lef6ail.)nl. - nrE uxcrr; wovcxgrzx 12,1977 pnnse edasion. No dnailed'aecount of the relauionship bnvxen plt -ge:and smoking within agr-gmups - gi+rn (table v). The focus of currenu concerts islhe ovn-40 wvm.n.who uses oal wosraeeptwn and smokes hea.ilv; buu the R.QG.P. dita do.npt relita lo this craiucal quntion lpccihwDy, despite Ihe reeommendations from Dr Kuenssbergand SirJohn Dewhuest (Oet. 8,;p. 757).' On theane hand, the handling oCtlata ussder- slatn llk vast diHneaees bnweea the hca+y-smoking or.l~ truepsi.e Issen andssonrsnalliug onFcrontraceptive users o.n 40. Le4unanewered, on thk otherhand, is the question of telf- sdeuion. ls it possible that the wosncn who smcke anduse oral oantrasapsives (particularl y,the oidc group) ansimplyn reila- tiwe ofa more Rambbyane lifestyle which nmy.wcll indJde more stress, mon akohoi, sootme medikatton (induding „downers'• and ••uppen"),a marihuana use+ Were those fae- ton looked at and adlusled for in the.dataanalysis?Theresenombe noth'nngin thedata which would answer onee.andfor all the qyeniuoe-is it the smokn or the snsoking which enue. riskk diRerence and u:it the oral contraceptwe uaer or ora}eontaacrysive use which is at the bonom of il a11?.Some- ho.wepdemiobgisal and Iaboratorystudie mustbe ddsigned to distinguish among the possible stiologies. lst us not Iel pre- conec+ad ideas alfeat nbiecsinry - eaamining the possible hyposhesa.. The nu: of Canan (ineludiog gennie) , mayy be more complQa than we think. Until these quesunsss see. resotied, however, we would hot disagree with the recmmiucn- dasions of Kuemsberg and Dewhum.. 11w.enasA,en. . IaMrre6 tllana aesvea Inu„ne l'.nn+n.dCm,ucks:. ta,we,a,. een.ck. tefM. L~ SA. D. G. HAACa: H. F..McKua Sla,-Thn papgr fromthe R.C.G.P. givn age-spcei6e dnth,ntq: It would be:inoerestingao see howthe obser.ations of, . Profnsor t'esser and his colleagues : cnmpare with . these naes, hut they do not give yean of:nposure tiv age-groups so this nnnot be done. I should like fo see this estra infornution nsade availi.ble for compvasivepurposes.. Ia.,.en d Canq,une ena SIaiN4: . a,m aeoree a<one. Nu,w,Afmmsullvl I. D. Hsu THOMSF1t-FRIEDENREICH ANTIGEA' IN HAMOLYTIC-DRlEM1C SYNDROME Sia,-The hewIrJc-nrz.mic syndeome' is usually.sern in infanu and ft is ofien fasal- lu ause is obscure.: McCOv ena).P found IgM deposius in the glomerulij usfng a 8'uoresttncetech- nique, but thq did not sa,y how these immune<ompleaes might be fomned We have- mvesugated two caan and! found some eviddnce for pathogmic fanors. Two 11yearold childden- died frsan p.nemnococcal psseu- mnnia wish seppis..Both showed the typical ceurse of hamuly- tie-urame syndrosne with harnolYtie ammina and acute kidney failure. Neeropsy.rerealed necrosis nf,.the:renal'eoan andthromb'osis of the glamerular, anerioln andd enpillaries. Lising an immuno9uorescent technique,' we . found Thomsen-Frio ddnnieh iT.F:) antigen onn erythrucpea and inlhc FlomeruG: Fluorecein-Iaheilid peanvl-agglutinin (Amcho •-r.,esa).has a highaifiniuy.fortheT.F. amigen' and'u a usen r.arker for this nonnally neunminic-acid-coated crypunucen. We rno- dude thst this aneigen wy eapored by poeuowcoccal neura- minidise. T.R antigen has been.n demonsuntedin formalialued and' t iG.wv. C. G.u,c, E- S,Kl, A- Sp-'----- R.. E.. n.G,YiA L.kA...- b. 1911. aL 9nf. 7. M< [u., L C. Ab.,v.kr. C a., ar„eEr.. a. JP.ti...'t114, tf. I ia. f, Kkq P. 1.. Newan. R. A...NNIv. P.. l'4kbbrY. G. G- Im,.n; L 1-FsAC.L I:eseMdtl. .. t-k: G. G. rtrex a. t. m6:0.w. C. z tr..,l-A. uN: ssS.n.
Page 110: azo61e00
gvl06SE0 I I I
Page 111: azo61e00
1! R PVXLJYi~~P MSMUXwf'.L p ~~2" - 7 4VZOGqcO b ~ f ~
Page 112: azo61e00
I 17 eVI06SE0 w_. ~ ~, 2 - .. _:, r : I I I
Page 113: azo61e00
Page 114: azo61e00
4olossco I 1 11
Page 115: azo61e00
J II 03S90151 .13 ~s8~~IR~~~1 ~ I xs,srasas a i~~~~5~~.~~ L_5 a~sxy.a~ao: ~ ~..x... x......~
Page 116: azo61e00
112 Ixs V. ytCAN J0UW-~AL ~01 OSS`ETRIC7 AND GYYECOLOGY: fc 14nr (~,~,.ra'I m: ui.~,•.~ I" C. r. ]Ir.CSbrnwr:) AmericanJournal of Obstetrics and ~'iynecolo~y! rolulnt 126, mmmber 3 OCTOeER 1. 1976 CLINICAL OPINION CiJarette srnoking, use of oral contraceptives, and myocardial infarction SVRCDH K. JAI]: Pn.D Srx Yb.i: Vr.rb'nl4 Thr r,rr.r n.4 ,r/..wn/nr.d ,nr.. n.,l+..l .../rr,rrn•n ,r.,., ~ry , n.nl ,.,.~rrnr rp,rr. .JArnrd.,o F.u,r;(m,.l'nud II'..1r...,r.lrr, raJ{ln,,..r! or,r.rrn,../ rh:,hr;hJ»..µ+lnn,.). ..nn4.r n rhr ,unAr p...p.dnm.n:. .I,nnng ........... Pr... th, rrhnrrr rnk na.+.N..A :r,rh. drr v.~'nrn( runhrur(drrn.n nAmenrA rr. hr 2 ra J, ad,n'h' n,mt .mturrrnlf. .Iqmfi'rnnr . I P~ r):18. brr . mrt pmbrMlin rotl. Thr Jpr.,.nr nnnh,u .,re;..tr rhnt .,en4ioq hr .nn:rnl.r..l n, .m,nhrnrnnnnnnda.rnwr J..rih.. prr rnjxnrn r,f,..rn/ uwlrnrn0trrr,~. Tlir r „rlia ~arrhrr ,rnGnnrr rhnr rln.,r :, n,n. :,Gi, .,n..4r,n ,rll n. .r.r •„nl r.nrrrm ryn:.. mrd nrr ,.nrrrrrrd nl. rr+LN.- rhi, u.i ../, n n,/nrnf rn,...n.l..ul n,fiu. nnn •hn.d.F7r .,.rrn.rngr,l rn rr.e np, Th},d,h,rr r„k.,~ ., n+/nrne,nr,nn,rl„rl ,n/n,rnn.. ..,.,ntrd:.:«k rhr~..r.1 ,..nl,.,nnmrM:;..,6..~,d'u n „~. . . .-brr., rrn.O;y. ..rr nrt te l nnd'11 ':. 1. , l. ,drJrnrlnrr u/w.n dv /,.n/,ndu.,r .•/. .mnkr... Fsn..r th, phblrrhnnhh . p.rnr..J th, rrdivon..r .n rlv r. . rn4 „/ r,nn/nrn!'..n- r..lrrd ur/n,rrrnn nrth,rrr.l !rr .lirnrnerrur,p , nnbq~ ,. ..1r.nnr.rl Yn lu, m,n h..nn.r rhh,r . nn b.. uhrr;r,l hr r/r.n,.,nnrr~. rb. n.r r./ nml'rnnrrnr+/,rn,.. F,n.n hl R-..-dn..t D.~... .r. Tl,r Gnpp..rlrA.w.l.r.l M \n nn/~/n4,n, r-/'H.n/rh r.uwr' V, P•.l HD-osA7 .md nndrnrrlen nrpnd ~~.thr ne .~...,hnnr.Ahi rh., . , r R,..a..rA.y.T.> c..n-dR.p..•.r 8n.•m.arnl.D.r..r.n...Tlrr P.:bnl.mo.n R,+k.lalnl~o.l~lv.,in.l.nnRr H. cs~~ 1'InM. \'- ~ 1-1 ItM'+?I ~ TNnCaiE-C/l\TROI .rrDiE, a+ndhued fu Ei„ylrtnd ~nd 11-alcs h,seindicalat that IheaDnndl rate~rl ..n6aallas.ucll a. (atulnn-rrlirl mlar.rt,.n .wu.. thoarr.nmon-hnrr,r nr:JlumtrjrcPcnrs.r..uh.ryn uwlla lu. :lrer th:ur rhe anr re~,-nehn•~ruoe. ..nunn~ wr,nurn uhu dn nou u.c rn.J aonuaacpu,r.' =(Lr.rcnarr~ n.l' catunate..ur .Ilutthi,:rhtn Ix- qcnrr.ilival n~ .ir.c.a INr..alct.. •d'..r:J r mrnucNri.rnm •~thcr 301 ~
Page 117: azo61e00
113 702 Jainn cotnstries. Fnr.eramplesr based ou the resulls.uf'thcse studies. the Food and Drug Administration had no- tihrd ducton in.the l-nited States - ... that patients over 40 be made thoroughlc aware nf the increased risk and he urged to utilize other forms of coutracep- tion."s The purpose uf'.this paper is to show thar.the generalizations of thcse excess risks to.mher countries or to di(ierent ugmenu uf the svnse cnuntrvcanbe misleading unless she differences in other risk factors-espesially the smoking habits of.women-are taken into consideration. This is.dnne byreanahzingy the data published in the above uw studies. T!heresults have far-reachitig implikationsior making appropriate recommenditions ab.ut the use of oral coutraeep,oises to.wnmen,in developed countries aswell asfix includ ing them m dte org:utizcd fam8tplannrogprugramssruund the.wurld., Plonfata Imyoca rd lal IAfaretlon >lano anel assaxiatcs' bast esunuatrd that IhelQlatne risk of nonfatal mvrkardial infarctiun ausoug.wnmenusing nral'cuntraceptise is 4.3 ta I in comparisurcto [hosenhn ne- used nral cnntraceptises..T7tis studrwas htsed uin two uf'.the 15 ha..pital rcgiuns in Enyland andl\ales. From.th'e publilshtd dau. the relative risk ass.xiated willi.the use of ural cuutraceptiso un~lie --/estimated to tse 3.0 to I lin athird husputal regiun.' ?Lstee the reginnal •ariatiune in thee relaoire risk estimates. \lann anrl ,ts.rkia[cs'aLa. estimated!the iuciilcuce rif txunfatal msucardiil infarctia,n :tmung users anrl notwscrs uf oral cuntraceptices- Huwever. these estS- mateswere h'ased:nn a sturls carrieefuutin un4v.one region..Inviewof,the regvsnal variations..it wnuld hardly seem appropriate to generalize the estimates of incidences based on onlv one hospital rcgion. Our interest is to re-csamine the assrxiati.ut he[s.eenn the use of oral contraceptibes and the risk uf'nnnfatal'mvocardial infarc[ian nhsened hc Mann and assnci-' ates' in two hospital regions. The main evidence seemss to be that the relative risk of nonfatal mspcardiall infarction ass.xiatedd with the use oUnral contracep,uives decreased from 4.5 to I to 3.1 to I after theeffects of'f other risk factors were considered. The standardiza-tion procedure manlc adjRsrs for the interrelation- ships bct..een vanous factors. Hnwever, the magniuudee of the standardized measure nPasxnciatinn between the. use nf nralmntraceptii+es and the riskof nonfatal mvncardialiinfarctinn wnudd dependlupmn the pre.a- lence uf other risk faea+re in the studsp.,psdatims. This aspect of the relatsbmhip becomes quite irnpnrtant when the comhined effece nf two or, nmre nsk (actors is :cnergistic. T!herefure. it has tu he.shmvn that among '-'those who are nnt cspnscd to aovother risk facenr the . Orra4x I, 1976 .~ . ~. or"u.. c-d, relatise risk.of nonfatal mvucardial infarction associ- ated with the use of oral contraceptires is statislicallr signific,tnt before the standardiud relative risktan be accepted as an indicatnr of an actualassncia[ion be- tween the two. This asxxiatiun is not supported bvy the analssis that follows.. Effect of other risk factors- From the data publishedin Table Xby Slann and associates.' it can be estimated thanin cumparison with women na known to.have any risk. factors the reLstive risk off nonfatal mnocardial infaretion in women.using unl'contraceptites alone is 1.3 to 1, which is one third uf the relative risk esnmated for allusers. )Inreoser, therelatiee.risk aur.,ciated with the use uf ural contraccptives.alone is also smaller than the cnrresp.mding risks asstxiated with other risk fae tors cunsiilered iiulisiduallp. For eaantple.,th~e rcLltire risk assd aiated a ith nlscsitv akone can be estimated Itu he 3.9 to 1, and dsarass.xiated s.ilh cigarctte smnking ,t.lune un he esumated to be 3.3 to I. Thc nuonhcrsare too small [odraw anv firm.cnnclusion. \eserthekss. the results do indicate dtat.[h'e use nf ool.comrxep- ti:es ,di,tx dnes nuc .ppreciah'bs increne the risk..f nunfatal mcrxardicd'.mfarctiinn. Thc reles:usce ofcrmsidering the interaaunn be- tween the use of ural'contraceptises and uther risk fauors for nunfatal mnxardial intarcnon can be further demoastratesl Is. anals rinvl the.cffect nf smnL'- ingpnt the estimataV reLltise risk assnx,atcd:.sutlithe u.c nf.ru.tl wmtr.,septises: Furthfs purp.sc. the elaw publishcdi.bs \lann and.uu.ci.ucs' in their Table V111 are further analszed.. Effect of smoking. It can hereadils seen from the results presented in Columns I and:4 of Table I that the relative risk of nonfatal' mrrxardiai mfarctiunn . among th'ose w ho only use oral contraceptiles is about the.same as the relative risk' arnnng thmc who oniv snenke. The risksamunq wnmen in these two groupsare.estima[ed tn he abr,ut.twice.arhigh as those who neither use onl cnntnceptives nnr smo,ke cigarettes. The relative risk assrsciaeed with the usc of oral con- traceptise alone is nnt statisticalfr significant (P = 0°8. Fisher's exact prnliabilitv testt whereas thereLiuse rukassociatcd with smoking alone is statiuicaRv.significant (P =A ~028: Fisher s eaact protnbditr[esn: The rela!ive risk among those w hu use nral comncrptists as well as smnke is estimated to he 11.7 I, tn 1. These compansons indicate that the use of oral cnniraceptisrs alone dltes n.a .ignificantlvv increase the risk of nnnfbtal ms.war, dial infarctirisn and dte presencc r)I:.hiah the risk factnrs simultanenuslrinereases th'is.risk.alxatt.n.elsefnld. The relative risk.n('.nmufatalirnsrxardial mfaruiun assnciated,wtth the use nf oral cuutracepuses amnngg smukers can he.estimated to he ii to I( I.I.b7tn 2.1i1.
Page 118: azo61e00
114 v.>a- tx8 Yum,b" 3 The. torresp,mtling nikk assotiated with smuking, among,usen can be cstinuted to be 3,g tas I(11.ti7 to'2.0?).. These comparisons indicate that, gissn smoking- the use of oral convateptives subatantiallv increases. thc risk of nonfatal msutardial infarctiun and. given, the use of oral cuntrataptises, smoking subu.tntiallvmcreases this risk. The increment in both cases is af>„utt the same. Effect of heavy smoking. The heavy smoken:.con- sisting of.those who fne,ke 13ur more cigarettesper dav. can.be treated sepsratelv fnum the light smukcrs. consisting of those whusmuke 14 or fewer cigarettes per dav. The results in Table 1I indicate that.iss cnnr- pa.risuu to those whuncither use oral'.cuntranepuses nor muke cigare2tcs hcavv smuking ah6ne inircaso the risk ofnunlatallmv,xardial iufhraiun ,dxnn li,ur- fuldl. rnd hcass suwling .durrg with the use nf ural tuntrrteptitcs increases this risk abuut.hftcentiu,o.. %li,reoser. the relatiie risk .,ssruiated w iih,heavssmul'- Ing alune Is..lbuut twice as high as.the.wrrespnutlii,g riskassntiatetl with the use.of oral cuntracepuses alune la,°- to.?A. or 2.1 to I). Huwcver, the difTcrcnte is nm sWUSWcaIlcenlnuflWrtt IP - 0.25. Fi.hcr'i.csatt prulxo- h'dltvtestl. Incidendv. based ou thc data published bv Vessey and Doll,s Frederiksenand Ravcnhoh' otimated'tliat in cumnparisun us the risk in nonsmukers anti m,uuscrs the relatne risk nftltnnnl.rembolism: in users whu were alv,:he:rvv,im,kcrl waa.^1 tn I. Theturrespnntl- ing rel:nsse risk arnnng uscn and nousmukers wAs 7.Lsto I.and amunK nunusers wha, were heavy smokers was 13'to I I Thus. the combi ned ef fect uf smukmgand use of oral contraceptiseson the risk nf thromin,emlM,bismappears tu be synergistic: snwking alone does nut seem toeletateo the riskappreciabls but the ase of unl convaceptices dnes seem to.increase this risk..ubstan- tiallv, even among nonsmokers. Interaction between.smokins and use of ural con- traeeptives. The.relati'veri.k.assuciated with.th'e use of oral contrateptivesis the ratio ufthese nsks amung userss and nunusers. These risks. in turn. are the weightal averages uf'.the cnrreepundinY risks armmng smokers andd nunsmukers-the weights being,the pnr purti„n of inuk!ers amunx ueers and nonuscrs. Thus. thee relative risk assutiated with the use nf'f ural con+tratepGscs in a pnpulauan cuttsisung of smokers and nnnsvnnkcrs can Ise otimated livv using the prupwruun of smukers amoug users aurl nunuusers and the rclnOVe risk.estimatesthnwmJn Culumnv Iland 4 ufTable I... Fnr c"mple- in the tuntrnl pnpulstinn oflthe Brnish. studv, aMmt 43 per cent :nf the uscrs antl itl per cent nl'thc nonusers.were claosihcd as.smukerss lt'iflirIie use nfth'ene prnpnruons and the reldtne rnksshnrvn,tn. Srmkinq, txal'conVatapliv.al and :myocardal ntarcroon . 300 Table 1. Estimated relatice risks of nonfatal mvocardial infaretitsnbr, use.oforal cuntraceptises antl cigarette srm,king 3„.+:n` /ar. SwnAin Total Heasr• lJiglsd 11 14 3 .67 4:13 .tl1 123 .38 0.7] Nunsmaken ? 0'! t.00 Heasv smnkcrs. 16 .^1 4:63 Light smukers or nunsmuken . ? .93 1.00 Smukers amd 3 .90 1.00 iwnunuken Bascd!nn duta iin fahle V.III Av V.nn aodasrwiates." •He,t.. s,oken: atiua IS uyarenes per dar.. rLight srnulen: le» rhan 13 cigarenes per dav.. Cnrrrs nrnP, ewnnrrp.en Table II..Estimated rekYtive riska uf nunfatal mvwardial infarttiun ,o a litauiun of prupnrtiun of snwkcrs amuorg cu.rrcut.tucrs..ind nonosrrs u6 ural lnnlralepti,e). Gu rnr ,...r.,,J P.. rer Rdnutv ,n1, nf n.x fmm! ,.w, n,- I ddstis~nn.no.. I A'., ±. Iq l 11167 5.43 J ru 10 0 9:15 1.00 1-es 3 0 , 6-6a 4.33 Vo 5 0 1138 1.00 Yes ° 3 4.43 3.43 No : 3 119 1 00 Yes 3 230 ?.36 v,, . 3 1 ~06 1.00 Yes 0 Sarl "'.0q.. Vn 0 1.00 1.00 •In cnmpuiirm tn nonsmoken and nomusers. rin enmp-fwn tu a0 nuuusen. Cnlumns I and'i of Table I I the relative riskassociated wlth thcuse uf ural cuntra:eptises in the unal'.popula- tiun can he estimatedd to he 6.16 to I.38ur 3:9 u., 1- iustead of 4.3 to I which w-as estiinateti As \lann and assrkiates.' The difference hctwccn the two csumates iss due to the differences in ~th'e numbers in their TLib'les 11[.IV,and VIII.. The relatire risk nf nnnfatal! mtntartlial infarttionasuwiated with the use of oral cnnuratcptises tan bce staudardized fnr snmking dilff'erencn hvr,nsununq thatthe prnpnrtion nf.smo,kers amunq,u.sen u ryual to the prnpnrtion,nf smnkers amnng m,nuscn ucdie cnntrnl'
Page 119: azo61e00
115 J04 Jarn (kwbtr 1. 1976~ .tm. ) 01- C.n...A. Table 111. Estimatetlic%cess risl' of rtt/rtfatal tlnucafdral infaratiun:attributalile.to srssaking antl use of oral mntratepti.cs S..b. C6rr /..re! r~.na.r.pnv. 1. . P ro p.+ru. a../ u v... e. fnrwsenrr q/twe/ntnl n,rardinf mfnrrtw. S. Err..ue.d' 1. E.vr.. Co.Po+.+t1 u/ r.anr nd 1. fFe/don• Yes Yes 0.0390 I IIti7•1 10.1i7•t OAltil•t~~. 11.96 Yes >7u 0.4345 3.13•1 1.13•1 t1.5?^; • 1 ~~ 32.74) Vo Yes 0.03^6 ?.02•1 1.(YNI 0.0330• I 3-N No Yo 0.4541 1 0 0 0 Total I.Otw tI999 0.99'1 tl' l 11M1.tN1 Culumn I: Distribution of 13i.wutnen in the tuntrnl pupulatfu.n uf the British studv.l'CulJrnm Y, Table II-1 is the intideme uff nunfar I Im,ocardial infarcrium anu..ng nmtusers aud:n+nu,ur.dera. 1.ri'!'- : Culunm I•Culumn _. Culurnn i- - Gvlumn 1- 1. . Cdunm!1 - Column 11•Culumn y: Column i: - Culumn i . 1WY. 0.991tl•1 group uf.die British studc. $tandsrditatiun tirc smuk- ing li`htts intreases the relatice risk. asm.itiatcd aiils, the use ol0ral cuntratepti~o fruar3.9tts d.tu i.3 ~to. I.. This eliglit'int rcase tan b'r.caplainrdl in tcrm s uf a small's negatise currrlatiown 1-0.tHlldse0sceu the u.e of or.d'l tumtr:scrptites and smuking in the cuutrul group ul d,e British stutls. \lann.antl asxxiates' presurnablv tandardi,ted thce rclaticerisk ofmsnFatal imucardial inFarctiun asstki- ared with ihe usee of oral comraceptises ftur heasv stnokiirlt. Forr this purpne. the light smukerss antl nouauuwklrn..u'r tu.he treated u,ycther Ih die tantrol group af dte British stnd.. abuut °_91fi per cemnf the u.enwcre heasc smukers astumpared m^_0:per « nt uf.theuunuscrs. Standardization fur the differences in heasv smokers reduces the relltisec risk of nonfatal mstxardial infarceiusn associated with the use of nral contraceptives from 3-9 to I to 32 m 1. A' small rediectlinn in the rel:itise risk is nbserseal Actausc of the small correlninn IO.INiI/ br:n.aen.th'e use aforal ron- trateptives.and heas vsmuking nbsersed'in the cnmro4 group,uf the British studv. The reductiua wnuld hase been much greater than that ubserved:had'thc wrrela- tiunben.een hea•v snwking and the use nf'.ural ctan- . trateptices lieenhigher than 0.061. Imerestingls. these results can also be usad'.to hosu that'he.tvv vnukinq als.o sttbstanaiallsinareases the risk' nf mrnfaml ,nsirtardial infarctiun and that the increase is ind6pendcntw)f the use nf unal ctsntraaeptists. In thc tantrnl grnup.nf the British uudL. aMtut 12.5 per cent. .nf dre hcacc suu.kers and:dAsut 3.`2 per cent of the rest uscaluraf tuntnccptises. With the use ufl these pru. p-ni.,os and the relausxrrskesumates in CG:durnns Sand Ys uf Table I. the relauve: nsl uf nnntardial infarctinn anu,nK heasssmukers can he estiinated tn:he 5 .9tu 1 m cumpansun with he rest. Thts rauo fs reduceditu a:S tu I a l ierdte t1i Nr ntacs-irs the prupra:- tiunuf uscrsamun% licas. snrulersand the rest are traltalar(li/r(1. Delsrndiitg uptm the mterenu in tfie searth, for acausal factur.uf nntKardial infart.tiim. therestdts of the same stuth-cutdd heused toahrno thm the use uf ural'. . contraceptises. indrprndcnt.uF sutulinK„is masiittetl: with an,iinrcascd risk ul,nunfatal'.nntxardialI infarc. txm. Alternaticck. the restdts.nf thc.s.tmc studitutddl alsto be tued:tco shuu that snsnking. i ntle pendrnt'u( the use nf ural cuntraceptises... is .us.niatedv with an in-.treasrd iiak oU nunF.ital nnrncrrdial. mlktttiun. T!hr latter inerrpretatinn is accurate brcause the relJtise risk af nnnlrtal innu.ardiall infatctiun associated with srntuk-ing issuusticallvsignihcants esen antung nonusersuf ural'cmmracepeises., Hir.ccser.,the fnrmcrinterpreta-tion uf the results is:incomplete.tnd usisleading Ixtause arnung nnnsrm,kcrs th'e relative risk of nunfatal msroatdial'iisf-uiun-x.:iated with the use uf nral contrateptisrs is m,t ataauucalls stgnrfiwnt.. Effect of proportion of smokers. Tbe rdatGse riskufl nonfntal rmncardiall infarcuwn associated with the use uf ural lnn[r.lceptlYQs ubsersedin a pupulrtiun.uill tlepeud upun: 111.the assutiatiuu.bctwccn the use of nral.cnntratepti.cs.andl smukiitg, and('?1ih'e pruptr- tiun )t:srnuk'crs amnno; users atsd uasnuscrs. Tu dlus- trare.these puints. the rcbtiwc riskscstsusated from the Brmshstudl.husan iit Cubunurs.I and:a of Table Il.trc cunsidered: First. umsidcr a,pupulatimn in Wh'ich all usenand ntureul the nnnusers xmokecigarettes. In rlus pupula- tiun. therelatise riskaswxiatetl with the usc nl ural cutrtr:uepu.es watddl le uh,crsed tno he11.: tu 1. ticcund:tunddcr mrnthcr e,treme situau.m-a pupul.i- rinn iit ,whids nnue ul the users and all nunusertsmuke ug:trettes.,lu d,is IxrPulauun. the rclausc rtska,WKn
Page 120: azo61e00
116 ated'w.ith the use uff oral rnntraceprises woudd be obserred to.be 0.9 to L In 6uth cases, these reGtive risks.sandardired for the differences in proportion of smokers aould,.be 2 to I. because. in hrxh cases..the standardizatiunpnaedure s.uuldassume.that none of the users as we11 as none of the nonusers smoke cigarettea. Consider a third~ populatiun inn which there is no correlation between the useof.oral contraceptivn and smoking. In this population: the «laure risk assrrciated with the use of oral.contraceptiresailldepend upon the proportion of smokers. because the combined efkct of smoking antl the use of ural contracepti.xs is .ynergistia. The propurtiua ofwomen who smoke cigarettes dues varr hetweendi.fferent countries. For rtampk. in dcseluping cuuzurics like lhcti~a. Pakistan.and Bangladesh., verrfea uomcn' smokee cigarettes:. For illustrative purptnes. the uver-allrisks aserxiated s.ith the use of oral concraceptises ass a liutction of the proportion of smukcrs are shnwn in Tsbk 11. It can be readils seen that the rekttiic risk of uuufutal mnaar- dial infantiun aaxwiazcd s.ids the use uf'oral'cun• traraptives iu a coutrtrscur. Id varv bctween 2.0 to I and. 3.4 to LAlpending upon the proppartion of smokers among uaers and nunuscrs The results presented aM seclearlvtndicate that the relativec n k ( n nl r l u- ardtal mf trstnon ass.Kt- atcd Iwtth the use ~d nr.d icontracepusesvtlsse:ntd in the British studs can fsa espbinediin terms of the propor- tion o(smokers in.the studipnpulation and the fact that the comb'ined effect.nf smoking andd use of oral contraceptives is svnergiseic: This aspccrof the cotn- pks.relationship ua ubscured in the standardizatimn procedure used in the British studr. Eaeess risk attributable to smoking and usenf oral rnntracepti.ea. Let I be the iin.idancee of nunfatal mvocardial infarctit,m among nonusers of oral con- traceptires who dJ) not smoke. Thiss incidence in the total pupdauon of the British studbcon be estimated to fie 1.99•I. In other word!-.smoking andd the use of.oral cuntrarteptsses increases the incidence of nonfatal nrvrRardial infarctit,m bs.aMtut 99 per cent. Tltisescess riskcan be partitiuned in sarinus comp+menu that can b'e.attribumd to smuking and the useof oral contracep- tives.separatelr. Tihe results are presented in Table Ill. It can.lse.seen that abnut33 per tent of the e.xcess risk can be attributed tn.smnking akme:,aMtut a:'per cent. to emnkingandtheuse of nral cuntraceptives: atsd nnlr Sper cent if the -ns riskof non(atal invocardial infarctiou can he attributert tu) the use of oral con- traceptis'es alune. In uther, t.urds.. 95 per cent of the &nolony.. oral mneacopwes, ard:nryocardat nlireson. 305 Table IV, Estimated re6itise riskk of death fro.m mrucardial infarction among differene categr.ries of use of oral contraceptires .Iyr .1 pnfnnn (.rvfu.nr,v,r.nr.prn•n lO J9~r n ..u.. -VCveruu<rs- IAO 1.u0 ~ c.nrnr - 24 munths ur kss ?S nnonrhs or more 7.36 1.91 3.06 4.7a °:19 d:37 Esusers Y.18 0:33 "Ever urcri" ^-.9° 1.62 -Ytveruseri and eusen Current usen 1.00, °.tlu LtAI 4,33 Based un dau In Tables It and Ill fruue Namu -d Inmam.r e<cess risk can he attributeul to snwking as cum pa red. to 47 per cent.attribuuble to the use of oral cantracep- tises. lUamoking is cumpletelv eliminatcd,.th'en the csteas risk can be estimated oo be 11.1199•1i-a reductitm of aMaut 91per tent from the ptcsent46cl. On the utherhand.r if thc use uforal.uontrauepeises.is uunplctrls eliminatedi then the.redru.tiwn in the e,cns risk usu Ise estimated to beabrmt 42 ' per cent. from.U.99'1 m0.J . 1 Thu3 frunt the public health In n of ctlle eCcessTlsk'uf n nldlall nr 'RarrIL11 ularut n a akti eocin be reclur d u nltdet thl 1 lt t d. iuk- mg rather than, bs cltmduung.thcuac of -al con- traceptis cs. , Fatal myocardlal infarction Thee data are nut availabk to analvze thcrclatise iiskf of fatal msocardial'.infarctiasn assuaiatcd irith the use uf'.ural'contraceptis,esseparatelv anrong sntukcrs and nunsmnkcrs bccause. Mann and.lmnan' aere mot able to uMain the data on smuking habits u('.tlrcir studi- populauon. Howeser, it is doubtful that the results for fatal mstcardial infarction for.stnokers and'nonsmok- en separatels smudd be sufiswntiallr.diffcrent than the results (ornunfatal mvocardial'iufarction, cspeciallrin siew of'.th'e fact that anusng all users of oral conar:wep- uves the relatise risks nff fatal mvtcardial in(aruiwn havt been sho. .nto be umdar in magnitude to the correspomdiing risks of nunfaol mvncardial infarc- tion.r. . Incidenslv. the estimates uff retatirerisks of fatal mvncarrlial infarctionassnciated aith theiurrent use n(' nral tontncepti.es cnuldl hean artifaut n('t the disunctiim ben.een.tht current usen and Illre eeu.crs. Thelidluwmg anafssts indiuata+ that dus c.+pecLJls masMr the case fbr ~rnmen a11tn:4-1 scars nl aqc.
Page 121: azo61e00
117 306 Jarts Tlie results pre•entcsll in Table Il" indicatc that.ti,r the.currcnt uxrs thc relatise nunrtalils risks in wumetu il) tnaa st.rrn uf agq are highcr tlsan the cnrrrspund- mq reLluse risks in wontcrl 3tl,to 39 vears of age whcrc.,. fur the r.uscrs thc reGtise mortali.n.risk in wr,mcn70tu 44 scan nf age is lower than that in wunlerv 31) to .'39 sears uf' age. This is quite puuling„ antl onc m:t.v be tempmd to attribute the dif Ferentes to snull numtirn, lot.rhe.sante time, other pn+sibilities carntot he rukd' nut..It h:pu..ib'Ic.that.some uf the erusers,iu thr 40tn 44 vrar age gruup nught hase bcrudas+dicd as cur- rcnt u+ert..There is nnwav tu..ticertams.hethrcor mu wumcn actualOv used nral cuntrarcplisrs rturing the amaoh hrfore.death e-pt hvhstrmr.wing dlcm liut r, cannut mtersrewthm+c who dkd: Thiss is a.. urntplisaerd prnhlemJsasing nu:uniyue -Iutlun. Uite mrs b.ee tcmptad to cnmhlne the evusrrs wnh the rrurent ¢.rrs fur estimaing,the relative risks:.How- eser ttlis appruxh wuuld al.sn hiass the resudts, liecause the tiine when Iheesstscrs stuppcd using nral cun+Iracrpliicc is no( known..It.mavhe otue uwmtti. sientnndhx nne rcar. nrr murc hr6Sre.rteatlt. Furtherrtsnre. thr du.ratiun liar wlii¢h the nauscrsused oral cun- acaptisn also is nna.6'nuwn. sud the duntinn,afuse amuuq currrnt users +crnuto he asiokiatedd witti the rrFltisemuirditv risk nf rnsu<artlial udarctiuut (sce T.dilc 1 N), Theret.mscnuutalirs riskbm the "rsrruserti'bfural cuntracrpursrsm curnp.rrr.oo toi nesrr uscrs' in the 30 . tu39.s rar agc grnup is.qlute clnse tn nhe corrrsponding , estimatr forth'e currencusers. Hnwever, fonthe "everusari" in the 40 to 44 vear age group.,the cstim:uc is suh.ranuialfsluwer thanth'at fnr the current uscrs- The ahnenos:uf the two estinutesfur the 30 to 39.ear age grnupand the difference in.thcscestimates hn the 400 to 44 sear age.,g,roup indicate tliat there might he some prublem in the litter age group in classifving women as .urrcnt nrc<users.,This mav have artificiallv elevatedd the mnrtahav risk assrxlated with the use of.ural con- Iraaepnses Irl.r-ltlen nf 40 to.4a scafs of age, Implications F,r prupcrlv .rsessing rhe .rss-iatimn Isen.een thee use uf nral cnntraceptnesartd mstkardi'aI infarctiiua. itrs esscniul to understand:the cnmptes interaative ef- feclsnf stnnkinss..tSe nt the wnman: and the use uflural untrateptnes. For th,s purpse ,tVsimperative th'at the relatise ri+ls nf fatal .md'.d nonfaul msrnardial _iular. ann aa,xlated wnh.the.u,se nl'„ra1 contr.icepttves re separatehe+ttmatcd'.tbr. srnulern and nnnsmukcrs withmieach age grnup..\lureuser, the durauuns nl use Qrmfxr 1. 1916 .a- J. arwe, t:. nrsd.' and nunusc uf oral kuntracepti.es need to be taken into cunsideratiou; especiailY for the riskoFlFual mstxardiall infarctinn., Thex tspes of analvsis.ought to fie.carrieduut.in fumre snrdies of this nature. The resultsof the present analcsis. though limited in scope: have de:rr implicalions for wornen usiitg oral contraccplises. L The results of British, . studies cannsst be gen- eralited to woman in other cuuntries withouu cnnsid- eratinn uf their srnoking habits. 2. There seemstobe uo evidence to sugg-t that Ihe useuf oral contraceptivcs alune signifiwnthincreasesi the risk uf nmu fatal m.ucardial infarctiasn. The relaure risk assu,ciated'aith the use nfural cnurraceptices albne is LS to I in cosnp.rrisun with the risk in wonren nutknmwn tu hase .1r1Y uthenrrsk Iactor and 2 to I amuagnoa.rnnkrrs. u{tich isnutsutisticalls siti;nificant.md is less dtau 4.3 to I, estintatrdd hs. \fann and assnciates' tirr all.taers. It'.is quiteiikeh" Ih'at rsinnilar rclltiartship will hulul for svomen 30.to 39 aears,uf age as well ass hrr. IhnseA0 ur tnnre vcars nf age. Howcscr, a firm cunthui,,m in this rcg:rrd.ccmnot tie rcacttcrl withuurt linther anah"sis- S. There is stroug rsidence tuo cutge>t that the cnnthined effect nf emrtktng..md u.e nf ural cnutracepµ tiles on the.risk of nontatal'mnxardial infLrcuun ns tvnergtseic >(urestver, the relative riskk uf nonfu:tl s•wardi;d infirction :awKiated with.snu,lihw;• is st,dts: tic.dlv +tgnlfi ant esen amnng nonusers af nral Icun- trat:eptises. These results indicate tharthusewumen s.hn smuke ass.ell as.use unlconaaceptiaes.andhraut to reduce the nskofnonfatal.mvncardial mfarctiun should he enmuragcd to.give up smoking. Smoking- in other swrds. shuu.ld be considered as another, con- traindicatiotv for the prexriptinn nf oral cuntraccp; tices. This.mav esprcrallv applv'to wont,c n nver 40.vean , uf'age. Howescr, it is quice lik'elv thatthe relationship mav hold even fnr svomen 30 to39 yra.rsof age. jfirm . concluswn canriot fse reaahod without furthcr analssis. i:.It ispu.sible that the above con iusions mav nuthultl for thensk of fatal mctscardial'intarctmo, but this seems quite unlikelvv in view of the simil3rrrv% inmaq- nitudesuf relativerisks ofTatala,nd nonfistal mnxar- dial'infarctiun assssciated scith the use ot nralicon- tracepu.es among allusers. Nlureuver, therstinuted rnnnalitr risk amung users 40 . to 44 vears srf age mas h'a.e been a.rtlhclalls cle+"atetl fiecauseLf vnncinaccrr- racyrn classrfsing them.as current users. This paper has greativ beneficed frnnt ahe cr,rnmrms aud. suggcsunns n4 Sushd I:umar.Dale Ruhertsun.,andl Irs-ingjisin. ' k ~
Page 122: azo61e00
118 V.au~e IY \..~~Rr f ~ REf.EAFNCES I. \lann. J I..Vea.rl:.s1 P, Thur,,K-a. tI~. invl Dull. R-- \4srR rdi61 ml;~r ~u r n ruunq w~~ w~ihsPrrul rrlrr-e la orsl .nnv:n1rpnnr P-n.enrBr iIrd..J 2: -liL.. 19u Je.uti Fru~n m..~.~rdul iwtae.nn~n:. Br. ttrd. Jl '. !ii: 1973. ). F~D{ Dru{Bullcnn 3 '. IIIIDHEtI' PuM-ri.ul \u[HE\i'l l9i). F-I -d Druq .iJ,mululr~uum. R-k•dld•. ?Lr.r.land. Julr-Aurlv. 1973.. $molung, ora/ COntraceULVCi. anC myOCarCiai ~nfarCbOn 307 4. %tunn. J' I. rhuurr-ll Ne I~~rrrs: l~. E_ ~od Pn~rII1C.: l)rJl .r. nlrJ.cPln. , ~nrl ~ rth~L inlJranun i ru.n,l. wau ecius i tunhrr rep~on.r8r~iled. J. k rilll 197sS,r i. P.~Ird Dulll. R Imr,crGv~~~n ~d1 rrF~won.hip,hrn.ren u ~~f ,rc~l ..n .cy,ri.r.~mrl IhrnrnMxniM.li. 11nru+r. i.tueihrr rey~nr4f8r \/crl J' 2: r:it. Ihi" . ri. Frrdrrd,rn. H., ~md R.un.d;~,dl:. R FI_. FAnnmlwrur- h..h,ni. ..ral .nnrr-pncee. ~n~f. uqsrcu<s. Pubile Hralah RrP. Bi: 197. 1970 ..
Page 123: azo61e00
21 ~ 11 ~ ,it ~lit& . ~All ,51 , p~ ~~ 9r~ ~3 ~1j 11111r, a$~n•gY o ~s•:pi g~~"'~•~~~~.9 ~, .3e./~x~.~~~E~~j n lia ~xar 1-8 i!IV!1I a n••••s M11~ a1142 1~11il BI ~g ~~~.~~':~~ ~it . ~ j='~o~i~3~ ~a 's ~ ~03 s IrS~ai~ 11 r ¢~ # ~~a~~~ I 03590159 .
Page 124: azo61e00
035901G© v 41 ~ A O I a ArArlR d ....;«P4 R 0 ~j _ y~ ~;~~~ ~! ~fJf$JJjJIjJ1jj1ljlj r! ~ye 'i n~.~ ~~ ~9 1 Y 4~~J' ~~i~==,eb ~~ 4~ ~~ " ~~.~~ ~# J~~~~~ t . ~ '..g~gg===~ ~~t~ .: ~~ I ~s ~ .i ass ~~~ ~.~~~ ~ ii1i1k!1~ ~ iHJ ~j`iji~ ~ sa~~~~~~~~" #~i~~x~; ~F ~ ~ E "MP~-8'z~i ~ " 01 ~ J ~ ;- f ia~ ~~~~ a5f " 1il~,E~jie~ 114~e ~ ~~ 9~~ ~~~aa t~ciS4~~;3= a~ Ya e9 Et t.~~' ~~ # o Ea~ I~-~ z . ° ~ ~~ ~ .€~,3 F 1a~F~~`~j~~~J~a~ ~.~li~a ° g ~= „~ i '9Ii!! s~ g 9 % A I G S ~~ ~ "~ ~1 1 1 1 . a~ '4 ~`'Y ~~d ° ~ ~~~,QyOs~~ ~~ ~ ~~~~.go~~ ~ .S~1 ~'~X $~~~ ~a~ ~9.~ k ..a..~ri$~1 Aj E ~~s~e~,~~~~F~~ tj y8;~ ni~~:in.~ R"o~'CEv"sg.4d ~nSa~ ~aE ~sE~S' S~ $>a „ ~E. I ~'~,•; ~~ad sx~.~ p~ ~% 'SS 's~r _~g g 6005~1586.3Scox ~'js~'a$99~ii~s:~~p~
Page 125: azo61e00
9 03590161 ja~•! a . 3 ~ ~ • 'O C +aa'' .~ aa03~~~y~jj9j ~ ~~~a~~~1~ ~~] N I~~•~ ~ti~1~,x' 1 fa3~~~ ~~~~~~~~#~ ~'y ~~ I~iA~~ ~~.~1 ~~as '~~~~~, a~°3 g ~~k~• 3 ~~~~ c~ ~. ~ ~ 5 ; ~~~~ ~ ~ ~~ ~~ ~ n. ~ gill ~ » co U ~3~~~~ yJIE~~ Q a.S.o O 8 ~ g ~~ ~~ i11iiI Ia Q ==?Cg:t•~~~ ~19~1~~~~•~ . : ~itpx~ea a• 3 g a3~ ~~ '~.. a9 ~ 3 a • $ _~~`~~~~~g . +3~~~ 4 ~ ^~ •~~ SSe!ry ~ '~e ~d8 ` g i "~.9.~ ~ ~g 7r~~# 3~•~1t~i.+$ `~Si~• 7.~' cA~ixH a$~~1~~ :; ~7.. ~~.`~'"l~ja'~o 3~ C~~ • it A 8a ~'s $if.~~$o~ ~$~~~s~~~. 9~;~~f~~ ~ y: E i .. ~ . ~ . s. ~ ~ ~ 8~ ~~iA~! ~r " ° E z~~
Page 126: azo61e00
I z9tossEo ~ a~I I a r=7grM. a ~,Vf p $J4~g~~7~~~pseaJ$ a -~E~ ta ~.~ pF~s$&r~s~niE-, 5~.:. i~ C q v~ S~o^G~ci~ ~~,+~ 6 v5~.qg ~5.8 ~g8sa ~8 i~9 E6~, sPT'c.~a°~ _ g ~sGC$ § 6e ~ ~ ~~: 111M w~!. ~~ ~~g~~ er~x5:e , 0. ~ #~io~ees$~r.t~ F ~°3.F`~1 ~ oE ~~.gL 8~~~_ ?_C~B~R~.§9~$?~~'.psw ~ •d '1 Sw ~~~~~ ~~~a ~~= rw ~• Mw ~S sa o» ,. gff € ~!°a g; w~n.v~~$ r g2""& sCt , ^~ q~'~ ~'°I"6~'~S$'a.f,~~5. Z~Coe5~a yC.L~g,F-p'.~'C.f'ofa~' Sf~S ~tr .~wn2.R~,.~ Ae'jw~'~nE ° 6:~Sgo£~}„poCP7L.Ri $G e seS3~V7 xS€~~~ 'n~~~ ~,C..1 ~~~ ~'~e°_8~ ,€ °g.ea ,~ ~ °C~ ~e ~ a~Ra~ 6 r c ~~~ ! s€~~- 41 g 5~ ~ ~' aee 8 la.~'GSP°~$~~P fi
Page 127: azo61e00
~»~a gls aR~" .~ xI '°c"~• .oXc ~. $.~ ~. v rd6~~.~7i ~ 15pe S~g. ~~ E.BZ - $ ~j5 c R -, g ~ ~ ~ ~~8 . ~s'p' ~P~' ~~~' ~a~. "'8a Q$¢~e S?g.ap.~~ sge ~~- i aS. ~' S. sr~' e $B" S B ~a8_~~v 3 ~+~$ °~~pG~g, ¢ $S ~"rE•Ca RY _ 6~. ! ~y}y~~'p ¢g~'..7.~' ."6.~~~ ~~'~1g w 0 ~3 g g i$~~"~ 1' p$ ~. $$~~~&q r ~fSCclR~~ ae RExS,: s•'i~$ ~ il' C8 I 6' Ti 6l ~ !• B.1 B 1B.11 R~fl WI ~t;xjRj~Qj~".e~ ~~~ ~~ ~.~~aa~:''N ~- a a"~p$~ ~~~~~tt~~$R Sg6ta~ 8a~~•"~ ~- a grr F ~R!H1U! ffp~ p~ ~~n ~ill &QP gR~iO~yp~'y~~ ;g~.,~ Z...CiS.w3-5a~$S~'r»gg°'xICP.5g ,x~'~~,~S.c~~vsp;$~~gp 0~, 5, np~~.,j o~='~^5 Rg~~~Ssc~a"s~ a ~ sEE~~~~rsa 3~s-Ea~~sl. Q ~s. s ilS ~~,. ~, tg,&~Sss•2.1 ~'v I~, ~ T&~f n ~''~,~~ °~~'r plrr_ ~,s ~~~.~+a~ Big '~ g~ ~g~g d¢. h~ '~ R~5333"~~i°i~' ~~g b.~ p•~ a°l ~8:• n9 :., zs ~~ g~a 9~1 a. ~ a E!pp. 9 "~6egd~c r°:~'.~~~.~g nw S~ 2; $t ~~&rc~ ~~ i3 ~° ~. E RE ~~~~ ~.. ! R V~w~~E S~are rr w ~ '~ ~ '° tl~~:~~.P~"a ~P as,,~ s' ~!S B~E~~sr~E~~x~~f~s~~~?~.~ ~a~~'_ EsiossEo
Page 128: azo61e00
` p3590164
Page 129: azo61e00
0 ~ 1 ~it S3j06qEU © a 0 iit.ilii.bIF i R , = 7 g ~.~-y~_~°~8 s, SS @~.p0.~ Y o• ~ g ~~ga.~sg g gl~s~.~~~sy}j' , ~ ~ ' 1450E 8.~ers', ~~~b~~~~~A9 ~.6~$6ES~s~ I 0 . ~ © f I I I
Page 130: azo61e00
03590166 ~~; ~ ~ l~~~ ~ 1~ ~~ ~~ ~~ ~ ~~~ ~ yI a~~ . .: ~~;~ ~~~~~~~ .~o .~. ~ ~- ~ ~~A{ .~ .~. M`~ ~ ~~~~ ~ F'Jr'J~a ~e 1 $~e.~ ~al R~~ i~~ =5~= ~ A.~ i~ a ~ .c^ iIiifl!t1iUUi1! ~ 9 ~l3li~ ~ t~r sg ~ 5 # 9a M Q g Mi~g il ~~1 0 °°0°°3 ....... ~ p . 3 ' ~`~g"~ 'a a'~I1IllFii osa~~aa ~~1a !~ ~$~.~;1=~~~g'~.~ °~`~~~~~~~~ ~~~~~~~~3~~~~ ~ ~~=¢a98~~~rs~g~ 'Iiia b~ e~~'
Page 131: azo61e00
ij:I -l~ nIlIPlili! I5~~ .9 JJ m Y Ja ~~Cg~be ` R : 1 c. .9V..a ~ aB ~~. 11 J'~• ~~ ~al~l ~b "9 ~~9~~~~ ~'g. :~ ~3331^i1 ~ ~ ~ ,a~18~ o~ar~~` ~ ~d ~~O~-1-11 iU J I 9 03590167 a.t._ ~~ xaxx~as~~~~ I wj21 _,...,.. d # I r ~ _ - ~-- :!I 19199999199 21211111111 =$xaxxaxa22 tyt t.Cg_.CCCC / l .w.•.•w.n.w3 : a
Page 132: azo61e00
rkp~'@F' ~il•.~;se~,pA~B11,@ ~6IgIRA~0.~i~~~e ~Ra R ~~ € C 4" V ss 8 -. r:..~ . Ur--e_= . ir.Y ~w Y ~ ~ tw Y 0 R o~ Q~a svn~Q_. ~ a W ~w9.p.I 212 ~~ R 9 aRIL $ I ~ ~ga ~~~9c1 ~w ~Qprr ~9 Pia'~l sI MF~~ +09. A $ 8n ~~ R$F~~~ ~$ I ~~ ~6 ~ ~ ~ ~ ~~ ~ 77 i I -. ~ ~~~; ~ Y~ ~t G ~~~~ F, l ~M1 q ~~ ~ ~~Frfii. ~ gg~t~~go~ L PI1d1IIi nt E __ ~ ~ M ~ Or~-~~@. CS~~R~ ~g~. ~ 8~• 4~~~r ~ ~~ lSrr~~p ~ 8Lr x ~.A~R~RA ~ ~,~ E ro~~ Ea p~9eq rr k =~f ~aee3 6~r~-SF.~~~3RlC"e' ~ ~ l;~r$gR I g s~6 g y ~ R g~ r.~ ~9'l s!~ R p~~ s~Q 4 f~E y~ R _ RF ~ ~~s~~-~~s~~C•a~ARR -1 R . Y r~ C 6 T ~~O W ~6~'~ ~ ~~11C ~~.~ ' e t$~F~6~~~ a ~.A ~SQ°Pn~g~6 SYp:~aL~~ ~ 9~Bx~~~~ sr T g9TO6~Eo
Page 133: azo61e00
129. Mr. Fovrrrnrr. We thank you for your comments and your an- alvsis of the situation. I will ask you the same question I asked Dr. KastenbaumL I have tried to abbreviate these in the interest of time. Accordinz to your testimony this morning, each of the studies has defects which, you say, on an individual basis do not permit a scien, tifically valid conclusion to justify FDA's decision to require the anti- smoking warning in OCIabeling. What is your opiniom as to whether or nott collectively the results of the studies considered would provid'e a valid basis to support the decision ? • Dr. GisBOrs. Collectively, there are only two data sets, both of which are extremely small. First' of all, it is statistically improper t'o combine two differentt data sets unless they are taken under the same conditions and the same quPstions are asked. That is not the case with these two data sets. Second, even if they were combined in an improper procedure, I still think that the information~ would be too limited to draw any con- clusions either way. Mr. FouNTnix. Are you familiar with the recent papers by Jick and associates and~ Petitti and Winrrerd on the question of increased risk of circulatorv disord'ers to OC'users who smoke? Dr. GTnROtis: Yes. I have not studied them in denth as I did the .Tain and Beral and other papers. However, I havp harl a chance to look at them. Aryain, the data s?ts are extreme] v small. The conclusions are not exactly the same. Jick (1978), concluded'that cigarette smokingg is weakly associated with stroke but this waG based on a sample of only 14 nersons who had suff'ered nonfatal strokes of which exactly 111 were OC users and 3 were not OC users. Mr. FOUNTAIN. What is your opinion, if you ha.ve formed one, as to whether or not the data lend snnnort for the FDA decision to re- quire an antismokinz warninim on OC labeling? Dr. CTissovs. You, are talking about the new data? Or the new studies ~? Mr. FoIIxTAIIw. Yes. Dr. GrRRo*rs. First of all, these studies had'. not been reported at the t;ime the FDA made its decision. So, they could not have been con- sid'e.rerl' tihen, I believee if these studies were considered now,, we still w.onld' not ha.vee sufficient information to conchide whether there is: or is not a relationship between smoking and pilt use and the incidence of cer- tain diseases. . Mr. FnuxT.ar-,. Mr: Duncan? Mr. DUN-ca:,. As for the issue of the sizee of the samnles, we keen liearinry that: they are too small. How big should they be to give us some ic1Ra ofwha-t we are talking about?. Dr. GmnoNs. The size of the sample is a very importnnt st'atistical nuestion which mnst be decided separatelv for each particular sample that we want to draw. There are statistical ways to determine how larpre a sample should he drawn in anv particular situation. However. to rlPtermine this number, we must Gtate what kind of error we are willing to tolerate and' what probability we want to have to be sure thst we do not have a bigger error than that specified.
Page 134: azo61e00
130 Mr. DUNCAN. So, if I specify, let's say, a 95-percent confidence levely you would be able to give us some idea of how big the sample sizes should have been in the studies ? Dr. GraBOxs. Yes; with a 95-percent confidence level and a 2-percent error, we could calculate how many observationsyou need. Mr. DUNCAN. In making the statement that the sizes are too small, doesn't that imply that there is some intuitive notion as to how big they should have been ? Would you be willing to give us some idea as to how exactly too small these sample sizes really are? Dr. GiBsoNS. I would prefer to give you an answer after I went back to my calculator and my tables and was able to come up with a fixed figure. However, for an issue of this importance, I think anything less than, let's say, 200 would be too small. That is a number off the top of my head. Mr. DUNCAN. I understand. Thank you very much. Mr. FOUNTAIN. Mr. Waxman? Mr. WAgnzAN. Thank you, Mr. Chairman. Dr. Gibbons, I would like to call upon your expertise and ask you some questions somewhat beyond the scope of your testimony but elaborating on it'. In your testimony you specifically criticize the FDA for warning women over 35 of their increased risk of cardiovascular disease if they smoke and use the pill. In all the studies you reviewed,,were there any that related smoking and age to the use of the pill ? Dr. GiasoNS. There was one study that tried to relate smoking, age, and the use of the pill. However, the author made certain unsupport- able assumptions, in particular, that the age distributions for one data set were the same as they were for other sets because that specific information on age was not available. Mr. WAXMAN. Which study was that? Dr. GIBBONS. That was Jain's study of 1977. He assumed the pro- portion of smokers among older women is the same as the proportion among,younger women. Mr. WAXMAN. Would you agree that older women who smoke run an increased risk of cardiovascular disease ? Dr. GIBBONS. I would agree that the older you get the more risk you have of any disease. Mr. WAXMAN. What if you are older and'you smoke? Does that in- crease your risk of cardiovascular disease ? Dr. GIBBONS. I think there is a good chance that it does, yes. Mr. WAx:NfAN. V6'ould you agree that the Mann study prov.idedFDA with legitimate data to establish a linkage between the use of the pill and heart disease? Dr. GrnsoN-s. No,, I do not. The numbers are too small for any con~ elusion. Mr. WASni aN. Your testimony has been highly critical of the studies FDA used to justify a warning to women of the health risks associ- ated with the pill. Are you aware of anv studies which support the position that a woman does not increase her risk of heart disease by smokina while using the pill ? Dr. Gms.otis. No. I do not know whether theyy have been carried out and not published or w hether they have not been carried out.
