Tobacco Documents Online

untrles: a revlew~ ::. 46'.5~-81, osiso4 causesof verbal autops,v:~ ?' 70: ' 499-507' ~IeS~~ too estimate rnniqt.~. 1992: . 7: on nJ:aYurernnl ch Paper 88-1. -ondon$chool nrinC Cuute.t,nl :ins Universitv rJntemational •l6rm/rn• Rurrs ~IS ~4"euk.. )93), International Journall of IEpidemwlo4yy i ' ' P~,-q ' a./'' ' !' F C "' PY a International Eptdemiologicat Assoclation 1994 Vok 23. No. 2' Pfintedun Great Bntain i Childhood and Adolescent Passive Smoking and the Ri~sk of Femalle Lung, Cancer FU-LIN WA'NG,* EDGAR JOHN LOVE, NING LIU'f AND XUlDiONG'DIAI'f Wang F-11(Department of Community Health Sciences, Faculty of Medicine. University,of Calgary, 3330 'Hospital Dt. NW;,Calgary, Alberta. CanadaiT2N 4fN1); Love EJJ Uu N'and Dai X-d. Childhood and adolescent passive smoking and the risk of female lung icancer. lntemationa! Journal of Epidemiology 1994; 23: 223-230. Background Few studies have reported the relationship between passive smoking (PS) in early life,and the'nsk of lung cancer. This study was done'to evaluate the risk of female lung cancer from PS;,especially that dttring childhood and adotescence: Methods. USing househoVd lexposure to tobacco smoke as an estimate of PS, a 1:1 paired case-control study was conducted in Harbin. China. We:interviewed 114 female primary lung cancer cases, aged 30-69 years, and their hospital-basedlcontrols. The controls were non-cancer patients, selected from the same hospital as the cases, and matched on age (± 5 years), residential area and smoking status over theiriifetime: There were 59 pairs who ever smoked and 55 pairs who never smoked. Information on PS was collected by residence foreach; of the following periods: 0-6; 7-14, 15-22, 23-30 and i31 -69 years i Results. Household PS significantly'increases the risk of',female'lung icancerforthose exposed at ages 22 or younger, who have:ever smoked:,The risk was also increased for those non+smoking pairs when exposed underthe age of'15 years. Exposure to matemal'smoking at ages 14 or younger, increased the risk by about 170% (odds ratio, OR 2,7, 95% confidence interval [iClj: 1,.49-4:88), but not to paternal smoking (OR I1'.41D, 95%Cl: 0!92-2:5D). The risk was highest'forthose exposedlundenthe age ofiseven (OR 3.46, 9'5°f;Cl:1-60-6i65) and wasalsosignificant at ages 7-14 (OR1.0tt, 95%Cl: 1.62-5.57)'and 15-22'(OR'.3.1I0, 95%Cl: 1.52-fi,31) years. Under the age of123 years, the OR. increased with amountof PS(P< 0:001). Of inote, the O'R in all'five exposure penods fornon-smoking pairs weresimilar to those for all 114 pairs studied. Conclusions. Household PS, particularly that during childhood, increases the risk of female lung, cancer. The assessment:of PS should be done'by'ddfferent ~ penods of exposure. There is increasingevidence suggesting that involuntary or passive smoking (PS) increases,the risk of lung can- cer.14 However, the fondings have not been consistenti with respect to who are at:higher risk: adultsor children: Some studies have noted the importance of exposure to tobacco smoke in early life in relation:to cancer, risk inn adulthood.' 10: while others found no ~ associa'tion for non-smoking women exposed to tobacco smoke during childhood.4 In addition, the methods to estimate PS'varied by study. The estimate of PS fromionly the husband or the spouse, which has often been used. is in question.1° I' This~study;,using household exposure:to tobacco smoke as an estimate of PS, attempted to clarify further the relationship between female lung cancer and PS, . Department ~of Community Health Seiences: Faculty of IWedioine: Uhi-.ersity of Calgary. 