Tobacco Documents Online

. G M . A'atetuVAN .lorntvnu niv Ertuennut.rocr ("upvriKhl:- 19M!) hv'l1he.lulinrIl,afkinr Utii.'crnitvlch.Nd.n& flyKiene. nwt I'ulrlir IN.-:dlhi\41 riKhts reeervcdl V-dl 1'u-, `i.'t P:u,n"d u, L".ti''a. Ai~ P~IR1oSPECTIV'E' A'1*FALYS1& OF~ Pf1~Y.SICA~L AC''TIvL'TY AND CAPt1C1?J1'" R'ICHA9RD K., 5'F:VEIRSON," ABRAHAM M. Y. NO'MIU!RA, JOHN S. GROVE, ANn (7;RANT' N: STEMMERMANN Severson„R. K. (Japan-Hlawaii CancerStudy,, Kuakini Medical Center, Honolulu, HG96817); A. M: Y. Nomuraj Ji S. Grove, and G. N. Stemmermann. A prospective analysis of'physical activity and cancer. Am J Epidemio! 1989;130:522-9. Previous studies have suggested that increased physical activity may be protective against' cwalon; cancer. Most of these studies measured ac6vity indi- rectly. The, present study evaliJated several direct measurements of physical activity and also resting heart rate as determined from an electrocardiogram. Increased activity was consistently associated with a decreased relative risk of colon cancer for each measure of activity: The colon cancer relative risk estimates for the middle and upper tertiles of'ai phy,sii;allactivity index (cpmparedlwith the lower tertile) were 0.56 (95% confidence interval (CI) 0.39-0180) and 0.71 (95% CI 0.51~-0:99), respectively. Relative risk of colon cancenassociated with modpr- ate or heavy activity at homef recreaCtoni(compared with mostly sitting) was 0.66 (95% Cl 0.49-0.88): A similar comparison of activity'at work resulted in a relative risk estimate of 0.72 (95% Cl 0.52-1.00). Relative risk of'colonicancer'increased significantly (p = 0.027) with increasing resting heart rate. Relative risk's asso- ciatedi with physical activity were also evaluated for cancers of the stomach, rectum, lung, prostate, and!urinary bladder. An increased relative risk of stomach cancer was associated with severaC' measures of increased activity,, but thesee results are preliminary and additional research is necessary, to further evaluatee this finding. colonic neoplasms; exertion; heaMrate Within the' past five years, several epi'- demiolmgic studies have reported' that de- creased physical activity is associatediwith ani ihcreased: risk of colon cancer (11-fi). Received for publication September 1!5, 1988, and in final form Feb'ruary 6;,1989. "From the .Japan-Hawaii, Cancer Study, Kuakini Medical Center, Honolulu, HI: 1 Current address: Divisions of Epidemiololry and Pediatric Nncolbgy, University of Minnesota, Minne- apolis, MN. Reprint requests to Dr. Richard' K. Severuon. Japan-Hawaii Cancer Study, Kuakini ilYfedicaC'CenUer: , 347 N. Kuakini 5treet. Honolulu. HI 9681.7. Supporteeilby Grant EiUR CA 33644 from tite Na- tional Cancer Institute, National Institutes of HealUh,. Bethesda, M D: The atrt'hors thank the fnllowinR,institutions on thee island,of Onhu, Hawaii for their helpfiil!conperalion: Castle Medical Center, Kaiser Medical ('entier, Queen's MedicaliCenter„St: Francis HbspitaU Straub ClinicandlHn.cpital,'1'riider Medical'i1`entier; Wahiawa GenerallHnrpital! and the Hawaii'Tumor Registry. Most of these studies used occupation as a surrogate measure of'physica'l activit.y' (1- 4)- Two, other studies have also loeen, pub- lished,,ane which found no rel'at:ionbetweenphysi'cal activity and colon cancer (T) and another which reported no association he- tween physicallactivity andlthe risk for all cancers combined (.9). [ir a report of three occupationa[ cohort studies, Persky et al. (9) found an increased resting heart rate in workers dying from colon cancer compared with workers surviving' to the end of the survei'dinnce period. Increased physical' ac- tivity has been associated with a decreased heart rate (1O, 11);. Although •past studies have suggested there' is' a link between physical activity and colon cancer, many of' these studies have'relied onindGrect tneasures,ofaetivitt(such as occupation) and have not' cmn- 522'