Page 135: azo61e00
131 Mr. WAXMAN. In the absence of such data, would it not be prudent public policy to err on the side of caution? Is it unfair to conclude that the Mann,,Jain and Beral studies justifv a presumption of risk? Dr. GIBBONS. I do not feel that these studies justify the presump- tion of risk. Mr. WAXMAN. Based upon the data that you have reviewed, is it your , position that women can feel free of increased risk if they smoke while using the pill ? Dr. GIBBONS. Feel free of increased risk ? Mr. wA%MAN'. Yes. , Dr. GIBBONS. I do not think that the data show that a woman should feel one way or the other abouf an increased risk. I do not feel that they are conclusive either way. Mr. WAXMAN. You have criticized the FDA's data base for failing to distinguish between various risk factors such as diseases like dia- betes, hypertension, and high bloo& pressure. Do you know whether it is a usual practice for physicians to prescribe the OC' pill if the patient is suffering from these maladYes?' Dr. GIBBONS. That is certainly outside my expertise. Mr. WAaMAN. How would you propose the FD A design a study that would account for risk factors other than smoking? Dr. GrsBOxs. How would they design a study ? Mr. WAXMAN. Yes. Dr. GIBBONS. I certainly could not answer that off the top of my head. There are so many aspects to the design of a studw: I would havee to work it out. I would be glad to provide you with this information later, if you wish. Mr. WAXMAN. Thank you for your answers to the questions. Mr. FOUNTAIN'. I am sorry, but we have another vot'e on the floor. W'e will have to recess briefly. The subcommittee will stand in recess. FRecess taken.] Mr. FOIINTAIN. The subcommittee will come to order. Mr. Goldhammer ? Mr. GOLDHAMMER. Dr. Gibbons, I presume that the subjects in the Mann~ study smoked cigarettes some time ago. It does not represe.nt, current smokers, is that right? These sitbjects are persons who had MI's after a neriod of use of thP OC and aDparentlv conr,in-Pnt with smoking. I think the other studies all represented smoking habit's of some time ago also. We do not know when. The composition~ of cigarettes is apparently changing. There was a furor about Dr. Gori's statement in a recent paper in the Journal ' of the American Medical Association that there are now brands of ciaaret'fes which are so low in tar and nicotine that they are consider- . ably less hazardous than the ci~rarettes of the Dast. Dr. G'ori is n microbiologist, I believe, at the National Cancer Institute. ~ If these cigarettes are beins, used +odav, and if their use is increas- ina. what would that do to the appl'acability of the smoking data as set forth in the M'ann and other studies in which they showed a rela- tionshin between increa€ed ha7ard from MI when the OC users also smoked ? Dr. GIBBONS. It' would make those data not relevant to the situation today. That is one of the reasons why I believe that new data should be obtained. r
Page 136: azo61e00
132 Of course, since these new kinds of cigarettes were not available, let's say, 5 years ago,, someone who is smoking today and also was smoking 5 years ago would present a mixed bag in terms of the effects of smoking. However,, I think it would be much better if we had data, current data, which did reflect the use of the new kinds of cigarettes. Mr. GoLnxAaiMER. Would you say that any researcher in the field should make an attempt to obtain some information as to the kinds of cigarettes that are being smoked ? Then we could identify whether women were smoking medium, high, or low nicotine content cigarettes. Dr. GIBBONS. Definitelq. They should also find if they are smoking filtered or unfiltered cigarettes. I think it would be very worthwhile for a Government agency or a researcher to run a controlled experiment and get this information, yes. Mr. GOLDHAMDiER. Thank you. Thank you, Mr. Chairman. Mr. FouNTAiN. Dr. Gibbons, we want to thank you very much for your testimony. Our next witness is Mr. Thomas A. Budne, consultant, applied in- dustrial statistics, T. A. Budne & Associates. Mr. Budne,,it is a pleasure to have you appear before us. We would be pleased to hear your testimony at this time. STATEMENT OF THOMAS' A. BUDNE, CONSULTANT, APPLIED INDUS- TRIAL STATISTICS, T. A. BUDNE & ASSOCIATES Mr. BIIDNE. Thank you, Mr. Chairman. I am Thomas A. Budne of T. A. Budne & Associates in Great Neck,, N.Y. I have been an independent consultant in applied statistics since 1956 and have served Government agencies, professional organizations, and companies in a wide variety of indlzstries. I have served the United Nations as a member of an international team providing consultation and training in applied industrial statistics in India. My education consists of B.A. and M.A. degrees in mathematics from Mont.clair State College, and a master of philosophy degree in mathe- matical statistics from Columbia University. I'have served on the faculties of Montclair State College, New York University, the University of California, the University of Connecti- cut, and the Air Force Institute of Technology. I have published a number of papers in professional journals and presented many more unpublished papers at meetings and conferences of professional societies. I am a fellow of the American Society for the Advancement of Science and of the American Society for Quality Control: I am a mem- ber of the American Statistical Association,, the Biometrics Society, the Operations Research Society of America, and the Institute of Electrical & Electronic Engineers. My professional biography has been submitted to the subcommittee. Mr. FoUNTAiN. Without objection,,your professional biography will be inserted into the record at this point. [Mr. Budne's curriculum vitae follows:] . ~
Page 137: azo61e00
133 i - T. A. Budue &Associates lnc..3 Du,urer Ruad; Gaor Nerk, New York 11021. Area Codr 516. 4xL6374. PROFESSIONAL BIOGRAPHY - Thomas A. Budne CONSULTANT in applied industrial statistics since 1956. A partial listing of clients includes: Air Products & Chemicals Inc., American Safety Razor Co., The Budd Co., Brush Instruments Inc., The Coca Cola Co., Corning Glass Works, Crane Co., Defense Atomic Support Agency of the Dept. of Defense, Eastern Railroad Assoc., General Electric Co., Gulf & Western Ind'., The Hupp Corp., Lynchburg Foundry Co., Milprint Inc., Mack Truck Inc., The Martin Co., S. C. Johnson & Son, Johnson & Johnson Inc., Philip Morris Inc., The Rand Corp., Spencer Chemical Co., Texas Foundries Inc., Vickers Inc., United States Golf Assoc., Xerox Corp. FACULTY member of the following educationallinstitutions: Montclair State College 1945-1947 New York University 1955-1956 UCLA 1961-1963 University of Connecticut 1966-1973 Air Force Institute of Tech. 1965-1969 Instructor for over 30 training conferences in statistical applications spon- sored jointly by the Institute of Electrical & Electronic Engineers and' the American Society for Quality Controli. PUBLICATIONS of greatest significance include: "Reliability Engineering,: THE ENCYCLOPEDIA OF MANAGFTMEiT!, Reinhold Publish- ing Company, published 1972 (Rev. Ed.) "Dangers and Potentials of Sva1lIQuantity Testing, "Aug. 7.970, Proceedings of Product Liability Prevention Conf. "Basic Philosophies in Reliability," July 196119 IQC. "Applications of Random Balance," May 1959, Technometrics. "Random Balance," April, May 6 June 1959, IQC. "SQC Can Be More Effective," December 1958, IQC. Numerous unpublished papers presented at meetings and:conferences of various professional societies. ASSOCIATIONS with professional organizations include: American Association for the Advancement of Science (Yellow) American Society for Quality Control (Fellov) American Statistical Association Biometrics Society Institute of Electrical 6 Electronics Engineers Operations Researeh Society of America EDUCATION includes: B.A. in Mathematies and Science from Montclair State College - 1940. M.A. in Mathematics from Montclair State College - 1941. '?t. Ph. in Mathematical Statistics from Columbia University - 1947. Miscellaneous - Member of a five nan,international team for consultation and training in applled'industrial statistics Ln India under the TechnicaL Assis- tance Administration of the United:Nations during 1952-1953. Qualified as Statistical Expert in United States District Courts for: Southern District of New York, Western District of Kentucky, Eastern District of Virginia. 0 w ~. ~ 0 '.[ r
Page 138: azo61e00
134 Mr. BUDNE. I have reviewed the statistical articles cited on page 4233 of the Federal Register, volume 43, No. 21, January 31, 1978, as the main basis for the revised physician and patient labeling of oral contraceptive drug products, which includes the boxed warning that women who use oral contraceptives should be advised not to smoke. Three articles are cited : A Comment submitted by the Population Council; a publication by A. K. Jain, "Mortality Risk Associated With the Use of Oral Contraceptives,"'Studies in Family Planning, 8: 50-54„ 1977; and a publication.by V. Beral, "Mortality Among Oral Con- traceptive Users," Lancet 2: 727-731i,1977'. The Comment submitted by the Population Council consists of materiall prepared' by A. K. Jain, who also authored the second cited art.icle. The Comment and Jain's published article are very similar. They are comprised, for the most part,, of a reanalysis of two studies conducted in the United Kingdom by Mann and others (1, 2, 3). The first study (1) dealt with the incidence of nonfatal myocardial infaretion among women, and the other (2, 3) with fatal myocardial infarction~; both with reference to the use of oral contraceptives. In Mann's study of nonfatal myocardial infarction patients, com- plete information on smoking histories and oral contraceptive use was available for a group of only 57 discharged women. In this study group, there were only 3' nonsmokers among, the 16 pill users. In the conr t'roli group of women, who were also hospital patients with other diag- noses, there were only 6 smokers among the 14' women who used oral contraceptives. I mention these very small numbers because they are used to provide estimates of characteristics for the general population of women. In, fact, the ratio of 6:smokers to 14 patients or 0.43 is used as the propor- tion of' smokers in the general population of oral contraceptive users. In my judgment, these very small! numbers cannot be used to give reliable stat'istical estimates or descriptions of the female population. Essentially, all the tables contained in the Comment of the Popula tion Council involved such small numbers, and the critical results denended entirely uponithem. Furt.hermore, in Mann's study of nonfatal mvocardial infarction natients, the women came fromi only 2 of the 16 hospital regions in, England and Wales. Jain has generalized this information to women in the United States desnite the fact that the data were colliected' in a! localized region of the United Kingdom. There is a serious statistical question as to whether the sample studied can be used to represent American women. This verv point was raised in a recent review titled f6Contraeeptive Methods : Risks and Benefits," in the British Medical Journal Ii : 721-722;1978; which published the Mann papers. I quote : Furthermore, the data dii+cussed'here have been derived largely from~ British studies ; it would be quite wrong to extrapolate them to countries with a different' O W ~ pattern of disease andAifferent risks associated with child bearing. ~ In the second cited paner,Jain reanalvze& the dat'ain Mann's fatal heart attack study. It is imnortant to pay careful attention to Manni's ~"1 own language concerning his data. Mann stated that : N] »'b These estimates of risk ... still need to be interpreted with c-iution.,as wnum- her of assumptions have necessarily had to be made in their calculation and the margin of error is likely to be fairly wide. r
Page 139: azo61e00
135 And'these estimates then formed the foundation from which Jain ultimately projected his results to the general population. These pro- jections can be afforded no more reliability than that which Mann attributed t'o the initial estimates: In fact, Jain made additional assumptions because there was no smoking information available on the deceased women. He assumed that there would be no difference in the smoking habits of these d'eceased women~ and the surviving women~ in the earlier Mann study. He also assumed that the proportions of smokers were the same for both the older and younger women. There is simply no way to deter- mine whether these assumptions are valid. Therefore, it is my judgment that Jain's results are not statistically reliable. At the end of his paper, Jain wrote : It is essential to point out that the mortality estimates used in this paper are based on small' numbers and may be subject'to large sampling errors. These esti- mates are also subject to upward and downward' biases, which may not cancel each other out.... The net'~ effect of these various factors cannot, be estimated without additional research,,preferablS in different settings. Despite this cautionary language, the FDA apparently found no problem in accepting,Jain's conclusions. I find the following sentences in the third cited study by Beral par- ticularly reler ant. "Without more data it is not possible to examine the interrelationships of age, smoking, and duration. of oral cont'racep- tive use. . . . The ratio of the death rate from circulatory diseases in ever-users to that in controls was similar for smokers and nonsmokers (table IV). This suggests that the relative increase in mortality as- sociated with oral contraceptive use is indepe.ndent of the smoking habit." These statements refute, fhe averred support of this studyy for the warning label. As a result of my rev ie.i , my conclusion is that the references relied upon by the FDA do not support the requirement of the boxed warning. Mr. FouNTA•Ix. Thank you. The list of references which you submitted to the subcommittee will be inserted into the record at this point. [The list referred to follows :] REFERENCEB 1. Mann JI, Vessey MP, Thorogood M, and Doll R; MSocardiaU infarction in, young women, with special reference to oral contraceptive practice. British 3fedicalJournalll:241-245„1975. 2. Mann JI and Inman WHW : Oral contraceptives and death, from myocardial infarction. British Ilfedical Journal II : 245-248;,1975. 3. Mann JI, Inman WHW, and Thorogood M: Oral contraceptive use in older women and fatal myocardial infarction. Britiah Dfedicai Journal II: 445- 447, 1976. Mr. FOUNTAIN. We appreciate your analysis of the statistical as- pects of the studies relied upon by FD A for their requirement of an antismoking warning, for OC labelhng. We have the same question to ask you that we asked of' the other two non-Government witnesses. According to your testimony today, each of the studies has defects which, you say, on an, individual basis do not permit a scientifically valid conclusion to justify FDA'sdecision~ to require t'heantismok- r f
Page 140: azo61e00
-k- 136 ing warning on the OC label. Do you have an opinion as to whether collectively the results of the studies considered would provide a valid basis to support that decision? If so, what isit? Mr. BunNE: Mr. Chairman, the second study by Beral, in fact, does not support the warning label. Therefore, the only support cited is that by Jain, and I find that his results are not statistically reli- able. Therefore,,I can find'that the results are no different collectively then they are individually. Mr. FOUNTAIN. Are you familiar with the recent papers by Jick and associates and Petittii and Wingerd on the question of increased risk of circulatory disorders to OC~ users who smoke? Mr. BUDNE. No,,sir. I have not continued my study of these artlicles, beyond those supporting the labeling. Mr. FOUNTAIN. So you have not had! an occasion to statisticallyy evalhzate the data ; is that right ?' Mr. BunNE. That is right. Mr. FOUNTAIN. Mr. Budne, it seems that most scientific and statisti- cal studies are open t.o criticism. for a variety of reasons, particularly inasmuch as most Americans are laymen and are not in a position to analyze and understand all of the evaluations and conclusions which are reached therefrom. Are the kind of problems that you describe sufficiently serious to justify a conclusion that FDA'sdecision is not well-founded, or even unfounded'y at least based on the studies cited'in the Federal Register in connection with the patientlabeling? Mr. BUDNE. I find that in this particular case, Mr. Chairman, the support is as poor as one could find in~ any such situation. I feel that there is no justification based on these particular studies to support that labeling. Mr. FOUNTAIN. Mr. Waxman? Mr. WAXMAN. Are you aware of the legall standard by which FDA must make a decision? That is, whether to have a,varning label or not? Mr. BUDNE. I am not sure I understand the question. Mr. WAXMAN. In answer to a question by the chairman you said that the FDA was not justified in requiring the labeling warning. I was wondering if you are aware of the legal standard required of the FDA to make that decision ? Mr. BLTDNE. I would expect tQiatt such decisions made by a Govern- ment agency would be based on scientific fact. In~thisregard, the sup- port fails. Mr. WAXMAN. Is it. your understanding that a decision bv FDA would be based! on the same standard of! proof that tiou w,ould! accept! as a scientist? Let me elaborate. Do vou think as a scientist you look for certain O kinds of evidence to reach certain conclusions that are more rigorous (,) in scientific evaluation than is required of a public health agency ~~ trying toprot'ect tliepublichealth~'? ~ Mr. BLnvF. I see. I understand Yourquestionnon-. ~ I suppose my standards, personally, would be higher than~those of'a ~ layman. However, in, this particular case, the support is so weak that ~ drawing the line as to what is acceptable and not acceptable is not ~ difficult to make. ir
Page 141: azo61e00
137 Mr. WAXMAN. Do you believe that women should feel that they are running minimall risks of cardYovascular disease when they use a birth control pill and smoke cigarettes'?' Mr. BUDNE. Sir, I try to base myy own opinions and jiidgments on fact. Based on what I have seen, I find the facts supporting,that opin- 1011 arc ~M1: : C4 +hore.. Mr. W.,kxMnx. Mr. Chairman, I have no further questions. ' Mr. FouxTAlx. Mr. Waxman, do you want to continue? Mr. WAXMAN. Yes; let me ask one more question, Mr. Chairman. If you were to be asked your personal judgment as to whether it is prudent or not for someone to und'ertake certain kinds of activities, ` would you advise an older woman that if she smoked and took the birth control pilll that she should feel there is no greater risk of catrdio- vasculardisease than, if she did neither? Mr. BUDNE. If I or anyone else answered: that: question, it would be a: pure opinion rather than being based on any sup2)ortive facts. I«ould' prefer to feel that my statements are based on my expertise at this hearing. Mr. WAXMAN. Would you agree that an older woman who smokes runs an increased risk of cardiovascular disease'?' Mr. BUDNE. I have I:o evidence on~ which~ I could comfortably sup- port such an opinion. Mr. WAXMAN. I see. Do you f'eel the claims that have been made that smoking is dan- gerous to public health are claims that are not supportedscient.ificallv.'? 1Ir: BUDNE. In this particular case; I have not investigated that whole issue. Mr. Waxnz.av. You are not unaware of the issue, are you? Mr.BLnxE. I am certainly ajvare of it; yes. I certainly feel that perhaps the evidence which has been presented in general studies shows a correlation between smoking, and certain diseases, but cause and effect certainly cannot be proven by the statisti- cal studies. Mr. WAXMAN. «Toulkll it be prudent public policy to withhold any comment about the dangers of smoking unt.il all t'lieevidenceis in that there is a causal connection? I am talking about if'you were in a position to make public policy. Mr. BunxF. If I were in that position, I think I«ould spend a great deal more time studying the issues than I ha.ve as of this point in time. Mr. WAXazax. It seems t~o me t'hat what yotu are saying is that vou reallyy do not have an opinion as to tivhether there is an increased risk. You are attacking the FDA's n•arning because they happen to believe that smoking is dangerous and that in terms of protecting the public healthy women should be.rarned that tliere is an lncreased risk among pill users. Mr. BunxE: I have addressed the. particular cited studies which they have used to support their decision. I find the support that they are using is just not there. Mr. WAXMAN. Thank you: Thank you, Mr. Chairman. Mr. FouNTa1N. Mr. Goldhammer? Mr. GOLDHAMMER. Mr. Budne, have You made anyy consumer surveys t'~o determinetonsumer reaction tolabelingstatements?
Page 142: azo61e00
138 Mr. BuDivE. No, I have not. Mr. Gor.DHAMMER. You do not regard! yourself as an expert in that field? Mr. BUDNE. I have not had the experience of conducting such a study, no. Mr. GQLDHAMMER. Thank you. Mr. FOUNTAIN. Mr. Budne, we thank you for your testimony. Our next witness is Dr. Donald Kennedy, Commissioner, Food and Drug Administration. Commissioner Kennedy, if you will come forward, we will be glad to hear from you at this time. STATEMENT OF: DR. DONALD KENNEDY, COMMISSIONER, FOOD AND DRUG ADMINISTRATION,, DEPARTMENT OF HEALTH, EDUCA- TION, AND WELFARE; ACCOMPANIED BY DR. J. RICHARD CROUT, DIRECTOR, BUREAU OF DRUGS; AND DR. HOWARD WILLIAM ORY, CENTER FOR DISEASE' CONTROL Dr. KE_rr EDY. Thank you, Mr. Chairman. Mr. FOUNTAIN. Please introduce: those who are accompanying you~ Dr. KEN NEDY. I will be delighted to, Mr. Chail rnan. On my left it is a pleasure to introduce Dr. Ory who has been referred to earlier in the hearing. He is the author of one of the review articles and a member of our advisory committee: I know b.e will be able to supply you and your colleagues the answers to some of yourquestions. On my right is the Director of FDA's Bureau of Drugs, Dr. Richard Crout. With your permission, M'r. Chairman, as you willl recognize, I have been listening to the previous testimonv: Some of it concerns us very much because we believe, without having, had a chance to examine it in detail, that it misrepresents a number of the basic facts in this matter. We w.ould! be grateful if you .votild! be willing to hold open the record for us to comment on that testimony as welll as to supply an~. amplification which vou and y ottr colleagues would like to see us m~.ke. Would that be possible ? Mr. FouNTAir. Let me say then, that we will hold' the record open for you to submit any informatiom you have. We will also hold the record open for all of the other witnesses who have testified and who Ina v ultimately care totestify. Dr. KFN'hEDY. Thank you: Mr. Chairman. Mr. Foti ti TaIl N. Without obiection, it is so ordered. [Adclit'ional information follows:] I' DA. COMMENTS ON TESTIMONY GIVEN BYDR. KASTENBAUIM, DR. CirIBBO.ITS. AND MR. BUDNE In evaluating the comments of Kastenhaum. Gibrons and: Budne we note a uniform approach whic11 seems to denote notl only a nrisunderstanding of what map lie claimed from epidemiolozieal studies but' an oversight in dismissing the role of the theory ofistatlistical inferences. The comments of FDA statistficians and their analyses of the papers on smoking and oral contraceptives as risk factors for mvocardial infraction and deathi due to cardiovascular disease are referenced by the subcommittee. The real evidence that wehase is demonstrated by some of the "n" raluesand confidence limits that we have referred to. Thus we stand by thest'atements made in the Dast~. And' while there is agreement between FDA andi Tobacco Institute statisticians thatl the studies havesome deficiencies FDA
Page 143: azo61e00
139 statisticians show that there is some strong statistical evidence that cannot be overlooked, as indicated in~the Commissioner's statement. Individual case control studies and surveys can present evid'ence that an association is present i.e., an event, myocardial infarction, is found much more frequently among women who use oral contraceptives and who also smoke than among similar womeni who are, exposed to only one of these factors or to neither of them. This does not mean that such a study demonstrates or proves a cause and effect relationship. However, in this case several different investigators have conducted studies of different designs using different groups of! women and all report similar findings; i.e., the Dlann„ the Beral, the Jick and! the Pettiti and Wingerd publications all consistently report the same type of association. It may be observed that in each study the incidence of the study disease is almost non- existent in the young women who had been exposed neither to oral contraceptives nor to current cigarette smoking, so that the association may be considered to be strong. Furthermore there is a temporal relationship; referred to in the FDA statisticall reviews, that women who were exposed to these risk factors in the month proceeding, the morbidi event had a higher incidence rate than those whoo had stopped but had been exposed at a prior time. It seems surprising thatl Kastenbaum, Gibbons and Budne criticize the subject sthidies on the basis of the numerator data without taking appropriate note of thee size of the reference groups or of entire study group or denominator. Note that the. 3fann, study of non-fatal myocardiaUinfaretion includes 58:myocardial infarction patients and 166 controls or 224 study women and Beral"s'mortality data is drawn from a longitudinal study of 46,377 Nvomen ; 23,611 who used oral contraceptive pi11s and 22,766 women who never used them. Another important consideration is that all formalistatistical significance testing,procedures take the sample size into account as an integral part of the computation. Furthermore; these criticisms reveal that no effort has been expended to see whether the published calhulations are accurate or whether a more appropriate analysis would provide deeper or more reliable insight ; or whether appropriate confidence limits raise questions or shed further light on study findings. For example, in his paper Oryy stated that using the niann data, the incidence of~ non-fatal mSocardiali infarction among non- smokers was not statistically significantly different from thati for smokers. He made this observation by comparing the approximate confidence intervals for the corresponding risk ratios. Hosvever;, to him the data suggested an important dif- ference. Dr. Ory's: problem was that lie used an indirect procedure which does not directly assess the difference between the distribution, ofOC users in the 3iI cases and Controls among the smokers and, among the non.smokers. A suitable method of testing that difference directly 7sthe empirical logit method described by Cox (Analysis of Binary Data, Methuen, London 1970) and the results of using that analysis are shown to be statistically significant at p=.001b level of test (as noted in Bureau of Drugs Biometrics Division memo dated April 4, 19766 in table A-4). Dr. Kastenbauni raises four specific questions which are answ:ered belbw : 1. Is the study group large enough and sufficiently representative to allow con- clusions to be made with respect to the entire population from which these persons come? ZVitln respect to size, 224 women were studied by Jlani.u, and.46,377 in the Royal College of Generall Practitioners Studr of Oral Contraceptives. Moreover as the Mann study of non-fatal myocardial infarction includes the data from 2 of the 15 (13.3%) hospital regions of England and Wales and the General Practioners Study includes data froni 1,400 physicians there seems to be no question about the representativeness of these data for England. The subsequent publications by Jick and Petit and Wingerd demonstrate that findings in L'.S.women are consistent with the British reports. 2. Is the information drawn from~the data~described accurately by the authors? The problem is not restricted to what the anthors said but it should also include those findingswhich are present ini their data without further description or dis- cussion. In response to Dr. Gibbons' comment the calculations of Jain and Ory are presented in a manner that is clear enough to verify their arithmetic. Some problems,have been identified with respect to the transfer of smoking experience reported by Mann for non-fatal myocardial infarction to the fatali MI findings of women whose age is reported but risk factors are not available. However, the. Bureau of Drugs Biometric analyses of the 11lannistudies indicates that,both cases and controh in the unu-fatal myocardial ihfaretibn study have been expose& to some acknowledged riskforthedisease ; this means that thesepat'ients in both
Page 144: azo61e00
140 groups would be more likely to get an infarction than would women without addi- tional risk factors thus making it easier to detect an association if it were present Turning to the Beral paper on mort'ality*, note that the mort'ality findings aree presented in rates per 100,00Q women-years. As: the experience in each year is related to what' happened in prior years the use of women is more informative. Consequently, we look at', number of' deaths among these study women without respect to years,of study observation and fin& the following: Smoking hiitory OC takers Control 00 deaths Control' death; Odds ratio Exact test(p)' None------------------ 12,252 13,306 5 2 2.72 0.19 Smokers ............... 11,359 9,460 19 3 5.28 .002 Overatl........-- 23,611 22,766 24 5 4.63 .004 We see now that the approach taken by Beral in analyzing the data by women years is a very conservative approach~which makes it difficult to detect the differ- ence between the mortality observed among smokers and non-smokers. 3. Are the statisticall methods used to analyze the data appropriate? Are the, statistical assumptions made by the authors justified : In responding to the preceding question we have noted that' the methods used by Beral are: conservative-thus they are appropriate in the sense that they are unlikely to overstate the risks of OC use and/or smoking withi respect to death from circulatory disease. Mann in analyzing his data on non-fatal myocardial infarction described the distributions for various known and suspected risk factors for the disease. He then tested these distributions one at a time to identify the risk factors which were associated with myocardial infarction at a statistically significant level of test (p<.02),. He then looked to,see how many women with m,yocardial infarction had one; two or three or more of the significant events and conclrzded that there were not enough cases to analyze the data for more than one risk factor at a time and that the subgroups were too small to look at individually. Since statistical procednres exist which enable, us : to adjust the data: for 2 or 3 factors' and: others which permit direct evaluation of 2 or more factors simultaneously, such methods were used by the Bureau of Drugs'statisticians im their evaluations of these data. The analyses of Mann's data made by Jain are' open to question and FDA statistical reviews,have discussed the problem in detail-however it is important tb note that the basic risk ratios (odd ratios), calculated by Jain, for risk of MI among, the four basic smoking-OC: risk groups are mathematically correct and~ have a degree of validity which is supported by finding of trend and Cox type analyses referred to above and by approximate confidence limit statements about the relative risks: non.smoker user of OC relative to, non-smoker non-OC, user showed an odds ratio of 2.02 with approximate 95 percent confidence limits of 0.31 and 12:13 ; smoker non-users of OC relative to non-smoker non-OC user showed an odds ratio of 2.15 with approximate 95 percent confidence limits of 0.88 and 5.53; but smoker OC users relative to non-smoker non-users had an odds,ratio of 11.67 with approximate 95 percent confidence limits of 2.87 and 49.93. Thus while the numbers are a bit thin to reliably estimate the relative risk for women n-ho only smoked or used OC without smoking. there is no problem about estimat- ing that the risk associated withi smoker OC user relative non-smoker non.users is statistically significantly greater than 1 (the risk of non-smoker non-users rela- tive to themselves.) The "p1" values are given in, the table: 4. Do the authors discuss any statistical shortcomings of their analyses? Have they ignored any other statistical problems?' While the authors of the subject articles have not discussed shortcomings of their analytic procedUre they have been extremely cautious about selection of subject matter and of techniques: As a result more complex analyses that might have been more informative have been avoided in addressing, possibly contro- versial issues. THE TOBACCo~ INSTITUTE Washington, D.C., December 5, 1978. Honi L. H'. FOUNTAIN, U.S. House of Representatiuea, Rayburn House Office Building, Washington, D.C. DEAR 'Me, FoUNTAIN : During the Hearings before the Subcommittee on Int'er- governmental Relations and Human Resources on October 4, 1978, you mentioned that the recor& would be kept open for additional submissions. I respectfully re-
Page 145: azo61e00
141 quest that this letter together with the enclosed material be made part of the record of these Hearings. The first enclosure is my statement in response to some parts of Commissioner Kennedy's testimony. The second enclosure is an article that appeared recently in World Health„a WHO publication. It deals with oral contraceptives and contains the following statement in the second from the last paragraph : "Increasingly, however, research is identifying particular sub-groups at special risk, such as women over 35 who smoke. It would„ with~ the evidence available, seem prud2nt for these women to use other methods." This is in marked contrast to the FDA package insert warning advising women to give up smoking. Sincerely, MAavIN A. KASTENBAUM, Ph. D. Enclosure RESPONSE TOSTIITEME\1ti OF DB. DONALD KENNEDY Early in Commissioner Kennedy's testimony he stated that' it was his belief that some of the testimony given by Dr. Jean D. Gibbons, Mr. Thomas A. Budne. the late Dr. Gertrude Cox, and me "misrepresents a number of the basic facts."' This comment is both professionally and personally reprehensible and false. When~ statisticians of undeniable professionali standing have spent considerable time studying a complex problem to arrive at well reasoned conclusions, it'~ is incredible that a scientist of the caliber and reputation of Commissioner Kennedy would so casually Impugn their integrity. Further, one can only be shocked', and saddened when a government official with the highest responsibilities in science and health assumes an adversarial or defensive posture n•ith respect to a significant'agency decision instead of fulfilling his duty to,determine what the current,scienee shows. A decision (the new labeling of oral contraceptives) that affects millions of women of child-bearing age should: be given the most criticaliscrutiny: Views,from both independent andi government statisticians should receive considerate and respectful attention so that,no question of partiality could be raised about,the sponsorship by proponents or manufacturers:of oral contraceptives of some of the major studies relied uponiby the F.D.A. Later in his testimony, Commissioner Kennedy responded to a suggestion made by Dr. Gibbons that a properly designed controlled study would provide useful information about the possible relationship between smoking and the use of oral contraceptives. Commissioner Kennedy stated that'~ this was impossible because "we would have to get'doctors to violate their own concept of good practice in order to combine rii;k factors." This response betrays a basic misunderstanding of the design~ of epidemiological studies. Dr. Gibbons' advice was directed solely to the gathering of more relevant data (genetic, environmental, and psychological characteristics): which could be obtained from, among other sources, medical histories and personal interviews. The decisions of private physicians to prescribe or not to prescribe oral contraceptives for their patients would not be involved at aM Such a study could he conducted if Commissioner Kennedy desired. In Commissioner Kennedy's prepared testimony he referred to charts which were displayed before the: Subcommittee and~ irtilized! by Commissioner Kennedy im responding to questions. These charts;, which purported to analyze the various studies dealing with oral contraceptive use and smoking, were inaccurate and misleading in a number of respects: An example of the misuse of these charts relates to the emphasis given in Commissioner Kennedy's testimony to the so-called synergistic action of oral contraceptive use and smoking in regard to circulatory disease causation. Com- missioner Kennedy, relying on a prospective British study, stated, "The Royal College of General Practitioners, which is conducting a prospective OC study, found smoking and oralicontracept'ive use to be independent risk factors for death due to circulatory disorder:s ; the combination, of smoking and OC use wasfound at least additive, and possibly synergistic .. " This statement'~ does not make sense statistically or scientifically because the finding of independence between, two risk factors necessarily precludes,the possibility of synergism. These are the universally accepted meanings of the terms "synergism" and "independence`." : Further, the Royal College Study explicitly concludes on p. 730, "This suggests thatl the relative increase in mortality associated with oral contraceptive use is independent of smoking habit." (Emphasis added ) Consequently,, it is inconceiv- able that Commissioner Kennedy could find support in this study for his above- quoted remark and for the entry "Suggestive" in the chart' column labeled "Evi- 35-727 0 - 79'- 10 ~
Page 146: azo61e00
142 dence of Synergism" with respect t.o th2s'study in the chart displayed for the Subcommittee.. The entries in the chart in~ the columns headed "Relative Risk" are also disturb- ing, particularly those for the Royal College of General Practitioners study. Of the three relative risk entries in the chart for this st'udy, only two (4.7 and 3:0)) were presented in the original article (Table IV). The authors chose not to include in their article the third relative risk, 13.2, which was apparently calculated byy the F.D.A. In~addition, the F.D.A. placed no entry whatsoever in the last column labeled "Relative Risk of Diagnosis Among Smokers for Current vs. Non-Current OC Use" despite the availability of this number (4.4) in Table IV. This omission can perhaps be explained by the fact that in this study the relative risk of smokers who use oral contraceptives (4.4) is less than the relative risk of non-smokers who use oral, contraceptives (4.7). This selective use of the calculations presented by the original authors is indefensible because this very issue was discussed heatedly during the F:D.A.'s Obstetrics and Gynecology Advisory' Committee meeting In November 1977. By omitting a relevant calculation contained in a table in the paper and including a calculation not provided by'the authors, the F.D.A. has's produced a chart that is highly misleading and incomplete. Observations similar to the above can be made with respect to the chart entries for the study by Mann, et al. (1975). None of the relative risks that appear in the chart for this study were calculated by the authors. These calculations were taken mainly from the Jain and Ory papers and presented in the chart', in such a manner as to suggest that they occurred in the original Diann paper. To further confuse matters, there were calculations of relative risks in the Mann paper which were not referred4o by the F.D.A. despite their undeniable relevance. For example, the relative risk of admission for msocardiali infarction for users;, smokers and non- smokers, was calculated to be 4.5 by Mann, et all This number is: quite close to the 4.7 relative risk from The Royal College study and the 5:4'relative risk calculated by Ory. With relative risks differing so little,, it is difficult, if not impossible, to draw, any firm conclusions. This selective and misleading use of Information as seen in the F.D.A.'s charts'represents a severe departure from generally accepted scientific practice. During the question period following Commissioner Kennedy's prepared testi- mony Congressman L. H. Fountain submitted for the Hearings record severali memoranda prepared by F.D.A. statisticians, some of whom were present at the Hearings. It is striking how similar their assessments of the scientific evidence under discussion are to the evaluations of the four nongovernment statistitians. Tbe F.D:A. statisticians raised serious questions about the adequacy of the data and were especially critical of the small numbers of cases andi controls. In refer- ence to the Mann study of non-fatal myocardial infarction they noted r"Thus it appears to be misleading to suggest that'the data are sufficient to evaluate the effects of individual risk factor(s) or oral contraceptive use." (Memorandum, Apri14, 1977). In, another memorandum the same theme is repeated :"Too few non-smoking current users were studied to permit one to infer ... that the increased risk of myocardial infarction associated with the current use of orall contraceptives is greater among smokers than among non-smokers." (Memorandum, September 26, 1977) Again the F.D.A. internal evaluation is highly critical of the inadequate data base: "Jain has:not resolved the issue that thereare(nsufficient data to evaluate effects of individual risk factors for non-fatal myocardial infarction and almost no data: for fatal niyocardial infaretion ..." (\iemorandum, April 4, 1977) These conclusions are unequivocally clear and entirely consistent with the views of Budne, Cox, Gibbons:and me. Nevertheless in his testimony Commissioner Kennedy described these conclu- sions of his own statisticians as,the exposure of"`some defects" or the pointing out of not very serious "deficiencies" in the studies. It is nartieularly difficult to under- stand the Commissioners' statement that, "The very firm association between OC use and smoking in enhancing cardiovascular risks has now been established beyond al reasonable doubt." in light of the F.D.A. statisticians' carefully reasoned conclusions. The harsh language used by the F.D.A. statisticians in describing Jain's anal- yses: of the \Iann data is somewhat' surprising because Commissioner Kennedy relied heavily on Jainrs calculations im the charts lie displayed before the Sub- committee. The F.D.A. statisticians refer to Jain's work as "ruminations in the absence of facts" and they continue: "Calculations based on rumination or mathematical models are not verifiable except by additional studies. Therefore, I ~. ~ i'
Page 147: azo61e00
143 they should be considered with caut'ion ; the comments made by Jain~ in this area are open to criticism with respect to thedata, the methodology used and the way in which he draws inferences." (Diemorandum April4, 1977) These comments provide an excellent illustration of the universally accepted statisticali maxim which states that statistical analyses of data can be given no more credence than the data themselves no matter how sophisticated, esoteric or apparently illuminating the statistical techniques are. The strong emphasis given by Commissioner Kennedy during his testimony to two studies published after the F.D.A. made its decision to require the new warning is quite revealing, This emphasis is, in effect, an admission that the studies relied upon bythe F.D.A. in its decision to require the new warning (^_ite& in the Federal Register of January 31, 1978) did not support the warning, Further, these new studies, by Jick et ali (a study of 26 cases of non-fatal myocardial in- farction) and Petitti and Winger& (a study of 11 cases of a rare congenital disease) do not provide any additional justification for the F.D.A. action for the reasons pointed out by both Dr. Gibbons and me during the question periods following our prepared, testimony. It is severely disappointing to a statistician to observe the apparent inability or unwillingness of the F.D.A. to engage in a responsible scientific discussion of the merits of one of its decisions. [From World Health, Aug.-Sept. 1978, p. 12-15] HOW SAFE IS THE PILL?, Millions of women throughout the world find the Pill effective and acceptable; the overwhelming majority experience no ill~health as a result (By Ron Gray ) The use of hormones for contraceptive_ purposes is a relatively recentmedical advance. Only in the 1950s did researchers discover that two types of hormones, called estrogens and progestogens, could in combination aet as a highly effective contraceptive. These substances arethe constituents of combined oral contra- ceptives, commonly called "the pill", which first became available for general use in the United States in1960. The pill was rapidly adopted in manyot'her count;ies, and the growth ini popularity of this drug has been such that„accordingto present estimates, more than 80 million women around the world are currently using the pill. This is a unique situation in modern medical science, since never before have such a large number of healthy young women taken potent hormonal drugs over long periods. Not surprisingly„ the situation has caused some concern to the authorities responsible for drug safety in many countries ; and not unnaturally, women who are actually taking these drugs,or who have used the pill in the past, are legitimately concerned that it may affect their health. The evaluatiow of drug safety is extremely complex,, and although there are certaim health risks associated with the use of the pill, it is important to place them in a: proper perspective. They must be balanced againstthe potential hazards associated with the unwanted pregnancies thatl might have occurred if the women had not been using the pill, or against the risk associated with other forms of contraceptive or abortion. This weighing of risks, which is still the subject of study and debate, is~ a difficult matter of statistical and medical judgment. Jiany drugs lead to so-called "side-effects" which, though troublesome, are not life-threatening medical conditions: With the pill, many women experience relatively minor side-effectis such as nausea, weight-gain, changes in mood. problems with their complexion, and so on These may respond to medicali treat- menr, or may be so unacceptable that some women stop using, the pill. But it is very difficult to measure the impact of such side-eff!ects oni health. Most of the questions ofsafetytn be considered here deal with much more serious medicali complications. Many of, the short-term complications that occur during or shortly after starting to use the drug are associated with the heart, blood vessels or blood pressure (the cardiovascular system), or changes in the body chemistry. Numerous studies have assesse& the potlentiali risks associated -%viththese short-term complications,, and these will be described later. However„ there may he longer-terni problems which may result from either very prolonged pill taking, or which do not occur until many years after a woman has stopped taking it. The pill has only been widely available for 18 years„ and
Page 148: azo61e00
144 few women have individually experienced both prolonged use of the drug and the intensive medical follow-up required for proper evalUation of any long-term effects. Therefore existing evidence can only be applied to a relativelq limited life-span, and it is not possible to predict problems that may arise in the futhire. Before considering the medicaL evidence for healt'h~ risks, associated with the pill, it is necessary to describe some of the safety screening that is tmdertakeni prior to the release of drugs for general use. In many countries, the drug regnla- t'ory authorities lay down stringent rules for testing drugs, both in animals and ini carefully controlled studies on humans, before such drugs are released for use cn a wider scale. Over the years,,the requirements for drug testing have become more and more vigorous. The pill was introduced before the present stringent standards: for drug safety screening were adopted, but over the years the various brands in current use have undergone extremely careful and continuing testing, Despite the careful testing, there are still many relatively uncommon c+omplica- tions which cannot be detected untili the drug has beeni used on a very large scale. For examplejf an illness usually affects one person per thousand per year, and the drug increases the risk of this illness fiVefold, to five cases per thousand, it is: highly unlikely that the association between the drug and this illness will be found until it has come on to the general market. In addition, many complications: could not have beemaccurat'ely predicted from the screening and testing of drugs in ~ animals or in clinical trials. One difficulty about research on drug,safety is that drugs seldomi cause ill health in the absence of other predisposing fact'ors; they usually interact with, other conditions to cause disease. This,meansthat only a certain minority of women in a population may be at serious risk of complications and, equally, it may be very difficulti to generalize the results obtained from research in one country to another,, if the background risk factors differ considerably. One of the most serious,problems encountered with tbe pill is thatl the estrogen componentl tends to increase blood clotting. But it', was not until the pill had been in general' use for eight years that researchers ini Oxford first showed an associa, tion between the use of oral contraceptives and an increased risk of blood clots (thrombo-embolic disease) affecting the veins of the leg, lung and brain. Subse- quent research both in the United Kingdom and the United States has confirmed these findings, and has also shown that the use of oral contraceptives is associated with am increased risk of heart attacks (myocardial infarction) and of high blood pressure (hypertension). This catalogue of caidiovascular conditions linked with the pill is daunting, and to it can be added other illnesses such as an,increased risk of'gall bladder disease, migraine and diabetes. Balanced against these increased risks are the pill's protective effects. For in- stance, benign tumours of the breast andi ovary, severe menstrual disorders and rheumatoid arthritis: all tend to be less common among users of oral contracep- tives. We also have to consider the risks of Illness associated with other forms of contraception, pregnancy or abortion. With so many competing, alternatives of risk and benefits, it is very difficult to draw up a final', balance sheet for the pilll particularly since some illnesses such as heart attacks or strokes are far more serious than others such~as migraine or menstrual disturbances. In the end, the relative hazards of oral contraceptives must' depend uponi the deaths which can be attributed to their use. It', has taken, almost two decades to arrive at scientificallpbased estimates of the excess risk of death, (mortality) associated with, the pilh These studies have mainly been eonducte& in, Britain: and the results: may be summarized with a: few statistics. Non-pilltaking British women aged 15 to 49 years have an overall death rate from cardiovascular dis- eases of about 5.5 per 100.060 per year. By comparison, the death rate among orali contraceptive users who do not smoke is estimated to be about 13.8 per 100,000 per year. and among those pill users who do smoke, the death rate is thought to, be around 39.5 per 100,000: ner year. The risk of death increases markedly among women over the age of~ 35, especially If they are also smokers and have used'. the pill for long periods of time. These statistics are, of necessity, an over-simplification of the position, but there is now no reasonable doubt thatl, in industrialized societies, the pill leads too an excess of deaths frcm~ cardiovascular disease. However, two mitigating con- siderations must be borne in mind. Firstlp,, some of the current evidence applies to hrands which contained very high doses of estrogen and are now no long~er in widespread use. The new lower-dose formulations are likely to have a reduced health risk. Secondly, although adverse effects of the pill are severe and should
Page 149: azo61e00
50J ~ a 40 w 0 30J 20J 10~' 145 ~r ` ~ J , .- . I~ ~`1 I ~- ~ i ;h. - ~ ~ ~' ' % i %,1i,.01Y, I V p?er 4 _. DrGreqorv Pincus (190.?-1967) of the USA. ; 'Fartiaro/ tAe PiIA':'. (Foto Italia MiMn.OJ. , Oeetlhs in UK due to arcularory.diseases associated wiNh'poluseenQsmok/ng .. y BANG4ADESH t
Page 150: azo61e00
146 not be dismissed lightly, the conditions causing death are relatively uncommom The medical' journal Lancet which published recent studies on pill-associated mortality commented :"Little in this week's Lancet is going to dismay the many women who regard oral contraceptives as a blessing which carries a minute risk of premature death. The doctors whose task it is to interpret this news will have to say the danger now appears to be greater than it once seemed, but it is still, in the absence of other risk factors, very smalllindeed for younger women." However, both the Lancet editorial and the British committee for the Safety of bfedieiries' emphasized the very marked increase in risk among,pill users over the age of 35 who smoke, are overweight or have diseases such as diabetes. Almost alli the studies in the association between pill use and cardiovascular disease have been undertaken in industrialized count'ries where women frequently have characteristics which place them at high risk of cardiovascular illnesses. They often have weight problems, they take relatively little physical! exercise: and many smoke cigarettes. In most developing countries, this is not the case and the: cardiovascular hazards may be substantially lower. In addition, the high risk of death associated with childbirth in most, non-industrialized countries may more than offset any pill-relat'ed ill health. However. further research~ is required, and WHO is undertaking an active pro- gramme to evaluate pill safety; especially in developing countries. For example, studies are being conducted on its effects on types of heart disease, such as rheu- matic fever„which are common in the developing world. The possible interaction between the drug and various infections or parasitic diseases is also under investi- gation, as is the pill's possible influence upon vit'amins in societies where poor nutrition~ is prevalent. It is welli known that many tumours occurring in, women are related to the natural production of'hormones within the body. So it is important to assess the significance of external hormones such as the pill in relation to such tumours. A great deabof research has been done on laboratory animals such as the rat, mouse, monkey and dog. Overwhelmingly, most of this work has suggested that the drugs usediin the pill today have no propensity to cause tumour development in animals. In humans, the situation is extremely complex, because it often takes a long time for tumours to develop and we have not had long experience with the pill. Furthermore tumours occur relatively infrequently during the reproductive ages when women use it, and we have insufficient information~ on the effects of prior pi_11 use on tumour development among older post-menopausal women,, in whom such tumours occur with greater frequency. In 1977 WHO convened a meeting, of international experts to consider these difficult questions, and the Organization is currently embarking upon an extensive research programme to evaluate the risks that, might be associated with the pill and other hormonal contraceptives. The experts concluded that pill use for more than, two years reduced the risk of benign tumours or "lumps" in the breast', and this protective effect appears to be associated with the progestogen component. However, the evidence available thus far suggested neither an adverse nor bene- ficial effect on the risk of breast cancer. After a very careful and extensive review of a large amount of scientific evidence, the group of experts felt that there may be "an increased risk of early forms of cancer of the cervix (the lower part of the womb) among women who have used the pill for a long period of time, and who have other characteristics, such as an, early age of~ first sexual activity; which predispose to this disease": There is evidence of a marked increase in the risk of certain benign liver tu- mours among women who used itl over long periods, but such tumours are excep- tionally rareamong women wtder 45 (in, the United States the diseaseoccurs at w rate of five cases' per million per year ). The chances : of' a pill user developing the dise;tse is estimated to be three per 100,000 per year for women under 30. The risk amongolder women may be substantiallyhigher but cannot at present be estimated:. What are the possible effects on the health of children whose mothers have recently used the pill, accidentally took itf during pregnancy or used it while breastfeeding 7 A number of studies hav.e assessed~the effect of the pill ou subsequent pregnan- cies, and most suggest it may delay the next conception for about two months but has no harmful effects. But this is a difficult area of research and the possibility of rare adverse effects cannot be ruled out at present. The effects of accidental pill use during pregnancy are also aub jiidicr. There is some evidence of a higher risk of abnormalities resulting ih miscarriages or,, very rarely, in live-born infant'sw•ith heart defects, but the research findings are inconclusive.
Page 151: azo61e00
147 In many societies prolonged: breastfeeding is of vital importance to the health of infants, and there is evidence that the pill' can decrease the amount of breast- milk. Some national authorities have advised women either not to use it while breast-feeding, or only to use it once lactation is well4established. The hormones containedi in the pill have been shown to enter the breast-milk and to: be absorbed by the breast'-fed infant, but the actual amount is : very small and no adverse effects have been demonstrated. WHO is collaborating with~ severaU countries in examining the effects of the pill both on, breast milk and on the health of breast-fed - infants. Clearly the question "How safe is the pill?" does not lend itself to any simple answers: Despite, the: actual and potential health risks;, many millions of women throughout the world find it an effective and acceptable method of fertility con- trol, and the overwhelming majority experience no ill-health as a result. Increas- ingly, however, research is identifying particular sub-groups at special risk, such as womemover 35 who smoke: It would, with the evidence available;,seem~prudent for these women to use other methods. For the larger number of young women who do not have such additional risk faetors;, the pill would seem, to be a safe and reliable contraceptive. There:probably will never be a final and all-embracing answer to the question The answer will vary, depending upon the health of the individual woman and the general health problems or health services in different countries. In the final analysis, it will be for each womani and her health care adviser to decide whether to go on;using the pill, in the light of current medical knowledge. RISKS OF PILL USE, BMOR:INO AND CIIILDBIRTiH Women, in the: United Kingdom who both smoke and take the pill run a much higher risk of death than women who do not smoke and/or do not't take the: pill. However, the risks of, pill-taking must be balanced against' the risks of unwanted pregnancies. Ini Britain, deaths due t'o childbirth are very few,, but' in countries such as Mexico or Bangladesh childbirth is muchi more hazardous. It must be remembered, however, that virtually no information is at present available on adverse effects of the pill inideveloping countries. (See p. 145.) . Dr. KENNEnY. Although I appear before you, as an agency head,. I have been listening to the witnesses list their own scientific creden- tials long enough now so that. perhaps I will begin my testimony in the same vein. Before I became Commissioner in 1977, I was professor of biology at Stanford. I was chairman of that department for 7 years and chairman of the program in human biolbgy for 4 more years. I am the author of about 75 original articles, mostlly in the field of phvsi- ology. I received a Bodich Award of the American Physiological Society in 1970. I am a member of the National Academy of Sciences, a fellow of the AAAS, the American Academy ofI Arts and Sciences. and before I came full time to the Government'. I served as senior consultant to the Office of Science and Technology Policy ill~ the White House. I cite this background, primarily, Mr. Chairman, to underscore a point I would like to make: Although I don't claim~ any special ex- pertise in statistics, I think I am familiar with the evaluation of scientific data in the interest of making wise public policy. I want to assure you and the members with you that I personally reviewed the record of FDA's decision in this case although much of it was made before I came to my position,,and I am prepared to support it fully. Indeed, I am glad of the opporttmityy this hearing affords us because it happens to be & ease in which an FDA decision is not only in our view correctly made in the: first instance, but is in fact being validated in the fullest possible way by the processes through which~ all
Page 152: azo61e00
0 148 scientific conclusions are ult'imatelly established in this world, namely, by open scrutiny review and repetition of analysis. Judged by that stern test, FDA's position has received overwhelm- ing consensus support and continues to receive it. Perhaps some background information on the orali contraceptive use and labeling matter would be helpful. Oral contraceptives have been marketed in, the United States since 1960'and are now used by an estimated~ 10 t'o 11 million women in this country. For some years after they were first marketed, these drugs were not associated with serious adverse reactions. More recently, however, various studies have shown that oral contraceptive use may lead to serious adverse reactions; thromboembolism and' other cardio- vascular conditions, including myocardiaL infarction; congenital anomalies in the offspring when used inadvertently during early pregnancy ; gall bladder disease and hepatic tumors. As early as 11966, the FD A established a; requirement for uniform physician labeling for all oral contraceptives: In January 1970, when the association between thromboembolism and oral contraceptive use was confirmed, the labeling, was updated to include this new infor- mation, Additionally, in 1970, "patient package inserts," or PPI's, were de- veloped and required' for these drugs. Women received with each pill packet a brief summary of the benefits and risks. Manufacturers were required by the FDA to supply physicians with brochures for their patients. These contained more detailed information regarding the effects of oral contraceptives. A nationwide survey taken by FDA in 1975 showed that women read the inserts and tound them useful and' clear, but that a large majority wanted more information on side effects and more details on warnings and precautions. In 1975, significant new information led to a need for a revision of - the labeling for both physicians and patients. To insure maximum public participation in the development of the revised physician labeling, a drai't of the revised labeling was circula.tedto: various pro- fessional, scientific, teacher and consumer organizations, and a notice cf its availiabiltitvv for comment by other interested parties was published in the Federal Register on October 24,1975. Many of these comments, as welli as recommended revisions offered by the agency's Obstetrics and Gynecology Advisory Committee, yielded changes. On December 7, 1976, FDA published physician l'abeing guidelines in the Federal Register to be put into use by April 6, 1977. In that issue of the Federal Register, the agency also proposed significant revisions of the regulations requiring patient labeling for OC's, and included proposed b ideline patilient labeling based on the new physician labeling. The new informa.tion, contained in t.he guideline labeling included the finding frorn two studies by Mann and his co-w.orkers, eonfimuing tha,~, an, increased risk of myoca:rdial infarction is as.sociated' with the use of oral contraceptives. I refer here to a study by J. I. Mann, M. P. Vessey, 'M. Thorogood, and R. Doll, regarding myocardia.l infarction in young women with special reference to oral contraceptive practice whi& was published' ini the British Medical Journal, 1:241, 1975, as well as a study by .T. I. _Nfann and W. H. IV. Innnan, with respect to
Page 153: azo61e00
•v 149 oral contraceptives and' death from myocardial infarction,, ibid. 1:245, 1975. One of these studies found that the larger the number of under- lying risk faotors for coronary arterv disease, the greeater the risk of developing myocardial infarction. Cigarette smoking was one of the factlors listed asenhancing the risks. After the labeling guideline was published •in December of 1976, newer information associating increased risk of cardiovascular dis- orders and death with smoking in oral contraceptive users appeared. These data were presented on November 17, 1977, to FDA's Obstetrics and Gynecology Advisory Committee, which then recommended that a warning be placed in the OC labeling. The warning developed as a resul't of this recommenda'tion~ states that smoking increases'the risk of serious cardiovascular side effects f'rom oral contraceptive use; that' this risk increases with age and heavy smoking-defined as 15 or more cigarettes per day-and that women who use oral contraceptives should' be strongly advised not to smoke. Two more recent papers, one by Jick and associates, and I refer here to a study by H. J. Jick, B. Dinan, and! K. J. Rothman regarding oral contracentives and nonfatal mvocardial infarction, Journal of Amer- ican Medical Association 239 : 1403, 1978, as well as one by Petitti and'! Wingerd, regarding't~he use of oral contraceptives, cigarette smoking,, and the risk of subarachnoid hemorrhage, Lancet 2':234, 1978', con- firmed the increased risk of smoking and oral contraceptive use for other kinds of cardiovascular complications. In the Federall Register of January 31, 1978,, the agency published new guidelines on physician and patient labeling ieflecting the infor- mation on smoking and OC use. FDA required that revised physician and' pa.'tient OC' labeling be dist'ributed for all OC's by May 31, 11978. As for risk factors for cardiovascular disease, over the years, many prospective and, retrospective epidemiologic studies have demon- strated that certain characteristics-cailed' risk factors-can result in an increased liability to cardiovascular disease. These risk factors include hypertension, hypercholesterolemia, diabetes, obesity, ciga- rette smoking, and the use of oral contraceptives. I think we can summarize most effectivel,y, for the discussion~ that will follow, Mr. Chairman, these studies by referrina to the charts that we have put up before you. Mv colleague, Dr. Crout, willl now undertake, with your permissiony a brief runthrough on how to read these charts and what the columns mean, so that you can refer to them as we move through the rest of this testimony. Would that be. all right? Mr. FouNTniN. That will be fine. f Chart shown.l l Dr. CROUT. This column shows the study under consideration [in- dicating]. We have put up the studies that hai•ebeen referred to by others previously. The.second column Dr. KENNEDY. Excuse me, the members of the staff may wish t'oo note that versions of these charts have been supplied as part. of the appendix to our prepared testimony so you can see them close up. Mr. FOUNTAIN. Copies of the charts will be placed in the record at this point. [The charts referred to follo.v :]
Page 154: azo61e00
I UII II:ILIII:IIS I IN111NG5 I. IIIIIOIIyIICII' AS)CS`axlClll. al Ilelatlvu Illsk uf the St LJy Illa9noses EvlJeecc uf Ilelallvu Nlsk of Nfaguosls nl- ' C S k s f MI11 lr In l liLnpiu;Is Ili>I. I 1//7 ~_ u41II II ~'1- /lf SIIIIJecls lor• d Lil~ups . . . I Ill;k nl V.u'luas Ages :w'rcnt UC Isc fn 11icme uf mnkfu9 Gn Snxlklug lu Alzulcc uf /16 Suuullaneuus fluYenl Ilsc uf NC anJ Suukloy Synerglsw__ ul urre er wl Auwng -_vs. Nue-Curreul l1C Llsu M..uo cl nl. Ny~li ~u,ll~l l Lnunul Lr a;u:rtaIocd lulnrcliun (uuo/.l l.l l ) 2.02 2. I!~ 11.7 Yes b..42 Itr Annlysls u/ Mann A. Assuupl.lun that Age Is IudepcudunL al Nclellvu I/nln Ily', Nlck ul Myu1 mAlel IIJ.url lue In Varluu s S ub- . Lllu _ _ Iruuys 11. A.~.Imyll.lun Illal Agu 15 Indcpr.udcnl ul lbe 111slrillu- Ilun o/ mulcla In Lhu IV' IP,ur "md,Nue .ISer liruups .112 . Hi 1.7 es 7. Ih y Nuuc 2.02 5.42 I/uyn 1 t•ni,~ I I N~nubcrs uf Ceses Cu l I r lc I'1'u'IIIII'loncrs Ilbr.rlv"d fn yerluus IL l.u~u"c.l /1. l.a I equr I r.s 4.7 1.N 13.2 Sugg usllve ©6t06SE0 I I I
Page 155: azo61e00
IIIIICIIfICIIa LvlAence uf SynerU isal I'ayc 2 Nelaflvq ItIsk 1.1 Illayousi/kllony Sllvlkcrs lur (.urrenl vs. Non-iaureul Ill: Ihr .IIckclaI. 1'1/fl PIISsIIIIe Illas Iu Selectlon ul fases and I:UIItr11I1 Iluc I.allll/ll IIII I./lllllllllell as tilel'L' are 1111 klyocarJlal Inlarc[Ion cases who ~[Fd I/un Inlnl f1yu~ arJlnl Inlnna lun lu Ilsc nl IIPIIA Ilnspllals Ubu wecc Illllluy lo f.uupr.rcalc and yulmllnry Cnnpcrntlnn uf1'hyslclans and I'etlr.llls nul snukc and alsn nul usu 04: Ilepor LeJ 12./5 I•~•.1II11 /l. /lldcr Ul.ncn (Avc ra9e a9u Is uld Hlnqr,rJ I//u '.nbnrarhnn6l _ - 1'1.2 lur I:ases; 4/ lnr . . f.uulruls) 11. Sm~lll Nnmber of Ceses /m' Slrnl 1111 al luu .69 1.9 es Ihauurl a.lc I6L06Sc0 I I I FIfII11NfS Ilelatlvu Illsk of lbo Study 1)layousos Currcol UC Cm•rcul SlxKlllanenus Use In Slanklu9 In Cwreul Ilse ALsunce ul Absenco of u1 111: anJ--' ~moklug 116 Salllklu9
Page 156: azo61e00
 152 Dr. Cxour. The second column shows the deficiencies in each of these studies as analyzed by our Food and Drug Adrninistration staff. You will see that there are known deficiencies identified' in the course of our review. The next major set of five columns states the findings in each of these studies. These findings are expressed as a risk ratio. I will point out that a risk ratio states the numbers of times that a user of oral contra- ceptives has of sustaining the event under consideration ; namely, myo- cardfal'infarction as compared to a: control group. So, if there were no increased risk, the risk ratio would be one. So what this number means here is that the author reported a risk of 2.02 times normal, or 2.15 times normal-that is, times the control group- or 11.7 times the control group,,and so on. The first column here shows the estimated risk for patients on oral contraceptives in the absence of smoking. The second coliimn shows the estimated risk for smoking in the absence of oral contraceptive use. The third! columni shows the estimated risk for persons using both oral contraceptives and cigarettes. Mr. FouxTnix. The figures go from 2.03 to 2.15 to 11.7; is that' right ? Dr. CxouT: That is correct. The next column reports whether there is evidence of synergism; that is, an interaction between the risk of oral contraceptives and the risk of'smokingwhich seems to be more thanadditive. In this study the answer to that question is yes. That is another way of saying that 11.7 would seem to be more than the sum of 2.02 and 2.15. Mr. FovxTnix. What is that based upon? That is, the 11.7?' Dr. CROUT. The 11.7 is the relative risk estimated by these author& in this study of sustaining Mr. FouNTnrti. The Mann study and others? Dr. CKOur. Yes. Of sustaining a myocardial infarction if the person uses simultaneously oral contraceptives and is a smoker. That risk in this study is estimated to be 11.7 times the risk that would occur if a woman is neither a smoker nor using oral contraceptives. That is the way one reads these figures. I think at this~point I will not go on because the Commissioner will run down each study, and as he goes along, I will point to the appro- priat'e figures. Mr. FouxTAix. Mr. Duncan ? Mr. DUNCAN. One of the issues was the confidence in these numbers. Is there any measure of the confidence in 2.02, 2.15, and 11.76 ? Dr. CxouT. That is an important question. It will come out more later. But I will take it up now. In all of these studies, most of' the women who sustained myocar- dial infarction are both oral contraceptive users and smokers: So, in all of these studies the number of patients in this particular column is fairly high. The particular columns where there are small numbers of patients are in these [indicating] columns: There are not very many women who sustained'a a myocardial infarction who are nonsmokers. There are a few more women-but not very many-who sustained, a myocardial infarction who are not on oral contraceptives.