3330 Hospital Dr. NW. Calgary. A'Ibertaj Canada. T2N14Nit. 'Department.ofEpidemiology, Heilongiianglnstitute.for Cancer Re- search. Harbin, The People!s.RepublicoflChina 223 especially that dttring childhood. Data were collected inn the city of Harbin, the People's Republic of China. during 11985'-1987. MATERIALS AND M ETHIO,DS The eases were female primary lung cancer patients:who were selected from the Harbin Cancer Registry. The criteria for inclusion were: Residentlof Harbin city, aged 30-69 years and diagnosed by histopatholo¢y during Ianuary 1'985 and,l*lovember 1986,,A total of 114!cases were eligible and alllwere interviewed''face-to-face. They included 55' adenocarcinoma. 28 ~ squamous cell carci* noma, 20 small cell and oat cell lcarcinoma, and I i other types. A reporting system was set up, to ensure quick registration of cases, and immediate investigation. The controls were patients withont cancer;,from ad- missions to the same hospital as the cases and matched on age (± 5 years), residential area and smoking status over lifetime (whether or not a current or ex-smokeri): Information on isex, age, residen tial area and the diagmo- sis was obtained from the patient file and the preliminary

224 I NTERNIATIONAiL JOU.R N/,L OF EPIDEMIOLOGY TAeLE I Charactrrirtics ojthr stud{~ .nebjeeta in a 1: I'matehed easo-contral stuQtt Harbin. China. 1984--148Z' Characteristics Cases Controts' P-value? No. i. No. i: A. Demographic Characteristics Nationality Han 107 93:9 112' 98.2 1 0.174 Other 7 6.1 21 1.8' Ethnicity tteilongjiang 57 50.0 49 43.0: Liaoning AL Jilin ~ 20. 17:5 1& 15:8~ 0.584 Shandon '? 19.3 29' 25.4 Other 16' 13.2 I8 1'5:8 Education Literacy 59 51.8 65 57i0 Pr'nnaryaohool 20 17.5 2_' 19:3 0i684 Middle school 27 23.7' 20 I7:5 ;- College 8 7.0 7 6.1 Occupation White collar 9 7:9 5 4.4 Bluecollar 63 55.3 73 64.0 0.456 Housewife 37 32.4 33' 29.0 Other, 5 4:4 3I 2.6' Marital status Married I00 87:7, 103' 90.4 0.546 WidowedTdivorced 14 ' 12.3 11 9.6 B: Smoking Habits Eversmoked cigarettes Yes 59 51.8' 59 51.8 0.999 No 55 48_2' 55 48:2 Age ;began smok8ng (n = 59) Mean t SD" 19.1 ±9.56. 21.6 t10.0 0.1179, Minimum to maximum 5-59 6-58 Number of cigarettes/day (0=591 Mean t,SD 13.4 ± 7:72' 10.2 ±6.40 0.013 Minimum to maximum i 3-34 2-25' Years of smoking ( n= 59) . Iv[tan t SD 316 +_ 13.? 3'1.1 ±: 12,6 0:546 Minimum to maximum 3-SU 7-46 ' x2 test for 2he;ditTerences in the percentage ofdemograph'icwariables between cases and controls or Studeni i r test for the differences in means of ' smoking habits between cases andoontrolsd b Standard deviation. 'eligible' patients{i.e: age and residential arealmatched) were ascertained! The f rst eligible patient whawasfpund to match the,caneer patient for smoking status over lifetime was enrolled in the study, The distribution of'the controls was as follows: 18% had cardiovascular, disease, I'5"/b had digestive system disease, 22% had respiratory infections, 3C1%were in hospital for acute surgery,, l 0`%d had gynaecological problems, and i 5% had other mmn+malignant disease. Using a pre-set questionnaire. the cases and controls were;interviewed in hospitals oraTthe patient's h!omrby trained interviewers; some of the cases and all the controls were interviewed by WF. The questionnaire included questions on demographic informtartion, resi- dential history. smoking history. and the,indoor smoking , habits of'the regular family members. etc. Information on indoor smoking was collected for each residence in which the subject had,lived for 3 or more years:,In this way, information on PSduringdiffbrent periodswas also obtained. TABLE Studj•, subject, A. Expk Cax Con B. Expo Case Coni • Numbc In this s ofindoors the same li• or more; a rette or orr amount of similarly, I when1 thee changedl T cancer risk. tobacco x y . PSI-t9 (grams) by : Case-cor groups of e. 6'9 years. C pairswhosmoked! M• for calculat interval (ICI method. Th, anallfjsiswaf Info (versio. i

