Tobacco Documents Online

co.ua aa lun lo-0u Vna uirusrrvtla as1c; p1AN.1(;h@ihN1 _-I¢Ju07 28/09 '99 09:54 FAX 01705798939 ~- 506 Eranom and Heath Epidemiology September 1999. Vol- 30 No. •, TABLE 6. Age•Adiusted Death Rate for 1959 Current CiP,rette Smukers by Numhe.r of Cigaecttee lier Day Compaeed Never Smokera amoog Glifumi> CPS I Subjects ~~ - [.4eth Rate by t.`ause smd Fnllow-Up Period id b f . ~9 C3,nn L.+ng Cznr.er 1959 Smaldne Srams er o ,m Su6j- 1960-19be 1960-1997 1960-1969 1960-1997 Ma1e natenr smotcrs q0+ ciyS2dry 2,617 24.52 23.02 3.307 2.581 11-39 ctg~2dar 4.790 [318] 23.85 11.9431 11541 9 6 Z.t Z 138 0drJdvy ~ ,194 7 7~066 [3.4421 9.2 ~ p 1 L 767 ] 1616 [1,2891 ' 1 59 ] ~ ~ 71 10-19 cydday . 3,742 19.29 t 4 .98Y 0 1 -0 1 4 12,474) [421 1160 lAtipt7day 1,550 E ; 5 ~~ 0.4t~ 0.631' Mde never sma4sa 14870 1051 1 9.46 0.127 0.133 Femzk cunevt emokets (i,414] [5d271 [18] 1761 21+ eigfday 2,884 1133 14.11 0.785 1911 20 ugs/day 6.875 S M? I11I~88 0.284 7z1000 U 2] 13,3611 1231 f3201 10-19 cir,Jday Qb91 914 9.30 0.085 0300 1-9 ags+day 4,687 [756 12698 0.0797 0.249 6~~ U.6521 Ul [33a Femalenevetmwken 39,218 6.44 0.09[ 0.11 17,7121 [13.2491 1461 [1971 causes are presented in 5-year intervalsfot cunent and never smokers in Appendix Tables 1-4- Demiled com- parison of deach race ratios and prnportional harards ratios by follow•up period is presented in Appendix Ta6lc5_ As further assessment, the deadt tares far all causes and Lungcancer for smokers subdivided by the number of cigarettes smoked per day in 1959 are shown in Table 6. The death rate and ratio treltds ia ehiss table are similar to those ior amokers as a whole. The rewha in Tables 4-6 suggest that the impact of stnoking cessation on cigarette smokers as a whole in this cohort is quice limired. Furrhennore, ic appears chat the death rates for self-selected long-term former smokets as shown in Ta• ble 3 cannot be used to describe accutately, the impact of smoking cessation on cigamresmokets as a whole. Discusaion This study provides valuable new evidence regarding the relation of amoking cessation to dte pteven[ion of co. bacco•relared diseases in a large California population. First, it ctmfutms the wel!-established relation that the relative risk of death among fornrer saakers compared with never smokers deaeases as the number of years quit increases based on sel6seleecai fotmer smokers and ap- pcoaches 1.0 afterl5-20 years of cessation. It shows. however, that the impact of cessation is limited during the fitst 5 years of cesation among these former samkers. Ftutltermore, it shows rhat cPs.vtion has had little impact on the absolute and relative nsk of total mortal- ity attwng all 1959 eigarerte smokers compared with ncvcr smokers during the next 38 years. The relative lung caneer risk bas remained essentially const:mt among the males and has increased markedly among the fernales, although lung cancer is the disease most strongly linked with cigarette grnokir>e and is the one that should be most affectedby cessarion over a long follow-up period. These results h@Ip explain niic there has not yec been a substantual decline in che fun-- cancer death rate among older US. males asa.a•hole and why the lung cancer death laze among U.Sr females as a whole has risen so much despite a corniderable degree of smoking cessation during the past 33 .Yars (Fef! 3. 