Page 157: azo61e00
153 Mr. FouNTATN. This is on the Mann~stud,y? Dr. CROrnr. Yes. These are general principles throughout all the studies. The statistical reliability of these particular figure& is lesss than the statistical reliability of this figure of 11.7 because the small numbcrs problem is present. Mr. FouNTAi N—. Did Mann have these figures in his study ? Dr. Cttour. Yes: (jSee pp. 169 and 171 for further discussion of this subject.] Dr. CROUT. Let me point this out because manv of the statistical arguments you have heard made by both our staff! and by the previ- ous witnesses this morning, relate to the validity of these lower num- bers of 2.02 which appear in the first column of the chart. Those same statistical arguments do not properly apply to t'hee figures in the third column like 11.7. Most of the patients are in thatt column. Dr. KENNEDY. Mr. Chairman, with your permission, these matter& are going to be dealt with in my testimony. The main purpose of thiss was to lay out the charts for you~ 11ir. FOUNTAIN. Yes; I understand. We will have Mr. Goldhammer ask a question and then we will proceed. Mr. Goldhammer ? Mr. GoLDHAMMER. If 2.02 and 2.15 may not be statistically valid figures, then is it not correct that the 11.7 figure may not be? You said t'heconfidencefactor issma% relatively small, be,cause we do not, have sufficientt numbers. So, there may be a question of the accuracy of 2.02 and 2.15. Is that correct? Dr. CROt; r. Relativelv speaking, yes. I do not' know whether sta- tistical validity was claimed by the authors for the lower numbers. I woulrl' have to go back to our statisticians for that answer. Mr. GOLDHAMMER. Let's assume for the purposes of these questions that those figures are not adequate and that they may be, by some indefinite number, either lower or higher: Then the 11.7 figure, whichiis a measure of the svnerristic effect, is also thrown into doubt, isn't it? Dr. KENNEDY. No, sir. Excuse me,, Mr. Goldhammer, there is a fundamental misunderstanding here. Mr. GOLDHAMMER. Yes. Let's clear that up. Dr. KENNEDY. I am anxious to clear that up. Mr. GOIyDIIAMMER. I am, talking about the lower numbers and the 11.7. In computing the 11.7, the lbwer numbers do not come into the picture at al1? Dr. KENNEDY. That is correct. Mr. GoLDHAMMER. Theny in the course of your testimony. I would like,you to explain how Mann arrived at that 11.7 figure. I understand that that is a Mann value rather than an FDA value. Dr. KENNEDY. Yes ; I will do that. rSee pp. 169 and 171 for further discussion of this subject.] Dr. KENNEDY. Let me begin to work our way through these studies, Mr. Chairman, by saying that there is no si,-nificant disagreement in the scientific community that cigarette smoking enhances the risk for cardiovascular disease, particularly myocardial infarction. 1blann~ et ail reported that cigarette smoking may have not merely an addi- tive but a synergistic effect on risk in women with underlying risk factors. Ory, in an analysis of Mann's paper, concluded that the data
Page 158: azo61e00
154 show that smoking is a strong risk factor for myocardial infarction in young women. I refer here to a review in the Journal of the Ameri- ca.n Medical Association 237 :2619, 1977, by H. W. Ory regarding the association between oral contraceptives and myocardial infaretion. From their review of the paper by Mann et al., on nonfatal myocar- dial infarction in young women, our stat'istieians have reached the fol- lowing conclusions, and' I refe'r here to the study by Mann et al.,, reported in the British Medical Journall 1:241, 1975, relating to myo- cardial infarction in young women with special reference to oral contraceptive practice. After having performed an analysis, Mr. Chairman, which I think your subcommittee staff has had access to, we asked our statisticians to challenge as sternly as they could the statistical basis for these conclusions. Their conclusions from the Mann study are, first, that there is an increased risk of nonfatal myocardial infarction for current users of oral contraceptives between 25 and 44 years of age who already have some other risk factors : for myocardial infarction. That risk is esti- mated' to be 4.5 times as large as that associated with women who are not currently using oral contraceptives. Second, after making adjustment's for smoking habits, there is still an increased risk of nonfatal' myocardial infarction for current users of oral contraceptives between 25 and 44 years of age. The adjustment for smoking decreases the estimate of the relative risk associated~ with OC'use from 4.5 to 3.84. Third, among current smokers there is an increased' risk of non- fatal myocaTdial infarction for current users of oral contraceptives: between 25 and' 44 years of age. This risk is, estimated to be 5.42 times as large as that associated' wit.h women who are not currently using oral contraceptives. Jain concluded that the excess risk-and we are now talking about a study that is, in effect, a review and recalculation of Mann's data. There are no new patients and no new numbers. That. is why the num- bers in the columns are the same. I refer specifically to t'lie: study by A. K. Jain as reported in the American Journal of Obstetrics and Gynecology 126 :301,1976; regard= ing cigarette smoking, use of oral contraceptives, and myocardial infarction: Jain concluded that the excess risk of nonfatal myocardial infarc- tion observed by :12ann and asscciates is largely explainable by the high proportion of smokers in the study and concluded that the use of oral contraceptiR-es ini the a'bse.nce of smoking resulted in lit•tlee or no increase in the risk of nonfatal myocardial infarction. Although our stat'~isticianshave.certain c-rit-icisms about Jain's analysis, theyd'o not believe that those criticisms vitiate the conclusion that smoking con- tributes to increased' risk of myocardial infarction. Dr. Or,v, in his study as reporteA in the Journal of the American 1lfedical' Association in 1977 regarding tlie association between, oral contraceptives and myocardial infarction, like Jain, analyze& the 3lann studv and concltided~ t'hatt there was a strong association between myocardial infarction and smoking ofuormore cigarettes per day.
Page 159: azo61e00
155 In contrast to Jain, Ory interpreted the data to show that oral con- traceptive use alone confers risks. That is the small number column that Dr. Crout was talking aabout before. He concluded t:hat for women less than 45 years of age,,smoking is the strongest risk factor for myo- cardial infarction, and that this risk increases with increasing number of cigarettes smoked per day. The Royal College of General Practitioners, which is conducting a prospective OC study, found smoking and oral contraceptive use to be independent risk factors for death due to circulatory disorders; the combination of smoking a.nd OC: use was found at least additive. and possibly synergistic, with respect to this risk. I refer'here to the Royal College of General Practitioners' oral contraceptive studv regarding mortality among oral contraceptive users. This was reported in Lancet 2:727, 1977, and the principa.l author is V. Beral. Increasing age also increased the risk. Women who used OC's, but did not smoke were about four times as likely to d;e of circulatory disorders as those who did not use OC's and did not smoke ; women who smoked and did' not use 'OC's were about three times as likely to die. But women who both: used O'C's and smoked were about 13 times as likely to die of circulatory disorders as those who did not use OC's and did not smoke. Jick and his associates in their studv dealing with oral contracep- tives and nonfatal mvocardial infarction reported in the Journal of the American Medical Association~ in 1978, found in their case control study that in women between 37 and 4I6' years of apm who both smoke and take oral contraceptives the risk of mvocardiall infarction is high compared with that known for myocardial infarction in premeno- pausal women. These investigat'ors observed that the risk of develop- ing myocardial infarction among OC users of this age --roup was 14 times that of non-OC users, and thev noted that among the women experiencing mvocardial infarction, 92 percent were smokers. Petitti and Wingerd, in their prospective study on the use of oral contraceptives, cimaret..te smokinfr. and risk of subarachnoid 'hemor- rhage as renorted in Lancet 2:234 in 1978: observed that cigare.ttee smokin-- and OC use were independent risk factors for subarachnnid hemorrhage-a form of cerebrovascular disease-but that the effect of smoking and OC use was svner,-ist'ic for this risk. Again, more than the sum. Women who used both OC's and smoked: were 22 times as likely to suffer & suba.rachnoid hemorrhage as those who neither smoked nor used OC's. These~ last two studies which only recently became available lend support to the earlier findinzs. In addition, FDA is supporting, with the National Institutes of Health, a collaborative study nt the Boston Drug Fbiderniology Unit to determine the interrelationship between OC use. citrarette smoking. and myocardial infarction. The data are presentlv being analyzed. but we can say at this point that the findinas, based on a much larger number of'patients than were available in the Mann, studies, strongly supnort the previous findings bv Mann, the Royal College, .iick, and Petit'ti and Wingerd that smokinpr Qrea.tlv enhances the risk for cir- culatory disorders in OC users and that the relation shows synergism. r
Page 160: azo61e00
156 Mr. Chairman, you asked me to address the quality of the evidence leading to the smoking warning in the oral contraceptive l+abeling. The evidence relating oral contraceptives, smoking, and cardiovascu- lar disease comes from epidemiological studies. Such studies are im- portant-indeed, often the only-methods for identifying seriousrisks from drugs, chemicals, and environmental pollutants. Many of the most serious effects uncovered in recent years, such as the carcinogenic effect' in humans of vinyl chloride, estrogens, radia- tion to the thyroid, and cigarette smoking, have come from epidemi- ologic analysis and studies. Despite their value, they are also open to the kind of criticism that can often be leveled against other kinds of statisticali conclusions. In epidemiological studies, Mr. Chairman, we have to deal with the worU as it is and not the way scientists would like to create it. One has the opportunity in a laboratory experiment to design the problem one's own way. But in the real world, one has to take samples of pa- tients who have suffered a particular disease and they may not be as large as one would like. However, when several types of studies yield similar findings, as is the case with estrogensand endometrial carcinoma and with smoking, OC's and and cardiovascular di-sase, the conclusions are so compel- ling that no conscientious ~ policymaker could dismiss them, unless the defects are so serious as to invalidate the conclusions. FDA statisticians have egposed' some defects in two of the OC smoking studies, the Royal College study and' the Mann study. These are similar to those discussed by some of the industry witnesses. For example, im the Mann study on nonfatal myocardial infarction, the number of cases studied was marginal, there was incomplete assess- ment of some risk factors and there were no dat'a: on relative risk with age; in;the Royal College study there were a small number of cases for each diagnosis. In spite of't'hese deficiencies, we believe the evidence was sufficient't to warrant the action taken. Subsequent studies have fully confirmed the high risk of smoking, in OC users, providing further evidence that the action was justified. Mr. FouNTniN. You say the Royal College study, is that the same ass the Beral study ? Dr. KENNEDY. Yes. As for benefit risk judgments; during it's deliberations of t'he ques- t'ion of OC's and smoking, our Advisory Committee on Obstetrics and Gynecology discussed a number of factors, including these: (1) It is well known that smoking is a risk factor for myocardial infarc- tion and other cardiovascular disease; (2) oral contraceptives use also constitutes such a risk factor; and (3) the combination of smoking 0 W and OC use is at least additive, as shown in one study, and probably synergistic, ;is shown in two studies, with regard to risk. ~, These findings led the advisory committee to recommend that the ~ risks be brought to the attention of the physicians and patients, and ~.a we concurred in that recommendation. The labeling does not nealect, to call attention to other factors- M such as hypertension, obesity. diabetes, increasing age-that contrib-
Page 161: azo61e00
157 ute to the overall risk of cardiovascular disease. The labeling con- traindicates use of oral contraceptives in patients with certain known cardiovascular and thromboembolic disorders or a history of throm- boembolic disorders, but does not contraindicate use of OC's in those with obesity, diabetes, and hypertension. Neither does the physician labeling contraindicate use of OC's in ,f, patients who smoke. Instead, it tells physicians that they should strongly advise patients who use OC's not to smoke, acknowledging that' physicians lack ultimate control over how their patients make that choice. The patient labeling, on the other hand,, states flatly that women "' who use OC's should not' smoke. The message is clear that women who wish to continue smoking should seek other methods of contraception. The patient labeling does not' also state that women who have, for example, hypercholesterolemia, hypertension or diabetes should not use OC's. Present data do not provide adequate information on degree of risk associated with OC use and these medical conditions. Such data indeed is difficult to obtain, because physicians tend to avoid the use of OC's in women with the other risk factors when the latter are severe. If further studies do provide information~ on this matter, we will take prompt action in notifying, physicians and patients. But, in order for us to conduct a controlled study that at least• one of your previous witnesses has aske& for, we would have to get doctors to violate their own concept of good pract'ice in order to combine risk factors. Obviously, we can't do that. Because patients have asked for more information on risks as- sociated! with OC use and because we believe that physicians should have the latest available information on which to decide whether to prescribe OC°s for individual patients; we have included in the labeling a strong warning against smoking by OC users. Labeling information allows consumers to participate in basic decisions affecting their per- sonal health an& safety-in this case, to decide for themselves whether they would rather take OC's and not smoke, or smoke and use othec means of contraception. The very firm association between OC use and smoking inenhancing cardiovascular risks has now been established,, we think, beyond a rea- sonable doubt. No less than five independent studies agree on the as- sociation, and show risk elevations for the combination that are truly alarming. I would be astonished, frankl,y; if anvonc who understands the scientific basis for estahlishin~ the presumption~of risk would chal- lenge the conclusion that the lethalityy of' ciLyarettes is areatly amplified in women over age 35 who also use oral contraceptives. - Mr. Chairman, I have taken lonner than I had intended to with this. I thank you for your patience. We will be delighted to answer ques- tions. Mr. FoTn:cTAIr. Thank you. • I want to thank all of you for the brevity of your statements in view of the limitations of.time we have asConzrPss rnshes through and probably passes a lot of unwise legislat,ion. FLaughter.] Mr. Kennedy's prepared statement flollows:1 35-727O.- 79 - 11 N
Page 162: azo61e00
158 PREPARED STATEMENT OF DR. DONALD: KENNEDY, COMMIS6IONER,, FOOD ANDDRUG ADMINISTRATIOF, DEPARTMENT OF HEALTH„ EDUCATION, AND WELFARE Mr. Chaimman: I am pleased to appear before the Subcommittee today to discuss the scientific evidence whi'ch led to The Food and Drug Administration's (FDA)l order of January 311, 1978, to require manufacturers to revise the physicians' and patients' package inserts for oral contraceptives (OC's) to include, among other information, a warning an smoking by OC users. - Extent of OC Use and History of OC Labeling Some background information on oral contraceptive use and labeling would, I believe, be helpful. Oral contraceptives have been marketed in the United States since 1960 and are now used by an estimated 10-11 million women in this country. For some years after they were first marketed, these drugs were not associated with serious adverse reactions. More recently, however, various studies have shown that oral contraceptive use.may lead to serious adverse reactions: thromboembolism and other cardiovascular conditions, includingl myocardial'iinfarctiion; congenital anomalies in the offspring when used inadvertently during eai-ly pregnancy; gall bladder di6ease and hepatic tumors. As early as 1966, the FDA established a requirement for uniform physician label'ing for all' orali contraceptives. in January 1970, when the association between thromboemboli',sm and ora1 contraceptive use was confirmed, the I ~ ~ ~
Page 163: azo61e00
159 -2- il labeling was updated to include this new information. Additionally, in 1970, "patient package inserts" (PPI's)~, were developed and required for these drugs. Women received with each pill packet a brief summary of the benefits and risks. Manufacturers were required by the FDA to supply physicians with brochures for their patients. These contained more detailed information regarding the effects of oralicontraceptives. A nationwide survey taken by FDA in 1975 showed that women read the inserts and found them useful and clear, but that a large majority wanted more information on side effects and nwre details on warnings and precautions. In 1975, significant new information led to a need for a revision of the labeling for both physicians and patients. To ensure maximum public participation in the development of the revised physician labeling, a draft of the revised labeling was circulated to various professionali, scientific, and teacher and consumer organizations and a notice of its availability for conment by other interested parties was published in the Federal!Register on October 24, 1975. Many of these comments, as well as recommended revisions offered by the Aqency's Obstetrics and Gynecology Advisory Commiittae, yielded changes. On December 7, 1976, FDA' published physician tabeldng guidelines in the FederaliRegister to be put i'nto use by April 6, 1977. In that issue of the FederatRe ister the Agency atso.proposed significant revisions of the regulations requiring patient labeling for OC"s, and included proposed guidel!ine patient l!abeling based'onithe new physician labeling. The new infornation contained in the guideline labeling included'ths finding from t:•+a studies l
Page 164: azo61e00
160 -3- by Mann, and his co-workers, confirtning that an increased risk of myocardial infarctiom is associated with the use of oral!contraceptives.1-2 one of these studies found that the larger the number of underlying risk factors for coronary artery disease, the greater the risk of developing nyocardial Infarction. Cigarette smoking was one of the factors listed as enhancing the risks. After the 1'abeling guideline was published in December of 1976, newer Information associating increased risk of cardiovascular disorders and death with smoking in oral contraceptive users appeared. These data were presented'on November 17, 1977 to FDA's Obstetrics and Gynecology Advisory Committee, which then recommended that a warning be placed in the OC labeling. The warning developed as a result of this recommendation states that smoking incrEases the risk of serious cardiovascular side effects from oral contraceptive use; that this risk increases with age and'heavy smoking (defined as 15 or more cigarettes per day); and'that women who use oral contraceptives should'be strongly advised not to smoke. Two more recent papers, one by Jick and associates3 and one by Petitti and rlingerd4, confirmed the increased risk of smoking and oral~contraceptive use. In the Federal Re ister of January 31, 1978, the Agency published new guidelines an, physician, and'patient labeling reflecting,the information on smoking and'OC use. FCA,required that revised~physician and patient OC labeling be distributed for all OC's by'May 31, 1978,
Page 165: azo61e00
161 -4- ct . Risk Factors for Cardiovascular Oisease Over the years, many prospective and retrospective epidemiologic studies have demonstrated that certain patient characteristics--called'risk - factors--can result in an increased liability to cardiovascular disease. These risk factors include hypertension, hypercholesterolemia, diabetes, obesity, cigarette smoking and oral contraceptives. There is no significant disagreement in the scientific community that cigarette smoking enchances the risk for cardiovascular disease, particularly myocardial infarction. Mann et al., reported that cigarette smoking may have not merely an additive but a synergistic effect on risk in women with other underlying risk factors. Ory7, in an analysi's of Mann's paper, concluded that the data show that smoking is a strong risk factor for myocardial infarction in young women. From their review of the paper by Mann et al,.l an nonfatal.myocardial infarction in young women, our statisticians have reached the following conclusions: (1) There is an increased~risk of nonfatal myocardial infarction for current users of oral contraceptives between 25 and 44 years of'age who already have some other risk factors for myocardial infarction. (That risk is estimated to be 4.5 times as large as that associated'with wcmem who are not currently using oral contraceptives.) .
Page 166: azo61e00
1 162 -5- (2) After making adjustnents for smoking habits, there is still an increased risk of'nonfatal myocardial infarction for current users of oral contraceptives b'etween 25 and 44 years_of- age. (The adjustment for smoking decreases the estimate of the relative risk associated with OC use fron 4.5 to 3.84). (3) Among current smokers there Is an increased risk of nonfatal myocardial! infarction for current users of oral contraceptives between 25 and 44 years of age. (This risk is estimated to be 5.42 times as large as that associated with women who are not currently using oral contraceptives.)' Jain6 concluded that the excess risk of nonfatal myocardial infarction observed by Mann and associates is'largely,explainable by the high proportion of smokers ih the study, and concluded that the use of oral contraceptives in the absence of smoking resulted in little or no increase in risk of nonfatal myocardial infarction. Although our statisticians have certain criticisms about Jain's analysis, they do not believe that they vitiate the conclusion that smoking contributes to increased risk f of myocardi al i nfa rcti on ~ Ory7, like Jain, analyzed'the Mann study and concluded~that there was a strong association between myocardial infarction and smoking of 15 or more cigarettes per day. In contrast to Jain„ Ory interpretated the data to show that oralicontracepti've use alone confers risks. He concluded that for women less than 45 years of age, smoking is the strongest risk factor for myocardial infarction, and that this risk increases with Increasing number of cigarettes smoked per day.
Page 167: azo61e00
163 -a- V The Royal College of General Practitioners,$' which is conducting a propsective OC study, found smoking and'oral contraceptive use to be independent risk factors for death due to circulatory disorders; the cambinat'ion of smoking and OC use was found at least additive, and possibly synergistic, with respect to this risk. Increasing age also increased the risk. Women who used OC's but did not smoke were about 4 times as likely to die of circulatory disorders as those who did'not use OC's and did not smoke, women who smoked and did not use OC's were. about 3 times as likely to die, but women who both used OC's and smoked were about 13 times as likely to die of circulatory disorders as those who did not use OC"s and did not smoke. ,lick and associates,3 in a case controL study, found that in women between 37 and 46 years of age who both smoke and take oral contraceptives the risk of myocardiatinfarction is high compared with that known for myocardial infarction.in premenopausallwomen. These investigators observed that the risk of developing myocardial,infarction among'OC users of this age group was 14 times that of non-OC users, and they noted that among the women experiencing myocardial i.nfarction, 92 percent were smokers. Pettiti and Wingerd,4 In another prospective study, observed that cigarette smoking and CC use were independent risk factors for subarachnoid hemorrhage (a form of cerebrovascular disease) but that the effect of smoking and CC use was synergistic for this risk. Women who both,used'OC's and sroked~were 22 times as likely to suffer a . f
Page 168: azo61e00
k 164 - 7 - subarachnoid hemorrhage as those who neither smoked nor used OC's. These last two studies, which only recently became available, lend support to the earlier fi'ndi'ngs. In addition, FDA is supporti'ng „ with the National I'nstitutes of Health, a collaborative study at the Boston Drug Epidemiology Unit to determine the interrelationship between OC use, cigarette smoking and myocardial infarction. The data are presently being analyzed, but we can say at this point that the findings, based on a much larger number of patients than were available in the Mann studies, strongly support the previous findings by Mann, the Royal,Colil'ege, Jick, and Pettiti and Wiingerd that smoking,greatly enhances the risk for circulatory disorders iin OC users and that the relation shows synergism. Mr. Chairman, you asked me to address the quality of the evidence leading to the smoking warning in the oraTlcontracepti've labeling. The evidence relating oral contraceptives, smoking and cardiovascular disease comes from epidemiblogical s:udies. Such studies are important--indeed, often the only--methods for identilfyi'ng serious risks fromidrugs, chemicals and envilronmental pollutants. Many of the most serious effects uncovered in recent years, such as the carcinogenic effect in humans of vinyl chloride, estrogens, radiation to the thyroid and cigarette smoking have come from epidemiolcgic analysis. Despite their value, they are also open to the kind of criticism that often can be leveled~against other kindk of statistical conclusions. 'dhen several types of studies yield similar findings, however, ! .
Page 169: azo61e00
165 -8- as is the case with estrogens and endometrialicarcinoma and smoking, and with OC's and cardiovascular disease, the conclusions are so compelling that no conscientiious policymaker would diismiss them, unless the defects are so serious as to invalidate the conclusions. FDA statisticians have exposed sane defects In two of the OC and smoking studies. For example, in the Mann study on nonfatal myocardial infarction, the number of cases studied was m3rginal, there was incomplete assessment of sane risk factors and there were no data on relative risk with age; In the Royal College study there were a small number of cases for each diagnosis. Im spite of these deficiencies, we believe the evidence was sufficient to warrant the action taken. Subsequent studies have fully confirmed the high risk of smoking in OC users, providing further evidence that the action was justified. Benefit/Risk Judoments Ouring its deliberations of the qpestion~of OC's and smoking, our Advisory Committee on OS/GYN discussed'a number of factors, including these: -- it is well known that smoking Is a risk factor for myocardial infarction and other cardiovascular diseases;, - oralicontraceptive use also constitutes such a risk factor; and -- the combination of smoking and OC use is at least addiitive (one studyJ'and probably synergistic (tNo studies),with regard to risk.
Page 170: azo61e00
166 -9- These findings led the Advisory Committee to recomend that the risks be brought to the attantion of physicians and patients, and we concurred in that reconmendation. The labeling does not neglect to calliattention to other factors (such as hypertension, obesity, diabetes, i'ncreasing age) that contribute to the overall risk of cardiovascular disease: The labeling contraindicates use of oral contraceptives in patients with certain known card6ovascular and thromboembolic disorders or a history of thromboembolic disorders, but dbes not contraindicate use of OC's in those with obesity, diabetes and hypertension. Neither does the physician labeling contraindiicat_ use of OC's in patients who smoke. Instead, it tells physicians that they should strongly advise patients who use OC's not to smoke, acknowledging that physicians lack ultimate control over how their patients make that choice. The patient labeling, on the other hand, states flatly that women who use Oc's should not smoke. The message is clear that women who wish to continue smoking should seek other methods of contraception. The patient label!i'ng does not also state that women who have, for example, hypercholesterolemia, hypertension or diabetes should not use OC"s. Present data do not provide adequate infornation on degree of risk associated with OC use and these medical conditions. Suchidata indeed' is difficult to obtain, because physicians tend~to avoid'tne use of OC's
Page 171: azo61e00
167 -10- in women with the other risk factors when the latter are severe. If further studies do provide information on this matter, we will take prampt action in notifying physicians and patients. Because patients have asked'for more information on risks associated with 00 use and because we believe that physicians should have the latest available information on which to decide whether to prescribe OC's for Individual patients, we have Included In the labeling a strong warning against smoking by OC users. Labeling information~allows consumers to participate in basic decisions affecting their personal health and safety--in this case, to decide for themselves whether they would rather take OC"s and not smoke, or smoke and use other methods of contraception. The very firm association between 0C use and smoking in enhancing cardiovascular risks has now been established'beyon&a reasonable doubt. No less than five independent studies agree on the.association, and show risk elevations for the combination that are truly alarming. 11 would be astonished, frankly, if anyone who understands the scientific basis for establishing!the presumption of risk would challenge the conclusion that the lethality of cigarettes is amplified in women who also use 0C"s. I shall be pleased to answer any questions, Mr. Chairman.
Page 172: azo61e00
168 References 1. Mann, J.I., M.P, Vessey, M. Thorogood, R. Doll: Myocardial infarction in young women withispecial reference to oral contraceptive practice, Brit. Med. J. 1.:241, 1975. 2. Mann, J. I., and W.H.W. Irsnan: Oral contraceptives and death from myocardial infarction, ibi.d 1:245, 1975. 3. JIck, H.J., B. Dinan, K.J. Rothman: Oral.contraceptives and nonfatallmyocardial infarction, JAMA 239:1403, 1978. 4. Pettiti, D.B. and J. Wingerd: Use of oral contraceptives, cigarette smoking, and risk of subarachnoid hemorrhage, Lancet 2:234, 1978. 5. Kannel, W'.8.: Recent findings of the Framingham study, Resident an&Staff Physician 24:57, 1978. 6. Jain, A.K.: Cigarette smoking, use of oral contraceptives, and myocardial infarction, Am. J. Obst. Gynec. 126:301, 1976. 7. Ory, H.W.: Association between oral contraceptives and myocardial infarction. A review, JAMA 237:2619, 1977. 8. Royal College of GenerallPractitioners' Oral Contraceptive Study (principal author V. Beral): Mortality among oral-contraceptive users, Lancet 2:727, 1977.
Page 173: azo61e00
169 Mr. FouxTAIx. Dr. Kennedy, according to FDA's January 31,1978, Federal Register statement, the guideline texts of physician and patient labeling for OC products were previously published as a pro- posed order in the December 7, 1976, Federal Register, and later amended in the May 27,1977, Federal Register. In reviewing these two notices, staff noted that there is a suggested warning of increased risk of myocardial infarction in the users of the oral contraceptives in proposed OC labeling published in the Decem- ber 7,1976, Fed'eral Register. The basis for including thismyoca.rdial infarction warning is given as the Mann et al. studies, about whii~h there has been some testimony at this hearing. However, I can find no warning against smoking while taking the pill, or an increased risk of myocardial infarction for smokers, in the December 7 Federal Register statement. Did the December 7, 1976, Federal Register require a specific warn- ing against cigarette smoking for inclusion in OC'labeling? Dr. KENNEDY. I believe it did not. The reason for that is that much of the analysis that we have been discussing here appeared subsequently. So, during the year and a month~ that elapsed between the December 7, 1976, and the January 1978 documents, new information appeared. This chaaig+ed the view of FDA staff on the necessity for that warning. Mr. Four TAix. But you do mention the two Mann articles and other articles in the December 7, 1976, Fed+aral Register notice, so FDA was familiar with them. Is that right? Dr. KENNEDY. Yes ; we knew about the Mann articles. Mr. FouxTArN. Does that show the synergistic effect-for example, the 11.7 figure on your chart? Dr. KENNEDY. I think the question is whether at the time of the December Federal Register notice our staff and the other published recalculations of the Mann data were fully available to inform t'hatt decision. I may have to get some help here because my institutional memory g(;es back to April of 1977, but perhaps Dr. Grout can help with that, Dr. CROUT. I am told by our statisticians that the 11.7 figure was not in the published Mann article at the time that it appeared. That is our calculation fromi those data. Mr. GOLDHAMMER. So that is an FDA-derived figure and not a Mann-derived figure? Dr. KENNEDY. That is what we were told. But my response to you earlier, Mr. Goldhammer, was the question about whether or not the 11.7 figure was derived' entirely from the 2.02 and the 2.15 fiaure and my understanding is that the answer is no. In fact, the 11.7 figure is a calculated risk from the numbers in the Mann study of people who did both. Mr. GoLDxaaiMER. That 11.7 figure-does that make allowance fo{-) other risk factors or is that smoking in the absence of all other risl~ factors ? (1; Dr. KENNEDY. Let us consult on that, Mr. Goldhammer: CD Smoking and OC, without considering anything else. O ' Mr. GOLDHAMMER. So you do not know the effect of the other risl.- f'actors? You do not know how many people have diabetes which is aig
Page 174: azo61e00
170 important risk factor. Also,, hypertension which is another important risk factor. Dr. KENNEDY. It is correct that there was an incomplete assessment of risk factors and that is why that deficiency appears in the staff re- port and is listed at the top of the chart. We would like it much better if that study had included that. However, the size of the attributable risk numbers convinced our st'atisticians-even the ones most critical of that'-that, in fact, it was a defect but not a crippling defect. I think we have to understand, -if I might embroider that answer a little bit, what we do in this process. Every time FDA gets astudy that we t.hink is likely to be the basis for a regulatorv action~ the thing I want our scientific staff to do is take it apart in as critical and harsh a way as they possibly can. We want to know all the bad news and all the things that various interest groups are going, to say about our study before we have to read about it in a less favorable context. So, in this case and in other cases, our statisticians are instructed to really tear it apart and to be as negative as they possibly can. In this case, the statist'ical staff feels that the assessment of risk fac- tors was incomplete. It' wishes it.were not so. But it also feels that'.t the conclusion~ that there is synergism between these risks considered alone is robust enough~to overcome that defec.t. Mr. FouNTeiN. Mr. Duncan? Mr. DUNCAN. Let me follow up. How do you determine whethel these deficiencies are criiopling or not? I mean that in al simple way. We have some people in the room, obviouslv, who feel that those de- ficiencies are crippliney enough that one should put no reliability on the studies whatsoever. Clearly there were differences or must have been differences of opinion on your own staff., What process do you use? Do you lock everybody in a room~ and say, °tBattle it out until you come up with something or we won't let you out?" Can you help us understand the process that you go through in making astatement that these deficiencies are indeed not crippling enough to kill the results of the study? Dr. KENNEDY. Yes: I welcome that question. My way of doing that is this. I know this accords with Dr. Crout's approach~ because we have done this together. Our approach is to as- siQn to people tihe job to be as critical and negative as~ the,y, can be. Then they get people in a room and listen to them, arg'ue it out and ask the best nuestions that we can ask. Then we. find where evervbodt comes out. If our negatives-thepeople we send ini there to attack it as hard as they can and the people who do not have that initial position-can argue their way through to a consensus. I supposeit is a little bit like lockinz neople in a room. The point is that we haven't taken an action at the time this takes nlace, and we don't have an ax to grind one wav or another. Sometimes it is a close calll I would want to emphasize very heavily here, though, that there is another process that, is at work out there in the world that is inde- pendent of re,(-rilhtion. That has to do with other people takin_a the data and evalnatinz and rPviewinq it as ind'ependent scientists: For example, the reviews by my colleague here, Dr. Ory, and by Dr. Jain~ were of t.hatcharact'er. There is another nrocess which takes place andl that is that othe.r Deople do stiidv it. ThPv are e.valnatinv either the same or different but clearly related cardiovascular risk factors in independent studies.
Page 175: azo61e00
171 When all of that comes together in a consensus out there in the scientific community that doesn't have a; particular interest in the outcome-and I would emphasize the latter-theni you've got con- sensus on a point. I am telling you that I believe that that kind of an external consensus producing process has taken place here and that it has vindicated the decision that FDA made. Dr. CROUT. May I comment also on a specific process used in this insta.nce ? For major warnings related to oral contraceptives, we have for many years used a specific process. That process is to take the informa- tion, scientific information, when it comes in and then write a pro- posed warning on the basis of that information and present it to our advisory committee. That advisory committee used to be called the Obstetrics and' Gyne- colbgy Advisory Committee. It is now called the Fertility and Mater- nal H'ealth Advisory Committee: In an open meeting we debate those data and the proposed labeling, and we allow anyone else who wants to debate that to come in. So, our own staff makes the presentation and any representatives, usually of manufacturers, can come in. Any person can come in and debate before the advisory committee, They can debate either the science or the warning itself. Following that, we will receive an edited'version of the proposed language from the advisory committee. We may edit it a little bit' further. Then it is implemented. That process has been followed for the warnings related to throm- boembolism. It has been followed for the warnings related to hepaticc tumors. It has been followed for the warning on estrogens, and en- dometrial: cancer, and followed precisely for this particular warning.. Mr. DuxcnN. The advisory committee was Obstetrics and Gvnecol, ogy. Are they also specialists and do they understand this hnd of research ? Dr. CROUT. They are an admiature of persons, some of whom willi be experts in epidemiology and the analysis of these kinds of data and some of whom will be clinicians. I would sav that they are all intelligent people versed in weighing scientific evidence, although they are not specifically research people in the field of epidemiology. Mr. Du.NcaN. Thank you very muchL Dr. GOLDBERG. Dr. Crout, would yow mind explaining how FDA derived the 11.7 figure for the concurrent use of OC and smoking from the 2.02 and the 2.15 which were reported in the Mann study, particularly in the light of the acknowledged softness of the latter two figures ? Dr. CROUT. Mav we answer that in writing after consultation with our statisticians ? I am not able to do it on4.he spot. Dr. GOLDBERG. That willibe fine. Dr. KENNEDY. Yes; we would prefer to do it that' way, if you don't mind. Mr. FouNTnIN. You are not prepared to do it here?DT-. KENNEDY. I think we will do a better job of it if you will let us do it in writing. Mr. FOr?~.TArN. Yes. That will he fine. [Thee information referred to follows :]
Page 176: azo61e00
172 With respect to the request for the derivation of the odds ratio shown in the FDA charts for the ManA data we submit the following explanation which'includes the formal tests of statistical significance (shown as p values)i along with approximate 95% confidence limits. Im this respect our reported •p• values provide a more reliable measure of statistical evidence than the confidence limits based on the approximate method of calculation. n Relative Risk of Non4Fatal MI among Smoker and OC User Gro p from Mann Data Formula for Condition Sampled Odds Ratio Subjects Exposed to Risk MI Controls (Relative a x d Not Exposed to Risk a b Risk) c x b c d Mann Table VIII Pg 243 BMJ 3 May 1975 No. of Cigarettes Smoked Daily at No. (!) of Patients Using Oral Contraceptive at Total Onset of Episode Onset of Episode MI Patients Controls MI Patients Controls None 3(18'.8), 8(1i0.3) 16 (10m) 78 (100'.0). 1-14 2(25.0) 2(',4.5) 4.5) 8 (10 44 (100.0) 15 or more 11 (33,3)', 4 (12.5) 33 (100.0)' 32 (100.0) Total 16 4 55 154
Page 177: azo61e00
173 Reorqanized Data Non+',smokers and non-+bsers MI of oral 13 contraceptives Non-smoker Users of OC 3 Odds Ratios, Exact Test Results and 95% Confidence Limits about the Odds Ratios Risk of MI among non smoker, Controls non-user of OC relative to 70' themselves is 13 x 70 = 1 13 x 70 P value and Confidence Limits are not relevant. 8 Risk of MD among non4smoker OC users relative nonfsmoker non OC users is ~ 3 x 70 = 2.02 8 x 13 Exact Test, P=.39 (two tail test). Approximate 95% Confidence Limits: 0.31 and 12.13 n Smoker, Non user of OC ~ 28 70 Risk of MD among sAoker non OC users relative non+smoker non OC users is 28 x 70 = 2.15 70 x 13 Exact Test, P<.05 (two tail test). Approximate 95% Confidence Limits: 0.88 and 5.53 Smoker and User of OC 13 6 Risk of MI amon Smoker OC userss rel!ative to nonsmoker non OC users 13 x 70 =1i1.7; 6 x 13 Exact Test, pc.0001 (two tail test)~. Approximate 95% Confidence Limits: 2.87 andi49.93 Mathematical Statistician - , Bertram D. Litt / ~/~ ~ 85-727 o - 71 - 1Y~ Chief, Statistical Evaluation Branch , . . Dubey Ph D. /'7- Satya D
Page 178: azo61e00
s 174 Dr. CROUT. If you are willing for a 1-minute departure, would it be helpfuI for you if I went over again the general method by which all these studies are done in a simple wa,y ? Mr. FouxTAix. I think we understand that.. Dr. CROUT. I see. Mr. FouxTAix. Dr. Kennedy, the Jain paper which appeared in the March 1977 issue of the Population Council's publication, "Studies in Family Planning," which serves as an important basis for FDA's antismoking warning requirement, has alread~r been placed in the rec- ord. To eomplete the picture, I am also placmg in the record Jain's January 26,, 1977; written comments, on the Population Council's let- terhead, on the December 7, 1976,, Federal Register proposals for re- visions in the OC labelin~. I think these comments also serve as a basiss for the antismoking requirement. [The documents referred to follow :]
Page 179: azo61e00
175 THE POPULATION COUNCIL 245 PAaR.AVeNUe Neti Yeet, N.:Y. /OOn. TLLtwqrl12121 eaf.a]]G CYCe: VoKOV.CiW Ntw. vo~~ January. 26, 1977 Hearing Clerk Hood'and Drug Administration Room 4'-65 5600iFishers Lane Rockville, Maryland 20856 To Whom It May Concern: The following coaments are submitted in,referenceto the proposed revision of labeling requirements for oral contraceptives (Docket nos. 76N-0487 and 76N-0304), as published in the Federal Register Vol. 411, No. 336 dated December 7, 1976, pages 53634-636; 640-641. 1. It has been correctlyy pointed'out on page 53635 that the estimates of risk of death for each method includethe combined risk of the contra- , ceptive method plus the risk attributable to pregnancyy or abortion in the event of inethod failure. The implicaoionis that the estimates of pregnancies resulting from method'fai~lure will affect the estimates ~ of the total risk of death for each method. This being the case, the same estimate of pregnancies resulting from method failure should be used both,for comparing the relative effectivenessofcontraceptiva methods as well as for comparing the relative risks assoeiatedvith these methods. Apparently, this iss not done. The estimates of pregnanciesresultYhgfrom method failure shoanon pages 53634 and 53640 are different than,.those usedbyTietzeet al. (Ref14'„ page 53639).inestimating the risks of death~ associated vi~thcontraceptive methods and shown on pages 53636.and 53641 inb'ardiagrams. Tiietze et.a1.(1976, p. 7),assumed a 99 percent levell of contraceptive effectiveness both for IUDa and oral contraceptives and 90 percent fortraditional methods, vhicti~aredifferent than those shown on pages 53634.and 53640.. The inconsistency should be removed. As a minimum,.the effectiveness of contraceptive methods used in estimating the total risk of deaths associated with the methods should be pointed out along with some clarifications for the diff~erences.. 2. Thedistussions of relative risk and absolute risk of myocardial infarction associated vitb the use of oraL contraceptives on pages 53635- 636 and 53640-641 are not entirely accurate. Consequently, the recommendation C on page 53640 is not correct. These discussions and the recommendation do not accurately, reflect the synergistic effect of other ri'skfactors,k especially smoking,on the estimated risk of myocardialiinfarction among users of oralcontraeeptives. As a result of recommendation C on page 53 640, those vomen aged 40 or more yearsvho do not smoke will be ,
Page 180: azo61e00
176' ,TME POPULATIONCOUNCIL -2- unnecessarily penalized and those women in the younger age groups who are heavy smokers and also use oral contraceptives will not receive adequate information about the synergistic effects of the two. If the results of the British studies are assumed to be correct and are taken as the basis of the reco®endation C on page 53640, this recommendation should onlyy applyy to those women who also smoke cigarettes. For those women, who smoke as well as use oral contraceptives, the recemmendation should reflect the choice between smoking and using oral contraceptives. Moreover, smoking should be included as one of the contraindications for use of oraL eontraeeptives, especially for women over 40 years of age. Even among women b'etween.30-39'years of age, heavy smoking should be included as one of the contraindications. The basis of my objectionsto recommendati'on C on page 53640 are brieflyy described below. The detailss can be found in thd enclosed:papers. a. The excess risk of nonfatal myocardiaLinfaretion among users of oralieontracepti'ves observed in England and Wales can be explained in terms of the high proportion of smokers in the study.population (43 percent of the users and 50percent~of the nonusers were classified as smokers in the control population..of the British study)~ For women in the 30-44 year age group, the risk of nonfatal myocardial infarction among th'ose who,eitheruse oral contraceptives only or smoke cigarettes only are about twice as high as that among those.who neither usee oral cootra- eeptives nor smoke cigarettes. However, the relativeriske associated with usa of oral contraceptives alone is not statistically,significanty whereas.that associated with smoking alone is. The risk.of nonfatal myocardialinfarction among those who use oral contraceptives as.well as smoke isabouo twelve times higher than those.who neither smoke nor use orallcontraceptives. In view of these relative risks, it cannot be said that.'brallcon.tracepti'ves, however,, were found to be a clear additional risk factor"'(page 53635). b... Unfortunately, the data about smoking.habits for nonfatal cases separately for two broad age groups were not included:in the published British studies and similar data for fatal cases cannot'.be collected with any.aecuracy. In the absence of these required information, the estimates of fatal and'nonfatal myocardial infarction in the users of oral contraceptives.separately.for smokers and nonsmokers can be obtained byus- ingtwo assumptions. The details aree shown-in attachments II 6 III. c. In comparison,to all nonusers of oral contraceptives,, the annual.excess case rate of myoeardial infarction (fatal and nonfatal)) in those oral contraceptive users who do not smoke and in those who do . smoke are -0:4' and 16.gper 100,000:women,in 30-39 year age group. The corresponding estimates for women in the 40-44 year age group are•7..7and 147.8 per 100,000 respectYvely.. The estimates of~excess rates.among all oral contraceptive users shown on page 53635 are the weighted averagesoff these rates for smokers and nonsmokers. For.example, 43 percent of the oral contraceptive users in theEritish,study were also smokers and the remaining 57 percent were not smokers. For women in 30-39 year age group, the
Page 181: azo61e00
177 Th7E POPULATlON.COUNCIL -3- average excess race among users wiLl be 0.43 *16.8 - 0.57 *0.4 or 7.0; and for women in the 40-44 year age group, the average excess rate among users.will be 0.43 *147.8 + 0.57 *7.7or 67.9. These estimates are similar to those ahown on page 53635. d. Another way to estimate the excess case rate is to estimate the excess from the levelsamong:those whoneither smoke nor use oral contraceptives. In comparison to nonusers and nonsmokers the annual excess case rate of myocardial infarction (fatal and nonfatal)) in.those oral.contraceptive users who do not smoke can be estimated:to be approximately. 1.1 per 100.,0Wwomenin the 30-39 year age group and 15.6 per 100,000 women in,the 40-4'4 year age group, The corresponding estimates of.excess rate of myocardial infarction in those oral contra- ceptive users who also smoke cigarettes are 18.3 in the 30-39 year age group and 155.8 per 100,000 women in the 40'-44'year age group. a. The comparisons presented in sections oand d,above elearlyindicatethatt the risk of heart attacks from pill use among smokers is higher than average and the corresponding risk among nosmokersislower~than average.This. is similar to thee statement on page 53640.. However, given the magnitude of differences, the statement on,page 53640, namely, "because of the increased risk of heart attacks, oral contraceptives are not recommended for women over 40",, is not.justified. This statement should: onlyy applyy to those pill users who alsoemoke cigarettes. f. Among nonsmokers, the estimated annual number of deaths associated with the use of oral contraceptives are 4.5,per 100,000 women in the 35-39 year age group and 7.0 per 100;000 women in the 40-44 year age group -- in comparison to 7.5 and 24:.9'for all oral contraceptive users shown in the bar diagrams on pages 53636.and 53641. The corresponding estimates for those who smoke aswell as use oral contraceptives are 13.4 and 58.8 respectively. These comparisons clearly showthatin the absence of,smoking, the risk of death frempill'use in women 40-44 years of age is less than average..Moreovery the annual risk.of death from pill use beyond age 40, in,the absence of smokingisalsosubstantially. less than that associated with no fertility.control method (7.0vs...21.9),.0 g.. The distinction between heavy.and light smokers.can be subject to some (unknown) degree of error in reporting the numb'ersofcigarettes smoked per day. This may influence the estimated risks of death for pilU usersclassified as heavy or light smokers in the Bri2ishstudies.However, the resultspresen.te&in attachment II2.doindicate a likelihood that the pi1L use among.heavy smokers in,.30-34 and 35-39 year age groups may,be more hazardous than nofertilityo control method, and that pill use.among heavy.smokers in 15-29 year age groups may.be more hazardous than anyy other regimen of,fertility regulation. For example, the estimated annualimortality ratesamong.those pill userswbo are also.heavy smokers are 16.5 per 100,000women in~30-34 year age group and 23.6 per 100,000 women in 35-44 year age group. The corresponding estimatesforannual mortality when nof~ertility.control method is used ar! 14.0 and 19.3.per 100,000 women respectively.
Page 182: azo61e00
178 TiTE,POPULATION COUNCIL -4- 3. It is recognized that additional research is required to establish the.synergistic effect of smoking and use of oral contraceptives buti thesame qualification alsoappUies for accepting the results of British studies and generalizing them for wemen in the United States. Given the possibility of synergistic effect of.smoking and use of oral contraceptives, the excess risks of fatal and nonfatal myocardial infarction cases in oral contraceptive users must be shown to be significant in the absence of other risk factors especially smoking,before receamending that "if you are over 40 years old you do not use the pill because of an increased risk of heart attacks fram the pill." A more.appropriate recosmendation,would be."if you smoke and use oral contraceptives,.you are advised togive.upsmoking." Serious considerations should:also:be givenn to include smoking as one of the contraindications for the prescrption of the pi1L especially amongg the older womeu. Sincerely., 2 Anrudb K..Fain, Ph.D: Associate Iaternational Programs AKJ:jab Encs, Attachment I Cigarette smoking, use of oral cootraceptives.andd my.ocardial infarction. II Estimation of nonfatal myocardialiinfarction rates per 100i000 women by age, use of oral contraceptives and smoking. III Hisk of death from pill, use after, age 40. 11
Page 183: azo61e00
179 ATTACHMENT II I Estimation of Nonfatal Myocardial Infarction Rates per 100,000 Women by Age, Use of Oral Contraceptive and Smoking Anrudh K. Jain Ph.D. Associate International Programs The Population Council 245 Park Avenue New York, N.Y. 10017
Page 184: azo61e00
180 This note briefly dascribes the estimation procedure used to dpcmpose the rates of nonfatal myocardial infarction observed for all pill users into its two components, namely, the corresponding rates for those pill users who also smoke cigarettes and for those who do not. The corresponding rate observed for nonpill users is also decomposed into its two components. The decomposition of these rates are done separately for women in the 30-39 year and 40-44 year age groups. The rate of nonfatal myocardial infarction (J) among a1Q pilil users is the weighted average of corresponding rates among smokers and nonsmokers -- the weights being the proportion of smokers (P) and nonsmokers (1-P) among pill users. The relative risk of nonfatal' myocardial infarction (R) associated with smoking is the ratio of the rate of nonfatal myocardial infarction among smokers (K) and the corresponding rate among nonsmokers (I). Thus, I - PK +(1-P),I; R° K/I or, K- RI or, J - PRI + (1-P)I or, I - J'/'(1-P + PRI) The values of I and K can be estimated from the values of J, P and R for users of oral contraceptives. The rate of nonfatal myocardial infarction among heavy and li'ght smokers can aliso be estimated by using the relative risks associated with heavy and light smoking and the estimated value of I. The basic formulas can also be applied~ to decompose the rate of nonfatal myocardial infarction among nonusers of oral contraceptives and to the users and nonusers
Page 185: azo61e00
181 -2- classified by age. The estimates of J, P, and R used in estimating the values of I and K are shown in Table 1. The estimates of J and P are taken from Mann et al. (1975) and the estimates of R are taken from Jain (1976):. The assumptions involved are briefly described below. Assumptions: 1. The relative risks of nonfatal myocardial infaretion associated with smoking among users and'nonusers of oral contraceptives are assumed to be independent of age. 2. The proportion of smokers among users and nonusers of oral contraceptives are also assumed to be independent of age. Results: The estimated annual, rate of nonfatal myocardial infarction by use of oral contraceptives, smoking habits and age of women are shown in Table 2. These resul~ts clearly indicate that the rate of nonfatali myocardial infarction among those pill users who do not smoke are substantially lower than the corresponding rates among those pill users who also smoke cigarettes (18 vs. 107 per 100,000 for women in the 40-44 year age group and 1.8 vs. 10.6 per 100,000 for women in the 35-39'year age group). In comparison, the annual rates of nonfatal myocardial infarction among those wh a neither smoke nor use oral contraceptives are 6.3'and 1.3 per 100,000 for women in the 40-44 and 30-39 year age groups respectively-. Thus, the excess rate of nonfatal myocardiallinfarction attributable to the use of oral contraceptives alone can be estimated as 12 per 100,000
Page 186: azo61e00
I 182 -3- for women in the 40-44 year age group and 0.5 per 100,000 for women in the 30~39 year age group. The corresponding estimates of excess rates of nonfatal myocardial infarction among those pill users who also smoke are about 101 and 9 per 100,000 women in the 40-44 year and 30-39 year age groups respectively. These results clearly show the importance of considering the smoking habits of women in estimating the risk of nonfatal myocardial infarction attributable to the use of oral contraceptives. Effect of Assumptions: It is possiblie that the relative risk associated with smoking increases with age or that the proportion of smokers increases with age. In either case, nonfatal myocardial infarction rates among, nonsmokers will be lower than those estimaced'for the 40-44 year age group and willi be higher than those estimated for the 30-39 year age group. Thus, for the 40-44 year age group, the estimates of nonfatal myocardial infarction rates among those users and nonusers of oral' contraceptives who do not smoke can be taken as conservative.
Page 187: azo61e00
k. 183 Tablie 1: Estimated AnnuaL Rate of Nonfatal Myocardial Infarct'ion Per l00,000iAomen, Proportion of Smokers, Relative Risk Of Nonfatal Myocardial Infarction Associated with Smoking by Age and Use ofOraIContraceptives Users Nonusers Items 3 0-39 40-44 30 •39 40-44 1. Nonfatal myocardial inf~arction rate per 100,000 5.6 56.9 2.1 9.9 2. Proportion of smokers .43 .43' .50 .50 3.. Relative risk associated with smoking: Total 5.78 5.78' 2.15 2.15 Heavy 7.33 7.33 4.23 4.23 Light 2.66 2.66 0.77 0.77 Source:. For items 1,.2- Mann et al. (1975). _ Forr item 3 - Jain (1976). • 1
Page 188: azo61e00
184 Table 2: Estimated Rate of Non-fatal MyocardiaL Infarction per 100,000 Women by.Age, Use of Oral Contraceptives and. Smoking Habits Smoking. e 30-39'ears e 40-4'4' years Users Non-users Users Non-users Smokers Total 10.58 2.96 107.45 13.52 Heavy * 13.42 5.64 136.30 26.59 Light + 4.87 1.03 49.51 4.84 Non smokers 1.83 1.33 18.59 6.29 Smokers & Non-smokers 5.6 2,1 56.9 9:9 * Heavy.smokers.: at least 15 cigarettesperdag. + Light smokers:.less than,.15 cigarettes per day. r
Page 189: azo61e00
185 REFERENCES 1. Mann, J. I., H. P. Vessey, M. Thorogood, R. Doll, "MyocardiaL Infarction in Young Women with Special Reference to Oral Contraceptive Practice," British Medical Journali„ 2:241-245, 1975. 2. Ja.in, Anrudh.K., "Cigarette Smoking, Use of Oral Contraceptives, and Hyocardial Infarction," American Journal of Obstetrics and G}•necoloov, 126, (3): 301-307, 1976.
Page 190: azo61e00
186 ATTACHMENT III Risk Of Death Froin Pill Use After Age.40 Anrudh K. 7ain Ph.D. Associate, International Programs, The Population Council 245 Park Avenue New York, N. Y. 10017
Page 191: azo61e00
187 In an excellent analysis, Tiet2e, Bongaarts and Schearerl in 1976 have shown that the risk of death from pill use after age 40 is higher than the corresponding risk of death from pregnancy and child- birth when no fertility control method is used. On the basis of this particular finding aliong with the results of the British scudies,Z,3,4',S the Food and Drug Administration in the United States has proposed to revise physician and patient labeling for oral contraceptives. The proposed revision for physicians staces that "the use of oral contracep- tives in women in this age group (40 and over) is not recommended". The proposed revision for patients states that "if you are over 40 years old'do not use the pill because of an increased'risk of heart attacks from the pill."6 The puroose of this no teis to show that the smoking habits of women strongly affects the esti.mated~risk of death from pill use. It is shown that if the results of the British studies are assumed to be valid, the risk of death from pill use after age 40, among those women who do not smoke is substantially less than the risk of deach from pregnancy and childbirth when no fertility control method is used. There are some indications that even among those pill users in 30-39 year age group, who are also heavy smokers the risk of death is higher than the corresponding risk of death from pregnancy and childbirth when no fertility control method is used. In their analysis, Tietze et al.l'compared the risk of death associated with various fertility regulation mechods and the risk of death from pregnancy and childbirth when no fertility regulation method is used. The estimated risk of death for each fercilicy regulation method included the risk of the method itself and'the risk of death due to preg- ancy or abortion in the event of the method'failure. The method reliated
Page 192: azo61e00
188 -2- mortality estimates were based on variousstudies conducted primarily in England and Wales and in the United States. The estimate of mortality related to female sterilization used by Tietze et al. varied between 10 and 30 per 100,000'proceduresbut these estimates did not include mortality associated with hysterectomy which varied between 150 and 230 per 100,000 procedures. Hysterectomy was excluded because "it is clearly a more hazardous procedure than the sterilization" (Tietze et a1.1 , 1976; p. 8). Similarly Tietze et al'. used 3.3 per 100,000 procedures as an estimate of mortality from abortions. This estimate was based on all legal first trimester abortions without concurrent sterilization that occurred in England and Wales during 1968-73. During this period about 619,000 abortions were performed in England and Wales.7 About 70 percent of them were first trimester abortions without concurrent sterilizations; 7 percent were first trimester abortions with concurrent sterilizations; 19 percent were late abortions without concurrent sterilizations and about 4 percent were late abortions with concurrent sterilizations. The mortality estimates for these four groups were 3.3, 47, 15.3 and 61 per 100,000 procedures respectively. The overall mortality estimate for all abortions in these four groups was 11 per 100,000 procedures. Clearly, had Tietze et all. used 11 per 100,000 as an estimate of mortaliitry rate from abortions instead of 3.3 per 100,000, they might have concluded differently about the safety of abortion procedures. It can be argue& that all, abortions should be considered and hysterectomy should be included in mortality estimates related to . ~
Page 193: azo61e00
189 .< -3- abortions and female sterilizations respectively for comparing the risk of death from various fertility control methods. However, such argument is not valid because the comparisons of method related mortality esti- mates clearly indicate that hysterectomy is more hazardous than tubal aterilization; abortions performed beyond 12 weeks of gestation are more hazardous than those performed during the first trimester and abortions performed with concurrent sterilization are even more hazardous. Impli- cations being that hysterectomy should not be recanmended; abortions should not be recommended beyond the first trimester; and concurrent sterilization should not be recommended with abortion procedures. For obvious reasons, while selecting the mortality estimates for abortions, Tietze et a1.1 did not consider the increased risk of death associated with abortions in the presence of additional risk factors such as concurrent sterilization or abortions performed beyond 12 weeks of gestation. However, in the case of oral contraceptives, Tietze et al. did not exclude the increased risk of death attributable to the presence of additional risk factors such as smoking. In order to be comparable, Tietze at al. should have also excluded the effect of additional risk factors from the estimated risk of death attributable to oral contraceptives. Since the British studies have shown that the effect of oral contraceptives and smoking on the risk of myocardial infarction is synergistic, the mortali'ty estimates for pill use among nonsmokers should be considered for comparing the risk of death fromm various fertility controli methods and from pregnancy and childbirth when no fertility control method is used6 The presence of smoking for the risk of death from oral contraceptives is parallel, to concurrent sterilization for the risk of death from, abortions. If one is excluded, the other should also be C) ~ ~ 0 ~ ~ ~ 35-7270 1- 79 - 13 r
Page 194: azo61e00
190 -4- excluded. 1 The main problem, Tietze eti aL. might have faced is that the British studies did not provide the estimates of excess mortality attributable to pill use in the absence of smoking. Based on two assumptions, the excess mortality attributable to pill use among non- smokers are estimated in this note and are compared with the risk of death associated with other fertility regulation methods. METHiID OF F.STIMATION'. Mortality rate (J) among all pill users is the weighted average of corresponding mortality races among smokers and nonsmokers -- the weights being the proportion of smokers (P) and~nonsmokers (1-P) among pill users. The relative risk of death (R) associated with smoking is the ratio of the mortality rate among smokers (K) and the morttalityy rate among nonsmokers (Ii). _Thus, J-PK + (1-P)I; R-K/I or, K-RI or, J-PRI + (1-P)I or, I-J/(1-P+PR) The values of I and K can be estimated from the values of J, P and R for users of oral contraceptives. The mortality rates among heavy smokers and Uight smokers can also be estimated'by using the relative risks associated with heavy and light smoking and the estimated va11~e of I--ttie mortality rate in nonsmokers. The basic f'ormulas also apply to the nonusers of oraL contraceptives, to the users and nonusers classified by age and to the mortality rate A
Page 195: azo61e00
191 -5- from myocardial, infarction and from pulmonary,embolism or cerebral thrombosis. The estimates of J, P, and R used in estimating the values .• of I and K are shown in Table 11. The estimates of J and'P are taken from the British Studies and the estimate of R for myocardial infarction and J' thromboembolism are taken from reanalysis of these studies by JainB, and Fredericksem and Ravenholt.9 The assumptions involved are briefly discussed bellow. ASSUMPT IONS 1. The estimates of relative risks associated with smoking and proportion of smokers observed in the studies of nonfatal incidences are assumed to hold for the fatal incidences. This is done because the data about smoking habits for fatal incidences are not available and it is also difficult to collect these data with any accuracy because such information does not appear in medical.records and one can not interview those who are dea&. 2. It is further assumed that the estimates of relative risks associated with smoking and proportion of smokers are independent of age and are equal to those observed for the total group of vomen includpd in relevant British studies. This is done because the relevant data for nonfatal incidences for broad age groups are not available from the published reports. RESULTS The estimates of annuaL mortality rates per 100,000 women from ' myocardial infarction and thromboembolism by use of oral contraceptives, M '
Page 196: azo61e00
. 192 -6- smoking habits and age of women are shown in Table 2. It can be seen that the annual mortality rates fr ao myocardial infarction among those pill users who do not smoke are lower than the corresponding mortality rates among those who smoke cigarettes. For example, the annuaL mortality rate among those women in 40-44 year age group who use oral contraceptives but do not smoke cigarettes is 10.7 per 100,000 in comparison to 7.4 among women who neither use oral contraceptives nor smoke cigarettes. The corresponding rate among those piLl users who also smoke is about 62 per 100',000 women in 40-44 year age group. Excess Mortality Since smoking and use of oral contraceptives are two risk factors considered here, the estimates of annual mortality rates among alli nonusers are higher than the corresponding estimates for those who neither use oralicontraceptives nor smoke cigareFtes. Thus, estimates of excess mortality attributable to pill use will depend'upon whether the excess is estimated from the mortality level of all nonusers or from the mortality level of only those who neither smoke nor use oral contraceptives. For comparison purposes, excess annual mortality attributable to pill use is estimated from the mortality levels of all nonusers as well as from the levels of nonusers and nonsmokers. The estimates of excess annual mortality from myocardial infarction and from thromboemboli'sm attributable to pill use are shown in Table 3. Tietze et al.l estimated the excess annual mortality attributablie to pi1Q use by subtracting the annualimortality rate in nonusers from
Page 197: azo61e00
R 193 ., J . the annual mortality rate in all pill users. However, this metho& over- estimates the annual mortality attributable to pill use because the annuaU mortality rates among users as well as among nonusers are affected by the annual mortality rates attributable to smoking. The excess mortality attributable to pi1S use alone can only be estimated accurately by subtracting the annual mortality among those who neither smoke nor use oral contraceptives from the annual mortality among those pill userss who do not smoke. Thus, the excess annual mortality attributabl= to pill use alone from myocardial' infarction is estimated to be 0.6 per 100',000 women in 30-39'year age group as compared to 3.5 estiuaated: by Tietze et all The corresponding estimate for womenin 40-44 year age group is 3.3 per 100,000 women as compared to 21.1 estimated by Tietze et al.l It can be argued 'that the excess annual mortality among those pill users who also smoke should be estimated from the mortality level of those nonusers who smoke cigarettes, and attribute the excess thus estimated to the pill use among smokers. However, this procedure will overestimate the excess risk attributable to pill use among,smokers because the effect of the two risk factors for myocardial infarction has been shown to be synergistic. The correct way is to estimate the excess annual mortality among those pill users who also smoke from the mortality lievel of those who neither smoke nor use oral contraceptives and attribute the excess thus estimated to the presence of both the risk factors. Thus, the excess annual mortality from myocardial infarction attributable to pill use and smoking is e,stimated to be 9.0 and 54.6 per 100,000 women in 30-39 year and;40-44 year age groups respectively. The excess
Page 198: azo61e00
194 -8- mortality of 54.6 per 100,000 women in 40-44 year age group includes about 3.3 deaths attributable to pill use alone, 8.5 deaths attributable to smoking alone and about 42.8 deaths attributable to the joint effect of the two. Risk of Death from Pi1liUse Nhat is the effect of the above mentioned differences in estimates of excess mortality on the annual number of deaths associated with pi'1'1 use? How does the annual risk of death fram pill use compare with the use of other fertility regulation methods and how does it compare with the risk of death fr.om pregnancy and childbirth when no fertility control method'is used? To study these questionsr the annual number of deaths attributable to pill use alone and those attributable to pill use and smoking are estimated'by adding the excess mortality from myocardial' infarction, thromboembollism and mortality from pregnancy or abortion in the event of pillifailure. The estimates of excess mortality attributable to pill use are taken from Table 3'and the estimates of mortaliity from pregnancies due to method failure are taken from Tietze et al.1 For comparison purposes, the corresponding estimates for other fertility control methods and no fertility control methods are taken from Table 4 of Tietze et all.1 The result$ are presented in Tablie 4. It can be seen that the annual risk of death from pill use among,women in 15-29 year age group is lower than the corresponding risk from childbirth in the absence of fertility control methods. This is true for nonsmokers as welili as for smokers in 15~29 year age groups.
