Tobacco Documents Online

T'~!Dt~?C A, Involuntary Smoking and Lung Cancer: A Case-Controi{ Study" Lawrence Garfirni Oscar Auerbach,3 and Lou Joubert7- 4' ABSTAJICT-(n aicase-control study in 4 hospitals from 1971 to 1981. 134 cases of lung cancer and1402, , eaxes of coion-rectumm cancer (the controls) were identified im nonsmoking, women. All cases and controls were confirmed, by histologic review of slides. and nonsmoking status and exposures were verifled 1 by interview. Odds ratios (OFR) increased with increasing numberpf cigarettes smoked by the husband: particularty for cigarettes smoked at home. The OR for women whose husbands smoked 20 or more cigarettes at home was 2.11 (95% confidence limits: 1.13, 3.95). A logistic regression_analysis. showed a significant positive trend of increasing risk with increased'exposure to the husband's smoking at home: controlled' for age, hosp6tal, socioeconomic ciass; and year of diagnosis. Comparison of'women classified by number of hou^ exposed a day to smoke in the last 5 years and in thedgst 25 yea: snowed no increase in risk of lung cancer.-JNCI 1985: 75:463-469. work or in, other areas. Conversely, some individua marriedi to smokers may suffer acute effects from inhalin smoke and consciously avoid such exposure. A survey c 38.000 subjects by Friedman et a1, (9) confi'rmed' th. hypothesis. Abouc 40 b of v>ornen nonsmokers and 50% (( men nonsmokers who were married to nonsmokers wet exposed to'uhe smoke of'others for some periods of tirr during a week, and, 47 0 of nonsmoking women marrie to smokers reported, that they were not exposed to tobaccc smoke at home. In the study reported here., we record tr smoking habit of husbands (total i`Io. of cig3rettc smoked'and No. srnoked!at home)l as well as the numbc of hours a day the subjects were exposed't~p the smoke r others at home; at, work. and in other areas. Other causes for concern are establishment of tl microscopic diagnosis of primary lung cancer an Much interest has been expressed in the past several years in the reported relationship of involuntary, or passive smoking and the development, of l'ung cancer. Eiirayama (1), in a prospective study in Japan, reported a 2:1 RR for nonsmoking women married to smokers as compared to the RR ' for nonsmokers married to non- smokers. Tricliopoulous et al. (2), in a study in Greece, foundi that nonsmoking, women with husbands who smoked, hadl an OR about 2.5 times as high as that of' women wiih husbands who never smoked, and rhe OR rose to 3.4 in women wlhose husbands smoked' more than one pack of cigarettes a day. tin another case-conitrol study; Correa et al. (3) found nonsmoking womenn rns ied to smokers with a lifetime consumption of 41 or more pack years had an OR 3.51times as high as the OR of. women married to nonsmokers. In an analysis of' data, from thc Ameriican, Cancer Society's prospective study, Garfinkell (q) found little iff any increase in RR of lung cancer fbr nonsmoking women married'to smokers (4). The RR was 1.34' for light smokers and 1.10 for those whose husbands smoked 201or more cigarettes a day. Kabat and Wynder ('5)1 in a pilot study found' no extra lung cancer risk in women and in nonsmoking men :exposed to smoke at horne, bur nonsmoking men exposed at work showed a slight increase in- risk. Kabat and Wynder rneasured' exposure both by the spouse's smoking, habit and the subject'ss report of direct exposure. fininvestigation bySa~ndler et al. (6) of relatively young cancer cases in North Carolina found an overall 1.6' OR (smokers and nonsmokcrs) fbr exposed vs. nonexposed cancer cases. Exposed cases were those in