PHV5Ir•At. qC•riVlitTVAvD o•Ah7cEK, 523 trullle<d Gur the potential confounding ef('ects uf loudy mass (1'?; 13). Also, the association of otlier camcers withiphysical activity has noti been extensively evaluated: We report here on alprospective analysis of the effects uf'physical activity and restiing heart rate on the suhsequent incidence of' colon, rec- tal, lung, prostate; stomach, and uriinaryy bladder cancer in a cohort of' Htiwaii Japa- nese men.. MATERIALS AND AtIGTI 110iDS A cohort of 8,0(1G ,Iapanese men who were living on the. Hiawaiian island of' Oahu in 11965 and' born d(nring, the years 11900 to 1I9119wereclinicallvexaminedand inter- viewed from 19ti5to1968 (14). In addition to the subject's demographic, occupational, andlsmoking history, information was also recorded on the extent of physical activity: A summary physicallactiivity index was cal- .eulated following, the~ method used in the rramingham, Study (15). It was based on the weighted sum of the usual amount of tinie the suhject spent per 24 hours in the 16lJowing, types uf activity: basal, as in, sleeping or lying down (weig.ht of 1); sed- entary, as in sitting or standing (weight of 1.1); sliglnt„as in walking on a level surface (weight of 11.5); moderate, as in mardening, or carpentry (weight of''? 4); andl heavy, as in shoveling or digging (weight of'5). Thee weighting factor was based on the approx- imate oxygen consumption needed for each level of.effr,rt. Also reeorded!at the time of interview were the subject's semitguantative estimates of the amount of physical activity on the juh and at home/reereation. V'arilousanlChrropomurphic measurements were recorded during,the clinical examina- tiion, including weight and standing height. Body mass index was caleivlated as the sub- ject's wei;ht, in kilograms divided by thee sctuaire of' his height measured in meters. RestinK heart ratie (beats per minute) was dhtermined from a cunventional 12I lead electrocardiogFam which was performed us- ing, an electrocardiograph at 25 mm per second. Because a complete or second de- gree atrioventricular heart block is _associ- atled with a slow ventricttlar heart'~ rate, the two subjects diagnosed with these, condi- tions, based' on their electrocardiographic f ndings, were excluded from this study. Subsequent to the initiali examinatiinn,, the cohort has been followed for the devel- opment of incident cancersthrough contin- uous surveillance of Oahu hospitals and periodiclinkage to the Hawaii Tumor Reg- istry which is part of the NationaC'~ Cancer Ihstitute's Surveillance, Epidemiology, and'i End R~esults(SEER')' program(Y'6). Oulythose cases that were newly diagnosed with histologieally confirmed cancer of the stomach, colon, rectumi litng, urinary blad- der, or prostate were.included in this analy- sis. Ilt has been estimated that only 1.8' per cent of the cohort has moved from Oahu (17),, so the cohort is geographicallyst'~ableand case identification has been, virtually complete. A t'otal of'81! menwith cancer at the time of examination (prevalent cases) were elim- imated from the analysis. The time at risk for each subject was computed as the time from examination to diagnosis with cancer, death, or October 1986, whichever occurred first. The relative risk of cancer associated with various measures of physiical actiNi'tyy was: estliinatled fromi proportional hazards regression models (!18') while adjwsting for age at examination, body mass index, and (for stomach, lung; and bladder caneers)) cigarette smoking: Statistical significance of the risk estimates was determined by observing whether or not' the 95 per cent confidence intervals included unity. Phys- ical activity index and resting heart rat'ee were recorded on a continuous scale and aa test for trend was performed based on these values. In addition, these two, exposures were divided into tertiles: (based on the distribution in the entire cohort'); and rel- ative risk estimates were cadculated for the middle and' upper tertiles compared with the lower tertile. RESULTS During the follow-up period~ there were 172 incident cases of stomach cancer:, 192 of coloni cancer, 95 of rectall cancer, 194 of lung, cancer, 206 of prostate cancer;, and 70