Page 199: azo61e00
195 -9- .,, .1 Smoking does not elevate the risk of death from pill use in these age groups because the method associated mortality estimates refer to those from thraoboembolism only, which are about the same for smokers and nonsmokers in 20-34 year age groups (See Table 3). The distinction between heavy, and light smokers can be subject to some (unknown) degree of error in reporting the number of cigarettes smoked per day and there- for the results of these two categories should be interpreted with caution. However, the results indicate a likelihood that pill use among heavy smokers in 15-29'year age groups may be more hazardous than any other regimen of fertility reguliation. The annual' risk of death from pill use among,nonsmokers beyond age 30 is substantially lower than the annuali risk of death from child- birth when no fertiliity controlimethod is used. For example, the annual risk of death from pill use alone among women beyond age 40 is about one- third of the risk of death from, pregnancy and childbirth when no fertilizy controL method is used (7 vs. 22). The risk of death beyond age 40'among those pill users who also smoke cigarettes is substantial4y higher than the risk of death fram childbirth when no fertility control method is used (59 vs. 22). However, for women in 30-34 and 35-39 age groups, the risk of death among only those pi1Q users who are aliso heavy smokers is higher than the risk of death from childbirth when no fertility control' method is used. It can also be seen that the annuali risk of death from pi11 use alone among women in 35-39 year age group is about the same as the
Page 200: azo61e00
196 -10- corresponding risk among those pill users in 15-29 year age groups who are also heavy smokers. However, the annual risk of death from pill use alone beyond age 40 is less than the corresponding risk among those pill users in 30-39 year age groups who are also heavy smokers. The revised estimates for annual number of deaths attributable to pill use only also reduce the total number of deaths expected during the remainder of the reproductive period if oral contraceptives are used for family limitation. Tietze et al. estimated that about 124.9 deaths would be expected during a five year period among 100,000 nonsterile women who start using oral contraceptives at age 40. However, it can be seen from the results presented'in Table 5, that the expected number of deaths during a five year period among 100,000inonsterile women who start using oral contraceptives at age 40 would~be about 35.4 if they do not smoke and would be about 294.4 if they do. The corresponding estimates for other age groups are shown in Table 5. Effect of Assumptions: It is quite unlikely that the relative risks of myocardial infarction or thromboembolism associated with smoking, or that the proportion of smokers will vary substantially with age, especially in 30-44 year age group. However, the relative risks associated with smoking or the proportion of smokers may increase slightly with age. In eiYher case, it can be shown that annual mortality rates from myo- cardial infarction among nonsmokers will be lower than those estimated
Page 201: azo61e00
197 6d for women in 40-44 year age group and will be higher than those estimated; for women in 30-39 year age group in this analysis on ttie assumption of age independence. Similarly, the annual mortality rates from thrombo- embolism among nonsmokers will be lower than those estimated for women in 35-44 year age group and will be hieher than those estimated for w anen in 20-34 year age group. Thus, annual mortality rates estimated in this analysis for those women (users and nonusers of oral contraceptives) in 40-44 year age group who do not smoke can be taken as conservative estimates. This means, that even if the relative risks associated with smoking or the proportion of smokers are found to increase with age, the conclusion of this analysis in regard to the safety of pi1i1 use among nonsmokers after age 40 is unlikely to be affected.
Page 202: azo61e00
.R 198 SUtR1ARY The choice of a particular method for fertility regulation clearly depends upon many considerations. The risk of death, no doubt a very important factor, is only one of the considerations involved in decision making. A clear understanding of the complex interactive effects of smoking, age of the women, and the use of oral contraceptives is essential for properly assessing the safety of oral contraceptives among women in the U.S. However, if the results of the British studies are to be accepted'y the present analysis indicates that (1') the use of oral, contraceptives, in the absence of smoking„ is safer than no ferciliiiy control method in all age groups including in the 40-44 year age group, (2) the use of oral contraceptives among smokers beyond age 40 is substantially more hazardous than no fertility control method, (3) the use of orali contraceptives among,heavy smokers in 30-34 and 35-39 year age groups may be more hazardous than no fertility control methodland (4') the use of oral contraceptives among heavy smokers in 15-24, and 25-29 year age groups may, be more hazardous than any other regimen of fertility regulation. Those women in. 40-44 year age group who smoke as welli as use oral contraceptives an&want to reduce the risk of death should consider to give up smoking. Sunilar women in the younger age groups shonldi aliso consider to give up smoking, especially if they are heavy smokers. Alternatively, these women can shift to another regimen of f!ertii~ity regulation but in this case, the total risk of death is likely to be higher than the corresponding risk associated with pill use alone.
Page 203: azo61e00
199 REFERENCES 1. Tietze, Christopher, John Bongaarts, and Bruce Schearer, "Mortality Associated with the Control of Fertility," Familv Plannine Persoectives, 8:6-14, 1976. 2. Mann, J. Ii., M. P. Vessey.,. M.Thor~ogood, R. Dool, "Myocardial Infarction in Young Women with Special Reference to Oral Contraceptive Practice"; British Medical Journal, 2:241-245, 1975. 3. Mann, J. I., and~ W. H. W. Inman, "Oral Contraceptives and Death from Myocardial Infarction," British Medical Journal, 2:245-248, ` 1975. 4. Mann, J. I., H. 1horogood,. W. E. Walters, and C. Powell, "Oral Contraceptives and Myocardial Infarction in Young,Women,"' British Medical Journal, 3:631, 1975. 5. Mann, J. I., W. H. W.. Inman, and M, Thorogood, "Oral Contrac.eptiveUSein Old'erWomen and.FataL Myocardial InfiarcCion," British Medical Journal, 2:445-447, 1976. 6. Food & Drug Administration, "Oral Contraceptive Drug Products: Notice and Proposal of Revised Physician and Patient Labelling," Federal Register 41 (236): 53630-53642, 1976. Tie[ze, Christopher, and Marjorie Cooper Murstein, °Induced Abortion: 1975~Factbook", Reoorts on Pooulation/Familv. Plannine, No. 14' (2nd Edition), IhePopulation Counci~l,. Neu. York,, Deceabe.r, 1975. . 8. Jain6 Anrudh~K., "Cigarette Smoking, Use of'Oral Contraceptives, and Myocardial Infarction," American Journal of. Obstetrics and Cynecoioev, 126(3):301~307, 1976. 9'. Fredericksen, H., and R. Ravenholt, "lfi romboembolism, Oral Contraceptives, and Cigarettes," Public Healith Reports, 85:197, 1970. 10. Vessey, H. P., and R.-Dolil,."Investigati~on of'Relat'ionship Between Use of Oral Contraceptives and Ihromboembolic Disease: A Further Report," British Medical Journali2:651-657, 1969. 11. Inman, W.H.W., and M.P. Vessey, "Investigation of Deaths from Pulmonary, Coronary and Cerebral:Ihrombosisand Embolism,in.Women,of Cfiildb'earingAge," British Medical Journal, 2:193-199, 1968.
Page 204: azo61e00
Table 1: Estimated Annual Mortality Rate per 100,000 Women, Proportion of Smokers, Relative Risk Associated with Smoking for Myocardial Infarction, I1u omboembolism, Age and Use of Oral _ Contraceptives Mortality rate Proportion of Relative Risk Associated with Smoking Causes of Mortality/Age/Use of per 100,000 Smokers -C11-Smok e_rs ~leavy . Zlgfi E Oral Contrace tives Women J I' R Smokers Smokers Myocardial Infarction: Users of Oral Contraceptives Age 30-39 years 5.4 .43 5.78 7.33 2.66 40-44 years 32.8 .43 5.78 7.33 2.66 Non Users of Oral Contraceptives Age 30-39 years 1.9 .50 2.15 4.23 0.77 40-44 years 11.7 .50 2.15 4.23 0.77 Thromboembolism Users of Oral_ Contraceptives Age 20-34 1.5 .43 1.13 3.14 0.53 35-44 3.9 .43 1.13 3.14 0.53 Non Users of Oral Contraceptives Age 20-34 0.2 .50 1.49 1.49 1.49 35-44 0.5 .50 1.49 1.49 1.49 Source: Mann & Associates (1975, 1976); Tietze et al. (1976) for mortality rate from myocardial infarction; Lm an & Vessey (1968 for mortality from thromboembolism. Mann & Associates (1975) and Vessey & Doll (1969) for proportion of smokers. Jain (1976) for relative risk of myocardial infarction associated with snioking and Frederiksen nnd Ravenholt (1970) for iclative risk of thromboembolism associated with snioking. a!•ZossEo I I [I
Page 205: azo61e00
201 Table: 2EstimatedAnnualMortaityRate per 100,000 Women fromMyocardial Infarction and Thromboemboldsm by Use of Oral Contraceptives, SmokinQ and Aee w Annual Mortality Rate per 100,000 Women from Myocardial Infarction 30-39Years 40-44 Years Users SonUsers Users Mon Users SmokingData Smokers: Total 10.2 „ 2.6 62.0 1519 Heavy 13.0 5.1 78.7 31.3 Light 4.7 0.9 28.6 5.7 Non Smokers: 1.8 1.2 10.7 7.4 Smokers and Non Smokers: 5.4 1.9 32:8 11.7 Annual MoralYtyRate p er 100,000 Women from Thromboemboiism 20-34 Years 35-44Year,s Users Non Users Users Non Users Smokers: Total 1.60 0.24 4.17 0.6 Heavy 1.46 0L24 11.60 0.6 Light 0.75 0.24 1.96 0.6 Non Smokers: 1.42 0.16 3.69 0,4 Smokers and 1 Non Smokers: 1.5 1 0.2 3.9 0.5
Page 206: azo61e00
202 Table 3: E'stimated Excess Annual Mortality per 100,000 uomen, from Myocardial Infarction.orThromboembolism by Use of Orai Contraceptives, Smoking and'Age Estimated Excess AnnuaL Mortality from.MyocardialiInfarc tion In Comparison to: A11iNon Users Non Smoker5.6 .Non Us.er s Risk.Factors 30+39Years 40-44Years. 30-39. Years 40-44 .Year:s Pill Use and Smoking Total 8.3 50.3 9.0 54.6 Heavy 11.1 6T.0 11.8 711.3 Light 2.8 16.9 3.5 21'.2 Pill Use Onlv. -0.1 -1.0 0.6 3-3 All Pill Users 3'.5 21.1 4'.2 25.4 Estimated Exee ssAnnual Mortality from TKromtioembolism In Cannarisonto• . All Non. U sers. Non Snokers 6 \on Users 20-34 Years 35-44 Years. 20-34.Years 35-44 Years Pill Use and Smoking Total 1.4' 3.7, 11.4 3-8 Heavy 4.3 11.1 4.3, 11.2 Light 0.6 .1.5 0.6 1.6 Pill Use Onl 1.2 3.2 1'.2 3'.3. All Pill Users 1.3 3'.4' 1!.3 3'.5 - ~ r
Page 207: azo61e00
203 Table 4: Annual Number of Deaths Associated vith Control of Fertilityy per 100,000iNon Sterile Women, by Regimen of Control and Age of Women r Regimen of Control Age Group 15-19 20-24 25-29 30-34 35-39 40-44 A. No Control 5.5 ' 562 7.1 14.0 19.3 21.9 B. Abortion Only 2.3' 2.5 2.5 5.2 9.8 6.6 C. Oral Contraception No Abortion 1.4 1.5 1.5 5.2 7.5 24'.9 Non Smokers 1.3' 1.4 1.4 2.2 4.5 7.0 Smokers - A11' 1.5 1.6 1.6 10.8 13.4 58.8 Heavy 4.4 4:.5 4:.5 16.5 23.6 82.9 Light 0.7' 0.8 0.8 4.5 5.7 23.2 D. IUD - No Abortion 1.1 1.2 1.2 1.4 1.6 1.4 E. Traditional Contraception - . _.:. 3.74'.7. 4.0 No Abortion 1.1 1.4 F. Traditional Contracpetiow - & Abortion 0.3 0.4 0.4 0.8 1.4 0.8 Source:. Except for Smokers and:Non Smokers, other numbers are taken from Tietze et al. 4
Page 208: azo61e00
204 Table 5: TotaliNumber of Deaths Associated with Piil Use That Are Expected Over the Remainder of the Reproductive Years, Per 100,000 Non Sterile Women, at Seliected Ages of Women When~ Starting Familv Limitation Regimen of Control A¢e at Starting Familv Linitation. 25 30 35 40 No Control 265.6 244.9 188.9 117.6 Oral Contraceptives - No abortion All Pill Users 194.8 187.5 1611.7 124.9 Non Smokers 74.8 68.0 57.2 35.4 Smokers: Total 421.3 414.5 360.7 294.4 Heavy 621.3 614.5 532.2 414.9 Light 173.3 166.5 144.2 116.4 Source: Except for Smokers and Non Smokers,oehernummers are taken from Tietze et al. ~ 1
Page 209: azo61e00
205 Mr. FOUNTAIN. In Jain's comments, he suggested an antismoking warning in OC patient labeling worded as follows, and I quote :"If you smoke and use oral eontraceptives,, you are advised to give up smoking." Did anyone else suggest such, a warning in commenting on the De- cember 7, 1976i Federall Register proposal? Dr. KENNEDY. I will have to look up some records to answer that question fully, Mr. Chairman. I am not aware of whether that was discussed in particular by individual members of the advisory com- mittee. It is a, possibility. Perhaps Dr. Crout or Dr. Ory, who has been on the advisory committee, knows. Dr. CROUT. The answer to the question~ is yes. The specific language suggested by the advisory committee is not that which we finally adopted. We can get you this specific language. Mr. FOUNTAIN. Who was lt? Dr. CROUT. I would' have to look back in our files and give you that later. Mr. FOUNTAIN. Yes; if you would submit that, we will appreciate it. Dr. KENNEDY. To make sure I understand this, Mr. Chairmany you want to know who on the advisory committee or who, by written com- ment, suggested specific language for the label warning ? Mr. FOUNTAIN. Yes; and also this particular language. Dr. KENNEDY. I understand. [The information requested follows:] Comments received in response to the oral contraceptive guideline labeling were reviewed by the Bureau of Drugs with results as follows : 1. The Planned Parenthood Federatiow of America Inc. suggest'ed a change in the wording in the patient package insert under the section "Who Should Not Use Oral Contraceptives." See attachment #1. 2. Dr. Ravenholt of AID proposed that the information provided both physicians and patients should state that women should avoid smoking cigarettes when taking oral contraceptives. See attachment #2. 3. Dr. Jain's letter of January 27, 1977 as referred to: by Congressman Fountain~ at the Hearing: In addition, the C.I.G.A.R.I.S.C: group gave test5mony before the Advisory Committee on November 17, 1977 regarding O.C. use and smoking and suggested severali changes In the proposed text of the labeling both for the physician, and patdent relative to O.C. use and smoking. See attachment #3. According to the transcript of the Advisory Committee meeting of November 17, 1977 Dr. Howard Ory suggested a box warning for both the patient and physician labeling at the beginning of the warning section. The wording agreed on by the. Advisory Committee was as follows :"Smoking markedly, increases the risk of experiencing serious cardiovascular side effects from oral contraceptive use;. especialls In women greater than 35 years of age. Women who smoke and who wish to use oral contraceptives should be advised to stop smoking:" See attach- ment #4. wHO. SHOULD NOT USE ORAL Co^TTRACEPTIVES' Attachment #1 ORIGINAL B. If you have had any of the following conditions you should not use the pill : 1. Heart attack or stroke. 2. Clot's in the legs or lungs. C. Although it is your decision~,it,is recommended that if you are over 40 years, old you do not use the pill because of an increased risk of heart,attacks from the pill. RECOMMENDED. CHANOES Based on your history or examination, there are other medicaliconditions which may increase your risk with use of the pill. Be sure to make your decision as to whether or not to use the pill after a discussion with Tour physician. $5-727 0 - 79.- 14
Page 210: azo61e00
206 It is also recommended that you do not take the pill if you are over 40 and have a strong family history of diabetes mellitus, a family history of early heart attack (before age 40), or if you are considerably overweight or a smoker. Attachment #2 DEPARTMENT OF STATE, AGENCY FOR INTERNATIONAL DEVELOPMENT, Washington, D.C:, March 21, 1977, FIEARSNO CLERK Food and Drug Adminiatration, Roch.ville, Md. DEAR SIR: In addition to the information which you proposed to provide physi- cians and patients with respect to use of oral cont'raceptives;,you surely should include information to both doctor and woman that she should avoid smoking cigarettes when taking oral contraceptives. In addition, you may wish to consider some of the findings presented in the enclosed, as yet unpublished,. paper on the "Comparative Side EfPects' of Three Kinds of Oral Contraceptives". Sincerely yours, R. T. RAVENHOLT, Director. Office of Population. Attachment' #3 TESTIMONYPREBENTED. TO THE OBSTETRICS AND GYNECOLOGY ADVISORY COM\SITTEEOFTHE I~+OODAND DRUG ADMINISTRATION BYC;I.G.A.R.LS.C. CITIZENB INSISTENT UPON: GETTING ACTION REGARDING SMOKING AND CONTRACEPTIVES November 17, 1977 A. TheRegulationa The regulations governing the patient package inserts for oral contraceptives (21 C.F.R. Section 310.501(a ))' specify a list of items which oral contraceptive manufacturers must inciude, as a minimum, in their labels. It is highly unlikely that a manufacturer will voluntarily add language which~ is not specifically required by the regulations. Because the proposed regulations do not'require a statement about the increased risk to pill users from smoking, we recommend that thee regulations for both the patient package insert and the detailed patient labeling require the manufacturer to include the following,: "A statement, that heavy smoking significantly increases'the risk of blood clots and heart' attacks:" The regulations should also mandate inclusiom in the detailed booklet of the relative risks for users who smoke in comparison with pill users who do not smoke. B. Pkyaician's Labeling There are five separate sections of the physician's labeling in which, it is appropriate to communicate the increased risks of death from smoking and pill use. The FDA's~proposed label omits these warnings. 1. Contraindications Seven contraindications are listed for which oral contraceptive use is not recommendedi Although the combined use of oral contraceptives and cigarettes should not' be absolutely contraindicated, the recommendations from the studiess support a statement that it! is "relativel.v contraindicated!" Physicians will thereby he alerted to the seriousness of the risk cigarettes present to pill users. 2. Warnings The section on warnings describes the major side'effect:s of the pill, sets forth~ tables which delineate the degree of risks, and provides a reference to articles in which the physician may find additional information. While this approachi is an excellent method of informing physicians about oral contraceptivec. the physician''s'labeling fails to make full use of this: medium to inform physicians about the increased risks of smoking for pill users. The reference to cigarette smoking as one of many factors predisposing a pill user to myocardial infarction and thromhoembolic disease is buried.,without benefit of any statist'ics,orcharts to bring this substantial risk to the attention of the reader. We recommend that the appronriate charts and discuscions of statistics isolate the incidence of mvocardial infarction and thromboernbolism for smokers' and nonsmokers. respectiveln-. and delineate the risks by age group as well. The studies which
Page 211: azo61e00
207 demonstrate the risk of smoking to pill users should be citedi Without these changes, the section on "warnings" is seriously defective. 3. Grmphs Many doctors who will not take the time to read the entire label, may make recommendations based on the graph. We recommend that separate bars be included on the graph to represent the respective risks for pill users who smoke heavily and those who do not smoke. Cold, impersonall statistics dramatized by the picture of a bar for smokers three to eleven times higher will leave a lasting impression. Visuall illustrations tell a story much more effectively than many pages of discussion. ¢. Precautions The section entitled "precautions"' lists the factors in a patient's medical his- tory which should be considered when prescribing birth control pills. Because cigarette smoking significantly affects the risks generally associated with the pill it is a necessary part of a patient's medical history. Failure to include it here would be a dangerous omission. We recommend the addition of the followingg st'atement; "Heavy smoking increases the incidence of thromboembolism and myocardial infaretion in oral contraceptive users. This increase is particularly significant for women, over 30 years of age who smoke 15 or more cigarettes per day and have used oral contraceptives for more than five years. Patients should be counselled t,hat'~ the effects are cumulative, and that even if they are under 30 years of age and have not taken oral contraceptives for five years their decisions now to smoke and/or use oral contraceptives will affect their risks in the future. They s'tould therefore be advised to evaluate these considerations now." J. Adverse reactions A physician~must be able to quickly Identify major adverse reactions from oral contraceptives, in patients: and, counsel them accordingly. The specific enumera- tion of adverse reactions omits the exacerbated effect on the incidence of throm- boembolic disorders for those pill users who smoke. Such a defect will cause many physicians to make what, may be inappropriate recommendations to their patient's. We suggest this section include a statement that the combined use of oral contra- ceptiives and cigarettes increases the risk of myocardial imfaretion and throm, boembolism. C. Patient package insert The package insert is designed to be a brief summary of the risks and benefits of the pilll Because the insert should be short, enough to encourage users to read it, much~ detailed information must necessarily be excluded. However, this goal is not well served by sacrificing vitallv important information about the in,- creased risk to pill users who smoke. We suggest the following brief summary be included :. "The risk of serious side effects,front the pilligenerally increases with age: Birth control pills are not recommended if you are over the age of 40, beause they increase your risk of! heart attacks. Smoking 15 or more cigarettes a day will also increase your risk of having ai heart attack or blood clot if you use the pill." D. Detailed patient labeling The detailed patient labeling is : intended to clarify and explain the risks listed in the package insert. The FDA proposed label contains only a generalized recom. mendation that women over 40 should not use the pill because of an increased risk 0 of heart attacks. Women under 40 are not warne& that they, too, have a substan- tial risk if they use the pill and smoke heavily. The following additions will (f; clarify all known risks associated with oral contraceptive use. (~ 1. The section "Who should not use Oral Contraceptives" should include the ~, following st'atement : ~ "Heavy smolcing siE~uifico-int.ly inereases: your risk of having a heart attack or ~ blood clot's if you use the pill. If you are over 30, if you smoke 15 or more ciga- ~ rettes per day and if you have taken the pill for five years or more, the risk is even higher. Because the effects build ttp over time, even if you are under 30 and have taken the pill for less than five years, your decisioni now to continue smoking and/or use the pill will affect your risks in the future. You sllould therefore evalu- ale these considerations now." • t
Page 212: azo61e00
208 2. The section "The Dangers of Orali Contraceptives," Part (a) "Abnormal Blood Clottiing"' should include a statement that the increased risk of: abnormal clotting t,o oral contraceptive users is exacerbated by heavy smoking in women of all ages, but especially for women over 30, 3. In the section "Comparison of the Risks of Oral Contraceptives with other Contraceptive Methods," the graph should include separate bars of pill users who smoke and those who do not smoke. The visual image of one bar towering over all the others may be the single strongest impression left in the reader's mind. The wording of the text,aecompanying, the graph should also be adjusted to fit this addition. 4. The "Summary" section should conclude with a recommendation such as the following : "If you are over 30 years of age. smoke 15 or more cigarettes per day and have used the pill continuously for more than five years, it is recommended thatt you either select aw alternate method of birth control or quit smoking." The FDA should be extremely cautious in amending its proposal not to over- warn the 64% of women who do not smoke against the use of this most effec- tive form of contraception. Oral contraceptives do have many beneficial effects for which no substitutes have been found': The convenience and effectiveness of the pill has made it one of the most,widely used methods of family planning in the United States. Therefore;, the FDA-sanctioned labels and inserts must strikee a delicate balance between cautioning those for whom pill use may create an unacceptably high risk of death, and assuring others who do not smoke or have any of the other predisposing risk factors that! the pill is acceptably safe. We believe that if our recommendations are incorporated in the package inserts and labels, women will neither ~be needlessly frightened away from~ the pill nor be giwem a false sense of security by undifferentiating statistics. Attachment #4 [Except from t'ranscript,ef Nov. 17 advisory committee meeting] Mr. LrmT. I don't think so. Dr. ARCHER. You would think ini terms of the mortality rate tables, etc., thatt that' would not be a realistic thing that would be important, then, to include? Mr. LrrT. We know mortality in the general population. We don't, know it for people,who don't have risk factors. Dr. ARCaES. My personal opinioni is that we need: some type of overview statement that can be inserted~ appropriately and where necessary backed up with a few-one or two other sentences that would indicate what is known at this time in terms of cardiovascular phenomena. Dr. KING. Dr. Ory has suggested that the labeling could be revisedi with a use of this warning. Dr. OsY. To begin the warning section- Dr. KING. To begin the warning section-this is the only change you're recommending. Dr. ORY. No, that's the overview he's asking for. Dr. KiNO. This is the sermon that would be plugged in wherever appropriate in the package insert. "Smoking markedly increases the risk of experiencing" rather than suffering, "experiencing serious cardiovascular side effects from oral contraceptive use, especially in women greater than 35 years of age. Women who smoke and who wish to continue oral contraceptive use should be advised t'o stop smoking." Dr. ORY. You sound like David Brinkley. Dr. CORFMAN. What's that last sentence again? Dr. KiNC: "Women who smoke and who, wish to continue oral contraceptive use should be advised to stop smoking." Dr. CORFMAN. That's cutting the mustard. Dr. PULniAM. Read the whole thing from the top again. Dr. Krva. "Smoking markedly increases the risk of'experiencing serious cardio- vascular side effects from oral contraceptive use, especially in women greater than 35 years of age: Women who smoke and who: wish tb continue oral contra- ceptive use should be advised to stop smoking." Dr. CORFMAN. Why don't you say "Women who smoke and wish to, use"-of course, to continue, what of those who are considering use?' Dr. KING. Okay. "Womem who smoke and who wish~ to use oral contraceptives should be advised to stop smoking." Dr. CoxFMAw. As should all smokers. (laughter)~
Page 213: azo61e00
209 Dr. CoRFnseN. Well that's what's implied in that really, isn't it? Because: the risk does seem to increase with age, if you're leaving out that- Dr. TYRER. No, he mentioned age. Dr. ORY. I got' 35 in there. Dr. KING. Okay. Dr. ORx. I think it"s a nice statement. It's the CIGARISC-it's their points 1 and 4 by the way, just rewritten. Dr. KING. So really that's been taken from the CIGARISC recommendations of 1 and 4. Also in a way 2- Dr. ORy. Right. Dr. KING. And in 3 it leaves that 3 out. Okay? Dr. ORY. My suggestion is that that be a boxed warning for both the patient and physician-you know right to begin the warning section. Dr. ORTIZ. If its okay with the committee we can work on the physicians labeling and then the patient! labeling can just ride on it-just a matter of trans- lating, if it'.S okay with the committee. Dr. ORy. Except I think that the CIGARISC and the ASH people made the point that it didn't appear that when you rew rote it that the physician-the patient labeling-some of the stuff wasn't translated there. Then I think their point's well taken as I look at it~. But that would change. Okay. Dr. TxReR. You w ould change that. Dr. ORTIZ. Yeah, we'll take care of that. Dr. OaY: Okay. Mr. FourrrAIN. I note this language that you finally adopted did not have the word "advised" to give up smoking. I note that the May 27, 1977, Federal Register amendment also failed to require an anti- smokir.g, warning although you had Jain's paper and his January rec- ommendation. We may have already touched upon this to some extent. `Vhy, did the FDA fail to give notice that it would require such a warning when it published the amendment to the proposed OC label~ ing order? Dr. KENNEDY. You are talking about the document of which date? Mr. FOUNTAIN. May 27. Dr. KENNEDY.14fay 27,1977 ? Mr. FOUNTAIN. You did not require an antismoking warning although you did have Jain's paper and his January recommendation and' the Mann paper. I have asked you why FDA failed to give notice that it would require su& a warning when you published the amend- ment to the proposed labeling? Dr. KENNEDY. I am guessing, Mr. Chairman, that our analysis of the Jain paper and the entire business was not completed at that time. Again, we can look i'ntothat history and try to supplyyott witlu a fuller answer for the record. 1ir: FoUNTArN. That willlbe fine. [Counsel for the Tobacco Institute submitted a memorandum on this point. See appendix, p. 257.]I [The information requested from FDA follbws :] At the time of the May 27, 1977 Federal Register published Amendment we were still, in the process of evaluating the comments received relative to the. I3ecember 7, 1976 publication as well as reviewing and evaluating the data regarding O.C: use and smoking in the O.C. labeling and how it should bee worded. Ac!tually, the Ory article and the Beral article were not published until, after May 1977. Taking all this new data int'o account, we subsequently reviewed the O.C. labeling and presented the revised version to our Advisory Committee on November 17, 1977 for their comments and~ recommendations. The labeling re- quired rather extensive revisions regarding O.C. use and smoking-i.e., new
Page 214: azo61e00
210 charts, new graphs, expanded wording on the, subject„ whereas the May 27, 1977 amendment simply required a redefining of' efficacy rates with a substitute reference. Mr. FOUNTAIN. When did FDA finally decide to require the anti- smoking warning? Dr. KExxEDY. A critical event occurred, as you know, later in that year. The Royal College paper came out. As you will! readily appreciate, the Mann papers and the Jain paper, which was an analysis of the Mann data, provid'ed us ~ with no new original numbers but later on in that year, I believe it was October, the Royal College paper came out. That obviously focused further attention on the problem an& it was, I think, only a month or so after the Royal College paper came out that our advisory committlee met., The issue was then presented to them an& the advisory committee was, I think, by the appearance of this second study, prepared to consider the matter with agreat deat more concern. Mr. FouxTAix. Mr. Goldhammer ?' Mr. GOLDHAbi3iER. Dr. Kennedy, you sayy that one of the key things in the later development's that led to this decision to have a warning. or at least consider a warning statement', was withi regard'to the Royal, College study. In your prepared statement„however,yousaythat that study showed~ that the two risks were certainly additive and possibly synergistic. You saU "possibly" synergistic. That is meaningless, be- cause when you say it is "possibly" synergistic, you don~t know whether it is~or is not. Dr. KExxEDi. We think additive risks are worth worrying about also. One of the things that the Royal Col'lege paper did, of course, was to reinforce our concern about some of those small group numbers alone. We are troubled by risksthat are merely additive. If they are syner- gistic, then~ they are more trouble. But you see, the effect of the Royal College study was to focus and enhance our concern about both risks taken separately and I think itt was our advisory committee's view-it certainly would have been~ my view-that that made the concern an even more serious one: This is whether otr not that study by itself con, clusively demonstrated synergism. I would make the point that I think a warning in this case is justi- fied even if all studies onlv showed an additive risk because the fact of the matter is that myocardial infarction in women is quite rare unless they either smoke or use the pill. Even i f the risk is only additive, the status of being a smoker alhvady elevates it' enough into the area so that to undertake an additional risk is something that should be con- sidered with the greatest caution. Mr. GoLDHArtniER. Let me ask this of Dr. Crout, as an M.D. I think your specialty is cardiac diseases; is it not? Dr. CROIIT. Yes. Mr. GOLDHAMMER. In your oplnlon,, as a physician and from your' experience as a physician in this specialty, which, in~ your opinion, is the ereater risk? I am ready to admit that there is~ a serious risk in cardiac diseases from smoking. Which is the greater risk-cholesterol and the low-density lipoproteins in the blbod, or smoking„insofar as a ft
Page 215: azo61e00
211 heart attack is concerned-myocardiall infarction or any other heart at'tack ? Dr. CROUT. There is not a precise answer to that. One has to hone down to certain age groups and whether it is males or females. However, in females who are premenopausal, which is the age group under consideration, clearly the biggest of those two risks is cigarette smoking. That is the point. Essentially, all of those women in all of these studies who are premenopausal who get a myocardial infarction. are smokers and OC users. That is why the quality of the dat'a on the combined column there on the chart is much better than~ the quality of the dat!a in~the other boxes. There are really so few women who get myocardial infarctions with- out being both a smoker and an OC user so that it is hard~ to get large numbers in the other two boxes. That is why there will always be some uncertainty as to the exact use of OC use alone and the exact use of cigarettes alone. The fact is that most cases involve the two together. This is a much more prominent risk factor from these data than hypocholesterolemia. The vast majority of these women do not have, disturbances of their cholesteroll metabolism. Mr. GOLDHA'MMER. How about diabetes? Which is the greater risk factor? Dr. CROrnr. That is not possible to tell from these kinds of studies. The reason is this. In order to do an epidemiological study of that type, persons with other risk factors are generally dropped from the study. The reason they are dropped' from the study is that physicians do not prescribe OC's in that group, and many of them have been educated not to smoke. So,, for purposes of doing a study, they are a biased sample. So, one drops them in order to get more pure study groups and control groups. So, I think we will always, that is, there will always be uncertainty as to the plrecise answer to your question. Mr. GOLDHA11iMER. Would you say the same thing about hyper- tension ? Dr. CROUT. Yes. Again, oral contraceptives are not ordinarily pre- scribed for patients with high blood pressure. Similarly, manyy of them have been educated'not to smoke. Mr. GOLDHAMMER. What can you say about obeslty, which is a risk factor? Dr. CROUT. The same principle applies. Mr. GoLDxAMMER> So, you are putting cigarette smoking ini a sepa- rate category.altogether? Dr. CROUT. No,, what I am saying is this: If you are an otherwise well woman, premenopausal, and if you get a myocardial infarctiony almost all! of these women, both smoke and use oral contraceptives. The figure in Jick's study, for example, was 92 percent. Mr. GoLDHAMMER. Isn't that a deficiencyy when,, if I remember my reading of these studies, the studies show that the women who do not smoke and who take the pill are small in numbers compared to t'~he«omen who~ smoke and fake the pill. So, you, donot, havea: com- parable body of data for both groups to be able to draw inferences with respect to the relative risks. r
Page 216: azo61e00
212 I think if you go back to the Jick study, you will see that he found that it seems as though only the women who take the pill smoke. You cannot find women who take the pilll and d'o not smoke. You do not't find many of' those. Dr. CROuT. Remember, he was beginning with patients who had myocardial infarctions. In a group, that is true. You do not find many premenopausal women with myocardial infarctions who neither smoked nor took OC's: It is very rare. But I am told that in general, in our population in the United' States, there is no correlation between oral contraceptive use and smoking as a general matter. This is from~ other studies and not t'hosee that are before us today. Dr. KnxxEnY. If I may, Mr. Goldhammer, let me say this: The bottom line is relevant to your question about those three columns on the charts and about the risks. We can only deal with samples in these kinds of studies of women who have had myocardial infarctions. We are studying the distribu- tion: of habits among victims and not the distribution of victims among people with a habit. As a consequence, our universe is limited because what we are saying, is that an overwhelming, proportion of women whom we find in hospitals having suffered heart attacks are oral contraceptive users and' smokers. Right away, that tells us something. It tells us that that is a very unwise combination of things to do. This is irrespective of whether we can achieve significance in these much smaller categories of numbers. Mr. GoLnxariMEx. Thank you, :1Zr. Chairman. Mr. Fou.NTniN. I am also placing in the record an Aprit 4, 1977, statistical review of the Jam paper prenared bv the Division of Biometrics and jointlv signed by Bertram D. Litt, Mathematical Stat- isticia.n, Robert T. O'NeilL Ph. D., and Satya D. Dubey, Ph. D., Chief, Statistical Evaluation Branch, [The document re ferred4o follows :]
Page 217: azo61e00
213 MEMORANDUM DEPARTMENT OF HEALTH, EDUCATION. AND tVELFARE PUSLIC'}lEALTHStRwICE FOOD AND DRUC ADh[I\ISl'RATIOV E. M. Drtiz,,M.D. DATE: April 4, 1977 Director, Division of Metabolfism~ and'Endocrine Drug Products . rROnt , Division of Biometrics (HF0+232) suaJDCr: Statistical Review of A. Jain Paper "Cigarette Smoking, Use of Oral Contraceptives and'Unpublished Attachment fI and III'. I. Introduction These papers are based uponiJain's recalculation of data on,Myocardiali Infarctioniand Oral Contraceptives taken from J. I. Mann's publications (BMJ 2;241 and 245, 1975 and 3:631, 1975),and from Tietze's article on Mortality Associated with the Control of', Fertility (Family Planning Perspectives Vo1.,8'Al, 1976). fhe essential!prob'lem broached by Jain Is the extent to which the Mann data may reasonably be "generalized to assess the safety of oral contraceptives in other countries"'. With this in mind it is instructive to refer back to Mann's paper on "Myocardial Infarction in Young Women with Special Reference to Oral Contraceptive Prattice" (BMJ 3 May 1975 pgs. 241-244). 2. Brief Summary of Relevant Issues in the Mann Papers In the paper on non-fatal myocardial infarction, (MI) the authors summarize the inferences from the papers on both fatal and non-fatalimyocardi'al infarction. They note that it is unlikely that the association between myocardaal,infarction and oral contraceptives shown in these two studies is due to chance. They find a strong suggestion that the combined effects of risk factors are synergistic. Th'erefore. they conclude th'at'it is difficult to conclude whether the association is causal or reflects the association of oral contraceptive.use with some other factor(s). AddStionallly. Mann and his coauthors are careful to point out that the number of women in each subgroup, see Tablp X reproduced on Attactanent 1, are not large enough to estimate the effect ofieach risk factor or combination of risk factors with a meaningful degree of precision. 3. Jain's Position Jain evaluates the data from Mann's Table X to obtain risk of - N©n-Fatal myooardial infarction,due t'o oral contraceptive use alone and'to cigarette smoking. He observes that association is much loeer among oral contraceptive users then it is among smokers. He theniproceeds to~reanalyze the data in. Mann's Table VUII simultaneously, evaluating effects of'smoking and concurrent oral contraceptive use without con;idcrationof aayadditional ris.kfec.tors. ~,I~ ~ W p W
Page 218: azo61e00
214 Following this he extrapolates the smoking history of women in the article on Non-fatal MI to those in the article on Fatal MI and suggests that the Fatal MI are probably largely due to smoking. His next step is.to extrapolate these event rates to health hazards expressed as risks of death per 100,000. Having finished with the Mann Data he then hypothesizes that the Tietze calculations should have included a consideration of smoking effects which he then~adds to the Tietze model. 4. •Our Evaluation of Jain's Position Before discussing the simultaneous effect of smoking and oral contraceptive use, Jain looks briefly at the Mann Table X data (See Attachment 1), disregards the warning,about the small numbers and overlooks the fact that.in Table X the No Risk Factors category may mean that women were either previous smokers or were current smokerss of less than 15 cigarettes per day - similarly women with, a single risk factor may have additional risks due to prior or limited smoking during the month preceeding myocardial infarction. The relative risk of myocardial infarction among oral contraceptive users during the month preceeding myocardial infarction is shown by,Jain to be 1.5 to l compared to women with No Risk Factors (as defined in Table X). However, when we evaluate the data, as shown on the first line of Attachment 2, we found that the 95% confidence limits include 1 being .0515 (for the lower limit) and 17.497 (for the upper limit)i. Thus the study data,'one oral contraceptive user with MI and'7 control oral contraceptive users, are not sufficient to reliablyestimate whether there is an excess risk associated with oral!contraceptive use in the absence of all.other risk factors. It is further shown, in Attachment 2 that the power of the test against the statistically significanU alternative hypothesis is less tfian 6Z (the power is shown in the right most' column Attachment 2 and the hypottietical distribution for the statistically significant alternative is shown in the preceeding 4 columns, labeled X1, X2, X3, X4; the actual distributions reported by Mann in Tab'les X and VIII are shown in the left most columns). Thus it appears to be misleading to suggest that t'he data are sufficient to evaluate the effects of individual risk factor(s) or oral contraceptive use. There were 7 MI who smoked 15 cigarettes or more per day with no other risk factors and 21 similar controls; they contribute a little more information. But the 95% (.9451 and 12.74) confidence limits on the relative risk include one and the 99% confidence (.75 and 16) limits clearly indicate that the true relative risk include 1. Therefore, . the study data are not sufficient to determine with complete assurance that smoking,alone, even 15 or more cigarettes per day, without additional risk factors is clearly associated with am increased risk of Non-fatal myocardial infarction compared'with women who had no risk factors (except as defined above). When we compared the risk of, oral contraceptive use alone with cigarette smoking of 15 or more per day (line 3 of Attachment 2) we found that this data were even less conclusive for making inferences then were those for comparing oral contraceptive use with,Nb risk factors at all.
Page 219: azo61e00
1 215 It seems to have been overlooked by Jain that Mann's use of the data in Table X was to evaluate the synergistic effect of 1. 2 and 3 known risk factors. We summarizedthe cases with 2 or more known risk factors and found 34 of the 60 MI cases and 21 of the 173 controls fell into this category. This distribution is statistically significant'at P<.001. It is important to note that Table X shows 19 Mi with two risk factors and 15 with three risk factors but Table VIII includes only 11 MI among women who'had smoked 15 or more cigarettes. The results of analyzing risk factors other than,oral contraceptive has been presented by Mann in Brit. J. Prey. Soc. Med. 1976 Vol. 30 pgs. 94-100. Mann showed that when data from all three medical regions are analyzed together that:there are several risk factors which remain statistically signifieant after adjustment' for both oral contraceptive and cigarette usage and that the relative risk of MI increased exponentially as the number of risk factors increases from 1 to 2 to 3. The author, Dr. Jain, then reanalyzes the data in Mann's Table VIII. The procedures used by Jain are conservative and'p,erhaps overly simplistic. More importantly the proceeding discussion has shown that the data are not sufficient to look at a bivariate analysis because of small numbers of persons with individual'.factors and:because the bivariate Table does not take other risk factors into account (compare Tables VIQQ and X.as shown, on Attachment-1). In addition Table VIII which shows the bivariate distribution clearly is in error with respecL to the nur.:ber of women shown as not smoking i.e. none and/or as smoking 1-14 cigarettes_per day. The bivariate display of cigarette smoking, Table IV (also shown,on Attachment 1), shows 14 MI'and 74 controls as never smokers or ex-smokers but Table VIII shows 16 Miiand 78 cont'rols. Also Tatile IV shows 12 MI and 50 controls smoking 1-14'cigarettes per day while Table VIII lists 2 and 44 these data are inconsistent and either Table IV or Table VIII,must be in error. Table VIII excludes two MI.and four controls tabulated in Table IV but there are too many non-smokers and too few smokers of 1-14 cigarettes per day in Table VIII: . Even when one uses the data from Table VIII as if it were correct a proper analysis would be bivariate. That is,one.should look at the effect of oral contraceptives after adjusting for smoking. Wp have done this during our earlier review of the Mann publ!ication (see S. Dubey's report to M. Finkel, April 14, 1976 page 14 and 15) at that time we noted~t'hat: " '1) After making adjustments for smoking habits there is still an increased risk of non-fatal myocardial infarction for current users of oral contraceptives between 25 and 44 years o.' age. (The adjustment for smokina decreases the estimate of the relative risk from 4.5 to 3:84. See Tables 2 and 3). . ~
Page 220: azo61e00
I 216 2) Among current smokers there is an increased risk of non-fatal myocardial infarction for current users of oral contraceptives between 25 and 44 years of age. (This risk is estimated to be 5.42 times as large as that associated with women who are not eurrently using oral contraceptives. See Table 3). 3) Two few nonrsmoking current users were studied to permit one to infer a) that for non-smokers there is an increased risk of myocardial Infarction associated with the current use of orall contraceptives, or b) that the increased risk of myocardial infarctiion associated with the current use of oral contraceptives is greater among smokers than among non-smokers." Altof the remaining material reanalyzed by Jain is a matter of rumination in the absence of facts. As one does not know the risk factors which the women who died really had, we can not evaluate the risk factors. Jain's ruminations about what the truth may be is not verifiable. With respect to Dr. Tietze's analysis of mortality,it is clear what (Tietze's) premises are; if Jain wishes to recompute the risks on the basis of other premises this does not make his results better or,worse - only different. 5. Conclusions 1. Jain has not resolved the issue that'there are insufficient data t'o evaluate effects of individual_ risk factors for non-fat'al,myocardial infarct'ion and almost no data for fatal myocardial infarction; moreover, as prior smoking or current smoking,of less than 15 cigarettes per dty 'are not considered'by Mann.as a risk factor, in Table X, his data are not'~ shown in a way which,would permit us to evaluate individual risk factors if there were enough:data to do so. 2. Mann's data suggest that all known risk factors for myocardial infarction including oral contraceptive use are synergistic; Jain has chosenito disregard all other risk factors except oral contraceptive use and cigarettee smoking and'thus has added'not'hing novel to the interpretation of Mann's data with respect to risks of MI. 3. Jain's computations are based on Mann's Table VIII, which,we believe to be an inaccurate data set. 4. Even If we consider the data in Table VIII as if they reported the risks of nonrfatal myocardial infarction t'o users of oral contraceptives and oficigarettes and if we also assumed that the women had no other risk factors, the proper analysis shows that after adjustment for smoking increased there is an increased risk of myocardial infarction compared with non-use of oral contraceptives; Jain has not taken appropriate rate of the effect of oral contraceptive use after adjusting for smoking.
Page 221: azo61e00
217 5. As shown by Tietze's article, the risk of myocardial infarction is greater in women over 40 then it is among younger women; as the risk of myocardial infarction is further increased among oral contraceptive users of any age it seems appropriate to question whether the increased increment of risk associated with oral contraceptive use becomes less acceptable in women over 40. 6. Calculations based on rumination or mathematical:models are not verifiable except by additional studies. Therefore, they should be considered with caution; the comnents made by Jain in this area are open to criticism with respect toAhe data, the methodology used and' the way in~which he draws inferences. Basically the increased risks of bothismoking and oralicontraeeptives are influenced by additional risk factors not analysed by Jain6 Bertram D. Litt Nathematical Statistician ~ Robert T. O'Neill. Ph.D:~ y f n Leader 6rou _ p - -Satva D. Dubey. Ph:D: Chief, Statistical EvaluationlBranch cc: HFD-230 HFD-232/Dr. Dubey HFD-200/Dr. Ruskin HFD-232/B. Litt Chron. BLitt/rab/4/5/77 . ~
Page 222: azo61e00
g~206SC0 ~ I Univariate Tables` TT~ ~ _ I ' I ' t- Bivariate Distribution~ ~ -- _ 1 r_~-`_i. -,. ~-. ~ ~ ~ ~I_/- r ~. --tl -.~_ ~ ~-l I ~ lll Ii I Multivariqte Distrtbution ' I i ~ r.-r„cl.uer~i,.ur.r.rr.rrw..rp.rr.,riu.:r..Ir.rrMU_ I~ ( ~ I I I ,~ ~.r L~ 1.. .C . I t I: I ~r~- I - - I ulu t- i I~ r ~ I ~I uiii ~ I ~ a i --~~}} l ~ I u I t t I I T I - ~ _ ~. {--=-- .L. I _ I I ~~= _ ~ _ - } f r J_ 2~ = ~ ,. ~ i '~-~ --~-I I ~ ~~ I ~ _J -- , -..- : -~ _ -~A-t-t-a-cb-ment-. 1.1 L _~~ ~~ f 1 ~ H' t ~-- I- I . -- ----' r-rr<ar.rr.-....Fxa u.nrr. .I ~hn..r.IW r~Ja..ri.:r ..4 C.rurd i r . . . _ fffSourtBMJ, 915 2, 2A1 245 ~. ~ -- --. - T~- -- - - ~- ~ r - ~ ,~ - r ~ i
Page 223: azo61e00
9 I 9 03590259 cOL /CT 91V LS%9' OL CT 9 CT 0000' LS69' 0009, 6005'i CbCC't C9CT'C9 L6Lt'i CLLL'66 CL06'L L999'TT (9 /CT ),'PY+lo S96S' OL CT LC CC 0000' S969' 0000' TLL6'T 69tb'i i0L6'9t 6*e1'T 0C0L'6T 6C0L'L • 6CS9'S (OL (+.C /CT ) W /CC epf •:0f1 = :v ~ (6L /CT ) = T0T0' SL 6 LC TT 09C0' 6C69' STLL' ib6T' C6CL' 69LC'C 5691- OSOL'L CTtL' L69L' (L~ /9 )C.I, •~SY ` ' " (0L . . . /CT ) 6699' OL CT ~ TT 0000' 6%99' 0000' 06T9'L T09S'C 999t'00T 6L9C'L C6TC'LL LL90'C LL06'6T 0 /TT )%sl <-o 96TC' it LT I C CSTO' T90L' 6SCT' C6LC' 6t6K' LCCS'00t SL62' [OLL'CL L006' 9e9C'i (ot (t /tt )- /L ~) . • (0L l6i )?J 0LL9' 0L 9L 9 CC tT00' '04991 CT00' C66L'T L969'T L009'St 9b6T'T LTt1'0t 0%66'T L9T0'S ~ (9 /CT ~ (9L /LL ) 9669' 19 LL ~ TT SCh0' 96%9' SC~O' 6' 9096'iL S99S' 8699, 665 690L'LT LTbt' OOOS'C (6 /TT 6LTC' C6 • S T C h600' 6%LL' LLOS' 6StC' LtlSS' STLL'LSt 6LCC' .6069'LTT *6S%' 0000'L (L /L ' ' ' ' ' (OL /CT ) ».rs 69CT' CL it I S 6110, 9599' 66Li' 0h9L' 999t;' 99L6'ST 00C: L6T0 [T 9LTC t 9SC6 (9 It (Tt /L '~fh7 /T C000' SL C C S 6900' 6SL9.' CL69' 0tS0' 99T0' S0L9'9 L600' M656'C 9CT0' 99L6' (L /T ) ~iri 9 A S609' STT it TL L•Ct0' 5609' 9CL0' 9CL6' SiLO'T 0R90'9T ZSSL' LttL'L1 TS66' 9If 6to 'C (SST (TL /Tt /L ) r -j :0, 9540' LIT 6 S C T%L0' 6TL9' 6LC5' 21110 1 SC7J' •66n9'et 99tu' Lbti%'L[ STSO' St6n'T (STT (L /TT ) /T ) h•e a0 ~ N +f< ily .*l 9 •/•66 •/•S6 •/•66 •/-Sb (aX/CX) I dAOh'D (LX/TX) Y dllJeD 8• 1 (83dun) 8•x^IH a 90Sl if Yc ii07 xsd 738isY su' 04 735: tU (xslY • lY tSas079 a :Y SI:iIT +i7:1:7 S:Ilil`I 3Ai3Y^.lY JO iS3t d0 73A17 iS2J d0 7p7YA iJL'3JIrOJ9 YDY30I+6'GJ a0ra0l:auo a1YNl.:S9) Y3L0d LI0ISYY09Ili0J 9IS Y3H0d . d 8Y807 C707S.2 OIlYY 5000 j ~OhN71Y . 0
Page 224: azo61e00
220 Mr. Fou:vTAiN. Dr. Kennedy, do you agree that these three indi- viduals have the qualifications and the competence to conduct a valid statistical review of Jain's paper? Dr. KENNEDY. Yes, sir. Mr. FouNTaiN. To me, as a layman in reviewing statistics-and I must say my experience is limited-their statistical review of Jain's position appears to be critical of' his use of the Mann data. For in- stance, at page 2 the FDA statisticians state that in using the data in table X of the Mann et al. paper, Jaim disregarded the fact that the authors were careful to point out that the number of women in each subgroup of the table is not large enough to estimate the effect of each risk factor-smoking being one-or combination of risk fac- tors, with a meaningful degree of precision. The FDA statisticians further state in their comments of Jain's use of the Mann et al. data, and I quote :"Thus it appears to be misleading to suggest that the data is sufficient t'o evaluate the effects of indi- vidual risk factor(s) or oral contraceptive use." That is on page 2. Two sentences later the following stated : Therefore, the study data are not sufficient to determine with complete assur- ance that smoking alone, even 15 or more cigarettes per day, without additional risk factors: is clearly associated with an increased risk of nonfatal myocardial infarction compared with women who had no risk factors (except as defined above). When we compared the risk of oral contraceptive use alone with~cigarette smoking of 15 or more per day (line 3 of att'achment' 2) we found that this data was even less concltisive for making inferences than were those for com- paring oral contraceptive use with no risk factors at a111 Other statements in the review point to other flaws or potential' flaws of Jain's analysis: For example, in speaking of the Mann data, the FDA statisticians stat!e, at page 4, and again I quot!e : 3. Too few nonsmoking current users were studied to permit, one to infer (a) that for nonsmokers there is an increased risk of myocardial infarction associated with the currentuse of oral contraceptives, or (b) that the increasedi risk of: myocardial infaretion associated with the current' use of oral contracep- tives is greater among smokers than among nonsmokers. In the review, under "Conclusions" the statisticians summarize the deficiencies of Jain's study. Among the deficiencies cited, an& one which I have not included in my quotations, is conclusion No, 3, which reads as follhw,s; and I quote :"Jain's computations are based on Mann's table VIII, which we believe to be an inaccurate data set." In conclusion No. 6 they state of Jain's study, and I quote again : Calculations based on rumination or mathematical models: are not verifiable except by additional studies. Therefore, they should be considered with caution ; the comments made by Jain in this area are open to,criticism with respect to the data, the methodology used and the way in which he draws inferences: Basically, the increased risks of both smoking and~ oral contraceptives are infltienced by additional risk factors not analyzed by Jain. It is significant,, I think, that'these three stati'sticiansdo not suggest a need for a«arning against smoking if the pill is used. If their review of Jain's study had definitely established that'a serious, reliably quantified,,s,ynergist'ic effect, e'xisted in the risks of t'~he OC pill and cigarette smoking whichi required, in their opinion, an antismoking labeling warning, as Jain indicated in~ his January 1977 written comments, would you have expected them to inform you of that need ?
Page 225: azo61e00
221 Dr. KENNEDY. Mr. Chairman, I think our statisticians did exactly what we wanted them to do. They were sent after that study to point out every one of its infirmities that they could find. They were told to attack it as aggressively as they could', There is no question that there are some adding errors in Mann's paper. These do some damage to table VIII. There is no question that there are some small numberss in those first two columns in the charts although there is legitimate room for some disagreement about how crippling that defect is. The passages that you have read to us are indeed very similar to thee objections of the witnesses that you heard earlier this morning to the size of those sampl$ of OC' users only and smokers only. They give trouble only from a single perspective, and that is, whether there is an accurately estimable and statistically significant risk associated with doing one of those two things alone. However, there is no argument, I think, that there is a profound risk of doing the two things together. In that respect, the analysis does not challenge the study one whit. Finally, and most importantly, it seems t.o me that what we have got here at this point is our studies out on~ the board and a fifth one coming in which clearly agree. They may, in one column or another, fall short of the sample size that one would like on which to base a firm statistical statement. To call' your attention to the language you read us again, "to deter- mine with complete assurance"'are the words used. Taken together, I think they show unmist.akabi,y the reality of the separate risk factors. They convince me that there is synergism. How- ever,, I would be sitting here before you, Mr. Chairman, defending a label warning even if they demonstrated' no synergism~ at all but only additivity. • Mr. FOUNTAIN. What, if anything, is to be inferred from the fact- and I appreciate your own opinion that you just expressed-that they did not mention the need of an antismoking warning in any of their several analyses of the Mann et al. papers or of Jain's studies based! on them ? None of the FDA statist'icians' reports that FDA provided to the subcommittee even mentioned such a warning. They certainly didn't recommend one. They had the authority to do so, did they not? Nothing kept them from doing so, did', it ? Dr. KENNEDY. It is not so much a question of authority but it is a question of what we ask scientists to do and then what we do with what the scientists tell us. To include or not to include a particular warning in labeling is a policy decision~ that we based on scientific review. What we asked the statisticians for was the toughest, hardest noseanalysis that they could possiblv give us of that study. We did discuss with our advisory committee the question of a warn- ing. We discussed it in the agency. However, we do not normally ask statisticians t'~hequest~ion, "Do you think we ought to put a label on?"We ask our statisticians the question, "What do these data look like after you put them through your massage ?" Mr. FouNTAiN. And you would not, expect them to make any other suggestion other than their findings?' 95-727~0 - 7q - 15
Page 226: azo61e00
222 Dr. KENNEnY. I would expect them ultimately to participate in the aiscussions of agency policy and to help advise us on that policy, but in a preliminary review, Mr. Chairman, I would not expect them to recommend a policy. I think they understand that that is a different stage of the process. Mr. FOIINTAIN. I am placing in the record, as a part of these hear- ings, an FDA memorandum dated September 26, 1977, from the FDA Statistical Evaluation Branch, on the subject of, and'I quote :"Stat'is- t'ical comments of 6 articles concerning the associationi between ciga- rette smoking and myocardial infarction among users of oral contra- ceptives." [The document referred to follows:] ,
Page 227: azo61e00
223 MEMORANDUM DEPARTb1END OF HEALTH, EDUCAT[ONi ANDWELFAREPubtic Health 5crvice ro : E. M. Ortiz, M.D: DATE: September 26, 1977 Director of Metabolism and Endocrine Drug Products THRU : Charles Anellb, Sc.Di f1r7~2 Director, Division of Biometrics (IiFD-130)~ f FxoM : Statistical Evaluation Branch (HFO-232). SUBJECT: Statistical Comments of 6 articles concerning the Association between~ Cigarette Smoking and Myocardial Infarction among users of Oral Contraceptives 1. Background This memorandum responds to your request of'August 19, 1977, to provide statistical analyses for the purpose of revising the oral contraceptive (OC) class labelling to take appropriate cognizance of the association between smoking and myoeardiall infaretion (MI), in users ofiorall contraceptives. The data to which you referred are largely confined to the two case control studies on MI and'OC reported~by Mann et_ al. (BMJ4 May 3, 1975). The Smok'ing-ML-OC papers by Jain and'Ory are reevaluations of these data and Tietze paper is a report whieh estimates mortallity associated~with various forms of fertility control from mathematical models which are in part based upon Jaih's calculations. The comments which follow are based upon the 6 subject articles and 3 earl!ier reports by the Division,of Biometrics evaluating them. We will also provide our recommendations concerning appropriate labelling together with the rationale for our suggestions. 2. E'pidemioloSic Evaluation of'Mann BMJ May 3', 1975 Papers On May,9, 1975, Dr. C. Anello sent a report to Dr. Ortiz evaluating th'e• strengths and:weakness of these two studies. Or. Anello stated, "The authors findings and'conclusions are reasonable based' on the study design and the subseqwent analysis."' And~he concluded, "..tAati the two British Medical Journal papers on myocardial infarction and oral contraceptives provide important substantiating evidence that oral contraceptive usage can contribute to the risk of disease and death from Late in 1975, Dr. Jain subYnitted p,re-publication copies of his paper, "Cigarette Smoking, use of Oral Contraceptives anTMyocardiallInfarction" to FDA and CDC for courtesy review. Shortly thereafiter, the Division of Biometrics was asked to consider: 1)'. Suggestions for relabe.llling Oral.Contraceptlives;Z.). T11ee inferences which could be drawn from the Mann dataa and 3) The amount of weight to,.be given to.Jain's article and thedraft.of the Ory.paper (which was laterpub'lish'ed in the. June 13, 1977, issue of JAMA.) Dr. Dubey's rep,ort to Dr.,Flinkel datedl myocardial infarction." 3. Statistical Evaluation of MI~OC data
Page 228: azo61e00
224 2 April 14, 1976, covered alli3 of these objectives by means of an exhaustive statistical review of the two Mann papers. The conclusions of that report are as follows: 'The two articles are deficient in~many respects. The major deficiency is the inadequate ascertainment and assessment of other risk factors for myocardial infarction. If the impact of these deficiencies can be ignored certain inferences can be drawn from the data: 1. There is an increasedirisk of non-fatal myocardial infarctionifor current users of oralleontraceptives between 25 an&44,years of age who already have some other risk factors for myocardial infarction. (The risk associated with current' users is estimate&to 4.5 times as llarge as that associated with womeniwho~are not currently using,oral contraceptives. See Tablle 2). 2: After making adjustments for smoking habits there is stilllan increased risk of non-fatal myocard6al infarction for current users of~oral contraceptives between 25 and 44 years of', age. (The adjustment for smoking decreases the estimate of the relative risk from 4.5 to 3:84. See Tables 2 and 3). 3. Among current smokers there is an increased'risk of non.fatal myocardial infarction for current users of oral contraceptives,between,25 and 44 years ofiage. (This risk is estimated to be 5.42 times as large as that associated with women who are not currertly using oral contraceptives. See Table 3). 4. Too few non-smoking current users were studied to permit one to infer a) that for non-smokers there is an increased risk of myocardial infarcUion,associated with the current use of oral contraceptives, or b)) that the increased risk of myocardial infarction associated with the current use of oral contraceptives is greater among smokers than among non-smokers. 5. There is an increased risk of fatal myocardial infaretion for current users of oral cont'raceptives who are less than 40 years of'age. (This risk is estimated to~be 2.8 times as large as that associated with.women who are not currently.using oral contraceptives. See Table 6). 6. Too few current users in the 40•44 and 45-49 age groups were studied~to permit one to infer that there is an increased risk of, fata4 myocardial infarction associated with the current use of oral contraceptives in either age group.