which the husband smoked. EYposedi nonsmokers had aa bigher OR than thatof'exposedt smokers. In nonsmokers the OR were elevated and statistically significant for cancers of the cervix, breast, and endocrine glands.. ... Occurrence of cancer in the latter two siites previously hat not been associated with cigarette smoking. Lumg cance also, showed arr elevated OR' but was not statisticalll significant; however, the number of cases was small. Two papers in Hong Kong by Chan andl Fiang,(7) an• by Koo et al. (8) show very little difference between case and controls with respect to involuntary smoking an, lung cancer. Except fbr the two prospective studies, all of thes studies were based on relatively few:lking cancer cases ir female nonsmokers; the -number of cases ranged fnorn, 2:. to 77 in various case-control studies. In a previous paper, we pointed out that'in any stud of involuntary smoking and liung cancer, caltegorizin nonsmokers by the smoking habiit of the spouse may leal to error in classification of'exposuref fn the United State particularly, there may be many women, married t - nonsrrnokers. who are exposed to the smoke of otliers < .AA6RiEVIATtIDNS USED:~ CL=ionfidence limiits:~. OR..=,xIds~ ratiot! RR=relatiive risk(s). t Received April 22. 11985;'accepzed'Jµne 20:, 1985. ' Department of' Epidemiology and, Stntiatit:s. American Canc Society: 4 Wesi 35' Sti. New York, NY ' Ioool. s Veterans Adtninistraoion Medic:ei Center. Easr Orange. NJ. aa University of Medicine and Dentistry ofI New Jersev, Newark. i`fJ. t We thank Dr. Robert V. P. Hvtrcer. St. Darnabas Hospital. Livin ston. NJ;, Dr. Herbert Derman. Riverside Methodist Hospital. Culur bus. OH: Dr. Jerry Rothenberg„ Morristown Memorial Hiospit. yiorrirtown, NJ: and Dr. DouglarSmith. Middlesex General Uiniwr siiry Hospital. New Elrunsnvick, NJi for granting us aceess to mediic records and pathologic materiaL' We also thank the,Goltowin; indivi uals ior•making vaUuahle suiqgestions regarding the manuscript: 1' William Haenszel, Dr: E. Cuyler Hammond. Dr. Geoffrey Howe. V Edward Lew. Mr. Herbert Seidman, and Dr. Steven Stellman. WC a thank, Ms. Nancy La Vcrda and ivbr. Henry Vasyuez (or assistance prncessing the data. 463 . INC:: VOL 75.,40.3. SEPTEMBER 1t,

t N. ...U...l.,r , -. . SiA2fE Involuntary Smoking and Lun4"dan=cwe'wr ' TABt.E 1: Lu»q cancer in women who never smoked- Re+cnrds of S' haspita4s: 12r1 -81 No. of women examined. Status At hospitals: Total % A B C D Nihcroscopic proof of lung 243' 93 276I 563 1.175 -cancer on hospital record • Smoker 200 70 182 440 892 Yonsmoker or smoking 43' 23 94 123 283 100:00 habits not stiated` Reinterview, revealed 15 3 41 54 113 39.9 smoker . Rr•interview, revealed' 1'8' 14 45 57 1'34, 47.3 nonsmoker. No microscopic proof of' 10 . 6• 3 12 36 12.7 lung, cancer ' 6896of the hospital records listed patientas nonsmker, in 3296 of ie rer.ords: smoking habits were not stated. i :view of slides by one of us (0. A.);, and, 113 (319:9%') were )und to be smokers upon reinterview; Only 134 (47:3%) ,ere lifetime nonsmokers with histologically proved' rimary lung cancer. Theyy were the only cases therefore witable for this stwdy. Among, the colonf rectum cases. -iere were many' fewer that were misdiagnosed-only .•1%. . The age distribution of the cases andiconltrols is shown -i : table 2. More than half were 70: years of' age or older, nd 2`? 4 were 80 years of age or older at the time of iagnosis: The histologic diagnosis of lung, cancer cases ras as follows: 65%-adenocarcinorma. 116% large cell. 8A lNarnous cell. 4% oat cell. 3 0' alveolar cell. 3b mixed. nd 1% ooo undifferentiated for classification by cell TABLE 2.