t Si{V`Iul(S(:)N~ H"{' A~I.. of hladdhr cancer. 'I°ahle 1 shows the risk of cancer associated ' with pltysical activityinc3ex catego'rizedli'ntrn tertilles and th'etest fcnr t.rend ('widh'physicaLat:tivity i'ndex ern- tered o'n a continuous scale)I Higher lkvets of'physical activity were associated with a reduced risk of colion cancer, although the risks did not decrease in a liilear fashion, and the'test fbr li'near trend''wa:s not sigrlif~ icant (p = 0.201). In order to evaliuate a threshold effect, a riiodel was fit wit'h thee inn+erseof'the p~hysicail activity index. A+1- though a sl'imhtly better fut was achieved, the relation was sti'll not si'gtlificant ('p = 0.091)'. It, should adso be noted that the trend for rectal and stomach cancers wass opposiA,e that for co'lon cancer. Physical activity index shoaved a signitilcant (p = 0:039) negative dose resporlse reliltion to llung' cancer, alk,hough the risks hased on' tertiles didrnot decrease in'a linear fashion. O'f the 1'9z colon cancer cases, 45 were located in the cecum, apperndix, and as- cendi¢lg colon, 3'1 were locatetl in t he t rans- verse' (iinclud'ung, hepatic s1n1('L spleniic llex- wres) ancl dhscentlilig ; c•nIun„ I 15w•ere cntietl iirr thesif;nlriitl cnhon, :uttd innr• wslti in an unlCncrwn I„cati/on: '1'aktle 'l shows' the risk of co~lo~n' eancer h yanrluotnilcsiite and phvsicai activity intdex. lutt:reasecl'' ph\•sic al activiity appea'ced t~oIneassol•ia~tedwi;th'a decreased riisk of colon cancer at each'suh- site, althoilKii none'of the trencU tests' was stati:stically signil-icant. Table 3 shows the risk of cancer assi ,ci- atedwiith~ restingh'eart ra'te.'I'herisk lol•' colon cancer increased signilTicantly 1 fu = QL0~27): wit'fi increas'ingheartr'ate: 't°he risk oflungcanucer in sultyectsw•ilh a henrt raie eyuail to or greater than 7'2'lueats per ntinulle was significantly higher thxtn the rii*k in sulljectl5 with a rate less than 72 iieats per minnt'e; but the test fctr trettt) wn:5 nnt sig- nificant. I*(orle ol1 the other can'cers ap- peareci to ha've a' stron'g, a5sociatitln with heart rate. "I'able 4 shcows'the ritik of co:lon' 'I'Ant.e 1 RefnGiue rl.h n/ cnncPr 6ti ph_v:.innCat'truitv iildrx' C~uccr 1'hycl nlnrli.,itv intlex terlilef' ~*aret Nont:wr~. RIIa uive ~ ri.skj , ~. ,.. rr.nlii6,n,p ~ irttrrRal I n•n l p vadi,ei Colon, lst 82 2;507 1:.00 2nd 49 2.608 0..56 0.34-0:R0', :3rd 60 2,5•52 0.71 1 T. 5 il -11.99 1)201 Rectum isb 24 2.555 1.00 2nd, 35 2.622 1.31 0.78-2.20 3cd 36 2;57/i 1.11_ Il.tlt-Y.:Sfi II.IIt6 Stomach lst 46' 215.13 1.110 2nd 56 2;6111 1(17 ll: 7:1- t.fi9' \ 3rd 70 2.54'2' I t1198'-4.9ri 0:1'lll Lung 1st 66 2,523 I;.IIU 1 2itd 75 '1.582 1,.(Ili 11'ili'-I.•IN 3rd 51 2,561 o:7n ~3~ n.tr;-t.nl /t.n:lsl Prostate Isti 66 2.523 2nd 67' 2.590 0:89 O.li3 1'.2ri 3rd 72' 2..540' 1.05 : 1).75-1,.41 (1.717 Bladder lst 23 2.566 ' 1.110 2nd 19 2:6.'ts' 0.77 0:42-1.42. 3rd 27 2.58b 1105 011i4)-LKat /116)t11' "1'he tolal nunrber of eases in this tnble may not match totals gii,•en in the test, becmuse nt' unknownvaluec : f Tertile levels: 1st„0-30.1: 2nd; :10:2-'W.2: 3rd, >343. $ Contputed fnom coefficients estimnted by Cox proportional haxardk mudelc while wnl rotting il'ur :ite; lnHl{• mass index, and ((for'stumach, lung, and bladder caocers onlys)eikarctte:cmokinK.