Page 229: azo61e00
225 7. However,.in view of conclusion #5 stated above, one may reasonably argue that the risk of, fatal myocardiallinfarction associated with the current use of oral contraceptives in the 40-44 age group should not'be less than that which pertains to the current users of OC who are less than 40 years of age, un- less other factors influence the final out'come. + B., On strictly medical ground, one may also advance a theory of higher increased risk of fatal M[iamong current users of OC who are in the 40-44 age group or older. 9. In summary, itiis important to reemphasize that the validity ofithe above stated conclusions can be questioned on 1P the ground of a number of deficiencies associated wiiththese studies. Nevertheless, the quality.of evidence, in several instances, seems to be pretty persuasive. The FDA ought to consider the total evidence and act with great prudence". From that memo it was our expectation that labelling changes would be limiRed to th'ee following: Labelling,Changes 1. The studies by Mann et. al. present evidence that suggest that'the use of Oral Contraceptives by women of 25 or older with other risk factors for Myocardial Infarctions would increase the risk of incurring FII by 4.5 times. 2. The greater the number of risk factors the greater the probability of Hl: in wann's data these risk factors inclUde current Cigarette Smoking, current OC use, Obesity, hypertension, Type III hyperdiipoproteihemia and diabetes. 3. After adjusting for amount of currentiCigarette Smoking but ignoring all other coexisting risk factors for non-fatal Mtithe Mann data show an increased risk of 3.8 for nontfatalll•1I associatedwith current use of oral contraceptives by women aged 25-44. 4. The risk of Fatal Myocardial Infarction among the women currently using Oral Contraceptives appears to be increased to 2.8 times that of wonien who are not using Oral Contraceptives. 4. Review of Jain's Analyses of the ManniData and Extrapolation of Health Hazard Early in 1977 pre-pub'lication,drafts ofithe second Jain Paper andIsuggestions for revising Tietze's 1976 article on "Hort'ality Associated with Control of Fertility" were sent to us for review. At.t~hat time we detailed the limitations of the Mann data and.thep,roblems that we saw.in attempting to extend'thedata asrecononended by,Jain - a copy of our report dated. April 4, 1977, is attached. - 5. Summary . The data originally, publlished by.tdann show.that:a. Among women over the age of 25~with other risk.factlorsforhil, the
Page 230: azo61e00
226 use of Oral Contraceptives may increase the risk by 4.5 times. b. The greater the number of risk factors the higher this probability of incurring a Myocardial Ihfarction regardless of OC use. The only risk factors analyzed jointly are cigarette smoking and current OC use; Mann's Tablle VIII (BMJ 3 May 1975 page 243) in which these data are presented contain some inaccuracies (see attached memo). It is therefore not elear exactly what the risk of Mb is among current smokers who do and do not concurrently use Oral Contraceptives_ Mowever„ if we take the Table VIII data at face value they show a substantial increase in the risk of MI among current smokers who concurrently used Oral Contraceptives compared to those who did not use OC's. c. The data on FatallMyocardial Infarction is less reliable then the data on survivors in that less complete data were available. In addition, as inadequate samples were available to study effects related to age, different sampling procedures were used for each stratum. Nevertheless it appears that theriske of Fatal Myocardial Infarction among women who use Oral Contraception when they have other risk factors for MI is 2.8 times that of similar women who do not use Oral Contraception. d: The extrapolation of health hazards following the methods reconmendpd by Jain are not supported'by the data. If it'is desired to extrapolate relative risks to health hazards the relative risk could be applied to the age adjusted vitallstatistiics for women of the 25'-40 year age group using methods similar to those reported by Ory. e. For completeness, we recommend reference to botb of the Tietze estimates of mortality associated with fertility control (i.e. 1976 and' 1977) but only by listing in the bibliography of the label. The graphs depicting his estimates arenot.reeommendede for inclusion, in the lab'el~ at this time. This recommendation is based'on the fact that the ti.oo mod@ls differ in thatithe 1976 model uses Mann's data for estimating, the risk of MIiper se but utilizes only legal first-trimester abortions without concurrent sterilization to estimate the risk of mortaliity following ab'ortion;,the 1977 mod&liincludes a more precise estimate of mortaliity due to abortion but partitions the risk of MI among smokers and non-smokers according to Jain's recommendations. Although',Tietze suggest that the 1977 modeliis th'ee more appropriate one, he has not demonstrated this using survey data. Inclusion of one of Tietze's graphs may be indicated if and when a separate study indicates that one of these models is useful. 4
Page 231: azo61e00
227 0 6. Reconrendations for labelling are stated at the end of Section 3 and'in 5e above. cc: HFD-230 HFD-232/B. Litt " /Dr. Dubey, HFD-200/Dr. Ruskin Chron. BLitt/sce/9/26/77 rab 5 BertrmD. Li'tt Mathematical Statistician Robert T. OlNeii1, Ph.DI Group Leader Satya D: Dubey Ph.D. le, Chief, Statistical Evaluation Branch
Page 232: azo61e00
228 Mr. FouNTAIN. In the opening, paragraph, the following statement is made and I quote :"This memorandum responds to your request of August 19, 1i977,,to provide statistical analyses for the purpose of re- vising the oral contraceptive (OC) class labeling to t'ake appropriate cognizance of the association between, smoking and myocardiall in- farction ( MI ) in users of oral contraceptives." The memorandum is signed by the three members of the evaluationi branch, including its chief. Under the caption "Labeling Changes" the authors recommend at page 3 that labeling changes be limited to four items based on studies which they evaluated'. They specifically omitted any recommendation for an antismoking warning. Since FDA's action is apparently based on the data in these six studies, does it not appear that if the data on smoking and oral contra- ceptive use were clearcut that the statisticians would also have included the recommendation of an~ antismoking warning under the section of suggested labeling changes g Dr. KENNEDY. May I ask for one piece of clarification? Mr. FOIINTAIN. Yes; Dr. KENNEDY. You have an energetic staff and they have given you a memorandum that I do not have a copy of with~ me. What was the date ? Mr. FouNTAIN. September 26,1977. Dr. KENNEDY. Well, you see of the studies that we have been talking abollt today, the only original study, I think, that was available as of that date was the Mann, et al., study. The Royal College study and' the Jick study and the Petitti and Wingerd study are all since that date. In fact, the advisory committee had not yet even met at that time or had the Royal College, which was the approximate trigger for bring- ing that matter before the advisory committee, taken~ place before that time. So, I think it is quit'e understandable and, indeed,, that provid'es a little confirmation of the statement that one reason why the agency hadnot quite reached~ that point yet was that it had not yet had access to this confirming study. Mr. FouNTAIN. I am also putting this document into the record to illustrate the fact that the FDA statisticians who evaluated the studies about whi& you have just commented, did not make such a recom- mendation even though Jain had done so more than 9 months earlier, and the FDA statisticians were aware of his recommendation. I believe you have commented on that. Dr. KENNEDY. Yes. Mr. FouNTAIN. To complete the record,, I am placing in~ t•he record two: FD A memorandums. One is dated September 30, 11977, , and the other October 18, 1977, indicating that FDA statisticians were as- signed the responsibility not only to evaluafe the data but also to snggest OC labeling changes based oni the analysis of the data. [The documents referred to follow :] . ,
Page 233: azo61e00
229 Satya Dubey, Pr.D., Cl:ief, September 30, 1977 Statistical byaluation Branch, HFD-232 Robert T. O'Neill, Ph.D., 1---D-232 vorl; to be perforred'on OC LaLeling Last Thursday, Septc.^aber 29,, 1977, Dr. A.-.ello, Fir. Litt, and I met t;ith Dr. Ortiz and r..embers of his staff relative ~ to Dr. Ortiz's recNcst for our review of the literature articles on oral contraceptives, r•.yocardial infaretion and s•Akinc. The result of thia -:eting is that Sert Litt is • cor_:nitted to review the oralicontrnceptive class 1abe1'inc in li;iit of t:ese lit~zreture articles and to edit, modify or suggest labeling cLanges accordingly. We have indicated to Dr. Ortiz that this effort *ill be completed b_r Crrtober 6, 1977. Since you have been in- volved in past reviews of these literature articles, you should be aware of our co.-~it_-aent and deadline. Robert T. O'Neill, Ph.D. cc: M'"D-200/Dr. T.tuski_7 EFD-233/D r. O' tieill 3IFD-2 3 0 'Chron, RTO'3:ai1l: jb
Page 234: azo61e00
230 ME"4ORANDUM nL'PARTWn'roB HEAIfrtl. DDCCA'TION; :CVU. %,'IU.rAUE PUe7.iC HiC.LLTH SF.kvice.. FOOD AND. DNUG ADM1Th I Pt.%T1[DY TO Fdwin Drtiz, M.D: DaTC: October iBi 1977 Feont : Bernard St. Raymond, M.D. susJscT: Status of Oral Contraceptive Labeli)ng.review 1. As per your request, 1 have outlined the current status and proposed timetable of my group with regard to revision of the labeling for the oral contraceptives. We have reviewed them.edicat comment's received referable totheo proposed regulation for Oral Contraceptive Patlpnt LabeiJng, Our comnents were forwarded toMr. Pacquln In t,1ay, 1977,. I-owever,, additional articles have a.ppeared.ln the Illterature (Jaln, Ory., ` Tietze) referable to myocardlal lhfarcttcn and the role that oral contraceptive use andret0arei'`e-smokhrg-TYaY"tn- lt. lf val id, this new data could require that major rev.llsloms be made in bothth'e patient and physician labeling for`th'ese-Crugs: -We have requested the 91ome;ry Dlvlslonito revlaw tlhls cata and provide os with comments and; 1IPIndlcatedy approprtate labeiing chances based on,their evaluation and Interoretatlon of tnadata. Addl'r~lonally, two other studies have recently.ao.peared In thelltterature (Beral, Vessey):whichisuggest that there Is amuch; higher mortality rate.dUe to circulatory disezses associatsd with the use of oral contraceptIves than iscurrentlystated in Dur labeling of Decemoer 7, 1976. The BlometryDlvlslon Is also reviewing thisdata, particufarlyln termsof wh'etherfurther labeling revisions would:.be necessary, We exoeet a response fnom the Biometry, Division regarding these matterss by the end of this week or the flrst'part'of next week.at the.latest. 11. We are currently ln the process of revlewlhg the more than 100 responses received which contain comments regarding the Oral Contraceptive labeling published lastl December, 7:e expect to '. complete the revlew of these com,nents by the end of this week and spend the following week revising the.tbbeling~.in accord with thecommentsrecelved'that we consider to-have merlt:_. l
Page 235: azo61e00
231 Page 2 Ill. It ls our Intentlon to present the newly revised Oral Contraceptlive labeling at Scientific Rounds on November 8, 1977 for addltlonal Input, revisions, and/or concurrence at the Of~fice and Eureau levels. - IV. The flnal draft of Oral Contraceptive labeling Is scheduled't'o be presented totlheObstetrlcsandGynecology Advlsor-y Committee onNovember 17, 1977. Any add.ltional changesreconvicer.ded by theCommlttee wlll be evaluated and ifdee¢ed appropriate and feasible made In the final version of the labeling. . V. Approxlmately l month Cor, perhaps, sooner) after the Advisory Comnlttee meeting, the final verslon of the Oral Contraceptllvelabeling couldbe published ln,theFEDERAL REGISTER. The proposed regulation regarding patient labeling could~be finalized at the same.tllme. ~ Bernard St., Raymonaj M;D. ~~ cc: HFD- 130 HFD-130/BSt:.Raymond!ID-18-77/1!s/10-19-77,
Page 236: azo61e00
232 Mr. FouNTniN. Dr. Kennedy; according to an article in the Janu ary 25, 1978, Washington Post by Mr. Morton Mintz, covering a press conference held the day before, you announced that the FDA would soon require the antismoking, warning. You are quoted as saying, and I quote :"The new FDA message is both loud and clear : If yotu takee the pill don't smoke;, if you must smoke, find another method of contraception." I have found Mr. Mintz to be a very careful reporter. In fact he has been following our hearings for years. I«ould not expect him to usequot'ation marks unless,it was a direct quote. Do you recall whether vou said that ? V Dr. KEN-, EDY. I am sure I did, Mr. Chairman. Mr. Fou:r TaiN . Do you subscribe to that today ? Dr. KEN NEDY. Yes. Mr. FoUNTarti. Did you subscribe to it oD January2-t,,1978 ? Dr. KENNEDY. Yes. Mr. FouNTniN. From what we have heard today;,t:he evidence sup- porting the warning requirement appears to be "soft'." Wouldn't you agree ? Dr. KENNEDY. Soft? Mr. FoLNTAi.x. Yes: Dr. KEN NEDY. No. Mr. FouNTniN. Assuming for the sake of this discussion that it is conclusive, «hyy did y.ou not use that expression in the boxed warn- ing instead of the blunt warning, and I quote again :"«lomen who use oral contraceptives should not smoke." The second part of your statement "if you must smoke, find another method of contracep- tion"-does not appear at all. The warning required' by FDA ap- pears to allow only one course, and that is to stop smoking. For many OC users, that may be a little bit difficult to do. Why did vou not advise them to use an alternative method of contracep- tion if they must smoke? Dr. KENNEDY. That is a good point, Mr. Chairman. I am not sure that my answer on it will satisfy you. I have, the feeling that in a labeling warning, you have a limited amount of space to capture people's attention to make points to them. You use argument,5 that are a little different than the ones that you use in explaining things in a meeting or a press, conference or whatever. I do not remember the situation that existed when I gave Mr. Mintz that quote, which I am sure he has renroduced' accurately. I tliinkthere are also some sensitivities involved here. It seems to me that to tell women "If you must ~moke: findl another means of contxacention" does this: In patient labelinz associated with the product that'theyhavealreadypurchased indic.atiha t-hatthey are committed to a particular course of actiom it isa little bit different than talking to people generally who may or may not ha.ve a commit- mentt'o a particular courseofact~ion. It seemst'o methat theR-omen who read! that warninL), ouRht to beawareof the hazard that comes from the as=ociated behavior of takina the pill and smoking, and realize that havingembarked a]ready on a- course of taking t'hepill- else t:hey would not have the labelinh-t'he main obligation in patient, labelin~ is to infernr them of the hazard, of the other hazard,
Page 237: azo61e00
233 Whereas if you are talking generally to people who have not made that commitment, you may say rather different tlhings about their options. My own guess is, Mr. Chairman, that most women who are selecting a method'of contraception and who are facing this choice on a compli- cated decision about risks and benefits, would perhaps prefer to con, sider first, smoke or don't smoke, and only secondly, "Should I change met•hods ?" Mr. FoLNTAIN. It just appears to me----and it has been argued also- that the required warning as presently written might well mislead OC' users. It is boxed in. It appears first in the labeling. It appears to indicate special significance above other labeling. It is seit apart. I think it could well be argued that it might imply to many users that if they stop smoking, thee danger of tihe pill is sufficiently reduced~ to render itpractically safe. Would you not agree? Dr. KENNEDY. Ib is a reasonable position to take, Mr. Chairman. I do not agree. «"e clearly did not. take t,hat view of the matter at the time we were planning the warning. It seems to me that it properly identifies a hazard associated with a combi.ned behavior of taking the pill a.nd smoking. It is aa sufficiently serious hazard in all of these risk elevation estimates which are rang- ing, on the order of twel vefold for the combined! use to 25-fold for the c.ombined! use, so that I think it certainly deserves to be highlighted and brought to people's attention in the most arresting possible way. However. I must admit t~hat it had not occurred to me that women who read that would realize that if they just stopped sn.boking, thatt they would return to some plateau of absolute safety. It, is my guess that most people might draw that conclusion, but we ought to think about the problem. Mr. FouNTAI;v, Those who are extremely sensitive to the language which you have used mizht feel that is the implica.tion that the pill is safe if you don't smoke, and consequently, smoking is the primary; cause and the only cause. I am talking about the way i't is boxed in t.helabelhng. That is one of the questions the witnesses raised. Dr. KFN N FnV. May I ask Dr. Crout to amplify ? Mr. F oIINTAIN. Yes. Dr. CROUr. I recommended the current language and format to the Commissioner, so, I would like to comment on some of the think- ing behind it. It was ~ our view in the Bureau of Drugs that the other contraindi- cat.ions and warnings with regard to the use of oral contraceptives are largely known t!o the medical profession. It is the physician who g enerally prevents women with other risks from taking oral contra- ceptives. By and large, the woman with high blood pressure, with hypercholesterolemia, with a previous history of thromboembolism, and liver disease and so on, are not permitted to go onto oral contra~ ceptives by their physicians. So, the real risk factor under the patient's control is whether or not they add the risk of smoking to their use of oral contraceptives. So, we, felt that t'hepractical risk that should bemosthithiighted to the consumer is the one of smoking. That was the thinking behind the current format.
Page 238: azo61e00
234 Mr. FOUNTAIN. Of course, our committee is more concerned about the potential danger of the pill if it is not safe than we are about the pill combined with smoking. The sensitivities about' smoking are in a number of'areas. However, I wonder if in view of that intenpretation by many, if it would not be wise for you to make a survey to find out how, many women get the impression that if they do not smoke, the danger of the pill is sufliciently reduced to render it safe? Dr. KENNEDY. That is a timely suggestion, Mr. Chairman. We are presently planning a variety of studies on the effectiveness of patient labeling based on conclusions that patients draw from patient labeling. I t'hink we can incorporate into the design of that kind of a study just the sort of thing that you are bringing up. Mr. FOUNTAIN. Particularly since we have the labeling, where you have cigarette smoking boxed in,,and all'of the other things which may be harmful are included in the body of the label. However, in political advertising, I know that those of us who run for office like the boxed- in approach. It' seems to attract attention. Sometimes people read the headlines of the box and don't read the contents. I am placing into the record the label' and the brief summary patient package insert required by FDA for OC's by the January 28, 1978, Federal Register final ord'er, to illustrate the prominence of the anti- smoking warning. [The documents referred t'o follow :]
Page 239: azo61e00
235 21 TABLETS 15-101-25-3 NORINYL 1+50 (norethindrone 1 m . with 1 _DAYneslranol0.05 mg~ TABL.ETS 0 81.11. rr...1.r.G. 1rrMr~GaO, Ale..0 nlGo eOre1 CizwetM snoldne ilcnases the r(sk d ser(ous adrerse efrects on the Msrt aeE blood vessels from oral r.vnbaaqtl.e use. The risk Yuxeases rritk stle and.Hk 6s.y smo(dns (15 or more cr{- sretles per day) and n auisl n.rred n woatn orr 35 yesrs at aP. Momen wtao usee eral eontracepti.es sriouN eot snoYa_ tlrat. eontraup6res takbn as directed re aboet 99% eneeMie in.pe.antina preanancy. (The mfni-pW, b.ewa„ Is seneshst Irss eH.cti.e.) Faiure to take 1'our triAs increases eKchance nr pregnancy. Women who Nre or hare had do(Gna disorden,.ancer of.the hreast or.ser orgNs. uneetrlaned va{nal hlledln(. a stroke, heart anack, chest pain (angina psctoris) or who suspedttrer suy be pre{nard shoeld not use orat contraap- ti.es. ~ Yosl side eHects of the ple ere not serioui. The.most . common side eeeas are nausea, somitinS. bleeding be- Neen menstrul peri6d,, weight Pin, and hressl kMw- nas. Howeser, proper ese o/ oraltxetraceptisa requires that tlley M INitn under Fon ddCUr's GonlinYoas super-.fsion, trecsose 6reF can he sssoda9sd wiU serious side efrects which rrurkad to disabiFd7 or d.ath. Fertanardy, . Mese.orxur sery inrraYUeney- Tar serian sideeNectsare: i. stood tloh In the leP, lonp. Ors3n, kan or else- where, and hemurhaP Into Hne brain dveto Gunt- irK of atdood vessel.. 2. lise tumors,.hich mn ruptoraand nuse se.ero h(eedin{. 3. Birth defects If the ple Js taYendorYq prepranq. 4 - Hipr blood pressure. , 5... Ga®4dder diseass: The symptoms assedated .pa 1MSen safous s(de dfectsaredlanseedin thetkla9ed].anetttire'rrpouwilh your supplyof pi9s. Notilry your doctor Yyou nolice any unusual physical disturaanuwhie.talorK trr piY.. The estroem in oral cont0acep8ves has han found b cause breast nncrs and other uncers in certain animals. Tkese findings suuest the orall contraaplnres nsr s(so _rasecancer in hen.ns. . Ho.erer, studies to dafe ke Wanee hkln! Ge-* . asarketea'onl'contraeepuaes hwe not.confuned that oral contraceptives nuse cancar in huarss. The deWkd keltd demriEes non oompkbytlr haaefds and ris(n of oral anusceplMS. a also txonldaa Mermetian on oekr forms of eontnnp6on. Read It aretdy. N you haw any ouestiom, oonsua yaua doctor. CAUTION: Oral wntncepEves areof eo ealua in the t>resentlon or trealmant of renereal diseaw. USUAL DOSE-Consider the first day of menstrual i fiow as Day I of the menstrual cycle. Beginning on Day, 5 of the menstrual cycle take one whute tablet daily foc 21 days. Wait 7 days, then,whether bleeding has atopped or not, begin a new cy- cle of 21 tablets on "Your Day." Continue in this manner, three weeks on tablets, one week oil. Even If spotling or breakthrough bleeding should occuri continue according to 4hls schedule: If, this Spotllog or breakthrough bleeding ehould persist, notlfy your physi- clan. In the space marked "Your Day"' write the day of the week on which you take your llrsUwbite tablet.
Page 240: azo61e00
236  BRIEF SUMMARY Patient Package Insert Cigarette smoking increases the risk of serious adverse effects on the heart and blood vessels from oral contraceptive use. This risk increases with age and with heavy smoking (15 or more cigarettes per day) and is quite marked in women over 35 years of age. Women who use oral con- traceptives should notsmoke. Oral contraceptives taken as directed are about 9996 effective in preventing pregnancy. (The mini-pill, however, is somewhat less effective.) Failure to take your pills increases the chance of pregnancy. Women who have or have had clotting disorders, cancer of the breast or sex organs, unexplained vagi- nal bleeding, a stroke, heart attack, chest pain (an- gina pectoris) or who suspect they may be pregnant should'not use oral contraceptives. Most side effects of the pill are not serious. The most common side effects are nausea, vomiting,, bleeding between menstrual periods„weight gain, and breast tenderness. However, proper use of oral: requires that they be taken under your doc- tor's continuous supervision, because they can be as- sociated with serious side effects which may lead to disability or death. Fortunately, these occur very in- frequently. The serious -side effects are: 1. Blood clots in, the legs, lungs, brain, heart or elsewhere, and hemorrhage into the brain due to bursting of a blood vessel. 2. Liver tumors, which, may rupture and cause se- vere bleeding. 3. Birth defects if the pill is taken during preg- nancy. 4. High blood pressure. 5. Gallbladder disease. The symptoms associated with these serious side effects are discussed in the detailed leaflet given you with your supply of pills. Notify your doctor if you notice any unusual physicalidisturbance while taking the pill. a r
Page 241: azo61e00
237 .. (continued fran etha side)'. The estrogen in oral contraceptives has been found to cause breast cancer and other cancers in certain animals. These findings suggest that oral contracep- tives may also cause cancer in humans. However, studies to date in women taking currently marketed oral contraceptives have not confirmed that oral con- traceptives cause cancer in humans. The detailed leaflet describes more completely the benefits and risks of oral contraceptives. It, also pro- vides information on other forms of contraception. Read it carefully. If you have any questions, consult your doctor. CAUTION: Oral cont'raceptives are of no value in the prevention or treatment of venereal disease. Directions For Use 21 Tablet MEMORETTE7® Tablet Dispenser Start the first cycle of medication on day 5 of your menstrual cycle (counting the first day of menstrual flow as "day 1"). Take one tablet daily at bedtime for 21 days. Start with tablet #1 and continue in numeri- cal sequence through tablet #21. Wait seven days- then begin a new cycle of 21 tablets, whether menstruation has stopped or not. The tablets should be continued in this manner-three weeks on tablets, one week off. Even if spotting or breakthrough bleed- ing should occur, continue according to this schedule. If this spotting or breakthrough bleeding should per- sist, notify your physician. Remember: you will always start a new course of medication with tablet #1 on the same day of the week, which is your regular starting day. Please write this day in the space provided. To remove a tablet; press the plastic bubble down with C ur thumb or finger through the bottom opening of the EMORETTE tablet'dispenser. At the end of a cycle of medicatiorr, put a new refill'in your dispenser. RefilNs for your dispenser are available at your phar- macy in accordance with your, physician's prescription. Your next package should be started in exactly one week (seven days). Be sure to renew your prescription in time so that you always start on the same day of the week. (See Other Side for Important Information.) SYNTEX (F.P.) INC. HUMACAO, PUERTO RICO 00661 19-101-15-1 217e 35-727 O~.- 79'~- 1fi~, . 1'
Page 242: azo61e00
238 Mr. FouNTnrN. Would you care to comment any further? I think you may have covered t'hat, However, I hope t'hatt you will take a look at it from the point of view we just discussed. Let' me proceed. The pill, of course, has its own separate risk of ca:rdiovascular adverse effects independent of thee many other risk factors. Is that not correct? Dr. KENNEDY. Yes. Mr. FoIINTniN. Mr. Mintz reports in his arficlle-I am not one who ordinarily quotes from newspapers because they sometimes get limited information and they have to write in a hurry, but they do the best they can-but Mr. Mintz, who has followed FDA operations for a long time, reports in his article on the press conference of Janu, ary 24!, 1978, and I quote again from the January 25 Washington'. Post :"Smoking aside, a reporter asked the 'Commissioner, what ad~- viee wouldhegive his tivife or daughter if either asked whether to take the pil2 ? Kennedy noted that he is not al physician, but said his advice would' be to find another method." Did you say that, Dr. Kennedy ? Dr. KEN NEDY. Yes. Not everybody is terribly glad that I did. [Laughter.] Dr. KEN N EDY. Including two daughters. The suggestion that I have responsibility or might have responsi- bility even for consultation on thatmatter caused some local annoy- ance in the Kennedyy household. What I had! in mind in response to Mr. Mintz' question, which I can assure you I did my best to evade [laughter], was some compli- cated issues in comparing the benefit:s and the. risks of various kinds of contraceptive technolagy. . I did make some other caveats that he did not report, such as "It depend's alot on the. kind' of life you choose to lead" and "on various sorts of personali preferences." I think it is such a complicated issue that it defies the kind' of easy generalization that escapes the newspaper story version of a response. So, I would appreciate it, Mr. Chairmany if you would be a little kinder to me than Mr. Mintz was and let me off the hook. I think that is a complicated matter that every young woman ouglitt!o take up with herself and with her physician. I think she does not need any advice from me. I am especially glad that _1Zr. DZintzmentioned in his article myy lack of medical qualifications, because I surely would not want any telephone inquiries on this subject. [Laughter. ] Mr. FoL:cTnrN. I am inclined to agree with your generallobservation. Let me ask you this. Why was not the same prominence or emphasis given in the OC labeling to the other serious risk factors contributing to cardiovascular disorders as given to smoking'? I think this has been covered to some extent, but I want to ask this for t'lie record to be sure. I am talking about high blbod pressure, diabetes, obesity, high blood cholesterol, or excessive low density lipoproteins in the blood. All of' these are really serious risk factors for circulatory disorders; are they not? Dr. KENNEDY. They are important risk factors, Mr: Chairman. It is not possible without saying much more to compare them in seriousness
Page 243: azo61e00
239 with smoking because they are all quantitative. It depends on howw much. A little bit of elevation in blood cholesteroli is not' nearly as serious as a pack a day. But, of course, an enormous level in blood cho- lesterol is more serious than five cigarettes a day. We have to supply so much quantitative specification to those risk factors that we can'tt say more than that they are all risk factors. However, there is something unique about cigarette smoking. That is I that it is uniquely a personal choice as opposed to a logical matter for physician advice. So, in our physician labeling we give much more~ evenhanded treatment to those different risk factors. In the patient labeling we do emphasize the one that seems to us to be not only very, very serious but also especially eligible for personal choice on the part of the patient. Mr. FOUNTAIN. You mentioned the number of cigarettes. How can you set the number at 5, 10, or 15? How do you arrive at that figure when every human body is different in so many ways? Some people have amuch stronger resistance to some things thani others: I have an interesting storv. We have heard a lot of noise about smoking, and the Secretary of Health, Edueation, and Welfare has made a: number of statements, some of which I thought were ill ad'- !' vised. Some I thought were too broad and not objective. II I got a call from an old gentlemani from one of my counties who was 92 years of age. He said, "Mr. Fountain, for wha.tever it maybe.worth to you7'=and,course, it wasn't worth much except that it was interest- ~ ~ ina-`1I just came back from Duke Hospital and I have been smoking cigarettes since I was 10 years old. I had' an operation and there were spots on my lungs but they were not cancerous, and I am in, good shape now. I am still smoking. But the man in the bed next to me had'to have one of his lungs out because he was fillpd with cancer and'he had never smoked a cigarette in hislife."I relate that only to indicate that, the nature of our bodies is sovaried. I t'hink, Dr. Kennedy, youlhave commented on that. In any event. there was no boxed warninrr renuired in the other risk factors which I have previously discnssedl The discussion of these risk factors is buried in t'heextensivebodvof' t.hedetailed natient labeling.. There must he millionsof women on tben;ll who have high blood nrees- sure, or diahetes, or excessive blood cl+olesteroll, or low density lipo- proteins. or who are obese. I know a few who are pretty close to me: fLau-hter.] Mr. FovxTaix. Z'Vhat is the rationale for not warning these women more bluntly to shnn the pill if they have any of these conditions? Dr. KENNEDY. Mr. Chairman, thev are mentioned, but we think tliat the more effectilre rrnite-and indeed there is considerable evi- p (7PneP of that-that physicians are persuading such womeni not to Lro C,,j the, pill. The nhvsicians are well aware of this problem of combining C!' risk factors, As a result, the best information we can ret suggests that CD it is ai nonproblem becausethephy.sicians are nersuadingwomen who, have those predisposing risk factors not to uo on the pill, ~ They do not seem to be as successful in persiadinLy t.hem not to ~ smoke. That,is a decision that peonle have to do for themselves: Their doctors are not very successful in doing that for them. 4 r
Page 244: azo61e00
240 That is why we think the primary route to that problem is through the patient whose individual choice it is. Mr. FoUxTaix. I suspect you would' be willing to admit that by being blunt and direct with regard to cigarette smoking a person may well overlook the other things. Is that right? Dr. KENNEDY. I would hope they would not, but iQt anyy event the evidence is that doctors are suffiicently aware of their problems so that they are he.lping women avoid it much~ better than they are helping women avoid the combination of smoking as a risk factor in pill use. Mr. Fouxzaix. FDA has known, or maybe I should! say they have suspected, for years that cigarette smoking may be a risk factor for OC use. It is curious that it was not until January 1978 that the decision was reachedto requirethe antismoking warning. What makes it doubly curious is that FDA's final order practically coincided!witlt HEW Secretary Califano's January 11, 1978,announce- ment before the Iti ational Interagency Council on Smoking and Health in Washington that HEW's antismoking program had been instituted. FDA could' have proposed the warning, t~-hen it published thee proposed order in December 1976, and again when it amended the order in May 1977. The same data was available then, but now you say there is some additional data to back that up: My question is this. Prior to January 11, 1978, had you, or anyone iit FDA, received any suggestions, inst'ructions,,or held! conversations,, or had! any other form of communication, from Secretary Califano, or any of his assistants, or any other HEW officials, concerning an antismoking warning for OC pills, or to encourage FDA to apply the Federal Food4 Drug, and Cosmetic Act to discourage or deter smoking ?' Dr. KENNEDY. The chronology of these events, Mr. Chairman, is that, as we have already testified, the comment from Dr. Jain and the study by Mann et all, had made FDA aware of the problem early in 1977. In October 1977, the Royal College study was published. In~ Novem~- ber, the OB-GYN Advisory Committee met and recommended label- ing to us and reviewed tltose studies. The Federal Register document announcing the final order and addin,gt'heOC smokingwarning was alreadyiir draft foim~ at the time that there was to be a formal announcement. It was made known to us that there would be a formal announcement of the Secretary's initiative. So, the answer to your question is that the: Secretary's initiative did not condition in any way knowm to me, the timing of that announcement. It is possible that we would have been a little slo.ver about publishing the final version which appeared on January 31'f t~han otherwise: I say that only because, quit'e frankly, it is not possible for anyliody, I suspect, to guess how much stron~,r secretarial interest in an initiative fine-tunes the speed of other agency actions any more than congres- sional attention to the errors of our ways make us more prompt. about correcting those. . ~ ~
Page 245: azo61e00
241 We try to be responsive to both~ I would suppose that we may have worked a little faster on the Federal Register order than we would have otherwise. It could not possibly have made a difference of more than a couple of weeks. Indeed, the first I recall of having put those two things together aG all was in a conversation in which we were told of the forthcoming announcement on the Secretary's part and' were asked in a purely routine way whether there was anything underway at FDA that ought to be made a part of the general announcement. Of course, we said, "Yes, there is. There are data of considerablee concern on the matter of the combined risk factors of smoking and pill use." That was included, I think, in some of the announcements. We briefed the Secretary on the matter. He accepted the briefing with interest. He asked us to survey other possible kinds of interactions ~ as we went along in the agency between smoking and other kinds of therapeutic drug use. We responded to him that there was interest in the agency in that matter because there is evidence developing that these cardiovascular risk factors combine in unanticipated ways and' we wanted to be very alert to those problems. I have assured the Secretary that in looking at drug reaction epi- demiology, cigarette smoking will be one of the variables to which we give our serious consideration. Mr. FotrxTAir. I appreciate your statement as tlo what would have happened had'you not had contact, but I gather from what you say that you did have discussions with! the Secretary or some of his sub- ordinates about the timing of your announcement. Is that right? Dr. KExxEnY. But well after the proposal was in draft form and being prepared for final publication. The most it could have done was to fine-tune the effort toward the very end. Mr. FouxTArx. Would you refer to your discussions as routine discussions ? Dr. KENNEDY. I am trying to recall exactly what kinds of discus- sions there were. Mr. FouxTnix. Did you talk to Secretary Califano? Dr. KENxFnY. I do not remember if I learned of the initiative from him or not. I think not' from him. I think it was probablv from one of his assistants in the executive secretariat or perhaps his executive assistant. As vou know, what verv often happens in an agency of that kind.. a routine check of the subunits is made to find out if there is any relevant information on thesnbject. It may have been that I discussed it with Secretary Califano in~ a P,eneral briefina meetin_v nreparing for this initiative. I do not' recall. If it, matters, we could look that up and supply it for the record. Mr. Fou-,Tnrv. I asked the qAiestion onlv because I think there has been a lot of unnecessary emotionalism in this whole area. Maybe there is an~ overreaction on the part of Deonle who are interested' in tobacco. There may be an overreaction bv those who have reached the point where they have become mad with themselves, as Secretary Califano 1
Page 246: azo61e00
1b 242 did, and just blasted tobacco and said he was going to go to work on it. So long as he does it within certain reasonable, objective, educa- tional bounds, I think he has a right and a duty to do that. Dr. KENNEDY. Mr. Chairman, I would only add, if I might, onee thing that I think I did not say on that account. I do want to assure you that neither our evaluation of the evidence nor the final form ofl the proposal was influenced in the slightest by the existence of that departmental initiative. Mr. FOUNTAIN. Mr. Goldhammer? Mr. GOLDHAMMER. While we are on the subject of the Secretary's mitiative, I have a copy of his speech that he made on January 11, 1978, at the National Interagency Council on~ Smoking and Health. At page 5 of that speech he says, "Women who take birth control pills, for example, particularly women age 30 and over, are up to 50 times more likely to have heart attacks if they smoke:"' Fifty times more likely to have heart attacks---I have not seen any figures as high as that. What ,you have here on these charts is 11.7 and 13.2, and the highest is 21.9, which is the most recent study. Do you have any knowledge of the basis for that W times figure that Mr. Califano used? Dr. KENNEDY. We will be happy to try to fin& it and supply it for the record. Mr. GOLDHAMMER. Would! you also indicate its accuracy when you find out? Dr. KENNEDY. We will be delighted. [The information requested follows:] We are{ unable to give the basis for the statement that "Women who take birth control pills for example, particularly women age 30 and over are up to 50 times more likely to have heart attacks if they smoke:"' This statement! is not in accord with the Jain;estimate for any age group. Mr. FOUNTAIN. Dr. Kennedy, I have a few more questions. Your Federal Register statement indicates that you relied substan- tially on the recommendation~ of the OB-GYN Advisory Committee for your antismoking warning requirement. I am a little curious as to why you relied on this advisory committee for advice on a statistical question when you have a Committee on Biometrics and Epidemiology. I have read the transcript of the committee's proceedings. Maybe it is because I am a layman, but it is confusing, at least t'o me. There seemed to be considerable misunderstanding among the committee members about the meaning of studies whichi have been mentioned so often here today. Even Dr. Jain, who authored the most frequently discussed studies, conceded that there were some problems with his interpretations and use of the statistics. I am wondering whether the question was referred to the proper committee. Epidemiology and statistics are highly technical subjects and some question has been raised as to whether practicing obstetri- cians and gynecologist's are the appropriate experts for advice in such matters as this. Since this was primarily a matter of epidemiology and statistics, was there any reason why you did not refer it to the Biometric and
Page 247: azo61e00
243 Epidemiology Advisory Committee composed of experts in those fields to advise you ? Dr. KENNEnr. Mr. Chairman, that is a very fair question~ and a rea- sonable one. Let me point out this. First of all, we do have epidemiologists and statisticians on that committee. Dr. Ory, who is with us, is one of them. He may want to comment. I hope he will. Mr. FOIINTAIIV'. Can you tell us who they are? Dr. KENNEDY. I do not have the entire committee roster, but Dr. Orv can fill us in. The general rationale is that the committee on the drug,product area has that combination of clinical knowledge and ability to evaluate ex- periments which permits it to do the appropriate job. As to why this was not also assigned to the other committee, I will have to rely on Dr. Crout for an answer to that because I do not know. I think it could not have been taken away from the one and given to the other. I do not think that would have'been~ appropriate because for some nurposes it is very important to have those clinicians as "ll as the other types of folks there. Would you like to add something, Dr. Crout? Dr. CROUT. We typically gave to the Biometric and Epidemiolody Advisory Committee complex and technical problems over particular studies: From our point of view, these studies were not that complex and the review of the data by our own statisticians was perfectly ade- quate. So, we went directly to the nresentation of the issues to the committee that we customarily put benefit/risk auestions before and labeling auestions before. That was the Obstetrics and Gynecology Advisory Committee. Mr. FOUNTAIN. In reading the transcript of the proceedings of the FDA committee which recommended the boxed warning, I noticed that Dr. Orv and Dr. :Sheldon Segal are listed as members. Dr. Ory has already been identified as the author of one of the studies cited by FDA, and Dr. Segal is a high officer of the Popula,tion Council for which Dr. Jain works. Dr. Segal himself endorsed Dr. Jain's position as exnressed in~his napers on the subiect. I am a little concerned about t.hP obiectivity of the committee's decisions. . What assurance can you give us; if any, that the committee?s recom- mendations were in no way biased by individual interests of at least two of its members? Dr. KENNEDY. Mr. Chairman, it is not unusual to have on one of our advisory committees a couple of members with related institutional afliliationG; The Pornlla.tion Council, as you well knmv, has been a distinguished national institution nursuing studies and policy studies, as well as scientific biolo,(zical studies on reproductive biology and on contraceptive technology. So, it is no surprise to find there a. counle of people on this committlee -hich desls with matters of reproductive biology who have affiliation wit'h the Population Cnuncil. To inform you further abont the composition of that committee. here is a niece of historv that T did not know. I have iust been supplied with it. I am~ told that the Biometric and Epidemiolopv Commit+ee was disestablished in June of 1977 and that, in fact, one of its members
Page 248: azo61e00
244 was given to the OB-GYN Advisory Committ'ee, so that, perhaps, explains the decision about reference. I would hope that' you would be willing to have me ask Dr. Ory, who was a member of that advisory committee, to respond on the question of the independence and objectivity. Mr. FouwTnix. Yes; please do. Dr. Oxr. I finished my training in internal medicine and then went to work for the Ccnter for Disease Control as an epidemiologist. I was trained, as most CDC epidemiologists are,,on the job. I went t'o the Harvard School of Public Health and got a, degree in epidemiology. For the full' 7 years that I have been with C'DC, I have worked in evaluating the health effects of contraceptives-all methods. I think if you looked at my curriculum vitae, you would see that I havee published papers that show harmful effects of the pill and they show beneficial effects of the pill. I can only speak for myself, but by the time the November 1977 advisory committee meeting was held, I felt very strongly about the interrelationship of oral contraceptives and smoking and the potential risk of myocardial infarction. In fact, I believe that I was the one that recommended the boxed warning. I think, butI am not certainy that the record will show that. I try and call the shots as I see them to the best of my knowledge. I thought that by that point, especially with~ the Beral article having just come out,, was a very strong piece of evidence that warranted strong wording. That was my recommendation. 11Zr. FourTniN. I asked the question to get your response for the record. I had no preconceived notions. However, inasmuch as they did have a policy position interest in the matter, some of us were questioning whether or not it was proper or advisable for these two individuals to participate in the discussion and in the committee's vote. According to the transcript of the advisory committee meeting, they expressed strong disagreement with Bert Litt, FDA's statisti- cian, when he appeared before the committee. Dr. KENh soY. Yes. 11ir. Litt was one. of the tigers t'hat we turned loose on that study to give it as hard a time as he possibly could. He is one of the authors of one of the two documents that your subcommit- teehave. So, coming in with~ that message to a member of that ad- visory committee who had contributed to the study that was. being criticized, it isnot surprising that t'here was a little disagreement. Scientists thrive on~that. Mr. FOUNTAIN. Dr. Goldberg? Dr. Goi.nBExc. Was this policy instituted since you took the helm of FDA? Was this to instruct your gladiators to go in there and play their role to the hilt? If so, how was this charge communicated to them ?' Was there a written instruction, or were they given pep talks? Mr. Goldhammer and I have for many years, upon occasion, ex- amiaied FDA files. I think ,your statisticians have always done a credible job of examining the evidence objectively and,, as Dr. Ory suggested', calling, the shots the way they saw them. In reading the analysis that was done by Litt et al. I did not come away with the impression that they had been any more ferocious than in previous
Page 249: azo61e00
245 reviews by your statisticians in evaluating weak data. I wish you would clarify that point for me. Dr. KENNEDY. I would be delighted to as long as the record will show that I do not accept the nextto last word in the statement. It is sort of like the beating your wife question. Dr. GoLDBER(i. It was not intended that way. Dr. KENNEDY. I want it noted that I disa.gree with thatcharacter- ization~ and that I think it is not supported by the evidence that is he.i ng brought out at this hearing. Dr. GoLDBERC. What was that? Dr. ICENNEDY. The statement that they are "weak" data. I think if you take together the data, and if you particularly takee the data on a combine& risk, I would not agree to the characterization "weak." I wanted simply before I answered the question to note that I took exception to that adjective. Dr. GOLDBER(3. There is a difference of opinion between uson the question of whether data that are initially weak in individual studies can be combined to produce a strong prodtict. Dr. KENNEDY. That, has not been done in this case. Dr. GOLDBERG. That question, of course, was asked! of some of the previous witnesses. You know what their response was. Certainly the analysis by FDA's statisticians does not suggest that the data are strong. Dr. KENNEDY: The analyses of the FDA statisticians, in my view, suggests thatthe data are inadequate in, some respects. We try to in- dicate those on the charts. I alluded, I think, very conscientiously to those weaknesses in my own testimony. The question is whether the dat'a are adequate or not adenuate to support the conclusion that' was drawn from them. I think they are. I think the statisticians concluded that t.hey were and that theY iden- tifie~l'~ in the second of the two memorandnms that you have those con- clus;ons t'ha.t thev felt survived that scrutiny. As to whet'her they have new marching orders in the, agency. they do not have,new formal marching orders. But I think T have made it clear in brrean st.A.ff mPPtinqsand elsewhere, where FDA scientists -aa.ther to inform us on public policv decisions, that I want the science that is used to imderaird regulatory decisions in FDA to be examined in the most critical possible wav: I think one way to secure that critical kind of examination is to uree people to make a dist'inct efFort to attack weaknesses in a study wherever thev find them. I think that• puts a special resnonsihility oni people. That is alwa.=s the wav a good scientist attacks his or her own work by saving, "What in this experiment is most.vulnerable to at- tack." I think that kind of verv aw~recsive A Dnroach to analysis is very healthv for the nualitv of the final product. I am deliRhterl that you find that quality of attack in previous work from our statisticians. Dr. rOLDRF,R(;. I had alwavs assumed that w,sR the standFlrtl for alll professionals in thea.Qencv: Mv auestion was directed to whether or not there is a new policy on the character of scientific. analysis based: ,
Page 250: azo61e00
246 on new instructions, or whether you simply expect your professionals to do a good job in the areas of their expertise. I think you indicated that you have encouraged them at staff meet- ings to do the kind of job that they are hired to do. At' least that iss my impression. Dr. ICExxEnY. Yes; encouragement and help in defining that job. Sometimes there is a bit of a tendency in any agency that is reaching regulatory decisions as a result of some science to be a little too out- come conscious or sort of asking the question "Well, what do people want out of this?" I think it is our responsibility to try to inculcate the attitude that we look at everything very critically and let'the chips fall where they may. The policymakers can ta.ke the product of that analysis and do the best with it. Dr. GOLDBERGI I asked the question largely to clarify in my own mind whether your earlier observation this morning was meant to sug- gest that one should not take their criticism too literally since there has been an element of overreaction in the statistical analysis. I hope that is not the case. Dr. KENNEDY. Certainly not. There is a tonal quality that indicates a strongly critical attitude. I was merely trying to account for that. Mr. Fou_N~Taix. Mr. G'oldhammer? Mr. GoLnxAMMER. One last question. Itis an amazing phenomenon that with all of the crusading efforts which 'have gone on over the years to att;empt to discourage cigarette smoking, people continue to smoke. As a matter of fact., women, especially teenage womeny show an increase in smoking; there may be a decrease in the percentage of women in other age groups. But, certainly the number of women and the percentage of women smoking, is very su'bste,ntial; despite a,ll the efforts made to discourage it. It points up the fact that they want to smoke regardless. Now, the Government, in recognition of that, hasstimulated research in the direction of finding a safer, or less hazardous cigaret'te. Presi- dent Carter made reference to that in his speeeh in North Carolina. Would it be advisable or useful for the warning to have an added admonition advising women who want to smoke to seek the lower nicot'ine and lower tar cigarette? I speak in particular of lower nico- tine inasmuch as nicotine is a vaso-constridtor and may have some role in precipitating heart attacks. I wonder if the warning might be ampli, fied'to increase the admonition to say'that if you have to smoke, switch to lbwer nicdtine as a means of possibly reducing the hazard? Dr. KENNEDY. I do not think that the data are nearly adequate to support that addition to a label. That would be getting on much thinner ice than even the most'harsh critic'of the la'beling warning that we have already put on has said! that we are on now. Mr. GoLUxaMrtER. How about nicotine? Dr. KENNEDY. We don'•t even know w'hether nicotine or some ot.her active principle is the exclnsive agent behind any par't.icula.r epidemio- logically defined hazard of cigarettes. We have not been able to refine thoso studies to the point where we can identify it'. Indeed, one of my many arguments with the whole safe cigarette concept is that safety is being defined on the basis of a few constituents whereas : t'he evidence about cigarette risk is for all constituents. (
Page 251: azo61e00
247 I•t requires, for me at least, a substantial l eap of faith to say that a particular cigaret'te issafe. I, in fact, agree with Dr. Gori when he says at the end of his discussion of a safe cigarette, that the only safe ciga- rette is an unlit cigarette. In fact, if we did know these things, I would want to see some epi- demiological studies at least this good that demonstrate the decreasee in risk associated with the decrease in any particular constituent of t;obacco before I thought the data were in shape to go in the warning. Mr. GOLDHAMMER. Thank you, Mr. Chairman. Mr. FOUNTAIN. One further question and observation. What is your opinion as to whether or not we may be spending far more effort and time than is essential in encouraging people not to smoke as compared with the efforts that we are making in the field of research to find a safer cigarette, assuming that smoking is harmful and may cause the problems attributed'to it. We must bear in mind that since time immemorial people have smoked and will continue to smoke. It will continue to be a legal commodity. Communist China grows more tobacco than we do. They do not export it; they smoke it. They have so many people. India pro- duces a little bit less than we do. I think we are at' about 14 billion tons; we are right behind China, I think. The tobacco companies are spending a lot of money on research, but maybe not enouzh. We are collecting $6 billion a,year on tobacco in local, State, and Federal taxes. The farmers are getting about $1 bil- lion of that, and we are spending the rest of that money for a lot of other things which, in my opinion, are sometimes good and sometimes useless. Secretarv Califano has already mentioned the waste he has discovered in HEW. What is your opinion as to whether or not we are spending ade- quate funds in the research area to find not just a safer cigarette, but other safer substitutes for commodities which are considered haz- ardous to one's health ? Dr. KENNEDY. That is a complicated and difficult question to an- swer. It is a question of allocation of scarce resources, as you well underctand, Mr. Chairman. I think we ought to be pursuing resParch, on the hazards associated with smoking on a varietv of fronts: Epidemiological fronts, estimat- ing synergisms of risk of the kind we have been talking about, investi- gating ways of better informing the public about these hazards as welli as inst informing the public per se: I don't know what the budget in public information now is. I do not find it large in nroportion to the hazsrd. Quite anart from the cardiovascular problems we. have been talking about this morning;, ci:rarette smoke is the largest single source and, in fact, larger than all other combined! sources of environmental chemical carcinogenesis, accordina, I think, to the very broad consensus in the cancer research communitv. I would like to be in a position to receive research results on epi- demiologv on a variety of things: I think we spend far too little na- tionally on epidemiclo-aical research,, and I would agree that research on alternatives ought to be pushed, too. However, as to the propor- tions: between different kinds of research expenditures and as to the
Page 252: azo61e00
248 proportions between all research expenditures and public information expenditures, it is a 1itt11e tough to guess where we are going to get the most good for the public health. I agree with you, by the way, that we cannot change the legal status of tobacco as a commodity. But, I do think that our understanding atpresent of the hazards associated with smoking are enough to justify the expenditure of very significant public funds in public education about those hazards. Then, if they want to make the choice, it is up to them. I really think we have to inform~ them of those risks. Mr. FOUNTAIN. I ask that question because Secretary Ciilifano,, as I recall,,asked for about $13 million to put on a campaign to discourage people from smoking. It seems to me that a portion of that money might have advisedly been put in the field of research as well as edu- cation. I assume he is talking about education. I do not know for sure. Dr. KENNEDY. Yes, I believe there is a research component. Mr. FOUNTAIN. But, in the process, Secretary Califano recom- mended-or someone recommended-the elimination of a number of very important research laborat'ories which were costing, $2 or $3' million. They had been~ engaged in important work for a long time. Some of the best work done in Government, I think, is done quietly. People do not know about it, particularly in the area of research. For example, there is an Oxford Research Laboratory in Oxford, N.C., where they have a number of highly qualified scientists who can take a leaf of tobacco and do almost anything you want done with it. They have now discovered that they can extract from tobacco, a tobacco leaf, pure protein. That is the only green leaf that they have used where they can get' pure protein. They are notonly engaged in research to determine a better quality tobacco, especially for export purposes in view of competition and in~ view of the income to the farmers, but they are spending an enormous amount of their time now in health research. It has a limited staff ; it has a limited sum of money. However, if you were to tell them what it is in tobacco that is harm- ful, they have, a means by which they could take it out. People from all over the world, particularly Japan, and Germany, and others who are interested in tobacco and the tobacco we export, go to that la.bora, tory as one of the first places they visit. I must confess it was there a long time before I knew about it. When I found out what kind of work they were doing, I was tremendously impressed. But, it took a long time just to get the equipment which theyneededy todb that'kind of work. They recommended' the abolition of that laboratory. Maybe they were going to transfer it somewhere else. But when you break up a group of scientists engaged in a very important piece of work, it seems to me you are doing a rat'lier dangerous thing. Thatwas about to hap- pen, except for the action of the Appropriations Committee. Dr. KEN:cFDY. Was that funded by the Department of Healthy Fdu- catlion. anc Welfare ? Mr. FouNTaIN. No. It was funded by the Department of Agriculture. It allihappened about t'hesame time the otlier things were taking place.
Page 253: azo61e00
249 I think the tobacco people felt that everything was coming down on them~ at once. Before we conclude this hearing, which has been designed primarily to determine the scientific basis for the labeling, you have required in this case, let me say this. We have not had you up here often, Dr. Kennedy. We used to have your predecessors up here quite often. But, notwithstanding the neces- sity for this inquiry. I want to commend you for what I think has been a creditable job as head of the Food' and Drug Administration. I have not had a lot of contact with yon; but asyou know, your agency is subject to surveillance by our subcommittee. Our staff has been exercising that surveillance in a variety of ways. They have been in contact with~you. It is wonderful that we have not had to call you up here as often- in fact, very little-as we used to have to call up some of your predeces- sors: So, I want to commend, ,you for the work you are doing. Dr. KExVEny: With~your your'indulgencChairman, let me respond by saying that obviously I appreciate the fairness and openness with which you have conducted this hearing. It makes me almost, but not quite, wish that you would; call on me more often. [Laughter.] We appreciate the attentions of Dr. Goldberg and DZr. Goidhammer, whom I had a chance to get acquainted with earlier in my tenure and with whom we have had a number of usefull conversations. We want to fulfil!l.our part of the bargain of oversight, and we wantt to keep you informed about.our activities. Any time that either of your very, very able staffers want something from us, we.1vi11 give it to you. Any time you want me down here, I will be here. However, I enjoy not having to do it all the time, I must confess, because it,give.s.me time for other things. Mr. FoUxTaiN. Thank you. I hope you will seriously consider that survey to determine whether or not the implication is there. Dr. KENNEDY. I will send you a report on our plan to look at patient labeling across the board. I think that might i nterest you. Mr. FOUNTAIN. Thank you. [The report referred to follows:] O W ~ CD O ta OD CD ` .
Page 254: azo61e00
250 Ps:.;er.t Package Inserts: The FDA Approach Patient Package Inserts (PPIs) represent an innovative although controversial provision of the Drug Reform Act of 1978 (S2755, HR1161i). For right-to-kncw and patient education reasons, PPIs have beer, promotefi for the past several years as a means of directly informing patients about the drugs prescribed for them. FDA endorses the PPI conceat and wiil move forward with a program to require PPIs for the majority of prescription drugs. While wa endorse the PPI concept, we also realize that there is much to learn about how to structure communications to patients. Therefore, the PPI requirament shall be implemented in a gradual manner while we research their impact and consulc with health professionals and consumers on aspects ot their design, production and evaluation. The following briefly z:cplains the steps FDA plans for developing a PPI program: Reauiring PPIs: FDA's regulatory approach to ?PI~s shall soon be published as a proposal in the FEDERAL REGISTER (FR). This proposal snalil address issues such as PPI contenns, priority selection criteria, method of distribution, e::ez-..rticns, etc. ,
Page 255: azo61e00
251 Currently, FDA staff is addressing,health professional groups around the country to make them aware of FDA's inte nt to publiish this document and the reasons why we believe PPIs are desirable. After publication of the FR proposal we shall return to the professional groups again to explain the proposali and solicit their comments. Drafting PPIs: There has been some controversy as to who shall' draft PPis. FDA proposes that the first group of PPIss be drafted'by a contractor. The award of the contract shall be based in part on anievaluation of prototype PPIs drafted by bidders for five drugs which have d_fferent risk/benefit ratios and usage patterns. A group of FDA an&nongovernmental health professionals and consumers wi~ll help define contract specifications and award and monitor the contract. An open conference will:be held December 111-12, 1978 to provide input to this group regardir.g the most desirable qualities for PPIs and how PPIs should be constructed. As part of the•PPI, drafting process we shalil utilize small groups of consumers to review draft PPIS and provide co.arents and suggestions prior to issui~ng final requirements.
Page 256: azo61e00
252 Extensive prior pu;lic review of individual PPI~s by Federal, Register ps:bli ca=ion and other means shall be sought for initial PPI's and for all PPIs written on a drug class basis (i.e., PPIs encomoassing several or more related drugs). Approximately 18 such class PPis shall be issued over the next 3 years. PPI content will be based on the physician labeling for the drug for which the PPI is being drafted, modified to fit the specific drugs at issue as, for example, where it is felt that a particular warninc, should be brought to the attention of the patient. Specifications or g,eneral rules for drafting PPIs shall be issued for initial guidance. FDA shall retain final responsibility and authority over PPI contents. Although initially PPI drafts will be written under contract, eventually, the drafts willibe written by pharmaceutical manufacturers or some other group with FDA retaining a review and approval function. After the first 50'-75 PPIs, consideration willibe given,to changing to a system stressing industry furnished PPI drafts. Phasing in PPIs: The first two-to-three years of PPI prodliction shall be viewed as an evaluation period'. A major er,phasis during this period shall be placed on researching the i^pact of both prototype and required PPIs. Currently, over 40 published studies suggest mainly positive potentials for
Page 257: azo61e00
253 61 PPIS. These studies shall serve as a;basis for core Uiorouc: investigations of the impact of PPIs. Later this year FDA will contract for a series of experiments to investigate how different variations in PPI, form, content, phraseology, etc. affect communication, attitudes and behavior. • Two drugs and a drug class will be studied: an antibiotic, a drug used to treat sleep disturbances, and estrogens. A part of this research will be devoted to examining how variations of the estroger. PPI compare to existing approved estrogen patient labeling. In order to gain a full understanding of the direct andi indirect effects of PPI1s, FDA shall place a major emphasis on investigating both the im.m,.ediate and long term effects of PPIs. In order to assure that this research is objective, properly targeted, thorough, methodologically rigorous and planned in the light of all the best scientific evidence available, FDA has contracted with the Institute of Medicine (IOM) of the National Academy of Sciences. Applying PPI Research: In addition to~provildiing state-of- the-art evaluations, PPI research shall provide information on how to draft better PPIs and more efficiently utilize V 35-797 0 -R9 - lT
Page 258: azo61e00
254 them. In order to tie the research to the dra'_ting,process, specifieations for structuring PPIs shall be periodicalLy updated based on the best scientific evidence available. The IDM evaluation shall serve as the basis for soecification revisions. Initial specifications are being,prepared by • Dr. Philip Ley, a consultant.to FDA, and the leading authority, on structuring communications to patients. Interfacing•with the Public: During the two-to-three year start-up period FDA shal~l actively seek to solicit the views, opinions and experiences of health professionals and consumers. An initial conference shall be held to discuss and to solicit advice on "how to construct the best possible PPI." The T-nstitute of Medicine shall also have an oper. meeting on PPIs. Along with existing channels of communication, speci~al.focused conferences shall be considered on topics such as PPI use in hos7ituls, exemptions to PPI distribution, storage and supply of PPIs by pharmacists and prescribers. In summary, PPIs are viewed as a long term comniitment to provide drug information to patients in a way that will educate the consumer and improve drug therapy. a .0 - % r
Page 259: azo61e00
255 Mr. FbIIrrrAix. The committee stands adjourned, subject to the call of the Chair. [Whereupon at 2:15 p.m., the subcommittee adjourned, to reconvene subject to the call of the Chair.]
Page 260: azo61e00
.APPENDIX . MEMORANDIIM F$OM COUNSEL OF TOBACCO INBTITQTE . Counsell for the Tobacco Institute submitted the following memo- randum for the record. (See p. -.) September 27, 1978 MEMORANDUM FDA Regulations Governing Oral Contraceptive Labeling On January 31, 1978, the Food and Drug Administration ("FDA") promulgated regulations substantially revising re- quirements established in 1970 for patient and physician labeling of oral contraceptive drug products. The labeling required by these final, regulations differs substantially from that proposed in the Federal Register in 1976. These substantial changes in the proposed rules were'based primarily on information made public subsequent to the close of the period in which the public was invited to submit comments. The procedure followed by FDA in issuing the regulations deprived the public of an opportunity to comment meaningfully on them, was clearly unreasonable, and, in the case of the regulations governing patient labeling, violated the rule- c making requirements of the Administrative Procedure 5 U.S.C. S 553. (257). Act.