-rtps diatriulncGion of lung, cancer cases and controls"_ A Cases Controls .- ge: yr No. % No~ % 40-49 5 3:7 17 4.2 50-59 28 20.9 86' 21.4' 60-69 28 20.9 88' •' 21.9 70-79 44 32.9 121 30,1.! 80-89 M 17.9 82' 20;4' _90 5 3:7 8' 2.0 Total 134' 1100.0 402' 100:0 TABLE 3.-OR for matthed qroupe of women for risk of lunq ganeer /rrmn exposure lo smoke. as classified in 4 eaGeqpries Classification Risk of lung canl for women .~ . ©R' • CL .. . Exposed to smoke over last 5 yr 1.28 ' 0.96'-1.70 Exposed to smoke over last 25 yr 1.13' 0.602.14'. Husband smoked ' 1.22' 0.97-1L71. Husband smoked at home 1.31 0.94-1183- srnoke for the last j', years and' for the last 25 years. The women exposed during the last 5 yearss had an OR (adjusted for hr exposed per day) of 1.28~(95% CL: 0.98, 1.66) and those exposed for the last 25' years had'an OR of 1.12 (CL: (D.31. l.42): No increasing trend, wich increasing exposure was apparent in either group. In the 5,year exposure gToup~ the OR went down with increased eacposure„bt,ut the OR in each of the exposure groups was not statistiically significant. ( Table 5' and tea~igwsrs~-1-arcrd ,2`'show exposure classified by the husband's smoking habitjL.The:OR`:fbr ;Pc, women married to smokers was 1.23 (.'Tln1Q4-+_-&l~;:itl~ Table 3'shon.vs the OR andl CL for risk of' lung cancer, those whose husband smoked at home it was 1.31 (CL: ccc :ng to the 4 methods of classifying smoke exposure. 0 99, 1.73). Women whose husbands smoked 40 or'more. 'lte OR ranged frorn 1.13 to 11.31. Ml 4 methods resulted cigarettes a day iM4•• n~' t.99 .(CL: 1.13v 3.54).. Io 95'% CL f l h ' ' i M1Ver o ess t an 1 1 and werrnot statisticalll ,v -women whose husbands smoked 20'or more cigareties at gniif icant. Table 4 shows the average number of hours per day -ialt cases and controls were exposedl to other people's home had an OR of 2.1'l (CL: 1.13„3:95),..These were the only specific smoking 'groups in which the OR were statistically significana The Mantel extension oest for TaBLE 4.-Number of'cmscx and'conltola exposed lo smoke of others during 5«nd 25 yr before dia>nosia Variab l Exposure. No. of hr/day Tot;al , e None 1-2' 3-6 ?7 totai Mo. of', women Last 5 yr No. of cases 80 15 25 14' 54 134" No. of'controls 263 31 59 49 139 402 '. OR 1.00 1.59 1L39 ' 0.94 1.23 95% CL U0-?72 0.96-2.03 0.69-C,28 0.98-1i.66 Last 25 yr No. of'aases ' 42 17! 451 30 92' 134 No. o[ eonorolh 136 72 109' 85 266 402 ©R' 1.00 0:77 1.34 1.14 1.12' 95% CL 0,60-0:99 0.96-1.87' 0.83'-1.57 0.81-1.42' r•ri vnr --n r 1FaT-1AF4 1

Garfinkel, Auerbach, and Joubert TABLE 5:--Smoke exposure before lung cancer diagnosis. as claxsiCed by heuba4tdi arnoking habita Husband's total!smokling habits Variable No. of'cases No. of'controls OR" 95% CL No. of cases No. of controls Olt" 95% CL ;1 " Cigarettes/day Cigar andY Allitypes one <20 20-39 >40, or pipe of'smoking 43 148 1.00 11 45' 0.84 0.61-1:16 32 102 1.08 0.8'1~1.44 30 52' 1.99 1.13-3.50 1'8I 55 1.1'3' 0.78'-1.62' 91 254 1.23 0.94-1.60 Husband's smoking habits at home No ° Cigarettes/day Cigar and/ All'types ne <10 1049 j20 or pipe of smoking 44' 29 17 26 18 ' 90 1157 1.00 9'0 1.15 56. 1.08 ' 44 2.11 sa 1.17' 245 131i 0:84-1.5;8 0.76-1L54 1.13-3.95 0:80-1.70' 0.99-1.73 ° Figures include single women living alone. Cohabitants living with single women were classified as "httsbands.."' . antel'extension test for trend (one tailed): x= 2.31. P<.025. ":rtantel' extension test for trend (one tailed): X=? 