I Sh:NE1tM)N E'J' A1.. TABLE 4 Rplal(ur vr.ch nf cnlon cancer b:v anatamic site and rv..ifrrg heart ratn" tlrc1in • hrnc Anartomir•.ale r' t (':rsee Nrrnrn.~:~e<. ItrLrlice ' rale lwnlilhtl ri.kt Ascending 1st 12' 2;547 ' 1.00 2nd 11, 2;5261 0.97 3rd, 22' 2;G51 1.89 Transverse and 1st 9, 2,547 1.00 descending 2nd 8 2.526 0:95 3rd 15' 2,651 1.61 Sigmoid lst 3•5' 2,547 1. 00 2nd 411, 2,526 1.22, 3rd! 39 2,65'li 1.11 ,malidr•nrv . rnicrn~rdl •I•n•nd p.r.:d~u• O:-tal-2:2tr. (1:94~3!8a U:UfiB' 0.a7-2:46 0.7u-a:7fs' t/:1190 0.78 1.9'1, 0.711-1.75' 0:In9 • The total number of cases in this table may notmatchrtutal§ given in the text because of unknown values. tlTertile levels (beats per minute): 1st:0-T1; 2nd! i2'-81; 3rd„>_82:, t'Computed from coefficients estiinated,hy Cox proportional hazards models while cnntirolling for aRe and bodij mass index. TAnt.E 5 Rchr6iue risk nJ, cancer l..v plrysicaf artit•ifY at irnrlt and at ltunrrIrrcrnrrtinn' C'anecr L.oriatidrn Plrysicalarticity rati;ec Nintrasee Itelnlis. ri,k T W- ^; r.,tnlirleurr itttrrn•al Calon Home Mostlysitting1163;9;i7' L(N) MrKlerate or heavv 75' 31757' U!6C, 0:49-t/:88 Work Mostly sitting 511 1 57 7 7' l.tN)~ Moderate or heavy 142 6,142 0:72I Rectum Home Mnstliv sitting 5•t 3.999 1.1111 Moderate or heavy 40 :1. 7,92 u.7'l 14. I7-1_U-1 Work Mostly sitting 15 1i61rL LIM) Moderate or heavy 80: 6.204 I l'23 0.7 1-2.15 Stomach Home. Mostly sitting 7U 3.t.1R3 1-lN) Mhderateor heavy IUt 3.731 1.45 1.07--1.97 Work Mostly sitting 19 1,6081 f.tlt) h'4bderate or heavy 1•53' 6;1'3l 1.74 1.U8-2:81 Lung Home MostlysiltinR, Moderate or heavy 1l1 82 :t,9421 3175U~ 1.IIU U18U 1):60-1.Uti Work Mostly sitting; 411 1.586' 1.(xr. Moderate or heavy 153 6.131 0.88 ' n.Cid-t °5 Prostate Home Mostly sitting 1114 3,9:19 1_UO Moderate or heavy 91 3.741.1 t 0.77 n.5 t-l l U 1 Work Mostly sitting :37 1',590 ILtMO Mixlerate or heavy 1'69: 6.1'C.5 1.05 0.73-1.51 Bladder Home Mostlysitting 40 •t.U1'a 1_(N) Mtideraworheavy 30 :I.wU2 0.78 n;-1lt1 .l_2G: Work Mostly sitting 13' 1,614 1_tNl Moderate or heavy 56' 64229 1.00 1 IC5•I-1.83' • The total number of casec and nonca.ses in thi's table may not match because of unknown values. t Computed from coefficientR estimated by Cox proportional hazard!a mrKlels whilerontrolGnQ for age. Ixtdx mnss index; andi(Ifor stomach, lung, and'hladder cancers onlyJ cigarette stnukinK.