18, and Hnstrom, unpublished). Our finditv.s are similar m -[hose found recently among U.S. male vzcemn smokers over a 26-year period and among a repres~entarive sample of U.S. males and females over a l9-cear periai." 9ince the impact of cessation on the smoLers as a whole in this cohort is substantially less than one might ha.e cxFected based on the esnblishe-3 results for xlf- seleeied former amaket8,4-'124 we hace tried ~- -idzntify faccors that may explain this discrepanct'. The resula for all causes of death could be affecced bq iiie_n'le or ent-iranmencal changes other chan cessation rhat might have negated the benefits of smoking ce=sation: Euc it is not clear what these Chrenges might ]Mit e Eeen- li cessa' <ion is accotnpanled by other •-1•";:~ cltat aegace the benefit of cessation, duse changes x-,.,uld l7~ve to be considered when evaluating the over,i unpzc- of cessa- tion, fr is much less likely tftat other c:-~ea ~-ou13 have affected the mults for lung caneec. sirix ng=-eae smok- ing is by far the strongest risk factor r.- u4i Furthermore, the impact of cessa- ~,: m:_ 'n- be te• duced because most of the subjecs were e: "'e s^=wers wcre 45+ yeacs of age at entry and most long-term smokers who quit after cl:e s_e -_ 5;:Z veets. The heavy, long-rerm smokers with H-air, _arh raca . probably quit lacer than the lighcer, ;.--cr = smokers-

Smoking Cessation and Mortality Trends among 118,000 Californians, 1960-1997 ]ames E. Enstrom' and Ctark W. Heath, ,Jr. zs -....A-_.. We azu:.ed d.e impact of smoking cessazion on snMequent death rares an~g a cohorr of 51.343 men and 66,751 women in Catifr.nia ervolkd in lare 1959 in the original American Cancer Sociecp (ACS) Cancer P.evention Smdy (CPS I) and Collcwed for 38 years. We compared the age-adjusted death rate. expressed as deaths per 1,000 petsen-yean, among all subjeccs who smaked cioareues Ia 1959 but who had largely quir as of 1997 .with the death sace among never smokes over a 38-year paiod.llie all rauses death tare fnr maks decreased from 20.61 durina i960-1969 to 18.68 during 1960-1997 for smokers and deczm.ed from 10SI rn 9.46 fot never smokers- The lung cancer dearh rsre fer rnales maeased from 1.558 to 1 728 8 fer smokers and 'uureaud hom 0.127 to 0-133 fur ncver smoken The all ®rasGs death rate for feroa(es increascd from 9S4 to 10.14 for smokers and decreaced hom 6.95 to 6.44 for never arno4en. The lung cancer death rate for females in- aeaud greatly froom 0.208 to 0.806 for smoluus and increa-ed &oea 0.094 to 0.116 for never snmkas. These results indicatc there has.beest no irnportant dulinc in either the absolute or relative death raPrs 6am all causes and Lmg m'icu for ciga- , teae snmkets as a whole compared with never sowires in this large colart. in spite of a su6santialdegrtt dsmoking assa- tion. While cesationckuty reduces the mmalit, risk among long.term. former smokers, the population Impact of cessation appears cn be less than cumnrly believed. (Epidemiology 1999: 10:5c0 512) Keywnrdx cigaccrre smoking, smoking cessation, rmriality, lung