35: P<.025. trend' in bothgroups, wasstatisticallly, significant (P<.025, one-tai'led test). AnaiCvsis also was done for vears of smoking. There were statistically significant OR for those smoking for 20-29 years (2:21' for totall smoking habits and 2:17 for smoking at home), but no trend was apparenc Those who reported smoking, for 3'0-39, years and 40 years or more had much lower OR that, were not statistically signiificant, OR for exposure categories by age gmoup. hisooliogic vpe of lung, cancer, identity of' the respondent who wa tntervie.ved'n and socioeconomic status. Data are for aven gce~posure for the last 5 years, for the last 25 years, by ht~s~and'sh eotal smoking habit. and by his smoking habits at pme: OIt generally were r over, for those with h ter for those 70, years of • age adenoKarcinoma, when, sorneone ot in the immediate family was the responderit. and for tta lower middle class. There does not appe 064 of high OR for any of these subgroups in a s. categories. Some of' the OHt are statisticallly 3igniificainr, but th~ey us~uailly~~ carry very wide~ CL with~ them. " Total No. of wome 134 ' 402' Total ;. No. of'wome Tab1e 7'show$.Oe OR for classification of exposure women co stmo-k ac home; at work, and, in ocher areas, compared wit ihose wom~noc erposed arall. O,R'( exposure at ork dluring the last 5 years;-tiva5 0.88ffor`t last 25 y7>s. it was 0~.~3. The highesvbR obsefved w, 1'.77 for~eaposure during the last 5 ye~s in'"9fher ax6i: Nont-~of;- the`RR" sEtown= tn`t~'is aab~le az~:~tattsu~ significant. _ ' One of the questions in the interview was with rega to exposure to smoke in childhood. Those women w: replied, that they had been exposed in childhood had RR of' 0.91 (CL: 0.74', 1.12). LOGISTIC REGRESSION ANALYSIS An unconditional logistic regression, model was used which~ included terms for age. hospiital~, socioeconomstntus, and year of dliagnosis-to account for possir confounding factors. Testing was done on each of t: four exposure variables, three of which were used' in t: i4Enntel-Haenszel analysis: l) exposure during, the.last a:n 4.00 CASES' 43 CONTROLS 14s <a0 :C.•]s. I 45. r.IQo11f IN C dOT. e in the lower or toibe a pattern 4 expbsure 1:13 ~[1 al.l 3MOKCRi <10 a 1.06 1o-Nli 20•. ~ -- Ctpor.tt.y a GoT eR~r I 32 30 tt+ 91 CASES 44 29 I7 26 119 9C IO2 52 53 ' 234 CONTnOI.S' 137' 90 36 44 55 24'S TEDtT•FIra7RE' 2.-0(~ (or exposure to husband's smokink habits •rlr.trRE 1'.-OR ' (or ex TcxT posure to husband's total smoking habits., ._, home.

Involuntary Smoking and Lung_Caaq Teets 6.-OR Jor smoke rrpoeun cntievories. by aqe'9rouPe hiatbingie~" , o/ lxnp camoer, tio'ant{t1f . of res7wndrnG' and socioecoxo+nie statua Smoke exposure Specification No: of' cases ls5t 5 yr OR 9596 CL. Last 25 yr Husband's smoking habits Total At home OR 95% C1. OR 95% CL OR 9596 CL A qe. yr <60 33' 0.96 0.65'-1'.42 60-69 23' 0.82 0.57-1.19 70-79 44 1'.32 0.93-3.53 ?80 29 2.00 0.76'-5.25. 1.00 0.63'-1.51 1.1!9 0.72-1.98 1.30 0.75-2.26 0.55 0.41-0.73' 1.20' 0.66-2.19 1.42 0.70-2. 1.22' 0.7$'-1.90' 1.26' 0.79-1.99 1.43 0.85-2.39 1.75 0.81-3.78 1.28' 0.72-2.27 1.10 0.68w- 1,79 Histologic type Adenocarcinoma , ,87 1.•l3' 0.99-2.06 Squamous cell carcinoma 11 1.28' 0:52-3:19 Large celllcarcinoma 21 0:550:4'1-0:74 Mixedand'other152:291 0157-9:10 1.15 0:85-1.56 1.33 0.94-1.87 L48 * 1.01-217 U5 0:43-1.69 &00 • 1.28-1'9:33 5.00 1.43-20.18 0.67 0.47-0:94 0.76 •' 0.51!-1.13 0.62 .0.45'-0.86 2.67 0.4'li-1'7.35 0.81, 0.48-1.37 1.00 0.53-1.7T Respondent Self 16 1.96 0.62-6.17 0.91 0.51-1.60 0.83 0.50-1.38 1.00 M5-1.74 Husband 34 1.00 0.67-1152 0.4GI 0.38-0.55 0.77 0.56-1'.06 0,.92' 0:63-1.34 Da+-ohter or son 48 0.92 0.67-1126'. 1.41 0.85-2.36 3.57 0.84~-15.28 319 0:9'1-11119. 