Page 261: azo61e00
258 Fundamental*qifferences Between the Proposed and Final Regulation /s- On IDecember 7, 1976, FDA issued proposed regulations governing information to be provided to patients to whom oral contraceptives are dispensed. 41 Fed. Reg. 5360. The proposal was the first substantial revision of oral contra- ceptive labeling requirements since they were published in the Federal Register on June 11, 1970. 35 Fed. Reg. 9001. The 1976 proposal recommended that information about oral contraceptives be provided to the patient in two different ~ labeling pieces. It recoaaaended that one labeling piece be inside the patient's package and contain a brief summary of the most essential points about the drug and that a longer, more detailed pamphlet be either inside or accompanying the patient's package. Both of these patient package inserts were to list certain serious side effects of oral contraceptives (thrombophlebitis, pulmonary embolism, retinal !/ The extent of the differences between the proposed regula- tions governing the information to be contained in patient package inserts and those ultimately promulgated is demonstrated graphically in Appendix 1. The proposed and final texts recommended for patient labeling which, if substantially followed, will comply with FDA requirements, are set.forth in Appendix 2. These texts for physician labeling are set forth in Appendix 3. */ The regulations in force at the time of the 1976 proposal required that the patient receive only one pamphlet setting forth brief information about the drug and informing the patient that more detailed information was available from the physician. , , Ir
Page 262: azo61e00
259 artery thrombosis, stroke, benign hepatic adenomas, fetal abnormalities, and gall bladder disease). Aside from listing these side effects, the regulations proposed that the more detailed patient package insert include a comparison of the risks of death from various types of contraceptive methods. Interested persons were given 60 days to comment. At the same time, FDA published a notice revising the guideline texts for both patient labeling and physician labeling. 41 Fed. Reg. 53633. Although FDA allowed a 60-day comment period for the proposed physician labeling, unlike the regulations affecting patient information, the physician labeling revision became effective without delay on April 6, 1977. The final rule governing patient package inserts, published in the Federal Register January 31, 1978, and cur- rently in force, is fundamentally different from the 1976 proposal. 41 Fed. Reg. 4220 (to be codified in 21 C.F.R. S 310.501). Although the 1976 proposal did not mention smoking at all and did not require that any information be prominently displayed inside a box, the final 1978 regulations require that both the brief su*mary and the more detailed pamphlet to be given to the patient include:
Page 263: azo61e00
 260 "a statement in the form of a boxed warning that cigarette smoking increases the risks of serious side effects on the heart and blood vessels from oral con- traceptive use, and advising women who,~ use oral contraceptives not to smoke."- The 1978 final rule also required that the detailed pamphlet contain a discussion of the relationship between the occurrence of certain serious side effects resulting from oral contraception to age, smoking, and other conditions; and that the comparison of risk of death from various contra- ceptive methods describe the risk faced by smokers and non- smokers who use oral contraceptives. These requirements were not foreshadowed in the 1976 proposal. In the preamble to the final rule the Commissioner states that in making the revisions listed above he reviewed several documents. one of the documents was a comment submitted to FDA that presented a statistical analysis of data obtained from a British retrospective study of oral contraceptive users. The author of this comment had attempted to demonstrate that the risk of myocardial infarction in women , FDA has promulgated a regulation setting forth precise language which, although intended to be merely a guideline, if followed will enable any person to comply with the state- ment quoted above. 43 Fed. Reg. 4230. The text it has approved provides: "Cigarette smoking increases the risk of serious adverse effects on the heart and blood vessels from oral contraceptive use. This risk increases with age and with heavy smoking (15 or more cigarettes per day) and is quite marked in women over 35 years of age. Women who use oral contraceptives should not smoke." M
Page 264: azo61e00
261 who smoke and use oral contraceptives is greater than in those users who do not smoke. Additionally, the Commissioner reviewed several studies published subsequent to the close of the 60-day comment period which discuss the association of smoking and certain effects on the cardiovascular system among oral contraceptive users. FDA presented these documents to its Obstetrics and Gynecology Advisory Committee ("Advisory Cownittee') which met on November 17 and November 18, 1977. The Advisory Committee recammended that the patient package inserts reflect the variable risks for smokers and nonsmokers suffering cardiovascular side effects. It also recommended that the inserts contain a prominent boxed warning advising women who smoke not to use oral contraceptives. Subsequent to the close of the comment period (but prior to the meeting of the Advisory Committee and adoption of the final 1978 rules), on October 3, 1977, a petition was filed requesting that FDA reopen the comment period for 60 days because studies demonstrating the effect that smoking among oral contraceptive users may have on the cardiovascular system had been published subsequent to the closing of the formal comment period. Petition by C.I.G.A.R.I.S.C., Nos. 76N-0487, 75N-0304. The petition sought the reopening to "enable interested parties to assist the FDA in effecting the most accurate and informative warning." Petition at 1. FDA neither responded to this request nor reopened the comment period. 0 W C11 ~ ~ Cj 0 ~
Page 265: azo61e00
262 Promulgation of the 1978 Regulations Violated the Administrative Procedure Act The 1978 regulations governing information to be contained in patient package inserts for oral contraceptives were adopted under the authority granted to FDA under 21 USC 5 371(a) of the Federal, Food, Drug and Cosmetic Act "to promulgate regulations for the efficient enforcement of. . this Act. ..." Section 4 of the Administrative Procedure Act, 5 USC S 553, requires that for such rulemaking general notice of the proposal including either the "terms or substance of the proposed rule or a description of the subjects and issues involved" be published in the Federal Register and that "interested persons (be given] an opportunity to participate in the rulemaking through submission of written data, views, or arguments." This statutorily protected opportunity to comment on regulations before they become effective was denied •/ FDA takes the position that its regulation governing information to be contained in patient package inserts is a substantive one and is thus subject to formal rulemaking requirements of the Administrative Procedure Act. However, with regard to physician labeling, FDA has promulgated only a notice setting forth a recommended text. FDA contends that the notice setting forth this text is not a substantive rule, but is merely a guideline which, if followed, will comply with FDA requirements. If physician labeling is not substantially the same in content as that set forth in FDA's guideline text it will, however, regard the drug as mis- branded. FDA contends that because in taking such action it would not rely solely on the guideline text the notice is not a substantive rule. Accordingly, in promulgating physician labeling regulations, FDA's position would be that it need not follow the formal notice and comment requirement of Section 4 of the Administrative Procedure Act.
Page 266: azo61e00
263 G . by FDA's adoption of radically changed final regulations. The Administrative Procedure Act requires that the rule supply sufficient information about the "terms or substance" of the final regulation to permit all interested persons an opportunity to participate in the rulemaking proceeding. American Iron & Steel Inst. v. EPA, 568 R 2d 284 (3d Cir. 1977): Wagner Electric Corp. v. Vo~l , 466 F.2d 1013 (3d Cir. 1972). At the time FDA solicited comments the proposed regulations made no mention of a special association between smoking and the use of oral contraceptives as risk factors for heart and artery disease. In fact, when comments were solicited the principal studies used by FDA in making the changes in its proposal were not even published. Final FDA regulations that differed substantially from those proposed were recently declared to be improperly promulgated primarily because they were based on a statistical methodology made public subsequent to the publication of the proposal and the close of the comment period. Animal Health Institute v. FDA, No. 77-080 (D.D.C. Feb. 8, 1978). The fact that the studies rellied:on by FDA were published at the time FDA's Advisory Committee met in November, 1977, and that this meeting was open to the public, will not cure the vi'olation of the notice provision of the Administrative Procedure Act. The Advisory Committee has no
Page 267: azo61e00
264 statutory role in the rulemaking process, and the matters it considered were still required to be determined by the agency itself in accordance with the law. Federal Advisory Coffiittee Act, 5 U.S.C. App. S 2(b)(6)(supp. 1978). In any event, persons interested in the rulemaking proceeding would have had no way of knowing that the Advisory Committee was studying the issue of smoking,and oral contraceptives. The notice of the Advisory Committee meeting which FDA published in the Federal Register did not even mention that smoking would be considered -- it merely stated that the Advisory Committee's discussion would, among other topics, include "oral contraceptive labeling." 42 Fed. Reg. at 55647. Without notice that revealed the substance of the topics to be discussed, persons had no way of'knowing that they may have been "interested" in attending this meeting. It is true that one comment submitted to FDA during the formal comment period did discuss the alleged association of smoking and certain side effects on the cardiovascular system among oral contraceptive users. But the comment did not propose a special boxed warning, and only the most knowledgeable observers could have foreseen that FDA might develop the requirement contained in the final rule on the basis of that comment. Persons who were not ` ~
Page 268: azo61e00
265 s s so "knoWledgeable" were entitled to saae notice of the warning FDA was considering and an opportunity to comment on it. Wagner Electric Corporation v. Vol e, 466 P.2d at 1019. FDA's failure to provide adequate notice to interested persons was particularly unjustifiable in view of the relatively simple means available for the agency to permit public participation in developing the new warning. In addition to the notice and couanent procedure required under the Administrative Procedure Act, FDA's own regulations provide numerous procedures the agency can follow either before or after a proposed regulation is published in the Federal Register. 21 C.F.R. S 10.40. The agency can, for example, publish a revised proposal or a tentative revised final regulation in the Federal Register, or put tentative final regulations on display at the Office of the Hearing Clerk. At the least, FDA should have published a notice in the Federal Register stating that the agency was con- sidering the risks of smoking among oral contraceptive users and that it intended to review recent publications on the subject. At the same time, FDA could have placed those publications and the transcript of the Advisory Comaittee meeting in the public docket file of the rulemaking proceeding, so that interested persons could review and comment on them. 35-727 0 - 79 - 18
Page 269: azo61e00
266 These simple procedures would not have been burdensome. They would, however, have given notice to the public that the issue of smoking among oral contraceptive users was being reviewed. In aum, the final regulations governing both patient and physician labeling differ so substantially from the 1976 proposals that the public was unable to comment meaningfully on them. Surely FDA should have informed the public that it was considering the possible risk of smoking amonq oral contraceptive users and given the public a chance to submit views on such an important matter.
Page 270: azo61e00
267 APPENDIX I Requirements For Patient Labeling tia}or differences between the proposed and final ruLes have been marked, P FINAL REGULATIONS }710.~01 1•r.pxati,.nn L,rc,vnlrxepluon; labaunsK dirrrn d 1611m Pali.nt. la) C)r;tl [onfrocrptirra.. (1). The. Commicslonrrr o/h Food and Drugs t.ron- eludes, Qlat the safc and effcotlve useOf Oral Contrarehtlre dru0 products re. qulrAa thau patirnls be fulh' Informed of the benefits cnd r6k's hlrolrtd In lhetlse of lhese.dl'uss.. ]nformatlon inlay languase concerning efft•etlveness, . eontraindldatl0n• aarninr.s, preeau+tlons knd sdirrse'rraetlotts 1luall befurnlshrd to carh patient t'eeeivlns onl contracoplitcs. Thl:: Informationihall be¢Irento the paUrnt bythedixtx•neer in llin form ot a brief sum-mary of errtain essr rtLial Itdormatlon Included In each packaip'dlsuensedto eachDatient, and,ltlt n lonLCr, detalledlabelinR p/t•teLu or accompanying earlt, packaCe Ihspersed'to each. pa. tlent. . Patlent labclint; tor drug prod• ucts dlaprnsed iit . aeula•earee hospitals or lattlo-tcrm-carefacilitlcs will be con-sWrn•d to havr-tivrn ptutidrd In+2ccor• ttance, •.'ith.this seetion If provided tothe patlcnt before adlnintsAratton.of theflrsr onll fontract•ptlveMd evKy 30'days therratter, as long as lhe thor• alyy cuntitluei 121'The brief sunlmaryshallsOCeAl- y ea11y Include the fo4lowinr UT. Aslatemn~nt that oral contraeep- Uces are effeetive, but.that any failure totake Ihrm In ateord•1nre.with the rt'eommrnded dosace Increases tlle , ehnnlte of pret;nancy: ~<III A statrmenlof the speMne Itrms of history te bee told. llleYhysician that wotlW lead the pltysieian nut to prescribe oralltuntraceptivcs tl.e„ Utecont ralnAlctlona to use l; tltilA slatrment that oral edntraeep• tYct•s slloulJ bl•tak'en only und.-r tlle eonllnued supA•mt+hbn of a Ultyxielan.. Uv9 A IL+UnIr of theserious a6dc ef. feels or nral cunlracrL+Licrs,.t surhas lhromb)pAtrbitls.. puhnonary embo- Ilsm: nlyocarrliaL irlfaretton, retinal artery Ihnlnuixais, strok'e..bcnlcn he- paticadennlnax; inaYlrtion of 6•lalab- normaliuvc; an.l r:.llirtadel(rduraw•. ,e . I\1 _I..I:•llqarlll In ltll• 11.r111 ,.1 :1 bokcdw•nrninll that /:iaaretle timoklnL' iucrraa' sllieri..k'ss of serious siih• ef. (rtl.von thee hlarl alld bl.wJ 1-lsari: from OntL ContrneCptit"e Ilsl•• and tdtltl vIj0.•otaen LL'i10 uxr oral Contrarrp / il'C3 n[.1 to ~:nnkr. PROPOSED REGULATIONS C.1111.•'.I/11'n~ilarali1 f.,r ,ura..•1.. l i..u; .l.il.oli. „.If n rlnl 1w Ii,r lal ()rul rul4tru'ecyltirrf: - 11 ThC Conl- mu..tonerof Fiwd mld.DnlCa runrchitlnIhnt Ihe sllftundeticrlhc Weofor.il~ I.I1inlA:rbth't dllltr hrodllel ~t reftnlreY LIIat lcllk•nlxisc.ftdlvinfnrmutt of IIYm brne- tit, aud rLksinlolveJ!lu tlu u•c oblliu•s.• Jrtn:v. llltnnnation C.IIICCtnnik elllYaltr- lllh.s [o1ltr:tLtllllrutWll. ltanlllllta, tlrl[•11U. liolix. atul adrerse rrartlwlx ntrnl bt furninhrd to t:rclt.Imticnt r.roirlnK ornl runtraccUtiv- 'ntia, Infunnaliun dlall be Civen to 111o b:dicnt byvhe disp,etuor Itl thr.i furtn o( n brll•f sunlrnar.+of rertaln ,cnlial Ihfurtnalio- iudullyd Ine:iall IKtrkaCr dL Pcittn to Ca'.ali1p:nlrrll. antt. in a lunyer. I1rAailcdlldWt•hnq ult•rclu er .-Mi11/a11V.11111 LaA'tl 1':n'kaP:C (il~lk•11~r/1 1.> r.Irh4altkrlL. A_t '111er, brit•fFUnLnury sna11 ..prr li• rally III,hIJO lhl•ft•~14lv:ille: . A i l A slnienlvnl' I li:tt'. ur:..l Cutrlrarnl- l lvcr are C Ift•etltr. but'that any Jet latlun (rum.n•ronuncndrdd•e.aer uarrrners uir A'ltnlln'l' of Lret:1ta111'y. -ii1 1.tatcmcnl nf Illc'xverafir itenlx, o61Yr.tory.y tn In•. tnid'IGr• p1iTMomn9h::t r:InJJ IIA.:nI llis.l'hy'xn'i.ltl 11(.1 1. l/rrnl'ribe oml tulYlraCrntii~t% /i.1•, ihlP rnmr:.ill- /tKntlnnl' to Ir:rl.. tiill i A atuta•mont Ih:.t, 01.1 rontruct•n. Ih•t•n sh,xud bctaken mdy'under Ilterwl- tihutvl sttLeMixlon of :t nh)siCi:ul. _ lirl A Ii1~L1111t of'111e.A1•r1U1tY side llrt•etauf ortl culltrltt•ntltta stu•h nsthmlubox phl(•liilix; Pulmouary emb,nli»nn, rceh.cdi arirr)' tflrolnho~t5; stfoke, bCniK1111e4atiA" nllenunlar, htdurtion of.telal nbnurma.li- Uc+r onJ ¢allbladdcr nL •asa•. ~ *'No~comparable provision. 11
Page 271: azo61e00
268 FINAL REGULATIONS trtl .ls.]tvment of the most common Side ettc ts, such as nausea atld vomiting,,w•ciaht Clf,lnye. Chan'Y•e in menses. and breast tcndernt•u. (.rii) A statement that the estroCenIn oralcontracrpUves hos 6nt:n found to cause breast cancer and other can- errs in certain animals and that tlrese. findingesuaticst tUatoral0ontracap- Uves may also cause cancer in humatts but that studies to date. In women taking currcnllyy marketed oral contra- eeptives have not confirmed that oral eontraceptit'es cause eancerln humana. ( eiii) A statement that oral eontra- eeptices are of no valur Inlhr preven• tion or treatmencot:vrnereal dix•ase. t1riA statemcnt camngattentlon to the detallyd paU-nt labeling and a ree- ommrndation thatlt, beearefully.raad. U)The actaued patient taoelmgshall be a separate printed leSaet in- dependfnt oLany additional materials. It shallspecifically Include theJollow• Ihg: U).Name of the drug_ t1w Name and place ofbustntss of the manufacturer; packer• relabclcr or distri'butor.. tlll).A statement that oral eontracep- UYen areCftretii•e but'ean cause err-ta'.n serious sidr etfects. iivl A statNment that oral eontrarep- tlea should be taken only unda•r the eontinued suprrvisinn of a pitysician. (v) A statement of the effecuvrnexs; of oral rontraccptires. inciuding the differences in effeetlveness among dif- terent types and,the.relatiuiwhip be• tween effeativenesa aud estrogen dosage. tvil A summary of.tlx effectl6encas Of other nlethody of Contraception. .~, tnii)~A boxed warning stattng that. 'Cetgarette smoking ihcreases the.rlsk of. serious sidee e(trets on the heart and blood vessels tromoral contraceptive use andWvuing women.who use oral contrtlceptlt•es not tosmoke. AI- 2 PROPOSED REGULATIONS 1v) jL stattolenGot ute moln eonlmonside e6eets such as ttousea atld'vomitnrg. weight clutt¢e• . change hl ~~ mensrs; and breast.lendcrnea.:. tvh. A statement ttuta)though the estrogen In oral Contr•.tCepttves caUSeebreast cancer and other cancers in cer. hUn anttnals. It. hn't known whether or not oral contnt¢epttves can cntwe c•rnerr 61.humansl tvil) A statement thatbral contnu:epn tives are of no value in tJle prevcntion or trcotment of venereal dist~se. "vfill A atatement.ralliua't attention W the de4tiled R taent I:tUcliiq. atal a rec- onnncndatlon that.itlk• carcfulLyread. 1T)'I'he detailyd patiant'Inbeling anall be a sclu+rate prhnad Icailet Independett of any add illotuU m'aterfals- in a tyveface that is not roudcnsed and tto stnallcr Wlut 0-19ahlCtypo. ItsUatl..pecl6cnlly In- riude tite following: t a1 Name ot Uu drug. , ii) Name and pktce of butilnes, of the mamtQactnccr. packer. relabeler orr d)s- trlbutor: (til). A statetnent.t)tat oral eontraeep- tnes atc eRCetl4e but ean cause oertaln. .. serious slde effects. tivl. A statement that oral contruep- lives should be taken only underShe con- tinued'supervision of apityskian. Iv). A statement of the enectlveness of orat contracrpUvea,Inciudlna tile diRer- eemes Inn eRectlveness among different types aod the relattetuirr beta•een ef. rcetlveness and atrogtvt dn.atte_ tv))'.Asutmnary of thc.e::ecuirncss ot other metlwdsof conlnw~evtiotr. ~ No comparable provision.
Page 272: azo61e00
269 I/. • FINAL RECsULATIONS Q~'arnlha ret:arding the seri• am +.Ide of'eeta of oral contrarl•Dla'fa•Ineluillnt the rr4atlve risk (where known) facid by users compared to nontlsen and the relationshio of the ~. a ts o ake. sn~o mg, an tv i tu at4! !TIlL"[! men tmn s n tnc udc, thrombophtebutls.. Dul• monary embo4ism, retlnal artery lhrombosis. stroke tthe..relatlon of thesetuestrogen.dose aAall be men• tioncdl; myoaardial Infarctlon, banh.m hepaUe adrnomas, lnduetion,of fetal abnonnalllh•s. anal gallbladder disease. The followtnr shall be mcntioned: la) EstroG~rt•r liave been showntocause ranccr in animals; whlch show'tngytu• titlrs the Infeeence tltat estrogens maynusecancer in humans; (b) therc Is strong evidrnce Ciat' rstrogenuseIn- veases the risk of endometrial cancer In po.ctmt'napausal women: tc)'therei) some evidence tl+ett sequential oral coiilraceptivc (which are no longerr marketcd'may lnercase lhe risk.of en- dometrlalcancerin women:(d) atudics of an association between oral contra- ccptices and'brcast cancl•r arelargrly net^atiVe excrpt for aauggestion of In- creasedrisk (one study) in women with benirn brcast disease, and thrre Is no evtdrnce of an inennsed risk of, uterine cancer in users af oral contra- teptices other than sequentlals. (lx) A smti•mtfit of common stae et•f7ets. IneBudiug nausea andvomiting, wcightchange, darf.ening of the skin, changes in menses,.and a statement ofutherserioussideeffects, incluaing worsened miCraine, and; worsrnedheart or kidney disease due.to fluid re- tcnaion, growth of, uterine fibroid tumon, . depre ,llon, Jaundite. delayed return to fertlilty„bEood pressure ele- vatlon, decreased `IucoSee tolerance and elevated blood llpids,.. 1x1 A statement ofTeportedsldeef• feeta not definitely related to oral con• traceDllfea9C. (xil. A statement cautlonintt the Da- tlenb to cansult.her physiclan beforeresuminathe use of the drug after childbitth, especially if she Intends to brrsatfa•vd the ba.by, polhting out that thehortnonrs wl.he dru` are known , to aupear inn the milk and may dc- rrease the flUw: AI-3' g PROPOSED REGULATIONS Ivii1 A warninR rectp[rdh+g tlteserlonu aldteeIIeota oforal contraeeptlvea, luolud- ing the rclaUve risK twhere knoacn. faced by users compared to nunusers aud; the relatiotuliip of the aiile e(fects to age andlor ottierr eondltion.. The side eMetts meuth;ncd must ineludelhrombuphl!- bitls puhmonary embolism. retlnat artery Utrontbosls, stroke. 1 Ute rckttlon of these to estrogen dose.iSto betnentioned), be- nignhepatic adrnomas,. induetlon of fetat abnommlltita, and galJblddderats- ease.. Thee ability pfestrq,.en to eatne' malignant tumon In anmmts, endomet-. . rial cancer mwomen, and the evidence. thaosequentlal omi.contr,,cepticra may Increuae t1k risk of endometrl:d eanrcr lu wmnuf nntat be mcunened. ; Thrre sttisU atYO be a stat(Mient that studlrs.of :nt -mPkltibutxtweCn oral:.eotltrJrcp- tivcv and'bn-ut cattcer nrc 13rtw'ly neaa, tive exaept'fer a suC:esUOn of hurr,cd risek 11 atue'.yl. in women with bet+ignhroast dLrr.aee• and that there 1. uo evi- dc"uc of an inereased risk at nter,uer:+nccr ut users oforal eontrstcequicce utlter than acVuentiatt. IvIU) A statentetu of.coanmon ilde ef- fecta,, btclJtdittg nausrti and vomttinu. weight ch:u+CC, durkcnlnCf of the oktn, ch:u+uea In meuses„and a.statement of other serious side eQects, lucludmg wursened'miaralne, and wotsenedlieun, ur khiney disease dua too fluid rZtention, growtll of uterinellbhout tummn4 depres- awn: Jkundice: blood Dressure ekvation,, decreased giucosa tolerance attd elevated blood llpids. lix) A statement of reported sale et- (eeta nOtdetttutely related to Oral con- - lmceptive use. tx). A statement cauUonlns the Rttlent to cwuulther pttysic/an before resumhq the use of the drug after ehlld- bhih.: avpcciulir if LO tnteade to breent- rerd the baby. poinlmL out Utat t4e hormones In the dtvz.azc koowo to ap- pear in the milk and maY decrCasa the fiow. 11
Page 273: azo61e00
1q 270 FINAL REGIILATIONS (xll) A comparison ot the risk of deattifrom various contraceptive methodr(ora i cnntranrntivrs in smok- crs. oral eontraecoUec! m nonsm LTS. con u1Rt or aiaphralfm, con om or diaohn¢mwith abortion 1nthe event ofpreCnancy, no contraception but.abortlun in the event of preanan- cy, and no contracrption or.abortion):(xiti.)Astalcmvutr of the specific Itmns afhlstoryto be toldd the physi- elan which wouid+lcad lhe Physician not to prescribe oral conlraeeptives (f.e., the contralndicatlotu to use): (xiv) ~A statemen6 of specifie Items of history that might cause the pl(yxi.ciawt to suggest another method risk factors fur myocardial Infaretlon. family htstury of breast cancer or past history of fibrocyetic disease or abnor- . mal mammo¢ram, aallbladder disease) orr would: re¢uirethr Dhysitian's spe- cial attention le.a.- mitralne, uthma, epilepsy, heart or kldhey disease, , fi'- brotds, history of-0epresslon): . tkv> A statement that Jaundlae, de- prcaslon, breast lumps, and the par- tieular warnine altnals of thromboem- bolicc discase,. thronebotic dlsease, and ruptur.•d: hepatlcadenoms, should be reporeed to theptiyslciane txvU A statementot how.to take aral eontraerptices DroDefly' and w•ha6~ to do In the erent off oeue or, two mtved' periods., - (sciU A statementeautlonlna the pa-tient that thttdrut has been pre- scribed for tJle. partie(tlar Individual only and that the drug must not bcrlvcn to others. (xv1il) The date, Identlffed as such, of the most reccntrevision of the W bellnif protnhlentlypEaced, lmmedlate- lyafter the lost sectlon of such labd- ' Ine. AD - 4 PROPOSED REGULATIONS -i1 ~ A rom~lwrison of the risk of death fram various contraceptive Inethoda ,oml contraceptive. iVD., condom or diaphrbCm.. cottdom or diaphraCm with abortion in the event of pre¢nancy. no contraception but'abortion in the eveM: of pretmaney- and'no tontrac[ptiam Or abortionl . 'cil f: A statement of the splcnlc items vf Idstory to be told the Vt$+r(cinn svlticll would Ir.ld the physictan not to pre.eribc. oral contraceptives (i.e- the tootmbuii- eatimrs to usel. I xiiilA sthtement of speeifle.items of histoy that mi.-htcauRe the physiciattn to suagestt anotltermethod (e.C., rlsk fnctors fnr myocardiai ~u>tarctton, family history of broast.cancer or pnst history of Ilbrocystic disease or abnormal aram- moumm. Rallbladderdiuese) or would repmre the physlcfati>ispecial~attention (e.g.. mlgt•aine, asthma, eppetny. heart or kidneyy disease. f4brolds- hlstory of de- preslioo.'. ixlvi. A statement that Jaundice. de- pression, breast lumps, and the particular warning slprtals oftitromboembolic disease, thromboue d4casse, uW rupturedd hepaticadenoma, should be reported to thephysiafane txv/ A statement of how to take oml tontraeeptivesproperly and wlutt too do in the evcnt of oueor two mtsaed iurlods, . (xvil 'Phe date, Idrntlfied'as such. of tMmoettecent restlion of the labeling prominently placed lmmediatety after the )ost sectlon of such labeline.
Page 274: azo61e00
271 t FINAL REGULATIONS (e) ~For those oral contraeeptive dnlg products w,tthapproved new drug ap- pheaLions for indications In additlOn,40 contraceptlen, both the briefsummary and detailL•d patient labeling maY idcntlfyy these ather indiatlotls. If Ihe other indicntluns are identified. the la- belinCmust specifically. Includee a stalcmentthat'the information in,thr patient labeling relative tocontraindio eations, the dan.gen.of oral contratep- tirea.and the safe use of the drvQ.sre also applicable urhen ihesc druas are used for these other Indleatidns. tb)l The dltailed patlentlabcling shall be printed In accordance with thelollnuing sPeciflcations:. . (a)The:mittimum letter size (lower- case letter -'o" or Its equlvalent),stiall be not less than n'. Inch In heitht. (11).'PhebodY copy shals1.eontaln t• point leading andnoncondensed type. and shall not contain any light face type or small capital letters. (B), Pattcnt labeling for eachoral contraceptive drug product shall be provided to the retailer by the tnanu- facturcr. pack'er. rellbeler, or distribu- tor as follows (l)The brief summarY patient label- Ing shall be Included in each package. Intended to be dispensed too the pa- t1enL (111. The detatledpattentlabeling shalli be Includedlnor shall ueompa-' nyeaeh package intendedto be dis- pensed to the patRnL (111) In the eaae of oralcontraeepttve dnie products In bulk packages itttend- ed for multiple Aispetning, a: suffieienFnumber of patlent labelingp)eces shall be Included In or ahall accompany each bulk package to :ssure that both pieres can bee furnished with each package dispensed to everypattenL Each bulk Package shall be labeled with Insttuctlotu to the dilpenaer to. Incdudc both patlent labvling Pleces t:he brlefsummary to be in the park- agc and rhe detliled labraing pircee either in or aecotnpxnYing thee paek-lite) with each package dispensed to the PatienLShls.section dues not prp+eloae themanufaelurcre or latxlhr fram distributing additlonal pallent la• beling pieces to the dispenser. AI - 5 c, PROPOSED REGULATIONS r4U PatientlatiellnCtoreaenom)con,traceutive drug produet shall be provtded to the retailer by the manufsoturcr, Packer. relabeler. or dlstnbutnr ss fol- lows: tit The brief summary patieot labei- in4 snoll be iscluded'In tveh pctckage m- tended to be dispensed to the Pntient. (tU: The dctaUcd paticnt9abeling shell be included In or shall accompanr each patkage intended to be dlspenxM to tile patient: ' liil) In tlte case of.oral contraceptit-e druc products In bulk parkayes iuterded foc mulliple d)spetLSUte, each bulk pack- neeahale include a sulbcicnt nutnber a6patienL lubcling picces to pssure lhat bolh pte[esean be funiishod with each packapc disPensed to evcryy patieut- Ench'h bSdk.pnrkaae shnll be labeled -11t in- structlom to. thc dispetuer to tncludc both pucientt labeluiq, pieces (the brief summarytobe In the p.'tpka4e aod'the dttnued Iutiplingpicce.eitherin or ne- eompauTinLr Lltc Package) withl oaetl Inckaqe dispensed to the i>atient:. t.
Page 275: azo61e00
272 FINp+I, REGULATIONS (T).An oftt contra0ept(ve.dN[ Prod•' ttet.thatIs ttot labeled as reCUired byparacraph (a) of th(n sertion rnd'that 1s ellher introduced or delivered for in- Lroduct)an Into Intentate eommerce, or held'for sale after shipment In in•lentale commerce la mitbrandedd under section 502 ef the act. HoweYn, an oral contraceptive dmcproduat ptieka;ed before the effectivedate of ttLLs-paraaraph is not misbranded if ade0uale numbersof-copies of the de- ta0ed patlent'labe4uta reralUred by thisparasraph are furnished to wltalesal=era or retallers to permit any retall' purchasrr aflrr Lhu cffecttve date to obtaln aurh labeling withthe Prodttct. Then.quircmcnt'lhat any Oral contta- cepttre drug prtduct be dispetued: with detailed paticnt l,tbclina, as ap- plied to.pliy:icians, shall not be eff¢e• tlve far suppllLs In thelt possrmion on the effective date, but shall apply onlyy to supplies recelved1hereaftrr. (t) The Faod and Drug Adinlnlat7a• tion )tav avallablepntlcnt LzbrllnRfor oral contraceptLYe drug products that Includes htfornwtion.rexpondre to all the Items specified In Paraaraph (a). (2) and (3) of ttl'Ls s:ctlolt. The labeling hea been publishcd In the Frorlul, Raetsrta' and upd.dtd versions wllll continue to be published us guides whenchan0es occur. Any person may rely on tllenCa•est'version of this la- bellna as complying with para(fraph (a).(2) and.(3) of this aectlnn after the effective dataof this paragraph. ' -(9) Holders of newdtuQ applications for oral contraceptlve drut produets ~ that are subJect to paragraph (i) of . this section must submit supplements 'und.:r f D U.9(d) of this chapter to pro• ride for the labeling reCulred by'Para,gmph (a) (2) and (3)bf tliis.seotlon on or before April 3, 1978. TThe labeling may be put iutouse without advance approval by the Food and Drug Ad. mihlstration. EFFEETIYEDAT!E This resulatinn shall be effoctive April 3..1979. tSres. S0S.SOS. 701caA 51 Slet_ 1050-1003 u anuttded. LOSS ta U.S.C. 351.2/0. 77t(allA Dated: January 19, 1979 . DaNALD KLYNIDY. Conunlisswnrrol lood and Drnya AI- 6 PROPOSED REGULATIONS (S) All oral t:nbtraceptive drug Prod- w:L.tlutisnot tnbeled aereYulred by punantplt (a) of this seedpn and that u eltlier introrluced or dNirored for )ntro- ductlon htto Interstate cotamerte, or held fer sateafter slflpme-rlt lu intoratate commerce Um/sbranded utMer section 302of the act. However, anoml tron- traceutlve druR product p-ka(:e beforetheeQective date of thfi uoruCtuptt ii not.misbrondcdt if adeeuate nundoers of copies of tile detniled.patlentlObetintt re- eulred'by Ufis paraaraph ace.funt4hed to whelesalers or retailers to permit any retail putrhoser after theelfectivednta to obtaut such labeling with the product. (9) The Food and Drug Adminbtrr- Lion has svnltehle patlent labelfita for orel cantmreptive drug products that ln- ehldea informatlon responalve to aLl tlte Items s[w~ciPled in paragraph (a)(2t and t3t of this section. The labellna has been. - publl.hed In uta FrawtLRaararsa and updated versions wlll corltlaue to be pub- Ils[ted sa rmdei when cliaoCes ot.car. Apr person maytels an the newest versloa of this labdintf as complyhtC wlth para- praph (a1 ' (2) and (3) of tltias anctian after the edectlve date of thls paraarauh.. Nore* Per a pertW of Oe aays aner t4h- rturr.7..taa~.nr patnmt mar ssaa ra(r.on. Y romplylne.atn patnr,raph (st (21',and (al nf Lala +M.4tlon. tM 1aee11na In.tlte nou- tnWer DnekctNC. 15N-03y1 puanshN Inir FR 33603. t)lcembar 7, leTa. f7iHnklera of.new.dnt0 nnpllratlotu, for oral contraceplivednta producta Uuat at'e sublect to Paragraph fai of this see- Uan murt' submitstinplemenld andrr 1334.8 of thtt chapter to provide fir the tabellhitf tvqn(rcA brpnra`raph Ia). (21 and't3/ of Ulls secUon on or before Felr- natry 7. 1977. Thelatielln` may be put intC use without adva)tre aPproval by the Fuod a lul Drua A<Iminlftratbn. Interested pentota. tnay. on or before Februar77, I977„ntbmit to the Hearht'0 Clerk. Faod and Drug Admintxtmtioat, Rm. 1'-65,.5600 Fi{hers Lane. Ractrllle. MD 20857. writtencomment (preferably In anintuplicate nnd'ddentit7ed with the Hearing Clerk docket number found in brarketsln the heading of thias docu- meul) . rernrdinathis proposal.7:eceired comments..tnay: be seun (n Ute above On1ee between thehoutae uf~9 a.m- and 4 p_m Manday: through Friday. 1 , Tlie Pbod and.DntQ Admhllstraliottn has determinedUiat thia documetltdoes not.aontain a malor proposal reQUinnc. proparat(on of an 1nllaUon Impact aLste- ment-under Executive Order 11a~1.ntW OMB Cireulnr A-]07, A topy of tile in- AnnOn impact ssshament tr onflle willt . Utle Hearlua Clerk. Food and'Drua Ad- mthRlmllon. Da led : December 3,1976. SRxewtN Oucen-.. Acline0ommtrstoltiTOf FonR n..A n...~. t
Page 276: azo61e00
"t? 273 to. 0 G, t APPENDIX 2 Proposed Text for Patient Labelillq are natuea, vomiting. bleedlng between menstrual periods. weight gem, and breast teudentess. How•ever-prolxr u." of oral contraceptives requires that they be Laken under your doctor!s contlnuons supervision• bealuse they can beassoci- atedw7th serious side elfocts which may be fatal. tbrtutately. these oceurr very Infrequently. ?hesariotu side ertects ara: 1. Blood clots In the leps. lunps, braiu, heart Orother organs. 2. Liver tumon•. which may.' rupture and rau..e sevcre bicadinu. 7. Blrth defects I( Utee pill Is taken while you are pregnant. 4. Hlah blood pressure. 3. Gatlbladderdlsease. Notify your doctor If you notice any tmu.ual physical disturbance while tak- Itta llir plsa. AlllYOuah the estrogen :n oral coutra- cepUves causes brcasteancer and other cancers In certain anlmals. It Is not known whether or not oral contracep- tlves cun cause eattcer hl humans. At this time thore is no definite evidence that Uley' do. 'n1e leaflet givenyou with your supply of pills dtscusscs hl,mueh more detail the benefits atld risks of oral contraceptives. It alao provides mformattonon other forms of contraception. Itoad It care- fully. If you haae ally questtons. consult yottrr doctor. Caution. Oral contraceptives are off no value In the prevenuon or treatment of venereal dLsease. DatstttsPatcKt Lescuna \Vl6at tov atlOQto KCOW AaovZ osAtCOn7eACtt•Tn"ts Oral contraceptices P"the ptU") are the most effective way tescept for sterill - sallon, to prevent pregnancy. They are also convenient and. for mostt women. free of serlotu or unplea.ant.side effects. Oral~ eontracepth•es must always be taken under thacontlntwus supervision of a physiclan. It is Important that any woman whocolulders using an oral eontraceptivetmderstand the rtsks.involved: Although the oralcoutrnceptlves havr.unportant~ advantetes over other metltods.of cow-traceplbn. they have certain risks th.it. no otUCr method haa. Onlyy you caudeclde whether the advanta6es, are worth these risks. This leaflet w41 te11 you~. .abotuthe most Important risks. It wlllexpliln liow you can help your doctor prescrihe the pill as safelyy as possible by tellhta him about yourselfand:baioY alert for tlwe earliest signs of trouble.. And tt will'lell you Ilow to LLie the pill properly, so Ihat it will be is effective ass ppsslble. Thore Is more dethllod Informa- Uon avnilableIn the leaflet propared (or doeton.Your pharmacist can,ahow you a.copy: you may need your doctor's help In understandlna parts of It. WHO SnOVto Not [ls[ OsdL COets.ezrrvrta A. If you have any of the followths conditions you should not use theplll: 1. Clots Itl the lep or lunQa. 2. Angina pectorls. 7.. Known or suspected cancer of Ure breast or sex or;ans.
Page 277: azo61e00
274 - ;:1610~ NO7ICF5 /. Onmstulrngndl'bicedi0^l th'al silli Prove Itot go hare sottle of thn.ert-. aged:0M1o:44 it'.ISCstinutt'dtltal:about I liulyetbeellttitgneet<I: nltssnk.eflNeCuoflltCeSlfoeetlFto156ain.III 2:000'tl'Sttlg oraliCotltraCeptl've5 tviBl 5. Rnon'norSUStKCtedpregitancyr mC Vill rsCl.belom', bUt it if1vOtyCt' IKhoapitalixedtathytMnbecauseofab- D. Ifyotr h'are had an5'of the tot- tertain.that the mtnl-pi41 does in,.fact'nommal clotting. Among nontuen in the ]oa-in{ conditlons youslsouldnoouse the have.te•scr serldsts srde elfecls. Tlu, d9s- same: zCe. ,batno f in ^_0,000 -otrldbe pdL; Cussionbelom•, minle based nnOinlTl otr hosprtaliacdrachyear.Forwomenunder 1. Heartattackorstroke..k intdmtation about tlio combination pd1s. 35 ,ho use ot'al coptraceptices tite risk' 3~- C1oLs in the IeCa or hinCS.. shontld be censidered to apply as u'el1 too of death dlu t0 IunC clotss or stroke is C. Although it Li your dreisien. It LS t!le mmi-pol. about I in'M000 eatltyear; mt101re:nan- retemmetlded that il TOU a[e Ove[ 11 2. OthCrWags to PrevHit PregnanCy'- uners0f tmlf agC, Lhe nsk Of death -.,auld,vean old you do not use ttte p111 becnuse As. Lhiileaflet shll lerPlain. Osal Contre- be I in 500.000. each year. Fev womelu of an Inereased risk of heart attacks rrputcesharesevernlscrimta.ri.k!s.,Othrr, 35 nnd.Orer. users w'oultl Imve n 1-ih- froln the Pdl., rtteltiosls of . contraception hat•r le- :5.On0 risk'. n4 dea.th ench yearr compnurd DtYtpnlpTo vfe OeY. CODtRaeerTn'L1, r15k5 or noneaL 111.1 They arCalbOle3s a'Itill t-ILS 200',000.fpT L1or1W5ara.eITective than Onl . Contraetptwes• but. Even aithnut the Dill the risk of ha-, IitS Ityou do nQt haa•e.any ofthe condi- nsed''pnotrerly.maybe eReclive enough a luart attack imurases.tn6h age.andlfatsl4tedaboveandarclttink'InCabout. for.ntan3•. a'omen. The [Ollowing table Italso.mereased', bysuch heart.attack -vainC orallcoutc.ceptii~es, lo hrlp J'ou, . gtrea rrport.ed PrC.uatucyl ntce,~lhe risk.factotrsas hlgh,bleod pressuirr. highdecide, 70U iteed lUlormGLiEnabOut lite numtb'er af.w'omien.OmL Of I00a'Itp uroUld ch0lesterol• obesrtyl d4abetes.,and nya, adrantazes and risks of oral Cantracep•become pregnant In I ylael for these . rette smokmC.:Oral contraceptiees.tur- tivesand ofot2ier.contrneeVSisemethOdsf metltods: tlier incrensethe rLk af heart attavk 3 aa well...Tltis leanet de,ermRS lhe ad-perrv.eCrrta Pea.lo0 Werew Prw 1-w to 3 lime5, It is eallhvatCd that u5rrs ofvamtages and' risks 0f ~ oral [pntracCp-oral . tontratRptivcs: age 40 lu 44'halc tlvls. EacepL fer the2UD attd abortiom. P°rer*tabout a14in-1:000chauce each year of xhl0h hav! thClr Otcn . ettlunive rI5k5: e0eetlre Itaring a hesrt ntteek: romVaied u'Itlv a hururten>.e.dt.'IVe uupl• ]-]. 'tttat _ 1-In-5.000 chan<e m nonusers. Uscrs a LHe onlTy riska ol Othar metl od of ro os ' CR k. traeeption are tl]o], dllet0 ptegL nCJ ~ ~- ~, I, 5-t] ---- aT,9i M.ta 39alld Hn e 1 im•10:000 n, ah0r11d tlle metltod fall ar ot be psed p o~. ~I IL st.aan'hlle nottt4SCrs in tl aC.grOt p. t4d tonselentlmtsly. YOUr doctor catv answer. e0.tue I ttermptoe hrtthdra.vut. Itare:ab0lut a 1-h,• 7 000 isk.: TIt~•e afC qWestiOlns ypUma}•IYat'y n'ILh resPCCt t0S-I4..................... .......- ai~?S arel'J4e figufGs. If yol4 hatenontOf the uther methodf of contnceptton. He.cenfthrlpma 6-1'J---------------hCtlftat[iCk rLSk: (aeWra nICI1LiOt1Cd aLta anSWtr any qllestialta y0n may ha\'!Poaau eud lelliea Nwte. ]a-13---.--.. Sl:ea aberr..yPll wnl Ilal'e a Sma11Cr'r15k Ihan a[t!r rndin{ tHls leafkt on ora! cOmtra- Yoeonrnnpuao.60-e0L ia, llsted. If JoU haYeuveral riskk facters. CtpUS•H' The figtn'es /excepl'for the IUD- ran• you a•Illl have aaFea4cr.risk Wan areulCr. 1. WHat.Oral Contrncepttves Areand e.ialrly heesuupcoplodtRer in haw n~e0i Beenme of'L1tie inareased rifkk ot'heart How They Work: Oral Contl•acrlrlires Ute.'.use eard method. VerytaitlxluL users atmtks. Ont contrnttpttvef are noM1 ree- women at'er. 40, Tliey' arc of two Llyts• T7rc mostt comnton, of tltle: condem. truhbe, eOLttta inter- ommrntledtoce ottM~simpdy'calltd"'thepWl"Is.aentm- ruplusa~.ithdrxaalt. orrhyd.tm.maysltoWd.ne.erbensedaea.nyaCeb}tsnm• bilnatlopofanestrbecnand a progestha acltievte lower pre lhe ta~o kinds of . femafQ hproton<s. T7te CnanaY ratn th ts en n-lioha ehnQ a st. ek a lv~at t attack amaunt.of esirogen and progestih can tltmseclvenabOCe.whioharetheasen e an. atgIta peCtari. u'hohare. haAd but the amountef estrogen is ntoat rnUllis for large groups of n~omen..Con- blood <IuLS in the IcCS or IuniCS or tlse- nry: nt becautE both tile ef4ettlt'Eness . ecienllOtu ItSe Of'f the diaphragm alOng ~'hefe Imparta . nd aonne of the datuges of arnl cou- "'ih urenm or icilyy is.reportcd to be b-POrrrratian.of.fn:nors. Wfiencertm a w ..hormmnes tneeptivrs a relnted.to.the antountof ~g~' eRective. Ectelrt for tive IUD: e!ICC- anemalsare given fe'nale se^t re <stra5c:n. This klind'of oral coiuencep- tn'e useof these methoda requ'ires some- ronttihu ouslyfar Imngg peflod.., tauaer, lift ttork5 pYliteipalYyDCprePML1136 r!- uhat n.ore e!fbrt tha2 ply tak'~ng am'Lyll!oV'.~tllhebrCaSt.tel'It-ti. 'ngiila. Ica•eo(aUCCHItam~Lheo-ary:.~VhenthCsilugle oiLl e1t[Smon .4. S~-t.it.1s an 'srd 1ir. amount'otesatogcnis.somlerogt'antsor elTonttlat'matlyc0U11e3 ndertatie53u, No.p fenittn.at.Vlescntthatonl muee,n and the pill is taken as directed:cessllrlly'Yotsrdocturcanlellyouagreat', eomtr cept- es.cnu c .cer n ntnnaiu deal mprt about the5e.methodsOf 0 bUl'.-t' t vs possibletile}.niile be d- Otal cantratCpt 'es are m0[e tltaal 99^., effectire , ie:, tlese wou4d be Icss.than traCenlien:. Con r d'n ihe fut e to d s0 Sc-C['li'One peegnaney.il 100'rcontenO used the ' 3.Tltetha,ger'sofOr+lContrneell-. stitc/i~ h refouutdno ease b c~...t pill fOr l Tear,. Pllls. titatcolutain "0 a. AOr.O. m!'bldod Ct tt'na. I31ood I 4 ncc r ' l ilth ' 4l, st d> rc- tOJS mlerograms ot estrOCen 'ary.slBghtly ' ul vcrtota blood vessets ef Ihe bodyt te C Sted ar 1'co traeel l es Ig tLn,effectSrenRU: ranCl'ngfrOm:9g"'a t0tlte most:tenntvon of Ihe seetolu side al mcreaSC in, .b'reast cnneer in n'onrcn more than 99^.etTerli'vC, (Manufeeturer elCtclss of Ornl Cuntrafeptit'es. A clot'Can u~ho alreads Hare beNgn breast dtsea.•e . Plal iluertpregnanay rate far I1L5 prOd- D Intt, a henfiALUlrktlLflhle clut'is iuta 'e\VOmen~n~itha atronra Invrtly lu•to": of. nttIOUnA la elinital Lr111s., if~ PI'OduCO. IsacombimatbN. blood s'esstl of the heart>. or a: plllma- breeSAcancerorwhoHacebreastno0uln. lhe second tJyc o4 oral contra¢eptire. Ita1T, embnitn , a clot wluclh formsin thephrOcJatiOdisca.ue or abnonnal mmnlno- Often called the. "miini-pill'•', contnins . legs er pcNts. th-breaksplf and traveis, grann.s on trno were exna.setl tb the esrra- ondf a progestln. It'tcOtlc's in part bypre- to tht.ltmgs'. Auy af these Cann be fala0. Cen. slilbeslrol.' dmnng tueir molher; t!OSetS' eentingreltn]e0S an egg trmn.tlsee otary Clots al OcC r a ell n the b!l0otl vM- pacg/ tlt .K'must be f0110xed t-elyF buR also by keepi,n4 apetm ftom reaeh, srts of tt e e e tesulttngg in blindi,ess ar bly tl e doctors tf they ch0ose to uvc :1ngIhe rfg.andg bv naaktmgthe tlteP,la mpatrmt•nt:Of vaiot tn.tllat.e)'cTltete ort-.tl t0lrtraeepYfes instead of anm'hrtt. rwomht~lesa reCeplise ltlo arvy fertilized IS..e+idevte 11 at1llte~rl5kof clottitng itl- nsettMofOfrlracenliont alan SUUdles. . egg that reaeh<s it ThC - llvll le, cteaMS. ' li hClter estrogcn dosts.Il 'c hare l e,n tIV t-C t Liltg ' I.. • efLlCtli"C than the <Omhnl.l~llotn oral CO 3'. ll Crefo n:a'1 t to gCCll:ll e tip'e af t-'trCi tI I teIC .L04 Crlt e hC- ttartptire bout'971. ea'rce. . 11t rl t'o.Ct a lo a ito , ible ~a lon:: .LI c g 1 U n Vd.. c se illa ,tl. e. .1 o I e facturer m y. ert preC-, c t' . mII 0. IPaerlytne ased I . t atrrl t•t. oL d ol I t I I I .. Rr r'tl._ hi51 preduct fatmd ilu el-vlnll t'tals if hl I.rpnantv rate d d.n'A tatt•e •nnt 4. vdC ae h 5 6!d a tt prodLctiS aproae]titl•Otlly Otnl COntra• tn afreptablt chatljCs in tilemenstlval petu otte cCmtltanettt ot aral cotttlarCll- ct'nuvnr.In nddni0n. Lhe niroeesttn-onll•. Is'.nllern. urrs•, whengGven !or perlocL•s ofmore 1,111 hn• aa tcndency to ' 11 The k, of abu I. '101(b g" l.rn c yeav ldt -- ftcr tHC. UlClUl biecdingnhlch may tienuite IncOtn'evti tleaacr uh a_e:m HOth'h rs ond I101. I3 KC C11 e-k fa . te nr t;" oe tC t f bd d g. en[: I uverso( 0 1 c lracep[nes. but tl -n- utcru ntil. Thn . 1so c' TheP[OpC [ lypll is UEldAe.nut<i Ce'ta5[d r1.N" frOn1 a3e COntiairpLL C I3', del!CC tl L.a ki[~d'Of 0"ll C(•Ilt t nll~e IOn'ei eRettiteness.in the.tiope tl,at t pearatobepnesentatalnages.Farnromen w,hicl , no lortper markcted tle.e tcCrraa aeGIS[rt. YOLit,.NO. ]16-tt'Irsp.r: ofGfre6tt r. le.s
Page 278: azo61e00