nnceY. rrends, Cancer Prevention Study. it has been esumared chat cigmette smoking causes about ocre-£rfth of ail U.S. deadcs.r This figure represents over 400060 deadss.per year, mainly from coronary heart duease. lung eancer, andd other respirratory discases, rhat in principle can be sulsstantiallyy reducedLy smok- ing cessation_ Nevertheless, in spite of a substantial degree of cigarette smoking cessation and a large reditc- tion in tar and nicncine levels fn cigarettes during the past 35 ce-a.s in the general US- populatlon' the lung cancer death race remains pesistentlyy high.' Indeed. among U.S_ adults since 1966 the percentage of current cigarecte smokers has declined from4396 co 25% and the pemrnrage of former smokers has increased (rom 13% to 25%, and yet the Iung cancer death rare has increased by about 1..1'7%. Thus, it is important to evaluate the impact of smoking cessation from a petspeUive fiai mayez- plain these trends. fi.,n dx'idoo! d Publ~c HeaGh and I~un Cm.pRbenS.e ov,cerCmur. u,.i..,.,.. er eyrbm,. >~ weeda, cn, e,a 'n.oti;ran C~a s,riert. Mw.. c~. ~Reeu .dd,en Ard'mm~ {[frca Ac.eada FaurAariaa, 5-2 Hf7tryn. P.AS l5mama.,. Hemhims Gry, 7R-0615.Iwa AaAs,. «a«ama ro.l..+n E Fmwm Setaut ei tLN1c Heald,, t7xu- . Meoty d Cardumw. r.w A~/a. CA 9G095-171L .. ~1.6.nai..v ..i.,ued 6, ds Am.iran anes S,cieq anA e.au,ded (v1)o,. ~uy ~.. v.ry..md M_a .a.arl.8e^r 6eo. ,6eT.,~.r...n Dluase Aecud, Pq.amoJ rb. [-vsci,t nFC.M(em0. . s.a.e<..d 1.*r -. r996: Gn.l •.nw,. attrpvd Mnmy rr. 1999. o tvs9 br Epdvemtatr acwau n,c 5I70 The mortality benefits of smoking cessation are based primarily on nonexperimental observacional epidemio- logic studies, which consiscently indicate dtet the death rares of former cigarette smokers are lo(Jex ehan those of curtertt cigareae smokers; and $><uthe furrner smoker death rates converge toward those of never srrtokers the longer the former stnakers have not smoked.• Among former smokea whose srnokirtg stanu was determined at the time they entered an eptdeminlogic study, the de- cline inn ciskof death compared with never smokers begins during the first 5 yeats after quitting ahd contin- ues for at teasc 10-15 yea:s- After 15-20 years,. the risk of all•causee mortality recums nearly to that of never smokers and the risk of lung caneer mortality drops to about twice that of never smokets. These nortespuimental siudieshave thelimitarion - that rhe reported benefics of czmation are based on mortality patterns among petsuns who were already faemer smokers ac the time they enrolled and were classified by du number o( years since they last smoked at time of enrollmenc.lhe self-selected focmet smokers who enrolled in these studies are those mho vxse ali.c at the time the study begatt. A former smoker who stopped 15 yeass before the study began is by definidon one who remained aiive those 15 years. Pormer smokers who died before the study began would obviously not be included and tnighc be different frorrt .he self-selecoed 6mrer smokers who were inr..lvded. Also, former smokets who have only stopped for a short riruc petiod (generally 1es rhan S years) andlor former smokers who were at poor