0't 36 2.23 0.90-5.r1 2.23' 0.83-5.96 •:1.58 1.11-2.67' 0.77 0.57-1.03 Socioeconom ir status Goper and upper middle class 6 1.60 0.31-8.19 1.50 0:34-6.59 1.23' 0.36-4.18 1.50 0.34-6:59 Middle class 75 0.78 0.63-0:97 0.92 0:71-1.1!9 1.15 0.84-1.59 1.21! 0.87-1.69 Lower and' lower middle class 53 2.58 1.10-6:01 1.45 U6-2..14 1.23 0.83-1.84 1.45 0.88-?38 rears. 2) exposure during the last 25 years, 3)~ husband's moking at home, and 4) husband's smoking outside the torne. The latter variable was used rather than the tusband's total smoking as a check of the valid'iky of xposure to husband's smoke and was derived by sub- racting the number of cigarettes smoked at home fromm he total number of cigarettes the husband' smoktd per iav.. Each of, these factors was tested as a continuous xposure vari'able-the mostl powerful technique for 'etectinig any true underlying risk. T:+hle 3'shows the results of'this analysis. Exposure for ve .,and25y,earshadneptivecoefCicienits:Thetesrfor cigarettes smoked by husband at'home showed a posiitiv trend of increasing,risk with increasing exposure and wai statistically significant, with a P-val'ue (one tailed) "c. .0312: The test for cigarettes smoked outside the home wal not statistically significant. The table also shows estil mates of RR at the 10 hours per day exposure level and a~ 20' cigarettes per day smoked by the husband. The RF from exposure to 20 cigarettes/day smoked at home wa. 1.70; outside the home, ic was 1.26, RR from eaposurr during tlne la~st° 5 years and during the last 2'5, years wer less than 1. A separate analysis that included responden idetutity did not change the results materially. TA eLE 7.- Vumber ojcases d' ntr ls d t k TABLE 8.-Logistic rep7-rssion nwdel' jor involrccnta"j smoke exposure variabfes, on continuous dose-response basis , an co o expose o smo e of others ai horne, at work. and' in other areas Cuefficient ' . Smoke ~ Variable Smoke exposure Variable (,S~E) P=value exposure leveG::! • RR. None At home At work ln other areas 5-yr exposure to -0.0069 0.422 10, hr 0.98 Laat5 yr smoke (0.0035) ' 25-yr exposure to -0.016 0.303 10'hr 0.85 A',lo. of cases 80, 37 14 13 smoke (0.03'1) No. of controls 262' 99 52', 24 Cigarettes 0.026 0.032 20 cigarettes 1.70 OR 1.00 1.22 0.88 1.77, smoked atr home (0:014) 95% CL 0.92-1.62 0:66-1.18' 0.93-3.38 Cikarettes 0.012 0.127 20 cigarettes 1.26 L.1st25yr smoked outside home (0:01i0) No. of cases 42 73' 34 19 onomic status. anc socioer • Model includes terms for age. hospital No. of controls R 135 204 118 43'. , . year of diagnosis O 9 C 1.00 1.15 0.93 1.4'2' . 6One tailed 6 95 L 0.89-1.49 0.73-1.18 0.89-2.26 . Q1Relative to the nonexposed woman. JNCI. Vn't_ 75. NO. 3. SEPTEMBER 1985

,Ini a previous paper (4) the problem of dassifying ., : iRvoluntary smoking on t;he basis, of the:,:husband!s~ smoking habit w,as•discussed. It was pointed out that~ questions directed at ascertaining a quantitative estimate of'the number of hours a day that subjects were exposed might be a better measure than the total number of cigarettes that the h'usbandisrnoked, inasmuch as not.all of' the husband's smoking was done at home.. lin the present study we classified the exposure both ways: by the number of hours per day the subjects were exposedi to smoke of' others and by the husband"s smoking habits. We also recorded'the respondent's estimate of how many cigarettes a day the husband' smoked at home. In this group of women, husbands' who smoked cigarettes I srnoked' an average of' 2'7' cigarettes a day, of which 11.5 cigarettes on average (43%) were smoked at home. O'f' course: all cigarettes smoked at home were not necessarily smoked in