i iiVauAi. .ac°iuvrrrv ahn (Ar('i•at levels tseparately)' as nrustly sitting; mod- erate, or heavy: 5i:nce few sul)jects categw- riizeal either activity as heavy, the muderate and heavy groups liur buth exposures were comhined t'ar this analysis. 5ulij~cts with increased levels of either activity experi- enced a significantly reducedl risk of colbn cancer aindla sinmilicaintly increasetl'risk of stoniach cancer c:ompared with those who were mostly seKlrantairy. Nione of the other cancers showed a sig,niificant rel'atio:n to eiiher nieasure of physical activity. DIhCU~S;ION '1lhere has fueen' mueh discussion in the literature rebarding the difficulties associ' ated with measuriuim physieal' activity in aa meaningfullway ('lJ-'?3). Althougii t iis ca- veat applieato Uhe pr~esent study as: well,, we feei that our data represent a niore yuanitit.iative aud coinprehensive measure- ment n/" lpiiysiKraV activity t'han several of tlie pirevir,us epiilemniologic stud'ies of can- cer which have dealt mainiy with' occupa- tio:nal' activity. Otnrphys,icalactivit'yi~nKjex; was derived from individua'I in-person in- terviews iii which the suhjeet was asked his usuad amounC of t iine spent in various levels of activity on the jith, at home, and in recreational pursuits. In' additiion to this summary measwrement afphysical activity, we were also ahle to include separate self- asses5ed measure5 of physical activity on the jblr and at home or recreation. Further- more, restiing heart rate was evalwated. W'hile heart rate has 1-nmitations in bei'ng,aa surrogule nieasure of physical activity; it has the arlvantage of being, independent of any cdifliculties a:,sociaterl'i with subject re- call and reporting : Increaseci physical activity was consis- tently assuciuted with' a decreased risk of colun cancer for eaehi of tlie measures off phybical activity evaluated in qh'is study. This fandinl; supports and extends the: resubts of previous studles. In 1984„ Ga;rabrantand'cu-workers (1) firstreported' an increased rislc fbr men iin sedentary johs (lu;ised orr iocculpa( iion a, recorded im the 62,,7 medical record) compared with men with aetive jpbs. Similar results have been re ported im both case-control (2, 3, 6) and cohort (4', 5,9) studies. Studies in experimentaal animals havee sliown that p'hysical activity reduces the growth of'transpl'anted tumors' (h4)L I't was reee:ntly reported Chat 1,2'-dimethylhydra- ziine-induced colbn caneer incidence was significantly reduced in rats that were aI- Ibwed spmnrtaneous wlleel activity compared' with control rats housed in a standard fash ion (2.5) Fxamining spo:ntaneous; rather than forced; activity should help minimize any confound'ing,effect of stress associated' with forced activity. Also, 1,2.dimethqlhy- dra¢ine-induced rat neoplasms are specifuc to the colon and are histopathologically siniilar to human co'1on:neoplasms (26): It has been shown that physical activitiy stimvilates peristalsis ('?7)t and Wu et al. (5')' have suggested that carcinogens in fecaD materiiull may have liess contact with the eolbnic mucosa in active people because of a rediuced'stool transit Cime. In this study, there was npt strong,evidence that th'e de- creased risk for colon cancer was liimited to any specififc anatomic site. Ptather, it ap- pears that increasedi activity