Ull. ,LLe1,\:ryLJlla\1 I(/JUUD 28/09 '99 09:54 FAX 01703798939 ~,. Epidemiology September 1999. % 'ol. (0 No. 5 ALu), the cohon as a.vhole includes all chose who quit and all those who continued smoking, and healthy per- sons as well as unhealthy persons with preexisting dis eases. In addition, recent evidence inditaces the lungs of smokers sustain pennanenc generic damage_ chat would reduce rhe benefits of cessation."'d It is not possible to xnake a death eate comparison of coniinuing smokers with quitters over the entire fol- low-up period, since extensive smoking data were co1- lected only throtbh 1972. Other studies have shown rhac the dearh rate.s foc currenr =mokers and never smok- ers have divetgedsince the 1960s fot unexplained rea- som (Ref 27 and Enstsom, unpublished), as seen from 1960-61 to 1977-73 in Table 4. This aend would suggest that if there had been no cessation, the death rates for a 1959 cohort of conrinuing smokets would be even larger than tiwse observed in Table 5. On che ocher hand, such an inareslse might be moderated by the fact that the death races for current smokers converge toward dtose of never smokers with increasing age (Appendix Tables i-4). There are several limitations chat we believe do not affect the ftndings. The limited smoking data available on the cohore since 1972 are a..Taknessi however, they still support a high degree of cescation among the cohott even if the exact degree is uncercain. Substantial cessa- tion occurred in the cohort a of 1972 and the 19941 1999 findings are consiscent with numerous national surveyx that show most Anuxican smokers stop smoking as they age into their 70s and b'Os. The percentage of persons lorr to follow-up, 9% a> of 1997, is independent ofsmoking status and would noc substantially affecr relative comparisons, although it could affect the abso- lute death races in the later yean of follow.tlp. This paper has presenred data on rhe impacc of smok- ing cessadoaa on both the absolute and relative risk of mortal icy in one major cohott of Cali&7mians. Even with the reservations noted, the above findings. along with results of RGTs, suggest that rhe impact of cessation on mortality particularly lung cancer mortality, among cig- arerre smokers as a whole is less than currently believed. "fhe excess mortality risk associated with smokingcan be avoided by never smoking and can be reduced among smokers only by becoming a long-tetim former nnoker. - Acknowledgments We ,hv,k )n:cenn C,ahn9e< tar icrir ._. thc andr and fxiuntin.r the es,enSeS idh,..q•. aLJ,al Thvn for hdr:( vicwues. and I.~ E Kanin~ Michacl5ucar. and 6cnr hbea ivr cerhns_ anlonre. References 1. NcG:nn., ]M. inse wll..tauzl au•--• vF •lrarh m rhe Un:tad Svtes )t?Li 1993,27.':22e7-32L. SMOKING CESSSTION AND MORTALITV TREtUS 307 L Cenmr rw Disemr Conm•L Smelllance k:r sel.c,.d n•tw.x~v.hel.a. w,rUni¢d Sarc. \11(WR 199q:/3:1-i3. 3. L•nd;. SN, Mv,us T, HuMm S. Wipo PA Cancer Sunr.c,, 1999_ CT Gncer ] C1:, 19H:{9:8_3f.. . 4- 7f,e Nealth 8eae(m 915malroe C~si Rspm p ah. jurye•w Can- e,al, Pvblie Heabh Sen•let. DHI45 Put•. Na ICOC) 90-:416, RocL flk MD, 1990. S.PoscO,CeF.AL°._ _- •cormdlefaiadvn+-cmohaa3.:Rfivl (20 raar) ,edu. ] FRidemrol Csnmunits Heahh tA92,y(.i3-77. 