a room where the subject coudd' have been exposed. In this study, the husband's smoking, at'' home was related to the women's llung,cancer, whereas number oPhours of exposure a day to ad1 sources of tobacco smoke was not related. A potential source of error was the hospiltal"'s report of wl• rer the subject smoked or not. In this study, 40% of :he women with lung cancer. classified as nonsmokcrs (or ;moking,notstated) on the hospikal''record, were smokers ucsometime(tabde l). Another 13rod'id'nochaveprimary ung cancer. lirisapparent, cherefore. that in anystudyof histories ofs~ha sutbjecrs'have to be:confirmaif~ the exient~of their`:, tnvoluntary txpot~une zso~es o~rhusbamds, were ;i~btaut ed~fbr ~ ,. ~ ~ ' ~ ' u titio w ~1o were snookeis: The disirnb ri by si'noking'hagi . . shown in table 9: As we might have expected, smokers a:. more likely to be married to smokers than are rii smokers. The table shows that 43'of 134 women, or 32:1` of' the cases included as never smoked in this study ha husbands who never smoked; but only 21 of 113; or 1816~ of' women who smoked and were mistakenly classified s nonsmokers in, the hospital record had husbands who di nocsimoke: Among the controJs only 8.5%'of'womenwh were called nonsmokers (orsrnoking was notstat'ed) wel smokers. _ The table shows the effect on the OR, w'hen' or. assumes that 8.5% of the additional controls needed fc the 1i:3 maitch had, husbands with the same smokin ;..> distribution as the husbands of cases-who were srnoker: and that the balance had the same distribution as thac~. the4102 controlsincliud'ed in the study. The=OR for~:i~i husband"s smoking increase to 11.61 overall and ane'~ high as 1.631 fbr the 2q-39 cigarette a day smokers and2:3 for the women whose husbands smoked 40 or .mor cigarettes a day. For,exposure to'the husband's smoke P home, the OR are 1.66 overall, 1.53 for women who's husbandssmokr 10-19 cigarettes aiday> and 2.85 for thos whose husbands smoke `?IO or more a day at home. Thu the inclusion of'women whosrsrnoking habits have nc been reviewed greatl •v increases the O R. TAeLE 9.-FDypothetical OR resvlting jnom oomb'ininq women in study with women oriqinally classified as nonsmokers but who actually smoked Htiasband's total smoking habits Variable Cigarettes/day C'igar Adl Totals None <20 20-39' ?40 and/or piipe types of. smoking No. of cases ;fonsmokers (in study), Smokers (originally called nonsmokers) Total (unscreened) Vo: of controls Nonsmokers (in study) ''iitional concrols' Total (unscreened) (DR 43 211' 64' 148' 1'1'9 267' 1.00, 11 , 9 20 45 38 83 1.0i11 32 43 75 102 90 192 1.63 30' 24 54 52 45 97 2.32 18 16 34 551 47 102 1.39 91 92 183 254' 220 474 1.61 134 11'3' 247' 402 339 741 Husbandls smoking habits at horne Cigarettes/day Cigar All ` Totals None and/or types of. <1i0 1I0-19 jO_0 pipe smoking ;1m of cases Nonsmokers (instudy) 44 29 17 26 l8 90 134 Smokers (originally called I nonsmokers) 23' 22 22 30 16 90 11'3 ` Total (unscreened) 67' 51! 39 56 34 180 2d7 No. of controls `fonsmokers (in study) 157 90 56 44 55 245' 402 Additional controls' 126 75I 52' 39 47' 213, 339 Total' ('unscreened) 283 165I 108 83 102 458I 741. . JR. 1.00 1.31 1.53 2.35 1.41 1 1.66! ' Upon reinterview.8.596oCthecontrols were found to'besmokers:They were distributed according to thesmok'ingdistribution of husbandss eases who were smokers. The balance were distributed according,to the smoking habits of controls'in the st'udy:. tCI. vp(_ 75. Y©: 7.,SEP'TEMBER 1985