may decrease cancer ris!k throuohout the entire codon. Since peristaltic activity occurs throughwut t'he colon, tihese results are consistent wit.h the hypothesis tha't activity-stimulated peristalsis decreases the contact t'ime be- tween fecal cazcinogens and colbnic mu- cosa I'revious studies in this cohort (!28) and in' Scandinavian, populations (?9) have not shown any correlbtion lietsveen oral' amal' transit timeand co:loncancer risk. I+Iowr ever; these results are based' on measure- tnents made in a limited nvmber of people, and no colon-specifuc transit times were reeorded. Garabrantet all (1)' havesug- gested that it is possible- that the trransit rime of colonic contents can be aLtered without signilicant;ly affecting the total oral-an'aI transit t!ime. In aKlilifiOn to cnlon cancer, this stuoly

I'III1'tilCi\l, rU"!'IVI'I' Y'' ANU t"At.'CI•:I( Jf '. Kkmnel WIB, 1„rlic 1". Sonne benelits uGphysical' urtivihv: 'Clic, I tauuiint;ham Shwl,y. Arch hrtern Rlerl I617J:1':1J:tS:i7 -tiil. IIi: (:mcer incidence aud murt.rlity in tlhe Unitedl 5tuter, 197,t' 77. In: Y oung,Jl., I°ercy Cl:, Asire. A,1, eda. tiurvcillauce;,epidemiology, antl end re- sult,: incidence and murtality data„ 197J-77 Be- tlieu,+la, Rll): NaUiunal C'ancer Institute; t981L (N(?I monr~graph 5'7): 17. Heilhrtin I.K, KaRun A, r*lumura A, et at. The uurigins ut epidumiolugic studies ot heart disease;, cancer and osteutioru,is amnnK: Hawaii .JHpanese:, Hawaii R1ed .1119-t-t:Y".)d-7,. 18! Chx f)',R. Regressiun nwdels and Ili'e tctbles (with di~cussionl. JI RSbat 5ec, Appl 5tat (fBI 119T?;3a:, 1 t37-22(I. 19: l.aPnrte IIE, hlonllwye 11.1, ("asttersen,C:1.,Assess- tuenli ufl phyvical activity in ehidemiuluKic re- search: prohlems anrd prospects. 1'ulolic k{ealtih, Riep IJkl5;lANd!131--46: 2(1: Washburn RA„ 141wntoye 11:1! The assessment of physical activity by yuestlonnaire: Am,J Ispide- m iul I I 9H(S;1 X:1: 5'Ga-76. '-"L Sallis .1F, Haskell WL, Wood P1Y, et aL Physical aclivity as,essment methudology in the Five-City Pnuject. Am J Kpidrmitul 1I985;121:92-d1)6. 22: TavIor Cli; Coffey '11;, lierra K, et all Seven-day acliivity andlsell' repurt cnmpared to a direct mea- 5'29 sure of physical aetiivity: Am d Gpiilemiwl 1Ji6-3; 120i811t4-24l 2:1, Cauley JA'„LaPorte RE; Sandler RB„et at. Com< parison of inethWds to measure physical activity in postmenopausal women: A'tn J Clin Ntttr 1987;45:12-22. 24. Hoffman SA„Paschki KE;,DeBias DA; et at. The inlluence of exercise on the growt'h of trans- planted raU tumors. Cancer Res 1'J(iY;4:597-9. 25, A'ndnianpopulos G Neison RL, Bombeck CT, et at. The in9luenee of physical activity in 1,2 di- methylhydra:cine indueedleolon careinogenesis in tihe rat. A'nticancer Res 1J87;7:84'J-5'3. 26: Rogers AE;,Nauss:KMt Rodent models for earci- noma of the colon. Dig Dis Sci 1985;30(suppD: 87S-1'02S. 27: Holdstock D.d;,Misiewicz JJ, Smith T, et:al. Pro< pnlsion (mass movement:v) in the human ooli,n and its,relationship to meahc and somatic activity. Gut 1976;11:91-9: 28: Glblier GA, Nomura A', Kamiyama S, et al. Bowel transit-time and stool lweight in populations with ditTerent colon.cancer risks. l:ancet 1977;2:110- 11L 29: IARC' Ihtestinal Mioroecology Group. Dietary fbre; transit-time, faecal bacterta; steroids, and eolon cancer in two Soandinuvian populations. Lanoer 1977;2:207-1,1.