Cases 8'5 69.4 30i6 100!0 Controls 81) 68:8 3!1.2 100.0 TistE._' liredi.urihwmnolsrwrats.of~sid"srrnam.saroke.hr.espusureptriodinHarbin. China~ Study No. of subjects d Sources~ol sidestream~smoke(%')' ~~ :ubjciu otpose Father Mnthen. Grandparents Others Tbtal A. Esposure.at age'0-t4 years Caies 104 45.2 38:5 6.7 9.6 I00i0 Controls 73 5I1:2 26.4 9.3 13.21 100!0 g..Exposure acaee I5-699 years Husband Others Total i controls home'by j all' the' tionnaire ion; resi- smoking trmation dence in' ;. Inithis was also "Number exposed Ito specilic sourceltotal exposures expressed lass percentage.. In this study; PS'was de8ned',as I gram/day or more of indoor smoking, byany member of the family sharing the:same living accommodation as the subject for I yearr or more: a gramidav is roughly equivalent to one ciga- rette or one' pipe of tobacco per day. The cumulative amount of PS'was estimated using,the weightedimean, Similarly, we also calculated the weighted mean of PS when the smoking status of the family member(s)) changed. To measure'the eumulative effiect of PS on the cancen risk, a,passive smoking index (PSI) in grams of tobacco x years of:smoking, was.calculated! i.e. PSI = the average daily consumption of tobacco (grams) by family members x years of smoking Case-control pairs were classifledl into the fiive: age groups of exposure: 0-6. 7-44; 15-22, 23'-30andI30- 69 years: Of the' t'.l4 case-control pairs, there are'59 pairs, who ever smoked and' 55~ pairs who never smoked. McNemar's analysis13'was used'throughout for calculating odds ratio (OR). The 95% confidence interval (CI) for OR was'calculated using Cornfield's method. The x''statistic'was used to test for trend. The analysis was done on a~personal computer using Epi Info (version 5), „; RESULTS The cases and controls were comparable with respect to the matching variables (age, residential area within Harbin and smoking status over lifatime)land the follow= ing: nationality, ethnic originl education and occupational status (last job'); the number of years that they'have'Jived in the city of Harbin and previous history of respiratory disease, i.e:,chroni¢ bronchitis, emphy- semaorpul}nonary'tuberculbsis.Table I summarizes the selected demographic characteristics~ and smoking' habits of the cases and controls. For the 59 smokine pairs, there are no differences between the cases and controls in age at which smoking began and',years of smoking. However, cases smokedI more cigarettes per day than controls. Tablel presents,the distribution of sources of side- stream smoke for two age groups: 0-14 years and - 15 years. In'the first group, although the sidestream smoke exposure was mainly from parents; the percentage from mothers for tihe cases (38'.5°/i) was higher than that for controls.(26.4"/r): Forthe;oldergroup, husbands became the most important source: of sidestream smoke: ac- counting for more than two-thirds for both cases and controls. TratE3'RiskqJfrnwlelungrannerduringthildlroodi (1-' lJ rears) -riposurrroparerna6andnraternaltohaceo.smokin;e1 hr Harbin. Chinn Exposure PASSIVE SMIDKING~,A V tD TtilE RISIC~QF'FEMl1.~L'flILIWNG,CANCER Number and proportionexposed Cases Controls No. ~ No. Paternal 47. a1L2 39smokintt. Maternal 40 35:1 I9smoking Oddsratio. 95% . P-valuee confidenee interval 33:3 1.40 • 0.79-2;50. 0.273 16,7 ~ 2.70 1 L 39-5: 30 ~. 0~.003'.