r 275 V Ir NOTICES nucRliOtoPalcontraeeVtit'e,rnaY:IncrCase not: knohn• Some a'emen dlvelop high the rt.k of calYCCr of'Ute./'.tlrlLl.: There blood prCSfufe whtCe takulQ Oral comtra- nnlaims no crlulrnr.c. hmrc.rcr.,tl.lat.thc rrl+tirrs. which ordhlarily'y rctunls totlle ral tun4racrplivca. 1 nilahl0 llt- lniglln,.l Irrcl.S wltrn Lllc oral rotLlmCCV- cft:ISC'Ille ri6k of 1t145 canGer. •irr ts , bLOpIKd.. OI'aI rentriCCptive3 de GSUKC. yc1'yOther reaclio113..altllouglt nUtprot'ed Inrely..ahrAiCn,nOn-m:dlenantr tllnlUr tohc ranved byy ornt fOntracCfrtlvrt., ar0' nf'tLIC llt'er. TheSC tunlorY No: LLalspread- ln:N!y' rr)Mrt[d. Thrsf ItncludL'e bnt:tltcy.may rtlVturc ansl caLsriYltcru:d , rltcn,u[nt ,1 talinn nnd s m0 dls;- btcedifl7:trhschmal'.bc(n1aI:A'fewcascs: rornfnrt.Mhcn uriumting. ktdtreYdisease•: of raneer.of the jiver,hat o liecn re:wt LL'd tlirr.iuess, .wnle.losso( scnils. m amlen v<ing or.llleonunerwtlee+- tu¢ hnlr.ou a•ei n body halr„an in- ic is lo't yct.knownwhrLhcr Lhr Uru.rrc:l~:cnlrr11JC1CS>etn1rn',dntc, a!p+etitc calurd t!rem:. rhmla.: entornr.ts. and n uecd', tor a C Uanorrstandc'rc!Gpil+pclh/di/.nrnE' r'h:.n~,t'iniculal','t Icrw. V'd"csrr111lurn nr CnntrGCrP(InCS ara'1rrCd in. prt'o nnnf n„ Ilt:lblllly' Lp.IL t Ca11E1':IL't Iclf.[rS: Onl conlratcpti'resI ahm,td nect be lltkcn i Alterr you: aop ItsulGoral conlr:+rep- by-. Vlcenanl'wonlen.bct'ause thcsnlayti.ue.lhercmay.bca.delaytscforcyuntarc damage~lhe dOrelopint' clilida TIreTy cal- alae lo In'conre pFe^nanG. As di.eussMl. an IilfCeaSed riX6' efhealit def¢[ls:andl VrC3iou]ly,yaoFhouldY'aitalea'1110t10ha lilnb defects lu tlie child. llt aedilion. lhe' atlrr. xlnVVul; tlle Lull he(ore ynu tryto flevelopingb lrnl0lee ctlild u'hose o1o411cr Isrcrvlnenlegn:nt. DurinIhetiO. (eu' h:uattciecd.DEB edicth/stilbestroll: an~l nsnlrthc, use a,0ther.foum of contracep- [.Strogrll..tlnl'il1Q prefnaney t1aS avWk of lianl.,You rhould collnllilt your It1113ici1111', FettirrQc'lllcFrof tlletaciomorCrl•vix In: hefore 1'l:umfna - e/, ora1 c(+11IrnCrp- nrr tttnsor youngndullliood. This ri,k. 11re..aftcr~, clelldblrth. eslleclally" it yau leestunatedto he not moce than 4' il I Vlan to uutac Your b'nby: Dmlys ltt oral 1.00(/: IC is polTlble that other estrogelts•, cOntfacrpllCea.arek11oN'n lo aptsenr IIY sucll as tile e9trogen3 in oral'colntrRCeV-ttrc]..couldl havr the faEnO elreettn thC f8nsale chfld if the rnolher takes thenl., dureng Vregnmlty. I( you stoptaking Ornl contraceVtices:o to become pregnant. use another InMllod', . or:ColL[ranfeption for up to J'motlahs. T.Cc reaaon; (or this Is th1tthere Is erldenee from; StLLdtes hl w.omenwho IlavE had; "ml<elrringls' 'rlUeolL after steppltlg tlleplll, that the lost (etusa aree more likely to be alsnorlnal. Whether there u an~ overall . in<reass 1. . "mi•carrlaea" hl ~ a-nmena'ho become pregoant soonatter stolApiulg tllepill as comparcd wttlr women a'hoo donotuse thepidl is not't known, but it' is posaiblae that there ma5' be. d.; Catibtaddrr d"vrrlae_ tvomen+vtso h1ves used oral CorltreeeVtl,ve3 fer 2 Years. or longer have twice as great a rt k as nonusers of harfng gallbladder., dliesse rceuiring surgery. e. Othcr sfde e0et•OS o/ oral eontraerp- Someu0metlusingOml eontra- ceptlves experlonG! mlVle:uantafde. ettt[tl.thatare not dmlg:c(olu and:arenot.likely'to damege their Gealttf.. Somcof tntse may'. he temVornry. Your bl'easls . may feel tender,, nause:e. nnd; . 1'Otnl[Incm:nr occur, 7- may galll or.loser weight, and your ankles may swell. A: spoctyy darkeuilug of t1Yelktn. VartIQtClarly.Of the. (acc,. u possible.and'may Iserxut. You ma.y uotkee unexpeetedva.ginal'bllee<fiuqd or cha;ng;rs.ln your mcn.Lntal periotl:. Nore seriutu eide elfects incluUC rrors- eninC of luigcnfne: auilcusy', and'.kldncy or heartdisease beesttse ofntelidencyf0r •xlulct tobcretulnrd in Ulc botlyIChct1' oral eotlcflleltAtiVrs are used. O[her sideelfeets. are growthoL preexistln'gfllbroiU tllmol's OL eheutetus: men[al dtp0esllon:'. and Itver.problems with )awndlce tlel+ IAwlne of the.sk'inl', Your doetor may fLld thac Ih'el9 of sugsr and fatty itlh-SLnncrsln youd blood are.elerated:.tlteIAllg-ICrln CRCCte etithhe3e eh;nlgtS.aCe thc m'alk'„ and tltle Imr4-rnnge effect on 1Et- fnnLSis not k'llo.n'nat tlu!5 ttrvu. Purther- mGre: ornl COULracCVh,rrS n1Ly ta¢~r G decrease 110 you[ lmlk sulfhly.a3wcllau, ulUleellmlily ol thc'nnllk. 4. Cdurpaneunl oh tl.tcitl,~k's.oQOrolQOIltrneCULn-alld Other Contn.vrylnve. Urlho:l- 'lTw ,na r.a/nrlirn t~ra Ihe n,k. :nlll ,'lILY1it'c[lc.::''ut I nv1 rpuel':,rt'}~tlva nnd otdlcr mcthods ul rontrncclnlon h.nr~t been.an:ldyxed LOrntunnte the 19sk.ut denthuuetatcd trdh tarlulu. mctliod..' nf rontrarrpUOn.. •rh,. r 1- la.l. Itarts:. - tI1Qru'k oE'lnCnlertlwUli4srlf. IcY-: the ri.ik lllat'-ul contru:ccVtlrr< win r:ulfc death due to abnennalclol- Iliy;,', ;,tld -b'1 tlie ri:k.of death ddle toI%1'r::nane~ or hbbrtlOn.IN t1SCe eCCnL~[hf nuthod.fads. Thc rtsuuaa u( tliis.analpsi• Irm rhotvllIAt tltC bar'crapli,6['low,:'fhe hrichtcf, tlre bars is the 1 nlibor. of deaths 1>Mr 100.000. e0rnen each yPar. Tlticrc are s,:sctso( bars• each set refer- ; lo a s!+rclficagc group of',wanrelu. t1'Itltlll!r:nal :~t ob'har5. tllere i5 n,SlnklC barfur cncls,of -, dlReren[ conlraceu- ttve nuthods. t"Trartlt,an:l (ontr:YCn- nont' L1Gnt1S hca:pRr]a;lll'e9 rotldolnl Inc n rol ot lurirt.,ana no connore<r Fgure 1. Amnual numb.r o/ desrRS.a .sd +• n 100~.000 nonstenl. wamen. bV'reyimenDOrrantral ,.d.a9e.cl .ro..1an Annea/ de.rns 75 N' 27 22 71i 20 19'. le 1]' f s6 IS 1,. 13 i 1]' 11 10 f e 1 ~I s I • ] 2 1 a ~ , t- - 1o:ZS ~ l9.}9 An. 1 :1 , ]Sae , o-Jl Reolmam oA control 0 N'a metaed El luosanty ~'Ab-icnanly[M Orsls.only a rr,a:ronol Toainon.al ~ connra<ev9ion ionly ~ c . <eMion anAreborii- rECEesI 1fGlSrEe, V04 al„MO• fI4-i1/tSGAr; UECEMxte r, Ier4 aJ-i'.. t l
Page 279: azo61e00
276 .~~ . 7711s ¢na7ysts Is based on present knowkdga and new Information cowd.. of eottreq.alter 14.TLLO anaLsfd ahowe that the rilk af death from all mathodsat bhth eontroliLt Idw.compored!.to the rlaks of chdldblrth, except /or orateon- trnrrptfrH fa romen.orrn {B..It'shoxsthae Ula lowcxt rtsk'. 01 death le t.orintcd wllh Ule.eonddme ordlaolnrag!n (tradl- ttunaf eontroecptidn)'baekcd uy b,r ear!y, abortion in oase of laiinrc of the eo!adom ord'laphrngm to prerene pregnancy. ALsu. at any age.the ruk of'.deatlildue to unexpeeted ' pre¢tlatuyl trom use nf teaditional COp.ttiIceptidn even withoUt' a backup af abortiott is generallrY the same as,.or less thnn, that front use of ornl eoncrleLptives.. HMP SO Us. ORrL Cp1aRaLLCTlVLS Aa Srr'LLT eno Errccnvm.st As Pos.cucc.. Oncg Yov H.vcDectnra TO USC 2TlLV. 1_ What'tn Tell your Doctor. Yatean make use of the pill as afrly. as Iwasible. by telling your doctor it y'ou have nny.of the following: a.Condhlons thlnt: mean you should hOt IlSe 0lal rOnLrlGrp4lt'e5: Clcts in the Iegs or Wng... CtoLt in the legs On IY+ngs in Llle past. A strpke• heart attack. or angilla pectorts:. Knoarnn or suspected calrcer of the breaat or sex or¢ans- Unu.susl vaginal ~bleedmg lhat hasnot let been diagnosed.. Known or suspected:.pre¢nanq-. b. Conditions that Your doctor will wanE to watch closely or u:hich-ight eause himtoullggest.anothermethod ot canoraceptiotl: A famil)".hisloq-of brcast callcer. Breact' noddles. 8b'racsstlc dLease af the breast, or inabnormal tllammogram. Diab<tes. High blood r'YCSpre.. High chuleslecul.l Cigas'ettesmokmig. M1tigraine hmdaclies. - Heart or kidney dlselae. Eplleper. ffienEal depression. . Fibrold tumonof the u0erus. Osllbladder disease. Np1tC':5 c..Oncr 7ou are usbYf,r:or.ll conafarrp- Lng OralConlraCeptire<:'At tlSat, time tives, yUu should be Mrrt.for Signt of'an alld about once a year thereafter. I1Q wdl sertmu adverse effect and!call your d!ic- generslly <snmine your blood pressur-, tor If they Oreur:: breaStS. abdOmeLL and peiviC Urganl I ln- Sllarp paut in the chest,.ceu¢hing rludinga.Patxslncolaousmearr, bloocL or sudden shortness of breath Sunsesar rindiaating ptltoible clbls in'.tlle.hulgsl.. Palnin the ralf Iuns.s!ble clotJn the Oralrn!otncentirrsaretilemostcRec- leg', llt'e melhOd.. CSrept StrtlllrSltcll: fdr Cr115Ning' eNhE hadll Or Ilrnl'illt:5 ' in. PrCrClttUl¢ Pregnallc)-. Othtrr metliod3.dirnlim4Possiblehrartattack/. n'henmRCd cotlseicntious!y,.are alto tTry'. Sudden severe.lieadache orr romilinie: effective andliave fcn-er riskz Tite.sert- dIA.L11e.6or.faLRthlg.; AISttVrbnncC of, ri- 0113 risk.a of'oral ContrIlfCiXivef.ars nin- Slan orSneectl, or t!'eSkness Or n1lnlbnCe41 fomnlmL and thC "plli':' is a rel9con\'en- in an nrm or Ieg. IMdic2tin¢ a posrih'IP Lent met'hod Of prerentifl¢ prCt;n]nfy. atn.ket. If )rott Ilave certain cotlditions or hace Suddcn Pnrt1..1 or.comvicle losa o[ ri- IIaA these eomlitPOrvs . mthe Imst: sou sion tuldicathlg a possible clot in tile ahoNld nu4 use nnl contnceplil-es be- eye! . ratn ethe risk Ls too.grene: ,7llc:c rondl- B.reasl Idnutu lyouu sh'ould; nskycur limns are listed InULebooklet. If )ou: dodortOr t0 show you hoA" lo exatncne' your not have these rondlRlons, and decide too oAn brea.stal. !Irc Ula pleaae read' tha booklet Severepaln in the abdomen~./indient- carefWly so that you ean tlse the "pill' im¢ npacsib!e rupturedtumor oU the Inostsafetyandrlfecevcly. IYvnrl. Oral cotYtraceptivts are not' reCmn- Seseradepression. YLllov'.tngnf Uieskin I]aundl<r'I mcndrd4Or Iomen Gvcr 40 )'ears of ^ge: :: How to.Take the Pill So Tliat :t Es 1nlCrestcd persons.ma.y. om.or befora AlostElfectire. Febnnarr 7..1977-bmit'totlu Hea.ringtafanuf:icturer to ulw>Iy.ul(drulallon C1erk.,FOOd and. Dru¢Adlhnnistnnau.~ on dmage and: atlmfnLSeratieu and what Rm. 4 of tRo~ Pm3. 1~ -68:.5600. Fishers fnne. Reckville: Ioda.if Palient has (orsotten co take one rtD :085 i. u'rltten eommenY . Ipralrr- At llmes there m'ybcr mm aleluEruai ablP m quintbplitate and LtlentiEied n-inh period after aa cycle Of~ pi11s. Tl1rrlfore. lhe Hearing Clerk docket Ilumber (ound IQ you mics olle menslrualperiod but in bra'wkets in the heading ofthis docu- Itlace taken the P.ills ezact(9 as'yeu rrrc nlentl ~ rrgardip¢ tha ehysician andd suppaordlod continue aa ttxual u4LL! thL pat!en1' la6plul¢•: Rteeiredcpmmrnls nast cycle. ]f )•ou haP! IIoCtak'en [1.10 Inay V! xell in the.aboYe Gmre bctnreH Pi1L correctlyt ypn !nao be preaAant and Shmlld stoPft takhfg GrUI contnlcepth'esthe haulsOf'9.a.lit and 4 p.m..,Sfollday, u'utili)-ourdoctor determines wlletller or through Frida7:, I3oEyou are Pregllant. Until )'ouirarl get. ISrS )a1: sOS. s] Stac.. IDSO-Itsa, a. to )ou'r dortor, use another (orm of con:- wca ~xl lfs.C. as_.:assl! . sna n etpLrp01(In. 11 tA'0COIIrtYUIIYC Mlllarn!al l f ~ It iy to the Commttr~o~.iera~t pcrt0ds.are nlumd. you sltould'~stop d]k-ame or r~'It'g•e in¢.pld[s'mcil ic'Is ddternliucd':r.:lleeher ra'pruy, I:1 ernstl Ireroe,raatiar)oa,areprr¢nant.lf7•onda.becolnenrrg• ptum~ub.a.ln tne rceu..LRrusrr..Mi~nnr nant w~hil[U.ingoral.rontrar<ptirrs.3'011, Is:in-e1~1!tR^-4..]1):should disclsss the risks. Lo the derelop- Daled: Decennber:3. 1976. 1na citild with Your ddctor. Sxrasvtv O.RO<rs:. 3. Periodic Esamutation ActfeVCarunlsslnncr.ul Your doctor.will tak'e x co!npiete medi- Food and Dn/0i. Cat'and(amilyllktOry.be(Onf.pnC.H'rib- IrRLk!C.ifaf0i0Fillalra-:6:1!01a!nil fEOEgAI iE6tSiLL, VOl 41, NO.Lr1-NISpAr; pECEa1LER 7. IerL A2-a ~ '
Page 280: azo61e00
277 APPENDIX 2 Final Text for Patient Labeling 1 t HKtcr SvuwAKr PAncur PAcK.toc I+sntr ~ Ch;arette.amoklpa IGcrraw•s thc.r+xof re: nous adver, r e/frets nn the heart and btnp.l re;tseis from onl eoptracrpnce u.e. ililss ri•k Inarreeee sittt eee anda•IlhIlaty smoklne t1i.rn more elkarett.•s pertlAyl nndu au4te marked In eumen' orer 71 yrus.of tue. Women 'ho tlle orTlirorltr]e^ptlCei should notsmoke. Oral contraceptivn taken as dlrect- ot are about'9!N'a eftccti:•e In prerent- Int prcttn.vtcy. (The minl-t•ill. hoa•er. er...Is somewhat 1csss effretlVr..1. ForKrt- tlng to take yourpills Incrraaos the ctt3ncc of pregnancy. Woiecn who have or have had tlot- LIIIC dl3ordcrs• cancer of thc hrrv.t nr tex or'dans. unexplained va-lnalblr:•d- In4l i strokc. hrort attark. a•l; tna, rc•torts, or w•ho sus7cct tL•ry tnrT he prca7nant ihouldnot use oral contra-eepLiees. Atost sldeeffects of the.pltl nrc r.ot sertous.•The most.commor.stJr effectsarenausea• vomitin¢.. bf•edlaK be• t'uecn menstrual perilll:, a•cteht lain. andbreast tenderness. Hure:•er. properr use of orat eontrarepth•c.~re- Quftee thattluy bee taken under ;•our do.tor's en::tin::ous xlprr:'islr::. ti:-.- causc they can t:•a;::cci:•tcd w:;It:crl- ouaside effecla xhlrh inayy he fatnl.. rortL'na1eIP.r th^aeoCCur very Inlre- rlnrr.tty. The srr:nut shlr effrcts arr:, ll DIcw1 clol:: inthr Ire?% hlntr:, braln. ht•+.rt orothcr orcans.acd hem- orrhn;r Into lhrbrain.dile roburslin` of a blusd ressel. . 2. Wrer tumor.. wadeh may rupture and caus: sorerc Weedta,.. 3. Blfth.tkfcrlslf the p:ll ts taker zhile;•ou nre p'rFnant. 4. Hiahhlood prr.rure.. S. Ca1lbMdder di_cise. The sytttptol. :vaocial cd ailh I hcz,• ser7ous Sidei:IMl3:are di:,rlic7rd In tF.ce detziled IeaO:t¢ivenyou lti?h yourr srpplp of rills. :7otif•vy panr dc:tor & you noll7ee anFunusualp:iynual dts• turbance:rhilo-t:+an` the pill. Thc cstroSCa, in orai'cor.tnrrpilrs hra been round t•j ra:<c br:uvcancrr and other cattrcrs in certaii a-amal5. Thaae findtn^s sl:4gcsl that.o:r1 con• traceptirc3s ma; a Iso cause canccr In hwr.utt. lIo-x-crcz studies to c!ate in women taking currontlpmancct-•d ornt t•ontraecpGres have not cor:firmedfltat.oral confra,cpth"cs eauso cancer In httmatt.. The detailed :ranet describes more eomnletcl•vtae bencflts and ri:ks of aral eosuraceptira-..tt al.so procid•:s In• fornttdlon on other forms or coolra•eeptfon. R,ad it carefully. If.,ou have any Oue7:;tiJns.3•onsuC yrour de+rlor. CuafEon. Orrl eunlrncrpli1'e; are t.f no raWe ttl the F"errntlnn er Iren:-ment ofvenerr-[ disrasc: Dr.ensa °.trtgrn• I_tssctY: v1nAT Yn0 SI.G•J•..D K`rC-•N A6Or17 OM.LL co::: r.u•u•rt rs Oral contr•r.r, r.t;ra• ('•the r:P') arn thee nvst effccttre a•ay. (nv,:•pt f->r s;crliliationl ln nrrr.•°t prr;:taroS• Theyarry alrr+cnn.rninu r.nd, for most a-omrn.. .'r•e, of f.rriJU•: or u!t- pletic+.r.r. side c'fecis..Oral ce••ireecp- :Irrs eu^:6 alic :ya Ar ilraler arr continUotlsdV I~C:': is/On Of :t pl iy'Xican. (If ardont contr-ce.+icr Is ap• prored fnr indi:allca! otncr t!s.n t•on• tr.ueptloa /i-•locid 3 my: Orrho- rforum 2 mC. Qrtho•t:orum !0mgl0 the manufmch:ror maymrnlfr•ny thn.r IhdteaLions in lilr lise Par.--mPh in this sectinn e'r:1c/1U• Ihnt lhr infur- rnition in t/ti.+ L•nfl.•1 ur.<t••r tI•r hred- IifmK "Who YOt. tll- ('rnlCon• lr:tecpllr!•s.'• '!-1•r b:cnr.•r.;. nf Olal Conlrn_I:vlirrs.°nrtl •^.fo•rw in t;- Ornl Co1:S::.c.t.licrs c; fr:4'. i..tl n<. •- p4lcablc w:N:.n IrrsclltuF; nrr u~rd:lnr. 01 hPr itxlit 8 t iu.ls./ .ItI L:4ltaturtlrt.that ans nr.r..•:.n. tntlaid.•rs u•ir.:t an.nrr.l r.3nlr^:•cqa;,r uadcrn'and: It:.• trvol:.d. Al- LhouF-'th the Prat conttacr^II•..•;; has•• Impor•~.nt. ar;ruua,t•s rt•-.otdar.. mathnds of c-^.'rlc' ?ttOll, . th.'TY hal'C certain rc:s •hnCno other method A2-5
Page 281: azo61e00
278 hat. OnlyYou can decide whetlter the advantages ax worth tite.e risks. This leaflet u al tc9l you about t1e moat Im-. porlnnt rtsks.It will ecp,a:n how yos can help your dar.or prrscr:be tl:e 9:7I aa so:cly as pos:ible bytr:.^t, about yourself a.M bet7g alert 1ir t:-z earilest atgv ot trouble. And It will tell you how to we t.^.e pill propet:y„ aa that It wal'be as ef:eetlve aa ppsl-ble. There ts.r.aares deta::ed Is:furma-tlon available hrthelcaf:et propared' for doctors. Your pharmarl.st caa show yau a CopY: YoYY maY noed your doe- torY help In understanding parts of IL Wxo BNOOtn Hot Use Oa-tt. CoIRLLAtrrrrvn. A- If youu have any of,the folCowln;g conditions Youahould notuuc the pill: Il,Clatf In thelegsorlun gs.. 2. Anglna oecturls. , 3. Y.I1own or ]u.ipeetld cancer ofttte breastor sea argans. 4. Unusual vaginal bleeding tllat.hne not yet beendingposed. 5. Hnown or sa.fpectld pregnanry. H.: If youhave had any of the fallpv tng conditlans you shoul0no6 we the pill: 1: Hcartatltck or Strokp. 2. Cibla Idt thc Icgsor ]tmgs.. .Y C. ClMretteSrtTOktne mCroasfa the rtii of TvNOUr advme efrecta on tne hrnrt.anbloud vevelt Iroan ural eontraerpulve uM.. Tnls rhk' Inrreeaea wlth aae aaad w4th hewry mekmg (1i or more elgarNtes perday) and b aume marked In woanen over ]S yean of age..WOmen shG use oral tontrxntlrn ahould npt smnke. D..If You hare s0anty or Irrrgular prrlods or are a youngwomnn adthout a regular cycte. Youshou;ld usee an• other method ot eGntrarcptlon be- cavse- If.you we the plll. Yuu may have dlfflcully'be<oming pregnana or mayy fall Lo have menstruall pcrtods atter discontlnuing the pill. Ogcmtra To Ust Oaet. CONTMCVrlvts 11 you do.not have any of the eam:1-tlons listed above and are trilnking about ming oral eontraceptava; to heip you deeide: you needInformatlon. about the advantages and'rlsks of,onl~ contnceptivrs and of Otlier contnecp- tlve methods at.well..Thlsleafletde- scrlbes the advantages and rtsk.s of~ oral contraceDtlvea Exttpt for sterll- Iz~ttlon, theIUD'and abortlon. wh/ch have thelr ow.n escSUSi•:e risks, the only risks of other mrthods of contn~ erptlon are t)iose duee to prcgnancyshowld:!he method fall. Your drxtor can aruwer buestlons you may hs.ve with respect tn.other methods af;can- llac<ptlon. He e n.aL'o an.twer any. Questtons you may have a(ter reading th.is Ieaflet On onl!cantrnrrptlves-l. what OnlContracrTatlvesAre and How They Work. OraL.Cuntracep-tivea are of tw'.a lypes„ The mnSt common, often sim;plY ca11Pd "•he.plll"' 1 NOTICFS ts a tom•binatlon of an estrogen and a progestngcn. the two klndsof female hormoncs. Thee amount of estrogen and'. pro¢estogen can vary, but thee autuun:.of cstrogen Is m;a t Itnporttnt becawe bothh the effretivrness and sume of, the dangers of onl contracep- llt<s are related to ttSe amountof ea• trog,en TTtls kind of oral centraerptlve works prlnlcpa::y by prerentingg n- lease ofan egg from the ovary. When the amount of estrogen Is 50'Inlcrw grams or ntore,. and the pillls taken;as dlrected- oral contracepnves are tnore than 99qe effective (i.e.. theree would be less than one pregnancy'. If 100 women Itsed the pill for 1 yev)..PL9 that can- taln 20 to 35 mlerogratns of eslrogen varysllghtly bt effeetiveness, rangtng frem 98 eto more than;09;e effeetive.. fhisnufaeturerM may Insert premancy nte forr the monufacturer's product foucd tn clinlcal trials, If, product I, a eombinatlOn). Thee second:type of oral cnntrecrp- ttve. often called the"minl-plll'-', can-. Lalna only a prosestogen. Itworks In pa[t by' preventing releaSe of an rgg from, the ovaryy buc also by keeping spcrns from . rcaehtng the egg and by makingtlte utenu./tvombl leas receo-tlve to unyfertlllaed egg that reaches It- The min0-pNl Is less effective Llsan thee lombinatlonoral contnreptlve.. about 97aeffecltva fManulacturrr may.lnsert pregnancy rate for the manutaatorers product.found In clW- cal1 trlti,ls. if product Is a progcstogen+ only' oralloontracrptlve.lIn addltion' the progestoren-oNY pill has a ten- deney tocaase Irreeularblcediilg wtitch mayy be QuttetnaunreNCnL Or cessation of bleeding enllrely.. Thee prugcstogen-only pill Is wed desplteIts lower effectlvenCsin the hope that' IC o.111 prove.not to have sama of the scrfaw sEde~ effects of' the estrogen- contaih,ing pill (see belo•e) but Itls notyeG eertnln thab the nlinl.plU doas In facthave fewer serlouu slde effeets The dlscur.fun below, wnile based mainly on Ilhforrnatlon aboutthe trom- binallon pllll:,sliould be considered to apply ss we41 to the minhpLll, 2,Olhcr Noruurg)cal LVays to Pre- vent.Preg~nancy. N thls leaflet will ea- plaln: oral eontrneepllves haee sRveral scriow rtsks. Other methods of con- traceptlon haYe lesser risks or none at alt., Theyy are slao.less effectlve than oral[ eonLrnccpLlveS, but, used properly. may be effeotlreenougtte for manyy women. The following table givcs~re- ported'~ prrl:nnncy rates ( l.he numtxr of vromrn outt of 100 who would beeome prrgn: nl In II sear) for these methods: 0 Yus:a.xein Yra iaa wotten Ita YGa Intmutenne drrteetlllDl.Ir ttnanl-6; phal•~tna[m ./lh apennleldal. produeu (cre.ime or k111r./- 3-2C; CunJ-rubbrr I,: 7-]a:. Artmd foamn. r-29: Jelure aand err.ma. (36:: d?3l . Perlodfb aCitltfenee.lmythm)all t)'pes.. leas taan 1-47- t... CqfndM.meth.ud. ,t(+.7: 2Trmperamre method. t :o: ]. Temperatwre method-mtenv eNr tn poSt.ovu4e4orY phare- lesa tnan. I-7t i.' V'.'w mcthod. 1,25: No cantneeptlDa..60-g0.. Thee flptres (eseept for the IUD) varyy widely becawe people dlffer IEt how well they use ca<h rncthod.l Very faithful Ilsen of . thee various methods obtalnvlrygood results. eitept'for wen o( the calendar method of perl. ed0eabstinence (rhythm): Escxpt for theIUD; effectlve.wc of these meth-s ods requires sGmCwhatmoreeffort Uian.slmply taking a single pill ~every morning, but It ls an effort that many touples undertake successfutly.. Your doctor ean tell'you a great.deal more aboutthese methods of ,contnccptlon-. 3. The Dangen of Oral ContnMp- uses. a- Cfrevlalorpdisorders.(abnormaLy btood clo0lt'npand sfroke due ta Acm- orrAaOe.). Blood clots (In various blood vesseLs of the body) are thee most common;of tUe serious side effects ofoval eontrarepth•es. A clot can result In.a stroke ((f the clol Isln the brLn), rA heart attack (Ifthe elot Is tn a blood vesel of the tfeart); or a Dulmonaryemboltlts (a.clotwhieh forms In the Ieas or pelvis.. then bretuts off and travels to the lumts):..Anyof these canbe fataL..Cl'otaalso occur rarely' Inthe blood veseels of the eye..reaulting In bllndnessor Imps)tment of vlslon In that eye, There Is evidence that the tisk ofclOtting Inaresseswlth higher estrogen doses.,It Is therefore Impor- tant tok'eep the dose of estrogen as low at porslble-so long ae the ora3 con- traeeptlve wed has an a[aeptabl! pregnancy rate and:dosen'tcawe uo- acceptable ehanges In the mensttual pattern. Furthermore.. clgarette: smak- Ing by oral contraceptive wen Uh- crcues the risk of seriow advcae ef- fects on the heart sndb4ood veucls Thlss risk lncrea<es with age and wtth heavy, smoking (13 or more cigarettts per diy)-d:beglns to oetome.0uite marked in women over 35 yean oYage: For thbreuon, women whouse oral contraeeptlves should notstnoke. The rlsk of abnormsl ciolt/ng In• creasea a'1th age In both wen and nonusen of oral contraceptlves,. but tltie Increased'risk from the contracep- tlte appears too oe present at all ages.. For oral eontraaeptlve wen Imgener-al, Ithas becn estlrnated thnt1n wromenbct:een the; aers of 15 and34 theruk of death dJlelo a clrculatery disorder is about IlIn .1 C'.000 per year. , V:hereLt- for narltlsCr] the rate is abClll I In 50:000 pcrYear. In the age group75to 44, the risk is estimatud t0 be about I In 2.500 pcr 1'eardor oral cen- lnceptlve usan and about I In 10;000 per year for.nomtcen. Even wfthuut thee plll the risk of hnvlnt; a hra.rt nttack Increases with rEDfkAL eECISrE2,.YOL /3. N0.31-TUEIDAY, IANUAEY41, 1971 A2-6
Page 282: azo61e00
279 427:- . HoTICS , age and Is also Increyed Dy.sueh heart attack rlsk facton m high blood prrs• aure. hikh ¢hol¢sterol..obcsityi dinbe• tes, and clgare4te amoking- Without asy risk factors presen4 the use of oral contra¢eptlves alone may double the risk of heart attack. However, thee combtnation of citarette:smok9ng, cs•Decfallyheavy unoklhC, and oral cpn-tracepth'e usee greatly Incrcasrs the risk of heart aftack... Oral Con 'tiye uxers':bo smnke.nre. cut 5 tIm:re mpte Ikcly to.~t.nce ahcart attack t an -3 c nocsma~c~ a~10t'uncs;more likely to have a heart attack than nonusers , who do notsmoke. It has been estimated that users betneen the agesof 30 and 39 sho smokehave ahouta I In 10000, u a nenusers who do not smol:e* the llccr.have been reported ln w•omen binre serlons slde effeets ln¢lude a chaucein a In the age group 10to 4i:,ttiee riskIs using oral eontraceptlves-.but It~ Is not worsening of ~mieralne, estDma, epilep about l In 1,I00 per.ycar fvr usen.a~ho Yct Y.'now.n wttether the druHcasued xy, and kidney urr heart dlsrtu be-amokt compared'to aDaut' I in 1n.000 tltrtn. caus[ of a lendency'for a•ate[ to ~De re~ lor uaen a~DO donot amnk'e and to r. Danern fo'o dcrrzfepinp child Utalned In the body when oral cont:a• dn not Ifeavy aDOn6l~ln Ie;000~per7•ear for nonusen . orol' conlracrprtuea are usedtn or Im- ceptfce< are used: Ott:er side effectswho smoke. . smoktnW, rardlatrlU p.cerdtny preenaney.~ Oral are, growthof precststingflbrpldtabout l5 clBarettes or mocea day) t•ontracrptlvr.: shouldnot be taken,bytumon ofthe uterus:.mentaldeprrs- fur[her Increxses the risk.. If. yp:.t- do pregnant women because they mxyslon: and llver problemswithJaundicenot smoke end havenonoo4tite oth~~r damaGe the decrlupdng child. AnIn- Lyetlovcingof tlie. skln):.. Your doctor, heactt utack dekfacten describrd'Creasee rlsk of birth defeetu. InchtdlnH mayilnd4hat4evels of sucar and fattyaboce, You a-ILL hase a fmnller rlak heart defects and limb defects, hns subslxnces In yo.ur blood are elevated: than listed. 7i you have severaV.hcart been aasociated ayttth the use at sexthc Iengterm effects ot'tnese chanRra attack risk facten.. tfie risk may'~Dehormones. Inciudingg oral contrscep• are not knoa^t: Some wpmao,develpID conslderaDty,grealer than tlsted... LU'cs, In DrCgnancy: Inadditlnn.' the hlRh blood presure whlle takinga onll ln addition to Dlood~elotling Jisnr• dneloplhH tcmale chlld a•hose rttother ¢nntracrotlves,, xhich ordlnarlly re- den„ Ithu been t•.ttmated that hasrecrlred DES idUCthylstllbestroll; turnstb theoriginatlevcls when the women talting onlicontraceptivex areanestroaen, during pregnancy'hSS.a omltontnrepth'e Isstopped:s tw'1ce as tlkcly as nomisers lo have a rtskt ofgee[lnRcanter..of thevaeiita'or Other reactibrs•e altfioughnofh atroke due lo rupture of e. blood vcxsel cervix Ih her teens ar Young adult. proved to De esusrd DYorxl contr;[cey Inthebrain hood...This risk tsestfmated to be. Llves..are.occasionallyaeportedThese b. £,ormaaon eJlumora Studies about', I In 1.000 ctposurrs or less.. Ab- Inelude more frepuent urination and ]tave found that when certain animals normniifYes of' theurlnarye and' sex some dtscomfort when urinating. ner- are.glven.ttre:female set.Dormone es~ organs have becn reportedIn malee vousncsa. dl'rainess..same loss of'scal..p lrogen. wn4ch is an Ingredl¢nCof'oral~ offsprinH so esposed. It Is possible hair, an In¢reae. In bpdy'ha`.r, an In- enrl[raeepth'ea,. cantlnuaLaly fart lanH that other estraaena..such af the es- crease nrdecrease Irl sexdrlt'e, aDpe• coo[crac¢ptlvea. could tlte chanBes,.cataracrs..and aneetl /or trogens In orall prtlods, caneen maYdevelop In the have the same effect lh tha chfld' if a change In contact.lens prescrlptlon brcastieervlx.:vagina,.andllver.L the mother takes them duringg preg- OrlnxDldlty'to.usetontacL'Jenses. These flndtngseuggestthat enl cen• nancy,. Nter yow stop ue[ng oral contracep- tracepttves may causecan¢er In If yort stopP taking orat tontr[cep- tla~esthere maYbea d¢Ixy.he(ore you humans.. Hawtvcr, stwdib to dat¢In olves to become pregtant; your doctor arc able to ~becomc prcRnant or before women CxklnH cwrrentty muketed orx7 may. reconunend that you use another You resume havqng mcustmal perlodse eontrateptlves have natconfirmed methodd of contraccptton for a slrort Thlsl! espcciallyy tme of women w.huo that oral eontra¢eDtlvrs causacnnmr wtiue.,Thescasom (orthis Is that tD¢re. hadirregular mens[rual Cycles prior to. Ift hurnans. Severalstudies F.ace found 1~ evidence from studlesln w•oman who the use af ornl contracepttfes As dis• no In¢reue In Drea.s6 cancer In users.. havee tiad, •'m,lsearstages" aoonn after cussed pcevlously, )'our doctor mayy although ane study suggested oral . aeappit[Rihe plll, that the Josc', fetuses recommcnd that you wait a short eontncepciV¢s might cnu;e an Incre:lr~e.e arC more t7kely to be abnorml[l. whlle after stopping the pill Deforte In Dreutcancer:in ~.vomrnwho alreadyLV)iettur there Isan overallIneiv, se in Yvu try to becomee pre6:nant. Durtng uho become thls llme. use another fomt of contra- have D¢n!rnr breast dL ea.¢ ce:R:. c5'sla). "misenrr/a8e" Irt women ' Wom,en wltha:stronR faml4ylilstory, pregnlntsoon n(ter stopph+g tha.pill crptlan. Yeuahowld cnnsultyour p:ul- a( breast <anecr oc w-ho kavenrra:.tns compared with women who do not CianDelorc resumutq' use ol aral con- nodules. flbroCystlo dise^se, oC abnor• usethc.pill'~ISaloeknown, Eut'it Is Dos- lrnerotices x(trr childbvrrn.. especiillymal mammnKramxs oro who w'creex:- aiblc thnt there mayDrt..Ify howc,cr; 11'you, plhntonurse )'uur baby. Drugs. posed'lo DES IdlethYlslilbcstroli: aa you do bccome prrgnnnt soon.atter ld[ oral con:rx¢eptice3xre known toestrogcno and do aVprmr in the nrilk, andthe lpnrtrange. duR thelr mochcr't preR• stbpping 0ral contracrptivrso nancymust De.folloa2d verY'closely Dy, nochas'c a mfscarrint.c, tticre fs no evh effrct on lnlan[s Isnot'knoam at'this ttielr, docton If they cLoose te usc.ornl dcnce that0:hc babyy has.an.lnerenaed' tim,c. Fltrthsr:nore• oril contrnceptives comtracrvtt.'rsIhstexd nf anothcr rt,k of hrh~g abnurmM. may cause a decrease In youc m9tk metnpd of eetltraCep[lan..~bnny.stud:d.,Ga[thlddderdlsVOae. Warnen who supplyuw[I1:SIn:heGUalltyoflti6 lee have showntliat wCmen taking Uae arall can6raceptfves have a greater. milk:.. chance each year of' havihR a fatal. nanti tumur of the liver, These tumors frrouently seen when using the minl- heart attaek.cn.mperedtok about a I Inn da not.spreadt but theyy mayrupture plll -.comDlrtatlon ural eontr.tceptives 50.000 chance Inusenn who doo nott and 'auseInternal bleedina. which cnntaining less than 50 mlerogranss of hpt a I Ih1000n0 mnyhe fatal. A few tues of cancer of estrogcn. n mak• d1 repeqAt aeGlatea,. VOl ah NO. 11-TLRSDAr, JANVABT 71„1er1 A2-7 S oral eontran-ptives hare less risk of risk than nonusers of having BAII't•tad-eelting bcnifn brcast disca<e than der.ddsraserrpufrinr surgcry. The in- those w'.tto have not uxrd oral tontn- crezxed rixkmal` fint appear within I ceptives: Rcccntly, strong evldenco has year of use..[nd may douDle after 4 or emarqed that eaeropens (one cnmpo• i:ean of use. nen4 oforalcontracrpth'es).f wtien, e. Othrr.side e.ryects o/'eraf'cpntra• gh•en, for perlods ofmore thanune erpticrs.3omevomcnuslnBoralcon- year towomen after themenopau^c, trncrptlvcsexperienccunplcasantsideInerrase ttiee rlsk of~ cancer of' thee eltecta:thacace notdnngernus and aree uteruslw'omDl:Threelsalsoaomeect• not likely to damaga their healtn- dence that a kindd af orxl contra[cpttve Somn- of theSe may be temporary: which Is no longerr markCM1ed- the s!- Yourbreastsmay'feelttender:nausea Quen4lal oral Lrontraeeptlvv, may In- and vomlttng may occur, you may ;afnn creasc thC risk Gf Cancer vf ttte uterus., o[ Ibse. ifiBht,. and your ank'les Rlay Thererrmalns no evtdence, however, sa'ell. A spottydarkentng of tite skln, that theorat conttaceptlves now avail, partioularly of the face, is pos ibleand able Increnae the rlsk of thucancers may'. Drrslst. You n>zy' notice unc.pcet• Oral eontraceptls'es do cause, nL+ed vaginalDleeding or.ehaneesfn y-our though rarely a Denign . lnor.-mallR- menstrual perlod:' IrrcNlar bleeding !s.
Page 283: azo61e00
280 4. Comparison of tlte RIJY.S of Oral Contraelptlves and Other COn4rarep- tive Method.s.. T!u manyy studies on the risks and effectiveness of oral ton-. traceptl0es and other methods of con- tlaeeptlonhave been anal.2ed to esel- mate the risk of ddath assoclaled with various method.s n!eontneeption.; This risk has two parta:(a) the risk of the method Itself (e.Q., the risk tha.t oral' contraceptfve. willcause dcath due ta abnnrmal'~Olottlne)- and (b) the rl3kk ot death due to prclmnney or abdrtlon In the event the method faltis. The resultS of this analszls an shown NOTICES In ~.the bar e~raph below. The height of the ban is thc numbcr of deaths per. 100.0d0a-omen eaeh, year. There are stx lets oRb^n,..earh set.referrlne.to ipecifie at:e [raWp off women. . LVlthln eachsetofbara there Is a sinale bar for each of the different contracepuve methods. For oral eontncvpuves, thereare lwa ban-one for smokers and the otlter for nonimaken. Tilee analysia ls based:on present knowledge and new Infurrnntlon eould; of course. alter~ It. Thr analysis shouathat the risk of drnlh from all methods oQ birth cenLrnl la ]bw and below that associat- 4233 ed' with' etiildblrth- except forr oral tvntracrattves In,.a'omenpver 40 who smoke. It ahow-a thattlte lowest risk of death ls tuociatcd wilfs the condem or dlaphrat`.a (traditlonal tontraccption)~ bbeked up by early abortion ts ea.se of failure of'the condont or diayhramto prevent pretmancy. Also, at.anyaee the risk o4.death (due to unrcpected prCgnancy) from the uoe of traditional' contraceptlGn. even witlioul a backup of abortion; Is tencrally the same as or less than that from use of oral eontn- cepth'es. ry,.... t..rr ....+..~. rw ...,..w r,w a....~ w.,.+,. w...w.r s. ,r am+..w ...r ....r..w w.s.... .,.,,.n...,.. u r u M L n r a a 11 iY'sA1. Arr Q....,,w 0 u...,.« K ~~ .._. .:.« .. . rept.UL atWSTea,.VOL.47, 110,2I-rUtSDAT, JANUAlY 21, 1971 sa. C1 • ,
Page 284: azo61e00
281 ( 444 hovr ro Usk onAt..CoKrn..cerrncs ,4s S.retr neo Enccttvttr AS Posst- aw.Oncx Yoo Havs.Dactnsb To Uss '17tOt 1. What to Teit Your poetor. Youcaomake use off the pillu sa/elyas possible, byLelllna your doator.IQ you have ahy of the 1011ow- InC a. Condillons that mean you should not uae oral eonlrateptlves: . Clots in the IrIIS or lun{e,. Clots )n,the legs or IunCs In the past. A stroke. iteart attaak or anaina Dtttoris. Knowqor suspected cancer ol the breast ar sex ort;ans. UnusualI vea)nal bieedint that haa hot ye0been dlat;nosed... Known orsuspected pee;nincy. b. Coudttionsthat you doctnr will want to wstchclosely or e-hlchm/sht caux hlm toaussest attother:m,ethodOf e0n0raCeptlom A f:unlly history of breasceanar. Ureastnodulcs, fibeocystlo dlscase of the . breasti or an abnormalmammoa anm. Dlabetes.t ltish blood oressure.. . 1fi{lt eholester0l. ClParcttee smoking. Mi4ralne hendaches: tfeart or k)dney disease. £pllePSY: ~ Menflal depresston. F9brold tumors of the uterus. Gallbladder disease. e. Once.ynu are u:rine oral contracep• tlt'es, you should be alert for signs of a serious adverse effect and call you doctor If they,ocyur.3h'arp pain Ia thcchesG couetiln[ binod„or sudden shortness of breath ((ndicallnf pOssible cluls tn the IunGs). Paln latheo ealf 4poxtlble clot In the lee). Crush4ns chestpain or heavtness tln-. dicatlsta poasible heartutackl; Suddcn ttYTree tleldaehe or volttit- lnR. dirraneu or taintlnR. disturoance of risi0n Or. speMeh or a'ra/Jta'13 or numbness inan arnl or leg tindicalina a possible atroke/: Sudden parUal' or. complete loss of~ vision cindicattnt.a possib(e clot in the eye ). Brt'ast lumps (You shoUld'ask your doctor.toshow youhow toexamine your OuT brraSla). Severe pain In the abdomvn <Indicat- tne a possiale ruptured tumor of'the )iserl. Severe depresslon. YellowleR of the skln (Jaundiec). k How toTake the PLA So That it 1s Most Effective.. . (Manufacturer to supply lnforma- tfon on Ao.uee and administntion and what to dn It patlent.has forgotten to tOke one or two ptlls: Whereapnllt'a-ble; manufaeturers: should supply ap. proprlafe Inform,atlon reFardiiTCOae for other approved Indlcatlons.) At tim+•s there.may,be no men.strual period after.acyeieof pilla. Therefore,, if you miss One mrnstrual Derlod but have taken the p/11i uaetly aa trou seerr fappoaed 10 eOnttnue as usual Intotheo nest cycle.,if youu have not taken the correctlyy and mitsamen- atruA) period. Or If yOYare tak/nQ manl'pllls and It Is 4SdaYsor more from the.start.or your IastmCnstrualper)od ypuu may, bo Dresnant and should stop taklna oral eontmcepttres untilyour doctor delermines whether or not youare Orr9nsnt: Until you atn, get to your doctor„ute. anolher ferm, of cpntnceptlan.. It two COILSeeutlR taCnstrukll periods ars misaed, yob' should stop taking p101s untli It is de- trrminrd'whctheryou are presnant. 31 you do becomepreLnana: while using oral eOntnecp/tces. you ~should discuss the risks to the dcvelopina chLtd willl your doctor. 3. Periodld E.arninali0n. Your doctor will take a comNetemedicai'and fnmilYhlstbry tiefore pre- saribi:aC oraleontraeepUfra. AG that time and atiout once a year thereafter. fce xIU generally examine your blood Dreasure, breasta, abdomen, and pelvic or0ans (ineludins a Papanleolacu smear. I.eC, test for cuxer). SUStauar Oral eanlneepttvrs are the most ef- feetive method, exerpt sterillzatlon. for, preventlnapreenancy. Other methods, w.hcn usediconsclentlously. are aiso reryy effeclce and have fewer risks. The serious risks o1 oral icontra• ecptives ara uncommon and the "pill" Is as'erya eonveMent method of Drb Yen/tnCprcenaney: - ityouhave certain conditlotss or havelwd thcse conditions In th, paat;. You st)ouwld notuae Onlcontnceptiveal becnuse the risk ts too great. These condltlons are listed In the leaf]et: If you do not hac•e these eonditlons..and decide to use the "pilf." pleise read the leaflet earefully su that~ youean . use the "D1U' most.safely andelfee- tlrcty:. Dascdd on his orr hrr, asscssmrnt of your nlCdical nerd., your doctor has prescclbrd'thiA.drus for you. Do not atve.the drug to anyone else. LSen 502 s03.. tiSIaL.loS0-ldSS•. ae ameMed 1:1USC 152...:;500 and under lulnorltYdefrastedt0 tbe Caevnlsel0ner, of Food rnd DM1aY./y1 CFR S;i iJ Dated[ January 18. 197a. Dos+t,.o Kcxamx. CcmmissionerolFood and' 1)rupa IFR Doe'R-:301 l71e01-Y1-1a: 10:ti aml repfar.l 1EG1a1Ek. YOI. ea,. XO. I1 -TUESOar, JANWA4r ]L 19ye r I 95-757 0 - 79:- 19
Page 285: azo61e00
282 AYNl:;ND1X3Proposed Text For Physician Labeling. ORAL CONTe4rLPTrrr. Lr/v.rr.•r Dr.3c111Mt0'( •TO IIL sVrrLrrn er MANVrarlllrrr'' Uoaeriptinn atloldd ielrluefalh+o foltolc'iuq mfonnatlnn' r L..TIIC.UrO/)rietary nAlne?11ct IhC r- tahllshM nnmclf anl•. of tlte dnle urt: :.,7sle s~m0 rlualilnlisenud/irr tnr:lll- tltaUt'e lnercdicilt io1(ornmunn tc re- nulcrd for labels;'.. :l..Ttle pltanlmrolhctrrl ar IhCrnPe•1- lie cl:lcc of tile dtvc.;truduct: 4. The dtemiralname and strncttrrnld fnrmldn,, CLINIfAI. ("11 \t.11Aa'er(.lyt,Y FO/I 1nM111N,1Tlt1N. O[AL enNTLACI'r'fIl'e1. ONLY t'•rnlumutlou or:ll. rmltr.IrrnGacx, actt urannrttythronr.tt' tllee ntrehmnlau of, potltillvtt'ONin.sunPrCssInn rluCtb tile es- rroccnlc and pro^,es latlmtN: actB7tv . 0f the m:redlent3. AlthouBh tile prrmOr>' tuerhatusm~of actmn Ls inhtlultionof oVv-lation.:nilcrataotrs iil;lhe ¢Cndaltfclc4 itt-cdttdlhT chan6esLlt tile cen-teal m'uevs rrhicllIncrcise tile d'ofncnltT' ofsfxrmuenctrntiontand the endometrlum rwhicli reducetlulikelihood of Ila- uln.ulntiottlmnyalsocontributt to con- trocenttl•e ertecttvenecs. FOR.rrrrc:[sT1NORAL CONTnACI'rT(lrS ONLY The Drimnry mechanknt tlrroulFh uhrea . utsert narne of dritp,' ~ Preeents GonCrution is not knoWn. but vrol:Catiu- onlrr'ontrnceqtit-rs are kno:rn to niterr eho cen'tenl''tnueu.c. ecert a oroee_ctlonnl eRCCt ontno cndonlctriunli interfrrhlc actth unvllntntlon, and.. in somrRatlenlx.. -nnrrn.S OtVI:It1o1L''VTnlln(ZcRurer to includk informatiott oa :llisorvllon. Ills- trrbution, ellminntinn... nnd uhnrntaco• kltlettes lf vertinent:',alcoun druc.inlrr- actionc perctnentto huntan use. - INOIC.aT10YS. ANn fj3Anr. I Imert nntne Ofdrup.Is,Indicnled for llto vrCr'entlon of urCgnancy Ihn a'omen who cled:i to ua¢ ornlicoLlfYnceptires. n.. A mclhodof, conh-nrentlbn Onlirorltracerotiresnre IliRldy rdec- tlr6. Tlle preqnnory ratc inlF'Onl[n.uOille ronrcntional comhinatton oral.eontrn. rcPrlves 'containin; 35 me- or more of elhln.'I cstrndiol or 50 nice oritnore or IlleSlrntlolt' ISCenCrtillys rCportCd a,< Ic-rc tt1:111Olte hrr_r.nllrv IMr 1W n'nlinall- ccarA nf n>r•: SliOhtt}-y hiCllrr ratcrc •=ontr•- n•hat mnre than I Prr¢'»ann-. ncr 100 natnan-rcar3 of: lno' are renorted for <nmc combmatlolr urndtrcts cd•otamiue 35 mt•c nr Irca of rlhlm'1 c,tranroll.artd rstrs. On thee orc!rrr nf 3 prrcnanrtrs Itcr "on .rotnen,Teorsm aree rcnorteilfnr rlir'Innnrsl in-pntynrol eonlr0rr;tttr'e. Tllese nics arcdcrired fronl xp:u~:tte studlcs conducted by dilfcrcnt inrc.aar•a- tors in se+•ern.l poNtllatiott erouua and rannot be contqaredd preer.cly. Plrrrher- more. nrr^_uancr rates. Lend to be Ina.-er is rlinleal dttehcs are cnlnltnned• pns- •'drdue In .clrrRiceIcVcnliun in tile .•, I:rr .Iqthr. r.f 11NaePaI1CttLC whn :1. - ri'1.6 IhP Irr:1Ih1P'lt rCmnlrn anrl Couot rh.runl u,-.• n": .1 rnult., of, odorrnC, rrac- t:m:.. In'r•L•:In..n., y'. or otltrr rrnsOnF , Lt Nlnlt'^I Illst...t\itld•in,rlt nnntcaf :':,I: • m-Cl l nmrll,rr ':f •P:IllrnlArnnt. nl~ tc<t .,.,,. -Irr;lntl : Intal nf. 1rv•.:n:rurlr.+\rr^ rapnrlydi Thn rC10n- •.s14.alur.enalur, ':ale u( - - rcr Inn0 e'nrntan-vrnr< •monnfarttur-r loadd oth- rr tnfnrms!lon rc!:tlcd tn tile llrf~nanr: r:.'.~.•.rtlll lil.x 1•^Itir•:ht' I"l++tldlcl. I' nced• ctl tao prOCydnv atrcnllatl nrccCribm•: ut- fnl m:ttr^_.n to t!lr~hcalemn: •. Tho fo6lnn-amc lhblr rncv ralu:rv nU prrnnmlct• ralot rrpOrlrnliun stM1nrlarn teslbonk Ifit•f. U. Llr othr•r tur.ttlA M'f cnntf:trf(1tt0t1,. Alt nldn'fd11Tl I/altetlt tlraa% nflllrre111L•hcr nr lon-er 1'nle!lllth :ttlr enenunlrthod ~turvt tile ICDt. dcPrnd: ing utlwn tho drerer ef:adhercnrr to tile mrtlimd Plar,..rctu• 1'rr• 100:Ccrn1AN-\r.Ar.s IUD. 2-3: rOndnms. G,U: dinphtap•n \•~rU- crra.nt or ccL _•37: CoinLCiuter- ruuluc• G-10:rlnthm:, 6-17. fnnntn.. ere?ms or erls a.lonn. 3"2: nornul.l'N- ccution. Ii0~A0. EuldcntioloCloal stitd•icsi hare slioun ai po>(tire aASoc:nlten het6e^n thedofc ol cstro$ens iil:oral contraceptives nlldlthe risk of thronlbncmtio0aln I ruia- ^2 and .l r For thls rea.on:, itis prudent nnA In keeUlllt: n-1t11 Food'' Ur111cIPles OC thera- Pclltlc3 to Itlininli3cexPOSt1rCc to extm- pen. The onl contraceutire •nodurt Nrr- st•r.hcd forr anye•i1'e-laottent eliould h.•, thntpraluet °.hichcenlautsthe. Il'at atnOnnt of'f ettrnLrrln tltat L.ronln:llihle veilh an aeeePtable precnanrv ralcInd. palirnl.Occel+l2nre. It It rcronlmectdl I. t5:nt netr,arcrol ore nf Ioral cnnl: aroativr% sllouldbc .ctartnl on urenarationa cotl- rninlnc.i0 \tCG nr les.c of E.ctro¢rtl. Co.rr.rtvnacunnss Or.tl contrareutivcs,sldouldhlot lie u•cA follmautc t in a•otatcnl n'illt anyo( tile c011dIti0lt]:' 1. Thrmnbonhlrbllls oritirnntbocrnb"hr d'J,ortlCl"s. •. Apnstlucdorv of dccv reon Lhrmn- bot!htrbtl lsrfr throntbocnlbolic disorders 3. Ccrcbrnll t-asndnr or coronary ' ar- terv di.e:u.c. 4. {flton'n Or' "nFpCctFd I rnrclnum;% nt tile brcaat: S. Knmvrl nr xta•peclrtl c.troncn de- pendCUt neoltta.ra. 0~. t:nrlla::nOSeA .1blturtnxl ccmt;rll IYteedute. .. - . tillmun or auarrrlrn VrfirnnnrN.laaa n'arltllle Vdl 5,- . :VA.I:arvr.is 111r •'.'.r ~)f.nl'.11 t'Olllt':ICClntvr'. I.'.r ,Or1:IttC1 t':Irll Illrl'r.lsrrl r1:ak.OL .ac\'Crnll scrIOtIA rondnlona nl¢Inldlnc tltromlinrn\-hoGsm:, •trol:c.tn/a'arrdlal lnfarUrutl. heu:uu:' adcnonla, call blaJdcr dica.ce. 11>'pertllnston. Pi'aetlttonets ttre.•crtbincoral rontr:trcutil'ec. chonld be fvntllar wttlt Ilre fOllowtllL• Illlnnltntloll relltllle, Iq tI1nC
Page 286: azo61e00
i 283 NOTIfES 5=d:.S'. f , 1. Tl1+ .nbocrrehofre Di••>r,t.r nndt_. uuJ ial:emrfrmol4 Y Vrn;uwly au,• ~ c.tl.e dcv- oftltrolniuuanbolie r nrrll•r Vrtsrnl'vr Prnrytrms: An ittrrt, sed'~ pre[eM', :45sociatioN tTa01[a 3urd ~r. as+uetatcJ ,tteh I+rq:C+tin-onty orel corl• -:+k orilhrou:bOClnboLlt.ande thronlluotieThtse alntllr.- rolndOCLOd ht, the Untlcrll tiaceytiect Irl:ate [urL been ltertorntrtd. n ialrn:wrth'.tlJn rr nfor:rlF:irICUMIIa fottntia_.'1a e<PlctCd.,thOt llte CaaCS.ol:lllrmnlMVmiinlu'dts!•asE hat'e. !~n~lrNCCpttt'r. La a-CIl'e•d.JUlt+ltrd. T11rrer.rrt,l- lhr n~mibrr n/ nltdcrh'tnk n..k bcen r wt't,d'tn n urn u«ruL' tlul.e '•rhlcal'nlLttntltrxinCrCUtU:~1!nn~.rRt•I.. fdr[nr5lnl' artcry d4et.e . nroJtlct.c. a :vJ II:cY'~hou0d nuM1 bc nl~- 1 LllratIIh 0• ond three tn the UNllyd ::nLttC •urokL,"r hytlrrten.iont hyut•r- ar.d lo IM.fr¢r of rartt.. rL k. ?r; tes rRct+..7 throngh ~.10~ lurr Jrlr.nn• clrn!r=trrule;nia, nlrt: rtr. d'Jabepes. hulor-v' The:risk of thnmtn0e«llrotic atrd t6«om- .Intr[I atl inrsCnsutl rC•kk ut. falal tA Nnv!'l'larnlettr lovrrnuar thC 1i14hCt.illt. Iqtte thsordGre. rn Irutlt u vld u •nOJlai. ltr\ouf tl,«antllremltnll.,nland re..l-~ nf derclDp.itl'l tnt'ocor't1i:J int:Irctto:L. ers oPur:if ~6rnrarenUtt`rs., IIICrP3~1Vn: r-lroke: ttolh hemnrrlr]^tie :mn, Uvout- rey~..rrllr-s nt'o'hcll:er tlrr patlent'tf33 utthnCe..Oral eulttraravttrrn :1 e Im.c- •„Iic..Shcse..aodies caVnuate, lhnl tnt'n. r,.l rutltnrl9nivr, tPCr n not.. Oral evcr: an indet^rnnrnt n.k.ln-r.r flv rlur.,., .l ornl rontraerp,lirr are A lu.lf Ihnr, r011lraaCFtttH. Iunreeer. ,rrr^(nuud u, et'enl/r. `nortl likVlytltan n a todCvtlup b~e:lrlr:rrnelthlton~lrt~kT:rel~rr."L'IiC:,ne T+et.e.l ul• I t~e. nf r os. uitltm t. e-rd rtt rur.e Ilual c ~cx• r tr oLlynenrnr.d t.rbl••r. ld'i3!'. Ill.n1 :rle. Il/nnlr..rr> ta.rctlal. f t~ult:ultl tt ut t~tin rrutlCUn- rall nVr.•s IntnrlaWle dnC tn.lmhnnnr- tr «:rrrcl,live u rannatrd tm bc nA,- xn rarnve r.re....•+r•..... ' rmW 1 ill . rr sLrok : -s nlt,th'e M•drr I'rotu .t I IFr ]OOIMII -.0 tl ~ rnx 1It-I t0 3 ry deatlis3tlnWClly IrFr 100 Ollp x•I's In tllt• 3n 39K etr aEC Crnnlv nlld'G7 aetvt«e rua-. n.rr9.a11d Itterc3.tes. wttFl.aer •T'.,ble : i>Cr 100.1100 u'nrmen utor.. In'tl.te N/- " rr yrrart. lit a co11at1ora:tiveAnMr1CaC AtldY. (RCIn:. +1a4'r yIOt•nlrtenaathrcc tn onenUeu,rnrtlc dl.e sse___ 1- 1 rand10,ofcerebrora.•culardisytdenin.. r.,r.1..,::-illr-rrvn.r,r Lrr..u r..~urmnmornmnluc.comolt- . omcn alth 3tiS wetnoue p[etltsp0simu' r'r n,.,..n..rn..nn,rr,r.l.~rnt[„ner,rr.-f+n.r.. I 3usCi lt was CSthnalcd lRit~llYe.riikt o[' t.erntOrrha4iC strokC R9J. ?.0~ Lllllit'S xn~n•rnl/xr•.rlnrr,ynrlrrnixrw[tenwrrhn4len«ote__________________ ] :rr6ter Inuser5 Rr:111 lmnu>ers 2nd m 'eurnrtattnrk Imealrraae~oa91...._. v ,„,yr;,,,. r,,.,1 itn,rt atr.+ek Iptatl r,Ce+o +at...._ a.e iat of ttlromi)otlG sh•okr w e,+ tn 15 !rmee Crenter.in IISCI'w tlianiln nonert•R - ~-_ . Xn:rt Y!1Kk IrruretYtnlr (a9t•JOtit9r:. _.t 'T:rhIR•1•. Ueurtalnekllwntat.U rn:;-c.+n~++Y__ a.'r r..!. r ' AO.nurpn+an:nnu//Y .,t•'rv.~.-, _ oer.a-JJ . . . of Vralcuntrar:eptl- . .... +.t ht an analYs13 Of d31f1 dertt'ed: front .,.rn•.rr.. . -..._._...--- - ..... 5. .aet'Cra111RttUrlal adeer5e rCaGG01Y rCpDet~ I "' . n nfl rnre I r•r pwf... tn:5yS4e11As LRet: ?,• F$[tlnhnuY4'aLi.a• rnbwi . 'c6rnt te .nn.i• fl f lulat o cludett tltat thcrrsk, of throm- r n i.)~rl h. Ily p. Lr, x~•„ Itocmholt t rltctltdi«C.cOr0ltat)' tl«oat+ prrqrr•rxr rrnmrn b0.••ti i.: dtrectlyCClatCd to lilG dose of ektro;rR unrd ill oSlll eotntrac9plit[S. F'ren:u'aciona oonlairw1C.100 mcCor nuore .rooW.rr t••.a•a•rr ufe.trn4enwereossocualedwtthnhr-hcr Theavailablodat;r front avtnet)' nf sources liet•e beetl analyted',•Ret. t+r t,. cstiutntE the rilk of dtatli nn>OC1atCd „'ittt. vSnEt«. nlethGda ofCOt1t::iCGpllOfl. TItC [slllnileo of risk of t4roth, for eaclrl mrtlvttl'ttltlude the comloitucdr«k uf ithe contr:tceptive ntetnod/e.,.l throtnbocnu- bolio and thrOrnboUC dleasG Ih'ttie caac ot oral Contr.lcepttYest Ils the 5k. t-trrb t ble r tr e4nan y' or nbortio In tlxer ent ofn.cil odfait eru T1us I ltrr ria), v'lrles.u.nh tile eMCrtrve c.s ot'.ti)eM nntratcntn'e method. TihC ftndulys ut'. IIiLS arlal>'ali:are slto,tn~.inF'icure t belDlv. IRet. 11r'. Tlle study.cotncluded tlinr,Uw mortallt1" :1lsocta{Cd ,c!tll nll rnc•lllmr6s nt ui~rlh contral is lotv comtnlrcd.to the rok ot chridb:rUi. tvitit tile crcel+tion of oral untmrel+nvrs in q'onlelY,over 40~. and Anlite[eatrdTYltotatyGCOfdCdlntart- rdptor.htfolvrd. Tt:it.r«diltQhod bCetll lLat «iV10„est.mCrtality.ls.o.i5Orlatrd tion as.soeta[edwrtJt the Veeofooaliron• conf4xnled'd« t1leUnited;States (ReF. 31. , uith th'e:condotn or diauhra4m txicked trnaeptive_ hoa hcr•.n retv..r•cd. ' Rcta..l t. Cnrrful euiJeneinlnniral .sttudles tO delCr- IIp 0y early n.burtiorn.. rt.k of lhronllwr•mbolitim lhao tlraYe- colt- h1n111S. 50-bo ntrs od esttc4e'u Ttre«' ..• mml9'sL. ditl sulssest. 11owe.rrr, that the nuntutityof entr'u:etllaaynot:be tile solG rentut CeGitrFrM vOl. .1) MO ]li'~nn5;.lt..otLe•.tetr.r ,or.