JET LINCOLN Messrs. Murray Pte s nik ,:' 1i I could'~ take the author ''s reiSponse as a compliment; althoug;h,I very much doubt he m earat it that wa . ~E C€'f yEO SEW j j is8y. R. W!' h1WR-6,AY

• AMERICAN JOu Publi~ Health September 1989 C~antents ri (pontinued) Volume,79, Number 9 DIFFERENT VIEWS I pro riate Comparisons of Incidence and Prevalence in Epidemiologic Research ........................ 1301 Inap ~ MatM ` ana F/arrders and 7faomas R. O Brtenin Ote +I tUcal Mbdels and Scientific Reali t y tiarence Rates for Disease ............................... 1303 J Ali+an R. Feinstein, Charles K. Chan, John AJf: Esdaile, Ralph L Hnnwit¢„ Michael J. McFarlane„ and, Carolyn+K V'Nells ~ ~ NOTES~ FROMTHE~ FIELD! The AIDS Action, Committee Safety Neta Program: Small Group Parti•es for Safer Sex Education.............. 1305 Dorothy C. NVsrLZ, Anne Marie Silvia, John Dreyer, Jeff Epperlty Gary Sandison, and Chtis La Charite LETTERS TiO: THE EDITO~R~' r A Notefrom the Tobacco Industry ............... ........................................................ 1307 Response from Professor Sprafka............ !......................................................... . 1307 Determinants ofiStopping Smoking: Italian National;Health Siurvey............................................ 1307 Laws Regulating Tattooing ......................':...............,............,.,........................,._. 1308~ Contrasting View on Rheumatoid Arthritis ... . . . .' . . . . . . . . . . .. . . . ...,. . . . . , . . . . . . _ . .,. . . .. . _ . . . . . . . . . . .. . .. 1308 Comment on Changes in AlcohohReigted Inpahent Cace . . . . . . . .. . . . .. .. . . . . . .. . 1309 Response from RometsjS' and Diderichsen ........... . . . . . . .. . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. ..... 1309 Behavior of Clients of Prostitutes.......................... .......................................... .... 13U0 ASSOCIATION NEWS Call,for Abstracts, APHA Late-Breaker Epidemiology, Exchange Session................................... 11294 Annual MeetingiRegistratitin :..............,..,..........................,................................ 1311 APHA Constitution and By-Laws .........................,............................................... 1316 JOURNAL DEPARTMENTS Masthead ........................................,..,.................................................. 1206 Advertisers'Index........ ......................................................... 1214 Membership Form ...... ...... 1322 Agency Membership Form .................................................:.............,........._. 1323 Job Opportunities....................................................................................... ... ...............,...... 1324 Prospective authors should consult "Information for Authors" which appeared in the April 1989 issue of this Journal and may also be obtained from the Journal Editorial CNfiae: American Journal of Public Health;,10'15 1'5th Street, NAN;, Washington„ DC 20005. (202) 789-5600. The American Journal of Pu6lic Health, published monthly, is the official Journal of the American Public Health, Association, 1015 Fifteenth Street; NW. Washinglon,,DC 20t)OS. (202) 789-5600 APHA is a nongovernmentali professional society founded in 18T2, representing,aU"discipiines and specialties in public health. AP1iA OFFICERS Ihis R. Shannon, PhD, RfV„President Bailus Walker, 1r., PhD, MPH, l+nmediate Past President' Myron Allukian, Jr., DUS, MPH, President-Elecr Mabel HL Morris, RfNI MA, Vice President (USA) Jp4rn:M. Last, MD„MPH„ Vice Presid6nu (Canada)Carlyle Guerra de Maeedo, MD, MPH, Vice President (Latin,Arnerica) Caswell A. Evans, Jr., DDS; MPH, Treasurer Molly Joel Coye, MD„Clrair, Executive Board' George E. Hardy, Jr:, MD, MPH; Speaker, Governinq;Council William H..Mcf3eatti, IvfD„MPH, .ExecutiveDirector EXECUIITYE B'OARD Molly Joel Coyt,,MD, MPH, (1991) Chairperson Myron Allukian„Jr., DDS, MPH, President-Efect: Ted T. L. Chen; PhD, MPH, (1990) Brian A. Dobrow, MPH, (1992) Caswell A. Evans, Jr., DDS, MPH,:Treasurer Eugene: Feingold, , PhD;, (1991) : Frank Goldsmith, DrPH, (1989) MMargaret H. Jordan, RN. MPH, (1992) Allen N. Koplin, MD,,MPH„(1990) Joyce C. f..ashof. MD, (1990) JJudith Rae Miller, MPH„(1989) ShirleyReed=Randolphr MSPH.,(1991) Helen Rodrigucz-Trias,, M D; (1989); iris R. Shannon, RN, PhD; President Fnances:M; Veverka, MPH, RS, (t9rJ2) Bailus Walker, Jr., PhU„MPH, firanedirue l'asr Presiilent EX OFFICIO William H. MeBeath, MD. MPH, Eoerative Director Richard A. Wissell, MPH. Cltair Action Board Cecil G. Sheps. MD', MPH. Chair, Pragram, UerelnpmennRnnrd George E. Hardy, Jr., 1VID;,,MPH, Speaker, Governing Cmutril AJPHISeptember 1989, Vol. 79,, No,,91203+