226. 1NTIERNATIONAL JOURNAL OF EPIDEMIOLOGY TAet.E. 4. Risk oJfrnrale lung ranner.from household exposure to tobacco smoke br e:xposure period (age) in Harhin.Ch'ina Exposure Number and proportion exposedd period (age~in years). Cases No. % Controls No: .. Odds,ratio' 95% confidence interral P-value 0-6 75' 65:8' 48 42.1 3.46 I.80-6 65' < 0100 1 7-14 73 64~i0 48 42.1 3.08 1.62-5;87' <0i001 15-22 8'b 74.6 64 56.1 3.10 1.52-6:31, < 0002 ' 23-30 70 6'1':4 72'. 612 ' 0.91 0.28-2198I 0447 ' 31-69 75 6'5:8 W 70:2' 0.78 0.36-1.69 0:348' ' OR ftom McNemar analysis., OR'= Wc: x? =1 b - c - I~ h(b + c) As shown in Table 3, the difference in percentage of' exposure to matemal smoking,underthe age of 15' be- tween cases and controls was highly significant (P' = 0:003); the OR for'female lung cancer associatediwith exposure to maternal srrtoking was 2.701(95% CL• 1.39- 5:30):, ln contrast, the difference between childhood exposure to paternal smokingwas not statistically signif- icant (P= 0.273). Table 4 shows the OR for lung cancer when PS oc- curred for the following age groups: 0-6. 7'-14, 15-22! 23-30 years and 31-69 years. It was indicated that the risk for lung cancer was highest:in those exposed'under the'age'of 7 years (IO'R' 3.46, 95%CI: 1.80-6.65) ,and was also significant at ages 7; 141(O'R 3.08„95°Lo CI: 1.62- 5.57) and 15-22 (OR' 3. 10: . 95°1o CI: 1,.52--6'.31). Anali7sis of, lifetime exposure (PS occurring at any tinte)'showed that the crude'OR forltrng cancer associated with lifetiioe exposure was 2.67 (95% Cl;: 0.90-8.88), which was close to the significant level (P = 0.055). In reviewing these results:,it'shouldlbe remembered that mostichildren in the People's Republic',of China remain at home until age 7 whemthey start to attendlschool on a regularbasis: TAete. 5. Female lung cancer risk from household ~PS hr e.xposure.period ur smoking pairs and inon-.xmoking,pairs.Harbin, China Exposure. Number and proportion exposedd p eriod , , lagein Cases Controls Odds 95v,,,.. P-value.e years) No. %, No. .~ ratio" confidence interval A. Sinoking, pairstn , 59,) 0-6 7-14 47 47 79.7 79.71 33 34 559. 57.6 3_33 ?.86 f.17-10.76. 1_06-S4S 0.011 0021 1 S-2'- 23-30 53 411 89.& 69.5. 4'- 42 71 -2 71? 4.67' 0!89 1.12-29!42 ' 0!28'-2.8'2 0!015 0!999' 31-69. 43, 72.9 47. 79;7. 0167' 0!2'-'1'L92 0:502 B. Non-smoking pairs ( n = 551 0-6, 28: 50.9. 1 S. 27.3 3:60. 1.15•-13:33' 2 01012 7-14 26. 47.3 14 25;5 3.40 1.08!12:69 uJ019 15-22 23-30 3']' 29. 58;2 52:.7. 2D- 30 400 5'.t,6 2:43' 093 0$C-7.3? 0:38-'''S 0J(Xi6 0'999' 31-6'9 32 58;2 33 60:0 019,1 0:3'-2.53. 01999~ ' OR'from McNemaranalvsis: TAi e (aF~ 06 7-1- 15__ 23-:. 31-r To lexarr smoking', ot analysedlb~ in'all expost and non-sn- years (Tabh groups of ez to,those an The lum.• hold': PS w; signifcant - three younc 7 shows th cutnulative_ ticoably, an the younge, tiend P <'0! mISGIJS51 r Theeffecu c tanoer; ha. 1980s. This smoke, sides (a)ppb'rene, s *aesas mu, as manyset

PASSIVE SMOKING AND THE RISK OF FEMALE LUNG CANCER Tlawae6 RislcoJfamlelwg;ranrerJiomPShrkrefo(e:rpomrreruvlpesiadale.rpasure/n Har6in. Chsw, posed under 651 -dlwas '/o t . 1.62- 1)J Analysis me) showed ovithdifetime A was.close ewing these ildren in the : until age 7 r basis. Period lof Level of Cases Controls 95 ;' exposure (agr,in years) cxposure /gram/dayl No. "',of' total No: of' total Odds ratio confidence, interval. F-valJtr < 5' 54 473 82' 7,1.9 1.00 0-6 5-14 35 30:7 21 18!A 2:53' I.27! 5A6 0!007' a 15 25 ,?0 11 9.7 3;45 1.48'-8.20 0:003i Test for trend I P< 0.0011 <5 51 44.17 81 711.1 1.00 7-14 5-14 37 32.5 21 18;4 2'80 1.41-5.59 O:0U2' 2,15 26 "'8 12! 10.5 3,44I 1.50 -7!99 0 i002 1 Test for trend ! P< 0.001 <5 42 36;8 65 57!0 1.00 15-22 5-14 44 38;6 38' 33:3 1.79 0.96-3.35 Oi068i -v 15 28 24J6' II 9:7 3:94' 1.66-9;49 <01001 Test for trend I P< 0.001 < 5 64 56.1 64, 56.1 1.00 23-30 5-14 39 341 •i0i 35.1 0.98' 0:54-1.78' 01957' ; 15 1'1'. 