Page 287: azo61e00
284 riqnre.l. nnlw/al bnrcrber.ol r!r.rrbs.ns:o•:irted wirh r.nnttel pl /rtrr:blY andne ipnnnl ber. 100.0M nnnateril'e wnnren. hv.req:mwn nl cnnlMl nnd aqa n1 wmm~n. Annunl deellrs N - 1] ~ L r~ 1! 13 t7 11 I 10 ! 0 !. S P ] 7 p 'I ; .. . . '. . .~ .. : I f I , .. , N . : JS•79 .~e ppcl' menonnl~ual xmnan 1RCG. I4, 1.:, , nr4 19). One tluhltepti0n /frcf: Znl rC- Iklrlerl nfl llle, fnt'':1 r:IFe snbnl'.itlcd b't. pinsirians.ln a ccg/ntrf of aa ofa,dt•n0ca 1 mmoma of tile cndonml rionu in ~ v:mn<n undrr {0 011 nral <nntracenonrs.. Of tiv case3 Icund: In a:mn+•n a'~hont. nrrdintrrnlnn ri..:: factors.Iar :tdntvxar-' inl Lhr rnd.n.nrlrilllll rr r..; . LfrrQ-ltinr,rlaPrvlihn at IhrliinC rr,rtl.cnmin- rrl•tnr: .c'rrom lir,`_t'x+^rl'1. Inlvervnl,If ora- rlC:<r:,nenrl_~':+Il nrrllrili ill tl.mnCllwhp h.~d un.i a scnucnUal oral culnrerentive T7tvs0 nratluC_. nrcno IOrllter'r' ar::ctcd• Nn arinlencC Has be<n Mpnrlen supeestirtn 1 incrnsrtllri'k Of'c^domrtiial cAncor in usrrx of rnlvrcnLion:tl cunlbinat100 or I,rnurstin-ollOV ur:d conlru•rlai.vr..:. ScvCr al atUdlrs t Refs. 0 nnd ? I ltrrot~av 2, 3 hnve foluld 11o Incrcascilu.bl2a`•tcmlcer in amnnl Crkin; Ornl cmttrarep- Livrs or e.sLrn6Cnn: One aL11'OY IACf. =51„ Ilor: eycr• whtle alxu noLprnno a'rlnll In- cn•nsed rlvk of,U'rc:ut eanetr In wumru. Lrrnted.. w•itll oral colltra[cpli.ts.. fownd'n Icncc.~5 rtck in,. tha anbt;rolup3. 0f otal nn ' trllcrnlife llsers wtlll-documcntcd be- ,ru:n brcarLdL•rr.::r. A'rtt10eM rxenrrence of IMtri;tl hrrnnt ltlmcrs In nstrs orornl Cp11tIlCClltires lt•li b1Cnwell-dpell+rmcnl.Cd IIRrIr. 8. =11 :J. 26. AnA 27). In rY~mnlnr:.„lhlrr is ntprMeltL 110 ronlinmrlter-idenee from huuluan atudies of mL insreucdd risk:o6eanecr aseeuted nith oral caltnueptiyes. ClosOelinacat snR'otlldnao of nll c.onten ttk'tn4 Oral Cnnlface1rL11'rf IS„ntCt[thrLCC.S. RECnLIaf. hl , plt Cnce3 OI~IItIdinCne.erri''7rr.h,lelt Or rerurrental:nnrntalra:Itr..lUleeditlC.:nlt- rro rttte di•llino3tto ntcasures +hnuhl1 be n I Lahcn to n.~10 out ntmh6naneJ- 1VOmc)t st wiUh n stton^ famiil- Il'islorY of brtt entlcCr CF who harC brcnst nortulb, fl- tirntystit disca•c or, ab0orn1a1I0mm1co- r11m.a shonLd he nSCNlOrcd zitll paltiCtl- lar car0'if Ihev'r:lCCl to u ral ecntrn- Z,ptlvr'.sM insLead of other nter.haldA of ntrnerVtlon.. 4. flrn=r/iC rldl•'rmnrn. 0 Aborlionenly, eandireana eapliqn Only El The r oQ' lhron+bucr'I+hblic and thrombolleRduensr a.csoetated. wlth oral con'LncrpLlFrs incrt:up.a:lUh le'lalter, Ilpoaoxtmatcly a8C10 nnel'. fur m5•oear- diallnfnrctionl..'Ll furtlter in rn.ed:tiyeiprnte Smakifll:,.hyy-Citeltsian.,lEyner- Cholesteto~; Obesity. dia0elrs.o[ his, tory: of prerelamplLle toxonltn. Tile rukofmyoeOnital ilrtfaretlOn im mral eontrn- Ceplte< usars Ls subxinntlnlly'incteared in women.nCrC 10~ and ol'cr• r.,pcclnlly, those a•tth OLtterrts-r fuctors. The unc of oral canlrar.[ptirrt in u'Umen ill tlrts nqeMup Ls nn6 rttommcp<IM. Tile phyalNan n.nd Ihr pal irnt, ehnllld hC alert to. t)tc earlk.[ m^ nif ra ali0nn of Lhrolnboamh'nlic.andllhnvnbnnlC tW.rrr- den IC:a..,tlllrnnniulld11r1i.+lL':'. nulnnCna,1'Y MlMlirll. CCnCnru•~:+.rl'll:t.r"fIl•IlnfniCnrp.; cOnnnry IKClu.cion, rrlinal '. ll+ron+lnnni+: ma m .rntrriC.tllr:•mtlNnsi, r:.Ihrtllld alty, ef LIIPSe pCru'r C[~ I>C c/~cl,l'ilrtl•.114 'dn1T, sliollld be di e0n1 innc0 iiulnedt.+l r11': Afmw-.tosr<-lnlrl utrn•n.r•I ri>k nf Ivwt.rrrrcr.ry thromba•mbofic r mnlir:r- tlnlls.llas Ileen relrnrted ln ornl rrolttra- eepliveu en ! refs. 15 und 181. 1f ~rCaatble. oral euntraceptivt•s .houldbe Olatoptm- . Itenign hcl%+tic adellomvt nmxar to bc UM at li•a:ct.G'aYek'S,be[nre nurpary'o! svocinled, x',1ih Lhe+ttsa of'o1~~i Cun~tra- n/yl:c ax:OcintoJw'fUuan incrrascd rt.k ccntit-es-^8', ,8c alld 301. 1lllitnn,h of II1.rOmboemholLSm or prolonccd im- hrnlCn. nndr:lrrd hrnllic' adrnonlss mavmOblWrltldrL r1lnLure andd nAay. CatLe dpaih tIIrOllkh 2. Ocular. Lesiorts. TherC lnu'e bcen reInL[arpbrl0nuinal henlmrrhn¢r.. Tln,c . Iras. porLs. nf nCrlro.ocnlar lr:einlvs .ueh nn bernrepoetedn inshOrt-ternr 2.e.1 x:eltl a Gp41e.nCprltb or.retinal Lhrmuhonis as- Imns-ICntl', uKrSVf'aral CUalLraCrpLlvr.: sor.inted'n'tth theuse nforai contrncep- allLLteu^h ortr study rclatr•ss rlskwtth tlves. Dtscontlnne oral contraCepLiredpcati0n~ Uf u~enf ihe. M Clrrnticf mediaotlOrl If.thcre in wtranla,incd.:sud- rRCf.i 7nr..n'hilo brratlce ndclt4mA Ir a den orCradual.,IlnrtlAl or eonlpicte Ines- larc IC~iNI,~ it sh0uhl brcnn.iolrrrd. Inl of ri.-0tun: SudlliCn onset.n4 Ilrnnlttsiv.otwurncn'pse-:tine Abd.omntal'.nalnandl dinlopla: ))npiilcdrnnl: or rcLlllal rnsPlll- irndet'lless, oL/ftu,uiltal Ill- o.r u/'nTk. lr leclrnut ^ndflulitllLr o.nproiuL 4e dl- A fCw cn.aes nf ht^n!OeCllllllll[ t•arrnlo- aFrln+tlP alltl tliCraitClnll:mrnallrCT.e ltl:t ilAfC bCen n'1'arlCtl in trnrnlen L:rkine 1. Carei'nnnrtp.,lall-LCI'Ifl rn111illlnt111ns r:ll ennlrarz`nrive3.111e~rrlnlf0lY.hillnf aAllnlnL-Lrct1011~nI I•ifllPr tlalllnol nT splv- Ihr.c rITIIR3Ib Ihivl,nC Uf~ln+lielnanc;"i. IhelldCClrOxCltlincrrLain:r ul+nll!ncclrvrMknncnnU ll/-.tiolr, a,ct: 111CISCNIrrIV•J nf.r nQ. . t'rr 1 rrrmm'nr.r• R:rflr, nr:i•Y'/Y i Ihrtlhrnslirrn'Ix:nrtinl::ntttrlirrr^Crr- rlfl':nrilrn. nenl. ,tfoflTnnn•rivi.,. Frrnm(e 411n:.r'nl~httlcnrn.rstitur::lr •IUtlv' ftp-vprinn. :+rkY'lr.d. hat'r br•cn nmird IUrlnr'reasC. Tluc!r:.cUl lrntl.lo.•:.I~nrannne:.--bnen thelncidrnPC.ofln:rlmnl:+ryllo.l:dr.ae be•r.lrl+ecnlie.aud.lrro::estatlnn:~~l 'tnd111ullrnnnt.inrlc;:.. dnrtmL rtrl). nr,_nanry r i.ly. niln hninlans. tl.+rer:.rl rn4. nrl.. c.ICUnat:.C L1eCOftelnin.^•.,It li- l:hrrn rlirt-n hae.•'rnrortrd +n'.1nerCncrd rlrA mf r tllotfcm5le+e'+nutdf:r~utrrolndln.hyl- domcl.rinllcarcfnonun:use:i:itedaa..il;lehealllbcatru4analuteroidalalro:rn.Iiavcprolorlyd ttsC uf exn8elous Cslrogcn bln nn.hitcrcascd r1.akOf deralouin^ in latCr reare.cllfG:SIFr..VO[- ~. NO. f]b~rUr]e~r, prreM.tp S, I.Sb ®' O.els on/Y ® r^^a.,ioernrnn ana aliqrllon A3-3
Page 288: azo61e00
285 IplontF3 ( Ilre afortn.ol vagrnalor cernral cancer ruosn. ~Iany cliukinna recVmmHd 3 tlvouth bieetluug: siwtture and anlenor- ul:tt Is ordinarily exLteemely +mre fRefs. 1TOllllu. riltaa will probably be quite. \ riable. 31 and;3-:. Tluls risk hne beenntimated Tha ad~nliui.tiation . of prolMwt-onty6leeding irrcyularittc.OCCUr mnne :re- ..rsOt Creater thanAper I,000 cst+t~ures r VrOaeain-CatroCen combinaltotu ;u onenUyultlL thc use of Vrucrattn-ONS, ,I:Ct. 331. AIthOUClt LLere t1 nn evldCltl'e Udilrl' IT ithdra`.tA bICCding should nM be ornl Contraeeptlve3 th:un n•ttht11t cn- +tthepresentttmethat oral~rontt-acep- .ect:l.+atcnte6Ll'Cpnancy.l bIIlnLions and Ule:droVOUt rate du^to tivesfturtherentiancetzr:.k oUdevetoV- 8•,GnlIDfnddrr.0isense.. 1urhtmtdiuotts lslityhers inS tnt+ t)'peof'.lnalignsn,r-, s.uchna- Stu+~hes IRefs: 8 and 231retlort1 a women wttha Vasthlotury of oliso- t+rtlla sit0utd be mnnltOred ottlt ilnlrtiru- d•nlbli.lte Of thC rt k' Of suraeally ceu- nlennrncca or stcmfelnry amenerrhen 0 I:lr cnre if they elaet't0 LN'C Oenl lrnnlr:l- fmm-l qall I bladder dtsrnse in rs Of rOVnx~e:Omell'w'itltout reg\ulae cyFlc, m 1'Cnl'.IVe'}, 111atead'OfOLhCrmrtltntlt 01 1".111'OlntmeCVlilY•M.Or eetrOnl'11 f0f - or hal"talClldnt[yLOrtnnnn:uto\'1'lla'.toryor COnlraeeptiwl FOrlbCnnOrC.,u IJIuI,1 p!t- II'~ 1'raR.. to berolnCnnnCnOrrhelaaftCr:dl~rOllnnm- t•e1lta~e.nlattch~OSSrpsad w+nlen +frpm T..or! CnrlrrrA4drafc and Litnd 8fC(nbohc ationofertultontracrpllvrs-lYOusenLVIEh 70 in '!01L a haM- hecn l0\tnd t0 have Fd' !'t•. lhCt UfCCtLSIiuC Vrohterns should bx vd- Y!IAIa ialYd CCr- /\ draiCJSC In GluCne< fOLrt:U+re 1115s visld'Gf Lhi3 pOaSLbtlity au . e'enr0.u1'uE!YI eprlhcllnl rlfaneCR of tile t: vi<:Re/S.3+thrwlgh7nt.Alli+onlClllitl:e brctl'nlMrll'rd''in~aSictuflCantVerGe+a:.ea' 191te Otltlr Cpntr]rCpttre melh\rde.chanCes ure hlstolOgicxllylteniun'. IL Is noG of V:LticnPS. ot ral ConttnreVth'es.. FOr. I l. ECfuYicPrrOna+ t¢p. Idtow'n ahetlierthts:rondltinnl is a pre- thta rruawl. pledl3blLlc and1 d'+atrclic ErlOpiC as well as Intrauterlne prca- n+nnr of v-..elnat mell;mtnrp. 14ltILonCh Vaticnucshou4ds be csrefuuyy olssenedItulx), n v occur in contraeepttve IzLL-sin'lilar data are Ilot arvfiable n'ith'thr n'hlur rrrci\inqoral colttraceutlraa. ~ n. How•erer: uu oral contraceutive fa\d- lut.of oLlter eatrMem. Ilrnn'.nnt be.pre- :Cnn lltt•rease in b'igi)'ctr7des atW total uveri.the ratio Ot'eCCluVie to bttraunetvue sumed:Lhat'the)'w'VuldnOtinductTuml- VlYO<IBholllY,itlxt1116 beln olOCCVedd in..ly,rrClmncies is higher tlian.lnwomt- Inr rhan:b. Intienta rP,'eiviua OI31 eUUlrareullrtti , wtt0 are nOtreCeivtne OrSI COntrnCep-sever0i reVOrtE suggat a;ras.orlitimn ~ Ret: 1+1. Tlm e1lnlca4 sieniltcuncr'nf dir=, tivn, s e Gte erues are mone eRecu+e. between Intr3uterine caL>D tlrc in femnie funbng remnlPC to bedefihed- n' V[lclenUnS.intralLteriRe thanl ectoma .Vx hormonel nnd eongen t 1 1tOmalie . 8 Elrrafrd'DLOM Pr s r. Vrcalv Ltes. 1heIuBlter eetoVle-inOrau- nellhdinq tangtniti\t Lten.rt dt.cetsalid \I nr n.e in.OLUtd V OS"ure Iuubaen tertne ral:a.has 4celu reVorted rvith 06t1+ linlb-retluetlondefeeta fRefa,]Mlhreugh 1elarrtedrlnl untletllx'reeeiritlg oral <on- rnmbmattaLturoducttandnroteslu+-sin+li. i' '1.Oueta~ecOntrUiatudylRef:l2i.lia3 tnlCrf+tilYS- Iu ]omE w ut0: tryVerten- O[Otcortttacepti\'M rvtimnted S{.7-Inld IltrrC.LS, in fiak of `lolu may oCeflt e•ithm a'feue 1nonUtv.Of. 1Y-Bnratt Feediaa: Ilmb-rN6rtiotndefecqln .hnls.e<pmtd olnl rontracepte sC. IL tl'tO I romr.LCeVh\ef 4' irtl:e VOet- In ulero toug Itonnan¢Ss ral ronu~ pRt)f:rCoCU.te.theVre lencenf:wo enptttulni peri0d Inte fe c 11 II~Ial't . I reVti.Y.. hOrmatol vfthd al'tertta for th hypt t n o t tc t ', srrc 1ndThere may be a dccm . i tl+e q~ntity. urrCnnney' or, . IlltetmHed't, tmctlt f' I '•ly' Ix no liigld r tha rtl c f a . nd q.uality oL tilc b ' t. n14...F. tleer- Ihtcalened alqrtiunit SOm<rd lhea e<- Vnraltle.grmm of nOnuxers. Tllt Vrevi- nsre- n ltnall LnchonOf Lhe hotnien:d noe.ure.cnere reryslwrtand intvolad bnlr 1@nre in, . usersn increases- however. ailli ~ apemsIuoral cancrocevtivea bcs becu fo.vd.'t~s of tre'ttment. Tlue dnta xug1 ec '.VO ue: antl nt the Illth yc of - t! Iilletl illl tlYe mtlk O[ ou' rs. eceiv- •bt that the k nf linlb-redur.tlolu dc uc. t:o ' d atiait,tb three ttwc tlc ULe~cdru;slRei ~l TI efRet'tt;af fectsin caposed~tetuaa- u somewuac Irn. Voltcd tne'alence irlttle fteatyear .i.e aln., wt the breast fed c•iild'Ita\e Imt . th, noneinl:000titrrUirUU: t,ata0xlnrnalycorrclatedaltll~ttie.d0- Irettdetlrmined..Iffea.tnble,theltre'of In thlVVt;.fcmnlt sec hOnl.on0a liave lt•loyn+nlt ofh)pe[textslo~m Orut ton1- Oml eontr.tCepcL.ves, shuuld be. defrrna tmCCpti\'C ILatPOmin 1~0 prCVlwUit', UIIr11LIYtilllnntItl3ilElnM1\'NI1Cd, hren 1Red dur7ng prrCtl-'mf. in nra at- itale had h,rptrtewlon dunnepreCnaluer'. LrmVt_ to treat /ldrCattr•ed Or~ hnbittlal'l nr-9v~ be ntOre IikelT to develop clcratlwl Petc.unOns I uOOPt:en. 211ert h I:OnsidttaU'4 <riilr+:ce o4 byo0d presurte. nhen gh•<n orns rontra- atnr.a 1' Ltvtestrogtru am IeuL/eenve for Lheae rcntlvesIRef.=B+. IndOentiotu..nud there u 10 eVldtntx9L flendatiae. 1i,ACOmpleterrtediWland fatmlyhuy frw w'cLL controiled studiecs ll)at In•q-'I•tte onsetorezacerb:rai0n'.Of.mi.rainu loryslioluldibt taken prioc.tq lile inutia- mLittl aft effeetlca tbr thOe usa., or devel~oVment OT headache tlf a tle+v lion Cf ~Oral CNntraceplives. ThEpfNreat- Ttltte is sOme eetdence that triuloid-+' I,nttern which in recnrrent. pcnutent ur tnent atLd nerbdic.physical exatnlnntiwu and pOraibly other types of p0lyplnidy are, strere.. renuitm dieeo0tittuntion of ornt sl+mdid tutclnde spet+alI refCt'enee t0 bleDd Inere.\Sed. among ab0etuab Irom womenontra~eptires, and crahutinn of the Vressurl., breaats;, ebd6men and ptlela 'rttobeCanle prCgnlntanOn afLle CeaSing rauYe. or[au3, inchldiliaPnp.an1colami sn1lar Oral e0ntraeteVtlvea. LReC .1]:1. E,Tubn'oa 10: Dlerdirlplr/emNariLlb. and.rtlec' nt laboc.+torytests-As a.4en- +rilh the.a altomnilb erevtRUal]S.adn'n',). Lii'ouktltroufh bleedL+q. aVottiin:: and. [ral rule. ooal ennLrircrVtives..dtould not al»rted spp+Ltane0uslY• WIrether ehera Is tOrrhenare trequent reasnns for pa- be Vtescribed for lonser tl+sll ) ye.v - an Over>ls irK-e Vl spOntarleoluw abOr- atltss di.eOntntume 01'at COnttaeepttvel: n'+lhOut another phyallal eSaminalinn tien of pre:nattcies eoncehTd so0n attee In EreOkl.ttroual bleedlne, a•s. m all casn Te+nC'/MrEOnned. stoVVimg 0-1 routtateV6ivesis ullknow:n. nf, ityraular, bleeding frnm tho engtnn: 3Unaler the 1n0uaiue qt estroCen- It'ii recommended that for,mtyPatienC nOmfuneti0nnl cauaes xhould'be bttrnefl1 urollmtogen . prepantiom„ Vrcexlattltg a•hoha•Smiseed.LcrOcartseC+utircVCriwl., iuld: tn w!diagn0aed'perslatlnt Of rr- SttcrinelnomyGm4tanlny.lilcrcaalianrre. precLUlnryy si9orud be ru1M out befols 'etnrrent atntormal bieedi+vg from the ra- 3- Pattentss clth a histon ef ps)chicl eon4Lmuutg the eontrnceptlce r Imen., g+nu:.adeqn>tediaCnostie meiwrce nre depcexsiwrshould be ca efu4tyancen•ed IftllelkatientltaanntxdltcrMtotllcucC- indicnlqtltoTllaoutUrrannn0y0emoliC- altid ihedruC disconttnued4f dcpreulotn " seribed schedJtle, Ure Vua.ibit.ity ofpree- +L+ncy..L patholOqyy haa,been e.clutled,. recurxto,as sonous.drtrre: Pslicnts be- ,altollld lir COt1lldCred at ltletilne tiltne Or a CI1nnCeLO an00her fOnn.ulatioll[oming sigilNCllttly dcpreSaed whlle of'Ithe flrs[ mLced'~penOtl. and turlhersoln'q tha Irroblem,ChOnCinCtOantukiua nrsllcelitracriulises sliould st+Nr ueen6otr.llconttuetlstleN.houlidbewltll- or:dc0luinrcVtiven•ithahieheru'ctrogrntlw medir-ltion and ur ..afccrnale held unu5 ure~unnt•y hes.bern ruled.oatU.d cV+ntenti whilepotcntially'lr<efld in min1- +nethod of centracr,)ttonur n nn,atlesnut IIprt.^aatley is COnCimted., theFttie'nt InlZilre mettstrual ii'regnlarlty: should bee lo dcternltlneiflhe.Tnuutam.is dntg ro- .vhouldlx OVprised ol EI:e VOtrntial risL,< do+uc only if necexvey sinrtethi3 may Ln- lall.tl- to thc fetua nud the xdvt5ability,af cou- ntOSC Lhe rlsk of . thrnmbbembOhc1.. Oval conttacep,ticmmay cattae swne tlnuaUan of the VCCgnaecysl\o~uldl.be dilt- dba:tsc_ tlegrce Of'fluld retetstlq::: 17iey sl:oudd be vaaed'm tile Ilght qf thbe rlska•: FOI10a':LlgVarapraph LO he ii+sertrd for. Vreseribed with cautiorti and on4yw'lti( C Itt ilnLW, rteommtndtd tIL t' x'Omen prOqtatln-ORI)• ot11 COlltraC<ptives: . careful m0tntqrmg, 19 p'Itientl with Cwl.- +c1Yodlscmnthmeorlll'eonEn<epthm3\vlth A+Lalter3tloninmGnetniatp6tten1i13 dltlons ahichimhht be 3C."LSyatld by the intent of'.becunling Vreunant u0e:an likelY LO Occur ut w:onun ualre prOgestm- ftuld relenltien.: suct+ as ronelnlsla dls- SItGnY•'\t!fCtm of /:OLILeatepOlCfl for 3 otlly oral CnflLrgCeptLyCa. The amount orders. u11gr>Me vyndrollle. or I.rdtlt Or / uertod:of ULlte befoN s:ttemptlne to con, and dllntion 0f'M1Oa•f cycle length. btrst- rellal inaulticlnlay. :e0sr.1 [e0181ts, v0t -1• n0. ylLtlliESOIS: 0!<EMIft. r; leyi , , TI
Page 289: azo61e00
286 3. PaLients with a. past hulory: nf. Jaundice during prtgtlancy hat•c an ln- t•rcased lisk o[:rcr.urrence of~iaundice wllit! fttHt•in4 ornl cotfiraerpllt'o tber-apy. It jaundice devclopshcany patlcnt recaivilitg, sucit druGn, the medicationshould be discontlnued. g. Sternid harmoitrs. nta.' bct+norlymc4abblized In na:'aenta with Inslt,tircd1 liver function and shnuld br aAtnmb- lered with Cnutlon in Suc1r palleRlS.. 7, Orai cOnencrptit'eLLce1s n 1.}~ liate' dbtturbnnces in normal trqutopltnn meL belicnt whlcit,may. result in u rcla- livc pyritloxine deficiency. Tltc elinieal ~ simifieancS of tliis le yet, to bee deter- mined. ¢.. Strum folalelevcls maybc dc- pres+ed by oral emitraccptive therapy. Since the pregnant woman is precltsposcd to thedevrlopment of fdlnte deRCteney and Ui,: incidence of folate dellciency in.- crla3ess with Increasing ¢cstntlon, It ispossiblethat i( a u:oman beconyespucg- nant shorUy n(terr ateppin¢g onl ton- ttt+ceptivm,sjlemayhave a. crcntee chance of developiutg(olate deficiency and complIcatione attributed'to ttyls do- Reltncy, g. Tiyrpathulc¢L+t ehould be ndvised of oral cnOtYdcaplivQ tl.llrOpyw.hen relevant specimens are submitted. 10. Certain endocrine and liver tune- llon tcsta and blood'componenta may be fefQeCted by estrogen-centaining oral ^.on- tracepllves: a.. Incresacd. sulfobrwuotoh4hnleyn r!- tention... b. Incrcased pvulhlvmhin and (nctots Vih Vllf• IX,.M nnd! X; d[cren.+ed;. nutl- thrnmbhs.J; tiureasedd norcny'neuhrtne- Indueed'plate let n ggrcg,'1 bil ity. . n Intreased!thyroldbinding cldliulin tTHG/ leadinCto incrc +cd circntatiug lotal thyroid hormone. asn mcnsured byy preteht-bountl {odine tPBII, T4 by : coi- umn,: or T4by radii>immunosts•ny..Frec T0resln uptakeis decrCasM• rcilectin¢ tlta elevated ~T73G. (ree T9 cancentrntien L+unatt<red's d.. 1]ecreaaed pne¢'naaleddol excrCi:ion. !. Reduced respomse tomeCyrayone Lost t•irO.MaSrON Ft,. T11LraTtrMT tgeePntient Package Insert belon, 1 : omc tnresenona IManufacturcrto'supply'InfonntnttonOn cilniCally Si¢n~cant drum-Erng inter+ sctlOnsJ esserrroeenrsts See tV'nrniings section for inforntation on Lhe t•arCino¢eniG pUOUnmUa1 Or oral emt, tixeptlvel.. rsrcneN¢v Pregnancy cntegoryX• SeeCon4raln,dkattona and Wamt'ings. Npsnrnn tromceS. See Warnhngs. ApVSnst RrACnoMs An tttereased risk of Uie follou•iug aeri- eus adVene renctloru has been nssoclated . wiN ILe use oforat coptraceptlvea.laeeWaeainysi : MQncEs ThrntnibophlchltL . P,Wnsonary nnlblL.m. Cnronnp th]ornbosas. Cerstyrat Uiranvbo.+i+. CercbraLhemorrhaae. HyprrlCnclom. (i:all b1:nlderdL casv.. CnngcniLlcl anmtyrallks: TliarGL+ cvidenc1 of alt Llc+nt•i'n I6•n bc- L•.aCCfr uu'following rundilicn.a andlho nsee of'ornf nontracrptives.: althouch ad- di L[nrtal i:olvRntta tory studllt's areneaded: Masen4erle thminbosLs. Itt•uimu ItellKttomas. . Ncurc-eeulnr, tt•siona.. cy.. rceinnl tllrombbsi+ and optic neuritis. The.foflonvillt adverse rCacllOtu harelyeen reported.hypnttents recelvinqoral cnnlraceplivess and are believed: to be drug rclated: Nouses, usually the most common ad- vcrsc- Vnmiting; re.'1¢tion: ocnrarinp In np- ptoatmalr,ly 10,^. orlCSi of pallents dtH- in~g llie R'rst.cycle,: Othcr rcactlons• na a geneml rulc, nrc accn rttntclytess: irt- yuently.pr on I y. occussian:d ts':. Gastrolnttntinall sy'mplnms ~s,tth as abdomin•,licramwl andbloathngi.. Dreaktllrough blecding. 3potting• Chenge.in menstrvai flow. Oysmcporrtian- A nlemnrrhea, duling and after trent- ment• Temiarvylnfrrtility'after discontin- unnce of treatment. Fdema. Chloasma on melasmnwhk•h ntnypcr- sist. Drea-+t rtiangen: tendcntess. cnffirge- nncnt• and stercl.ion. Changc in aeight'. tIncrcaae orr de- crcaxel. Clmnge in cervieal ero:lon and cen'iral i Eceretion, Pauiblt diminutinn in laention n'hen¢iven Imtnedlatel'ypoatpartu- Cholestnll0 janndlce: Migraine, I:tcreaae hy si3e of uteriile ICiomy'o- mal.'1: ILUIti ~alfc,^,ricl. atentnldcprexsion.. Reduced toleraneetofarbonydrlles. 4 aginal e3ndidiasis. he foldowine ad-ne rencciotts Iinve been repdrted in usen df oral conlrarep':- tive.+, :ultl tltlcsssoclal•ion has ottn neither conDrm<dnor renrtcd: Premelmtrual-tike syndrome. Ittlo(crance m cmuact Icn+CS. CtyanCe in corneal curvature 'sterpen- inc>.. Cataracts.. Changes in libido.. Cliorca (•lYanges in nip pCt ide: Cy-slitk.lika sytixiratne.. llrminrhm.. Nen'uiuucsi•. DiYlttess. Hlrsutisuy. LosS.o( etalp ttnir: EryLhenuntuttLfoeme• . F•rsthetna aodpsw,y. Hemorrttagic erupuun•. V:1^.initi+. Porphyrla: Lnttatfed renal functlon. !t'ct+r[ Oegaposc SGrlous 111 ct4cctshare luot'hten te-imrteti 'foltott'mg SCJ4cl$ge.+lloly:of'lar$edu.+et of eralieentraeCntlves.by ypunq: rhihlrrn:.GVerdosacc nray c:trvve nnusCa, and willtdda:+ral bieedintt.nvay ocntr in(Cmalc.c DU.t/r.GaMn .leTltr:laTRaaroM Tn achicve. maalmu'nl, eoutrr.erptive cRl+'Livcnrcq , iuscrt namo u0 drnW i. utuaL; be takrn eanntly ns direeLCd and at In-t!n'aU nOt C'(CCediltg.^-1 houra: -bCtunfaeturcr to sunply infonnntiot'n otti ruutiiue adminl:trallon :utd snOciOc In-struaLinns on Ital.tdlhle yroblrmx.such asb'ceahttyrou¢h bleedtn¢, amenorrnca. [IC., usc abomt eontmcenuvCS ,in ute Gvent: ef m micwd'` menstrual pcriodf ~ " 1. IL:lhe patiettc hasnotadlterrd tW' lhe prescribcd dacasc rteitincn, thc tro.+si-btti6y,of pregnanc0'sktould be CorLtidcred' n(tcr Utc fltst miiscd pariod:.,ee aftrr 45 doys.front the last m n+tru0lipcri0di(. . the proge5tln-Cnty, orall eonlcuep',tivGs t;e used t'. and oral contcaccptiaee snould witlhheld until'pnE.'nanqhas betrlruletl ont. 2.:1I tlle patieneitas adiucred':1otliepreBCribed --inten and. miMlS LY:OcOn- seculiCe pericds, preplancy sltenld. Ire ruled out:b'efore continuing the eontrs-erptive regymen_ Hotv. 3orruse -,M nnnfast:xento sunply' iu6orn-ti'un, on aca.ilablC daage (ornys. I:ucent'y: colur: , and pactaging.) , Rrrucwca 1.NUCa. E. R. fcV.l• Tesru.,nM. ul'.Osne-cNmry: lYtnllnr+nnd Wn:xt'na, Ia:3.. ]. lumm[ tV: H.. u•, ve+ces• Ll.. Wr+lrr. hnlnc, e U A. EnyetYnd.,"ILra~nAOem~DOlYc dlaeve nd tb. it.NIQ.L can neYM1 O! aral con- S.aciDU(eii. tA repnrtloIne gn,mut:re 0n r a gs," nrlr art•J J'S::n]:-EO^. ]sr0.. ]. Stolle+: P. D.. J: A. Ton~.e;:n. yl. 5. To~k- n, P. L Frra'H^,.1: It RYiI[dte. and. V. P. JMMS,'. "TnromlMFif wltn Inw.entrp:rn r ,es::.tm J f•r,~n:~~,..r:maasrt iue ~rv,`s.,n .: ttCrol'ennree er c n<„lil-..~tn~~nt.r:. ..orv evntncrDmm~ m,m I n . m•r.-ul mn. .P a cofrcan r.mrYl-t'm_app. tson s.. tmm.n, w..x. W. aYd 'a r: Y . r'Im- estl;;nulYn ar d , I„w vy.+nQ e rn/~t• mtao+,a a c~lnYenrlnG a,'•+:,'Lrl'~eQI 4 nrr !,i xtg~-t'Ya~ en".E.6. rc-or. yl ~ r. .nJ R rn,n, 'nnvc.rc'.,uon on rr c ~t o urptl,e. n„d 1tl mmnatnnnnt,e d er A'n,etMrre- . port:' ono JrrQ,J a:ecl!ia'• tecm. A. T•y~Ut.: F. l). .Vtl~ra.. Wn~"rn,p,~Y o LL^n4rrx:. . Urmmi,awal . ~rnlrnud7;' An,~J'fY~. s., Nwtnnb.Gn+nnormure un~e a,~evcunan. •.. tnntM ,onedl.~ ~ratrn .' ,o o ¢adlYtM1dder1adet.e ma'Erea.CW.nYr , L¢n- ter 3:13'10-tioi.'ln:a: •l. Con.ltwrnOnOmuD ' e StYLLT nt 6troke InYOYne \+'nnwe• "OrU..coi,trn~rut~on .nQ mer<nem rna oc'cereen,t i«nen,,.+ o. EgoEEAI gEOlfiEg, VOL• !1„N0. 1]t-rbEyoAY• bKl•weee r, IYre
Page 290: azo61e00
287 APPENDIX 3' Final Text For Physician Labeling 9 oaaL QoNT1UCa'rtve Gatiltre Dncurrton (TO ae 10rlitRD aT MANUrACfYapl) (Desertption should Include the f o I Yowth[.Intcrmulen.) I. The pnoprllLafy name and the es- tab[IShed name 11 any. r/f . the d[ug productf 2. The same.oualitatl.ve and/or Cuam tltatite ingrt'dlenl, lnformatlon,s: re- OuCed lor labcts;: 3. The pharmacoloaical!or lherareu• tic claas of'the drua p:ojuut: 4. The.cnemlcal rrame and structural formula. CtIilleAt ptiAAMnerLOCY e0e tOa(sprAnOrr OaAt colvTanci'.RtvL1 oat.r Combina»lon aral cr~ntra¢rplirrs act primarily through thee mrchani.mnftonadoltcplnsupprrssion due to the eatro6enic.andprDCeslntlonal activity o2 the Ineredients. Allhorrch lhe prl- mary m~echan•sm of action is inhltlt- tloh of ovulatlon, altlratlons int1ien eenlLALtrnct hnludlrc changes In the t)for.other nrram of rontracrottmt cen'Icat murus Iwhleh Jnrrea.u the dlf- The efflrnry.of these mentu of con:n- fteully.ofspM-prnetrnllonl,andthe ceplton (except theIUD) depends endomelrlum (whlch reduce the itkNl- upon the de6reeof'adherence to lhe hood of ImplsntatloN maynh•o ton- method: tribute to contracepllve effectlvenes. TARi l~ PareNaNCte•s Pex 100'WoMC..Ye.w ILID. less ltian Lt6:DIlrhnamwlth spcrmlefdal prod-' ucts tcrearns or 1e4l1es4 2-20;Condom. 3-36; Aerosotfoatnl.2-2U; Jellles and eresnns. 4-36; Periodic abstinence (ryttth®) all types. less than 1-17: 1. C1!endar metnod., 11-47: 2. Tempcrature mettiod; 1-2a 3. Temperature method-Intereoure only'.In post-on::atery phase. less than 1-7; 4. Bfucua method.. 1-23t No <ontraceptlon; 60-R0: Dosc-Rrurco. RlsNor TxaoNSOCVao- LtsM F1loM OaAr. ConzaACtartvct T!a'o studles havee sho.n a posltlaea::.;ociatlon.betwe^_nthe dose of estro• 6ens.ln oral CoriLraCeptb'e3 and thee ris4.of Thromboc:rrbolL•m (reda 2 and 3). For lhs reascn. It is prudrnt'and In keeolnewttt Roei principles ofttiera- peutles to minimize exposureto estto- een. Tite ornl'contraceptire produetpresc::'xd for any pven patlentshould bee that product which cont.Wts the least arrtount of' estrogen that b compatible with an acceptable pree- natlc)' rate and p¢nentacceptance. IL bs reeommfndad lhnt new aceen,lon of ornd corntna'pih'rs beslarted on D:"i sratlona containing 0.3 ma or less o:estrotten. CONTMIXotGAT1ON1 Oral contraceptlvess should not be tr.ed In womrn wltli.any. of the dollow• Inp, enndlllons: . 1. Thrombophlebitls or tnromboem- bolic disorders. 2 A pasn hlstory of dcep rvcin,thrern- bophlc`Itlsor thrombocmbolle dlsor- den.. 3. Cerebral vaseularor coronaryy arter, discate. e: Known nr suspected:carcuwma of the bre:uti. S: Kr1oaT on suspected estrogen de- pendent necpizsla. 6. Unduacnosed abnormal Cenilat blce'Inr.. 7. Known or suSpeCted pretnr.nncy', (cce warnlna No., 5). Crn:S 3 pre':flancy.f.tle'Of'.~ ptr. WAR.YI:LGr 103 e'om.tn-ye3rs. 1Nnnuta<turer t0 * addp.1li Illfolmat:an re::tCd te IIYCCi[nrctrr.moktnelnrfea6etttte.rlakef!u~ p:em,mney. mtt wi'th It:f p~rtllu!trr rro~.u carYt.nrc<nr slde elfrna from arA4 with pamL:=t: 11 needed tb pruvldTadcq.u.lic.o'rt. ^rro-l.e us . Th tnere urimsre prc::nb:na In:ormation to tlie pitY11- ` n. rtn hearr~ '.r 1.5tlottl • .lomen ocer IS'yean ofskar~t`'amen who T.ible I cj'c] rnnarn of prlan.tncy. eral~e raeeptrcer rn-d be ar+onatrratea reported h; '!fe IILCratnre (Rnf. advneo'nCl loalnek'e. rot raoepTOCtN oaAt. eDNrfUrrmrn Oltti The priman' mechnnllmm through whlchthusert namcof drtrR) prevents eonceptton Is not known. but yroecto- eenonly oontracepth'ea arclUlo- lo alter the cervlcal mucus.: exert apro- 6estatlonal effect on lhe endome- trfum. Interferlne wllh ImplantJ•tlon. and..ln some Datienta, suppnm ovula- tlon. LManufacturcrr to Includelnfor- mxleon on absorptlon... (Ilstribtutlon. eltminatton.: andpharmaeok7nctlca lf pertfnent: also ondruatnteraetlom perthient to 11um Sn u0e.) IIIDteAT10Nf AND USAOa (Insert narne.of drue)~Ls Ithdlcated for the preventlon.of preenar.cy.ln womenn v.'ho elect to use oral contn- centives as a method of contraccption: (Idanttfaalurcn who have other ap- proved Indleatlens for.oral eontracep- ttvrs (EuovldS ma. Ortho-Novum 2 ms. Ortho-Novum.10mg) should men- tlen those indlcatlons here.). Oralicohtraeepth•ea are hichly'•tfee+llve. The pre¢nmtcyrate in women using convenllonal. cornbinallon Oral contracepllvee (conlainin7 35 mcE or more ofelhtnyllestradlol or SO.mee ormureof inestranol).IS generally rcport- ed a5lCr.t than one prei'ILancy' per 100' -woman-y'enra oQuse.. Sliahty h10',hrr ra.les tsomca'hat morealran.l prcanltn- ey per 100 woman-yean o2 use/ are - parted fur somecombinatfonpro'-aets tontamin6 35 meg orl8sa of':ethinyt'as- trad:olf andd rotess on.the order of.3' preGnancies per 100.women-yearaare0 reported:for LSe proecstoaen onlyorat cJntraccptiva.. TheseI ratcs are derived frem sepa- rate studlra conducted by different ln-vestlia_ors In sevcrali populatlon groups and eannot be eomp;urcd pre- Usely, Furtrermore, pre¢nancyratesy tcnd'to be lorer as clInical studies are con'Inued, posstbly.due to selective re. tent:on In.lhe longer studics oUthose. pl. tents. Yhoaeeept the trcatmentrc3iinen and do not discontLZUe as a result ol'.adPene reactions. preCtl]ncy. or otlier rcacons. In cltn:cal trtalas a'Itlt ~Unscrcnamc of druR)(LVert'number of) palitnts tvm- p4cte-- c)'cles and. a t;l tal ot, -- precnaacics were reported. Thls repre-
Page 291: azo61e00
288 Thense of'oral aontraarptlves Is as- soe/atedwtth increasedri•k of several serloue condltloltl InGltudtnt tttlont• boembollsrrt• stroke. myoeardialinlnce- llon.l hepatio adcnoma, gall'l bladder tllseasu. hypertenslon.. f4aetlllonerss prescrlb6la oral contraceptives shonldb0 (amlllar wlttt.the following inror- maaion relating tbtRese ritks... 1. TArombogmbbAicDisurdrrn andOLlter Vmseufar Probfern.T. An Increased risk oQtllromboembo)IC and thromho- t1c disease aa•soclated with the use of eral eontraceptires b weil established. Tttreeprtncipal stud(n lft Great Belt. aln IRefs. 4 through 0) and three In the Untled States (Rrfs. 1.through 10)" ttavee demonstntedan tncreased rsk af fatal and nonfatal tvnotu throm- boembolism and stroke, both hemorr: hatle and thrombotle.: These studi. estimate.tnat users. or oralcontneep- th•es xre 4'to 11 times more llkely.thattnonacen to det'elop thcsedJatssese rlthoutevtdlna Cause (Table 2). Caaanovsscuutt DtsonOnts In re enllaborath•o Amerlean study (Rrfs.9and l0)~of eerrbrov:ueulxr dl!-urdenan women with Zndwlthout ore- dlspostnt causes, It wu estldlated that the rLsk.of hemorrhaglestrOk'e nas 2.0tllnes greater In usersthauT nonuscn and the risk of thrombotlc stroke wxs 4 to.9.3 timcs gveater In usen than In . nonusentTabl! 2). Tuta 2 sCrHl•tar.or atL<nra a1sN orrxaoMaorArta- Lre e aras oTn~ v.aCa4a rroe- Lrfta I~nML CONrate tva aW <GK- aAaaD TO NONOSr71f a.ntiH 1+~r taawtnutnraveeemeeikaueue... _'w.a~ rea ru.arrr tnrameeemaeue Hmmwtwu. .:e nnq.,aeur,~.ow ~~.._ .-as Nern~nn.rv scMe...__~__ r Mrr+MwIn(lnttan_ 1-I1 1 r+onees aa_3 however•.were found to be a clear ad- degreee ofthrombuembollt ri<k atsocf- dltlonal risk faetor.. pred with proaeatogen-only oeai epn- y1h terrns of relatlverisk, It'haa been trsceptlresraa•e not beenperformed. Tsllmatrd IRef. SAL~that oralcOntra. Cases of thromboembolle d6seasetfat•e crlltlveusent'S11fdonotsnloke(amak- been rcpurted In women using these Ing Iss considered a maJor predisposing prodJtcta and they.should not be.pre- tondltlon to myocardlal mtnrctloro are somed to be free af excoss risk. about twlce at likely to havee a fatal mypcardlal Infaretlon as nonwsen x•ho Esnaun or Eaxns MoarauiY Fhorea do not smoke. Onl contracvpth~eusen ~, C)ACOLAZURi D[aGSif v•ho are also smoken hlve about a S- foldl.. Increased rfik of fatal !nfarctlon A lar0e prospectlve study (Ref.. 53)' compa,-edto LLtera who do not smoke• Carrled outIh the.U.K. estimated the but about a 10- to 12-foldIncreased nroetality'rnte per 100,000 w.omen perrlsk eompared to nonuarn.who.do not Year trorn dse;uca of the clrculatory amoke. Furthermore, . the antount of sys°em. (or usen and nonusera of oral smoking Is also an hhport_r1t fa<tor. In rontr3cepth'es according to ace,. smok-dctermining the Importance of these mghabits.,and'dur.ttyon ofuse. The rclattve rtskl, hov,-eeer- the baseline ot'e~l excesl drttth rate annually ratrs for vprious ag0 troVps. LSlho4n from clrculatory discases for oral con-ln.Table 3, must be tiven serlous can- traceptlve usen.v•as estimated to be 20 slderation: The hnportanee of'otner. per 100.000 (aCes 15-J4-5/100,000;. prethApoaing condittons mentioned ates 75-44=32/100.000; agef45-49'- above in drlerminang relative and ab- 140/100,000) , the rlsk being conren-sohale rirk0 has nol. ivyet D en tluanU- trated In Older women, In those with a tlcd; It is quite likely that the same lortt durtion of~, ttse, and In cigarette sytlergistl0 xcllon extsts. but perhaps smoken..fcras not.posal~ll, h0w•rver,., L Ih V h th ' to.a lesscr extent. Turs I tdtanated annuel marlatttrnteCCr t00•000n fmm, mreeard1a11eWarrllon bY'uie oe ~ oral'. c ra<eptbrn, rmGkingg habtta, and« etmJeanr. ura..metmtvniom wa,nrn <rN' wom<n yN )s-sa. .a-H . SnoklAr n.Wts Vwn Nanu.en Vaen NonnHn torsam ne terre ntlons e lbs of f age„smoking•.and duntlon of'use, nor to comnare theef(cets of conllnuous venus Intermlttent ueC. Although the study showed a l0-fold IAcrease Inn death due tb. circutatery dtseues inn uacn for..S or more years, all ofhthese deaths oeturred In women 33or older.. Until larger numbers of women under 7Scith continuous use for S ormore years are~ avallable• it Lsnot posstblb to asses•s the magTltude of', the relative rldk for this younger age group: The atallable dsta ..from a variety of eourees hnve tcen aaatyacd'Lfte(. 14) to estlanSte the risk 0f dlath LvGCilal- ed:wlLh vartoua methodsaf contracep- ttOn: The~ estfmatts of risk or death (or each method InUOde the combined risk oftne contraceptive method'(e.t-• th'romboe.nbnllo and'thrornbctlc dis- eate.in the case o1'.oeal contraceptIVes) plus tlie.rlsk attrlbutuble to pret/nancy or aborttonIn the event of method fatluze. This latter rtsk'vartes with the effectlvenesss of the ezntnaeptive mcLhod. The flindints of this analy'sfs are srtownln Flgurc 1 below LRef. 14):, The.study concluded that the mortal- Ity assoclnted wlth, aJl methods of birth control is lowandbelow.that as- sociated wlthehildblTth, with the ex- cc7xion of oral ttntrnct•ptlcca InN•omen,ovcf 40 Who smokG.. (The ratea given (oc pill only/imoken for ea.cn age troupare forsmok'en as a e7ass. For "heavy' smokcrs (mare than 15 cigarettes a da6): the ntelatvcnc'ould be about double~ for "light•• slaukcl'1 tlrrsthatli 15efCarrttes a day). about.60 percent.) The lowest mortality Iss assoclated xlth the c0ndom or d)aphngm backed up by early abortlOn. AttameLm___ It1 1.e aaa u.e a.t Tn N<wnaien.~ l.t t.i le.t em r,.n.,n..<n._ a. ta 12a laa u.s s.r t.t tv •xHn..m.Lrrts .r ,.e.. n.L.atH xr a.r. rro,n J<N. AIL a1WIH N r\mer r`wiLn. aae•. tJll. Rrsx or. Dose M o I r uttosAa trr,ucrror ' An lncreasedrlsk of.m~yerardtal l.e -` farction acsoclrtted wlth dhc usr ofmrnl con!raceptiveshas buen, . repor•ted1Refs. 11- 12.and 1llt, enntir,nmg a orevlouslYauspcct.ri zssOCRatlon. Thesee studies, cur,dueted In thc Unltrd'Klntdom, found. as expertr•d• thi[ ihe tieatdi-th'e nambcr of unllerlying.rtsk faCton for cpronaryartery' dleei~! ( eltaretre. smok'Ing• hypertct~ ion•.. bypercholesterolemla., obcsity. tllabe• tt•a, htstory of preccLmptld toxrmta)tlie higllcr the Mk Of drvrinnl~nn nt)'?• cardlal.i.+tarctlon.re;ard)rss of wttetn• er the patient waaan oral contracep-tlve user or not. Oral contraccp(ivcs„ In an,anmlysls of.data derived from evernl nallonat adt•ene reactton re- oorti.ng systems <ReY. 2)„ Dritlah Inves- tlCators rnncludcd thstt the rlsk. Of thromboembolY.im Including coronary thron.ib0>ia. Iss dire0tlyy related tothe dau: ofvstrOttcnusrdln cral contra- eeptlves prrpanllons containing 100 mt•: nr.raore n( estrutrn were assoclat- rd allli a lnyhrr rlak of,thromeorntbo- Ilsm llmn dtiosr conuhtlnt 50-80 r,.cs of.estroyen. Thelr analysis did sugitest, hnwrn•r, Ihat thn mtanllty of rstrocen nmy'rtot bu'.Ilte salcfactor hrvoh•cd.. Thls fmdlnr:, ha.; been conflrmcd In thc United Statt•s (Ref, 3): Careful.epl- demi0ta0iralytudllcs to determine the rtCCiAl.at6itir9., vOL tl, /t0• 21-rliliioAy, JANVAtr 31„1e71 A3-7' .
Page 292: azo61e00
289 ,-:6 f1onCls r...l t...........,.,v......«.««.e_......,.rr..,..,........-NUao~.....w..-,~«....«. ........... ,. u  a r a a a a M Ir n N ar. ata sa. a ~. rww C1Iy...,.q ® ,...n+w U= © N0. .+. The rlskk of thromboetebolle and thrombottc dtsense associated xlth oral coxursaeptlveshrcreaaes wlth aQaafter appro:[1lnatelY: aee30 and, for myoeardl\I Ittfarctton. It further. In- creased by.hypeetrnslon..hyper--eha lesterolemla, obesity. dlibetcs.or his- tGry of preectamptlc toxemfa and Cipe• cleity by cigarette smokln(•, Based on thedata currently avall- able• the fol6owlhe ehart.pives a s:o.vs, eetlmue of the risk of death fro.m cG-tulilory: dlsorden aaocleted:wtth theuse Of ooal contraceptlves: 6NO[ma. H.ums .en Ontn Amisroftwa. Cenutsro.r_RUS A,toctwm Wtia Dae or Ow, Colm..esm.as Ax. axb. io•:r N~ lo C 8 De. NarNnalen Ina a/M~oer:D C.D Noe.auuenu.: M«a n sreGlrC GB Wnnx~rnnCnla,yA .-tr,-..,.~..:.a .Iw ..rr nixn r.a~ p-Uw..eci.ua.ntn nran r•~.. C-UU ...:f.W .an mo,knu ry.. D-cr .rwrnw.iu m..Nt. The physlc:an end the patlcnt sttould be atertto the earliest manldes- tutonsot thrombocmbolue =nd throm- Dot1e dlsorden (e:[-• thrombophlebltL^:. pulmonary embul4sm- ecrcbro-seular lnsufflelency.coronary ordus/on-. rn tlnal throIDbosls. and' mesenterle thrombosis). Should snyy of these occur or be suspected• the drug should be dleNntlnued hnmedlately: A four- to slx•/oldAherensedrtsk of postt surgery thnmboembollc campll:- wLlonshasheen.reported In atel con- traocp04ve user.,Gefa 18anA.16):.11fe:uib!c, onl rontrncepuvershouldbe dLUnnl.lnued et h-asc1 weeks brforesurgery off e typc assodlated with an • Increased risk of th'romboarabolLsm or ~ prolorrd lmmobtllsaUon. a..t' 2. txalar Lextans. Tltere have been~ nporu 01 nruro-ocular Irstan.sueh i. . optic nrurltls.or rctinsl itlurombo.:Ls ns• seoctaled with the use of oral contra-. ceptlves. DLcontlnue aral conCracep, .L H I Y Nww+.. w ® •~w w w~. tlve medlcatlon If there (s unes- D6ntned. sudden or graduaL partWlor complete loaa p1 Yl3ten: onset of prop- tosb ordlploplu paplHedemc or re- Llnnf vascular lestons and Wtltute ap- proprlatadLaenOstlc and tllerapeutld mcasures. 3. Cercinoma. lan3-term eonttnuousadm:nlstratt0n of ettRer~ natural or synthetic estrogen .. ln certain enlfial specfes Inereases the freQuenc7: of car- emoma nf tnee breast- ceniu,.voa'lno. and liver. Certain s)TllheLle proaesto- eem. none currentlyy contalhed In onl l . eontraeepnves„have been noted to 1n•ercase the incidence of mammary' nod• ules. benign and mallanantIn doF, In humaas. three.ca,Te control stud- Ies hat•e reported an mueesed risk of endometrlal cnrclnolnr assoclated with thr.pnolon"d u;eol exoeenoua estrc- je7l In postmenopaulal women (RefS: 17, 18. and.19)1 Onepubllcatton fRef. rH7e3At rtta3tfe, VOL. U.. MO. 3I-TUlfaAr; ,lAnuAe7. 31, 1974 A3-8
Page 293: azo61e00
i 290 201 nported on the flrst 21 caxes sub- mitted'by- physicians to a re6istryy of,o:aesofadenoearclnomaof the Cndd• mctrlutn in women under 40 on..or.il eonlraceptifes. Of the caxes found In women wnlloLLt presd:5pA5iil(:risk.fac- tora for adenocarelnoma oflhe;cndo• metrium (e.g.- irregular bircdinqat the time oral con9rueptivrs wrrc flr..t 81vcn., poq•cysllc ovarlrs)„ nrarly all oecurredln -xomcrt.whonad used a sr-vuenttal otnl enhtraccptlvc. Tht•seDroduees are no Inngermarkrtrq- No lvldlrnec has. beenrepartedSuCge.ctlnRan Increased risk ofendometrlal cancer In usersof'cunventlnnal'crsmbl• natlon or progestOgen-only oral con- lracrptlves. Several studies (Refa. 8and 21 through 2t)'.have found no Incremse Inn breast cancer Inwomen lakingoral contncept(veaor estrogens. One study (ReL25), however,.w'hWc a4o noting no overall Inc reased rL^k oU breast cancer Inw.omen treated with oralcontraceptlves- found an excess risk In the subttroups of oval cnrnrat;ptlve uun wtthdocllmenled benign breatst dL{ra.e. AA reduced occurrence of benign beeast tuman m users of oral eontraccptlvess hu been w.ellidoau- menled (Refs. 8, 21, 25, 28, and 27).: Insummary; there Is at prescnt', no. eonfirmcd evid,mee from human stud. 1!s of'anincrea[ad risk of cancer aaso• clated wlthoral contraceptlvcs..Qlose clinical sLLrtClllance of all women takingoral!contraccptives Lr, nrverthe- leal% es.entiaf. In all;ciases nf lntdlag- nosed persistent or rentrrent abnor- mal vaalhal'blceding, approDrtatef din g- nosttc measures ahould he takcn to rule out maDgnancy. Women w.ith a strong familyhlstoryof breast cancer or w.ho havebeeastnodules.e flbrocys- tle dlaease or abnormal'mamnvoenmsshould be monitored wtth particula.;r carelf they elect to use oral aontrn-cepttves Instead of other methods off eantraceptlon. 4:. ilepoltc Tumort.: Benitm hepatic adenomas h:ve been feund to bce assso• elatedwith,the use ofoml contracrp- tlvcss lRefs. 26.: '9; 30, and 461. , Ono study tRef. 48) showed that,orsl con- tracep4lve (armulatlons wlthhlch hnr- monal potency were azsoclated with a higher risk thandower potenoyy formu. latlonsAlthough benfgn, hepatk ad- enemas may rupture ar,d:may catl9edeath through Itttra-abdominal hem- orthage. This has been reposted' In shorGlerrn ns well as long-tcrm usere of oral contrsceptlvcs Two studies relate rlsk wlth duration of use of thecontraeeptlve.e the risk being mtmh greater after 4 orr more yeera of oral contnceptls•euse tnefs. 30 and 46). Whlle hepatlc adanotna ts a rarelesloa it ahould be con.sidered In women presenting abdominat paln snd tenderness,• abd0minal mans or shock.. A fea eases of:hepatoccllular earet- nonn rlava been reportedIn women t NOTICFS J2=7 taking oral'contraceptives. The rela. tepllres (Ref. 45). fimbryos wlth dhese tinruhip of thrsedrugsto this type of anomal!es are virtually al?*aysebortcd ms.iipnatlcyis not known at'IIJs.tlme. , spnntaneousty: Whether there la an. S. thein or fmmrdialcly~PrrcedtnD overall increxsr/nspontaneous aboe•Pre4anep..Btrfh DrJrctrs tn t7J/sprsn9; tian ofpre.-nanciesf concetvedsuon and.)foli9nanep in Frmafe OCsDnn4. altor stopping oral contraceptives is Tha use of female sex hormones- nnknocn., both~ estroxenle and progrstatlonal It.ls recommendedthatfor.any pa- acrnts-dnrlr.Dc e,vly preglnsney' may Licnt who htis mLS ed two cnnsecutlve. .crrlonlydimape lheaff.cpritrc...Ite has pertods: pregnancy should be ruled out bd:cn shown that femalrs exposed In before continuingtne contraceptive utern to dl~ethylsllibcstrn4 a nonstcrol- reCLnen.. If the patient has not ad- dsl eMLroCCn; have an InerCas^d rlsk of hrredito the. prescribed schedtlle, the deerloping In later llfe a form,of vag4- posslblllCy o4'preqnancy ahould be oan-l llal or cervical cancanthat la.ord!"-farily sidered at thre time of the Ittat mLSSed rxtrem;cly rare (Refs.]t anJ 72):.Thls pcrtodifor aftcr 45 days from the last risk hns beerf eslimatedtobe of the mcnstru^.l,period ifthe progestogenordcr of 1In 1.000esposuresor lesss onlyoralcontraceptlvesareusedl:and' (Refs. 73 'and 47), Allhouah thereas no further tme otoralf cnntraceptlves ev{dence attltet present.tPme that oral' should be withheld until p~regtuntYconlraceptlves further cnhance thee has been ruled out. If pregnancy ts risk of dcveloping this type of':ma14: conftnned, the. Datient.shouldbeap• nancy: such patients should be moni. Drtsed of the potential risks to the toredwithDarticularcarelfiheyelecL fetusandithe advisabtlltyof enntlnu- to use oral oonvaceptlvts Instead of ation of:the preCnaacy'should be dia- otlier methods of contraceptton; Fur• cussed 1n the light ofthese rislu. thermore. a hlth perecnlage of such lbtdnnfacturrr to.supplyapixoprlate exposud women tfrom 30 to 90 ;1 have Ibfortnatlon foruse.llv endomelrioslsl' brrn found to huve epyt:,cllal.l changes It Is also recnmmended that'women of the vagina and cervix (Refs. 34 who discontlnue oral. contraceptlves through 381. Although these changes wllh,the intent of becoming pregnantaret histolegically benign. It is not use an alternate form of contraeeptlon known whether this condillon is apre-fora period of t!me beforee altempting cursor of vaginal mnllcnaney.. Mole tocoetccive- 6tany. Clltticlans reeom, chlldren.so cxposed rnaydCvclop ab- mend 3 months althouph no DreelSe normalltles of the uroKCnltaltract /nforn:niion.ts available on which to (Refs. i8', 49, and:S0). Althou¢n sbnNar base this rcr.ommendatlon. dat.rare not avai,lablo w1t), the un af The administmtlon of proges6ogen-r ofher estrogetu, it cannot be presumed only or pTOR~rstogen-estrogrn combina• that tlucy would' not Inducee sbhllar ttons to induce wlthdrawal b4ceding changes- should notbeused as a test of prrb An. Inereased risk ofcongenitalf anomalles. Ihcludi'ng heart defects and'lYmb defects,. ha.ss been reported: wtttt the use of sex hormones, Including oral eontraCCptlves, in i preg'nancy. (Rels. 79 through 42. S1): One ease contra) study.lRef. 42/ ~has cstlmatcd a 4.74old Increase In risk oflimb•reduc- tlon defocts In Infants e<pueed'in alero to aax hormones (oral cnntracep- tlves. ltanuonal withdravral, tests for pregnancy or attenlptCd lrtatntrnt'for thrersten¢d'abortlon)... Su'me of thcse exposures weree very short and In- valvcd only a fewdays of treatmenL The data su¢gest that the rlsk of llmb- reductlon defcRa In exposed fetuses Is someu hat less than one. In 1-000 Ilve blrtlv. Inn the past., female sea hormones have beenuaed during pregnancy In an attempt totrcnttnreatened:or ha-bilual abortion. There Is cDnaldcrablaevldence thattstrogCnaarc Ineffectlvee fer these Ihdlratian.s, and there is no evidence from Well ~ controlled stud6es that progrstogens are effective for thesa usea. - There is some evPdencee that trl- ploldy and possibly.other.types of pa-lyploldy are Increaaed amang abos- tuses from w'omcn.aho become preg-; rtant 3oon aftel ceasing oral contrac, nancY. 6. GaaBEnddar Disease.: Studies (Refs, 8, 23. and26) .repprtan increased riskk of surgicaliy een- fitmed ¢all bladder disezse In usen of oral eontrxeptives and estrogens. In one study, an Inere~sed rlskanpeared aftcr 2 years of use and doubled after 4 or S:years uf.use. In one of:the other studics, an Increased'rlsk was apparent. Delocen 8 and 12 mon7hs of use. 71 Carbo,hvtrate and LtpidlUsfabolie Efhcts. A decrease in glucose tnderanre haa been observed In a sti'tlflcant,perrent- ageaf patients on oral contraeeptlves.: Fornthls reason „ predlabellC and dia- betic patlents shouldbe carefully ob- served:whue.recelving oral contraacp- th•es. An Increue In trlElycer:des and total phosptloltpids has bcen oosr.n•ed In Da- thcnLl -rccC/vingg oral cantraCe;lllres lRef, .l41.: The clinical slgnlflcance of this ftndlhg rrmattu to.CC defirled: 8. Elevaled Blnod Prfssore.. An Increrse us blood pressure has been reported In plllents reettving oral icontracrptfves (Ret,: 26). In aame women. hypertension may' occur wthin a few months of beginning anl! cUntrarOpttCeuSe-.Inthe tlGt year of: use, the prevalence of wemen with hy. ELO[ILt IEdISTEE, VOL 43, N0. 21-tUESDAY,,JANUAIY 71, 197! A3-9 ~ ~
Page 294: azo61e00
291 \{. S NOTICFS l4 Ecfopic P+r9nenrr.. Ectopic u well as Intrauterlne. prcg• naney. may occur ln:coaLLraceptlt•e f611-. Ilres.. However. . In prOgeStoaen-onty orai:contraeepuve lnllurrs, the ratio of ectoplctolntrau4erine prennaneles Ish(ghcr than In somen aho are not rc• celving oral contraceptlPCs- Since thee drugs are moce effecth•eIn prevent(ngIhtraulerlne than ectoplc pregnancies. 12. 8reasf Feedino, Oral conlraceptlcra given In the postpartumpericdmay interfere wlttsconves pregnant shortly after stoppinT lactation. There mayy be a decrease Ineral cnntraccptlves. shemay' has•e a the auantlty and:qttallty of the brtast• greatcr chance of developing folx[e de- mllk. Furthermore„a small tract/on of ffctenry and enmp4lattlon5 attributed the hormorul agrnte In oral contra• cepLlvei has been Identified Intheenllk of'mothcrs recei