6. The MRFIT RaaRh Gmqc Mmdirv a6a 16 yeaa rar tafcirqnu ,.,.dnmhd m d,r ayd:ipte Ruk F.cu. )nernenriua Teat Ci,euLim. 199694:946-051. 7. Mthuntcu NR, Csnnaa )E IOIq 3P, Alme A1D, Bv)ey WC, Bvie AS, Can.nYwAlr, Fmiehs PL, KeveRE.O'Han P,O.eo GR. Scnnbn PD. Ta.Lkfn DP. 1/ir RA. rAeas af aoka,t uuer*entivn arai rhe me n/ so Bdnl.d a.aiehoffitivyic hrmmhnd'Jav m, rk,a ddeirz,r d FEV S r.a,e Hmld, studr)• JAMA 19)t•27L3497-)sos. B. FbrahrmS,SmishOD.Syau,at<rtvieu•n(ma3emead(nmollcdu:zbof rwktpk K9r faRar, ircervenbm (¢ taa+antl nE mcanary heart ilueaae RMJ I997}141666-1674. 9- Lillm)ele AM L'Jicnk)8 L14: [awidaiau d EPHcnlnhyr. ~ york, . D.(ad UNresir Pwss, 199P, ta-u. 10. Doll R, P<w R. MwuliR in rebtim to.mvhr E9>4s nhecnabae m mak Brfaah~ 8MJ 19762LS2S-75]6. )1. Fasucm]E Tmds m motntiry.mongfalilomia Phlskievu ahcr dv~ng up smeki,>Q: Y950-79. RMJ l9Hk2di1101-1 105. ~r tl. Hararosd EC SnnYirK M rtbtv, m monatirc a,d marbidnr-Fadin9 in Rru r6:ry.fwr a,anths dfoltv.•y, In s pmpecri.e xvdg ,onsd ;n 1959. 1 NerlGrcarrct 1469,32:1161-1188 13. Ham,mn:rEC.Smcking4,rtlaEmwdeuh,msacmemilFOnmrnard .wmen. NC7 Mdn9aph 1966;19:IZ7.2W. 14 . GaeGd.el L Seler6m,, rdlaa~ aad ardlsis +n d,e Amc,ican Gncn Sackry Iv:mpccrt.e sudin. Wad Cana.r fnx )Innrgr i985:67:49 52. 15 . Ha,nm>nd EC, Ca,fvJ<I L. Q,avge m cs,c.,e mdnnq. t959-196~. Am J Pvblic Hedrb 196&36:3(1-45. 16 GaAin4d L CTanyn +n numbnduPrctre.xmt.d cmqerN ru rhm¢cs fn uram7 natnre•mmn acc a 13.rru{e.bad.6anb,vy Rsw, }. A Sa e Ciqarertt. Gad Spnnq Hariqr labw tar• 195l19-]R 17. IcvFA,CarMkd 1_MenaYqxaps?3 arvioldnm rhe~erP.nw fiun Swds (CP51)- CA CSncv I qtn 19tt`,I: a0-724. 1& Bwns OM, Lee L 5h.n LZ, GJym F,TMIer IiD, Vauahn ], Shxnls TG. Cgaxae saob:nF in rhe Un:rvJ Stares In Smokinl and Tab,orv Momgrayh S. Naaw,al Cancn lm,irure. NIH Pvb( Na 97- 42ll• 1997• pp. 1]-I12, and U. 5. 8umv d the Ccnws, Cnrxm ta,au.tn,v sa'er,. l9. Mawx. RR. Mah_•u oF.rbtio-crvrv:•al and troportiaeal mwoGry. G.mp Bbmed Rc 197f,iv2i-3R 7A Hmadl F.7he PHREO r+vadvo Ia SUOI Sv,lea,c..a1 Li6,ary lhc.S (3mda 1989 Editlon Gq, NC SAS Mcrnm.. 1989. 21. Metaaald;., )K. Hn,b« 2, 01- Q/J, rraumrni IF.1r. Smding and eanee, mw,e7:n• amm,g ti5. Vecer.,..: a 26yev fdb-vp tnr 1 Cerccr t995.bc*. t9a-19T 22. Emvnm JE. Smnka,e cemrlon anl ,norul:rr aends ar~wne Mv Uni,a.l Snta poPoL~raei ] qin Eptleald 1994;i2:9V]-999. 21. Hammm,dEC,Gran61 LTheh,Ilumcedhcal9,onamok,KlvhisNO Mamgnph 1966:19•?69485. 24. 7(avatlH I, Cddte GA, S.anrpfn MJ. W alea WC, Manwn JE• Raner 8. Hmuer D]. Hameksrn CH, Spe:>Q TE SmNti e.un m rdaaw m ew7 mumRn,arer m.woen. A P•oIDmiw rnhun rndv. Ann Isuoa Mc.1 1993;119992-1R17. ss. Mau L taJS, x,rrie 3M, Fuc m LipPm.n SN.I<e ]l, iblY. &ea+cn A. Yv R. Murtr< RC. K<mp BL IQ,v: FR. waW, GI., Hic tafinon Wy Hu,g smders. WK. Clmsl genaric ahrrariuw in r6. tvugs cf <ur,enr and /arme' ] Nad Canxr Ina 1997,ffi857-86L 26 W6n3n IL Lam S. 8d,erns C V:nra,l AK. Fpp ln.f, ILRidss ], Snar ]M• S,:wcsea9• Muxm )D, GaaiarAF biokcn6rdamaee In r6t trmrcrual epkMeltumdovrzrssaWf smokcrslNadGncerlr,ct997:&9:1366- 3. 27. Thun Ml. mY-laRr CA. Ouc EE, fbMe,e wr). Hexh CW. )n fxcea mmvllry a:bng ciGUnee v.ekes: d.v,yc in a 20•ynr imeml. Am j Pn61+c Health 19949"x1I21-12r0. N fil O 01 Ull OD Ull