The classification used in this study might be aitici~ed because some women married to ex-smokers could be counted in the same exposure category as a.woman exposed to smoke up to the time of her final illness. - - However, all patients who have gone through diagnosis and treatment for lung cancer had some period of time when they were not exposed to others' smoke,, either before or after treatment. We believe that the classificx- -- - tion we used was indicative of the "usual amount of smoke to which the person was exposed." To determine the experience of a"pure." nonexposed group, 17 cases and 56 controls in this study were identified who were, not exposed to the smoke of others durine the last 3 years. during the last 25 years, whose husbands never smoked at home or elsewhere, and who never were exposed to smoke in their childhood. These ese case•s and controls were compared with all other subjects. The OR was 1.14 (CL: 0.81, 1.59). In conclusion, we found an elevated risk of lung cancer, ranging from 13 to 31%, in women exposed to the smoke of others, although the increase was not statis- ticall,v significant. The women who were married to smokers of -k0 or more cigarettes a day or who were exposed to the smoke of at least 20 cigarettes a day at home showed a risk twice as high as that of women not exposed at all. This result is consistent with the dose- response risk of exposure to the husband's smoke shown in some case-control studies (2, 3). A dose-response relationship was confirmed in a logistic rEgr-ession analysis. The lack of a relationship when exposure was classified by hours exposed to smoke of others may have Occurred because this variable does not accurately mea- sure intensity of exposure. There is no consistently " ~ Involuntary smokIng and:Luii - hi h i k f o o g er r s or certairr age gr ups or by hisi -6ypes: loc:t or oy exposure at tiome or at'worlE.•;Exposure'in other areas carried a higher OR, buEthis,fir3ding is c3ifficult.to; interpret. _ ~ REFERENCES (1) HtR,tYaMn T. Non-smoking , wives of hea vy smokers have a higlr~;: risk of lung cancer: A study from, Japan. Br. Med J'1981: ' 282:183-185. (2) TRtCr+orout.os D. Krt.AneiDt A. SrAstttos LL et al.:Lung cw.nser' - and passive smoking. Int J Caneer. 1981: 127:1--1 : ; (3) CoRR.Ew P. FotrtHAM E. PtCC.t•.E CW ei al Passive smoking an lung ctnEer. Lancet 1983: ?-593 597 : .: ` " (4) GnRFlNkEL L Time trends in_ lung •Ganeer mortality: . among; nonsmokers and a note on passive scnoking. JNCE 1981 ;66c: 1061-1066. (5) KA6AT CC. WYNOER EL'Lung cances, in-nonsmokers. Cancer=' 1984. a3:12_14-12v1. (6) SAN©t,ER •DP. EvERSOrt RB, Wtt.COX Aj. Pauive smoking in adult= hood and cancer risk. Am J. €pideinio! •1985: 121:37-~18.- •, . (?) Cccnr~ W C• Fur+c SC. Lung cancer in nonsmokers iti Hong lEong.. ,, lns Grundmann E, ed Canea aimpatgn ;-VoL;,6 `Canee epidemiology. Stuctgan and New;York: FisF)ier.Vcrlag, 198?:'-°=~ 199-202 (8) Koo LC. Ho J1-1-C. Snw D. Active and passive smoking among female lung cancer•'paucnts and controls in 1•long r+.ong. J Facp Clin Cancer Ru 1983: 4:36i-373 - ;. . (?) FRtEDMAN CD, PE7Trn DS. BawOt. RD Prcvalence and correlates" of passive smoking. Am J Public Heilih 1983: 73:401-405. (10) AUERawctt O. GARnNcEt. L, Patus:VR; et al. l•iistologie type ef lung canccr and asbestos cxposure.'Canccr 1984: 5i:3017-SO21. (11) PIxE uC. MoRROw RH, -•SrtdtlsfiGll.,analysis of patient-control studies in epidcmiology FactQr under investigacion on all-0r- none variable- Br J Prev Soc Med 1970; 25:42-44. (12) MiE-rnNEnr OS. Esumability and estimation in• case-reFcn:nt xtud'tes• Am J Epidemiol 1976: t03:Y2-6-235. (13) BRESt-ow NE. DAY NE Statistioal methods in antxrresearch. Vol 1. Analysis of case-control studies, Lyon: IARC, 1980. M43E7.7.437,07.

---