9.7 10, 818 1.10 0.40-3J04 0!974 Test for trend P'= 0:87d <3 56 49;1 58 509 1.00 3ILb9 5-14 47 41.2 45 39.5 1.08' 0.60=1.95' 0:889' = 15 1'1 9:7 II 9.7 1.04 0_38-2:82 0175 Test for trend , P= 0:8'97 To examine and'control lthe potential effect of acti ve smoking on cancer'risk;the OR for lung cancer were analysed by smoking',status. It was shown that the OR in all exposure periods'were close between smoking pairs and'non-smoking pairs, except in the period aged IS-22' y,ears(Table 5): I't is also noted thatithe OR in all five age groups of exposure among non-smokirtg,pairs were similar to those among all! 114 pairs. The lung cancer risk in relation to the level of house- hold PS was assessed by exposure period. The highly significant 'dbse-response -trends',werrobservedifor the three youngesv age groups of exposure (Table 6). Table 7 shows the OR for lung cancer associated with the cumulative exposure to household'tobacco smoke. I*1o- ticeably; among the,6ve age groups of exposure, only in i the youngest is the OR'statistically significant (test for trend P < OAOQ ). DISCUSSION Ttie effect of PS on cancer, especially on female lung, eancer, has been of increasing,concern since the early 1980s. This is partly because, eompared,to mainstream, smoke, sidestream smoke has'three times as much benzol I (a)pyrene, six times as much toluene, and more than 50' times as'much dimethylnitrosamine;°` which may causee as many severe health problems'as active smoking.15 237. However, how to measure sidestreamsmokeexposure and its effecti on health remains a key issue. SeveraC'U studies tiaokxhe husband's smoking status'as anestimate of PS for wives but this is far from accurate for the following reasons: ~(;I ) other family members, such as married relatives, parents'and children'may be irnportant sources of sidestream smoke; in some oriental countries such as China, this is particularly true, (2) husbands may smoke outside the home, thus exposure does not occur,. (3) also, information about'exposure before marriage is not available. It is reasonable to believe that if PS has an, effect on lung cancerrisk, the nature and extent of that risk during childhood differs from' that during adult- hoodi In this study, householdl exposure to tobacco smoke, from husbands and other family members, was taken as anestimate of PS, and collected for eachiresi- dence„from childhood to adulthood. Asshownin Tablb 2, hpusehold exposure to other family'members' smoke accounted ifor abouYone-third of total sidestream smoke during adulthood, which should be'taken into account when estimating PS andlits effect on health. This study found that'the OR for lung cancer associ- ated with household PS varied by exposure period. This may be partly a result of changes in exposurewith respect to freqµency, intensity; and duration. It was indicated that cancer risk was highestifor those exposed under the

228 INTERNATIONAL JOIllRNA L OF EPIDENf IIDLOGY TA7<uE 7 Risk ofjennle fae8 raecrr,lrorrt PS b}• expo.vice period and eumutarive hoterebold esposure'ro ro6orno smoke. esrimated iby PS/ (St anes'fobaccolyearsiofexposwn) in Harbia Chtna Exposure Cases Controls 95i4 period PSI No. %of No. 7,of Odds confidence P-ralue (age in ynrs)', totat total ratio 'suerval <40 62 54.4 89 78.1 1100 0b' 40- 24 21.11 . 14 12.3 2.46 1.20-3.07 0.015 80!- 28I 24.6 Il 9:7 3!6'5' 1.74-7.76 <'0.001 Testforarend P<0.001', <40 63' 55.3 82 71.9 1.00 7-14 40- - 31 27.2' 23 20:2 1.75 0!94-3.29 0.079 80- 20 17.5' 9 7.'9 2189 1.26-6.63 0.012 Test'for'trend P=0:r20. <4T 68 59.7 76' 66.7 1.00 15-22 40- 34 29.8''. 28 24:8 1.36 0:75'-2.47 0.315 il 12 10.5I 10 8,7 1.34: 0!55=3:29 0.522 Test fortreed P=0.350 <40, 77 67,5 81 71.11 1.00 23-30 40 ~ 30 26.3 28I 24.6 1.13' 0.62-2i06 0:699. 80 - 7 6.1 5' 4.4 1.47 0.45-4J82 0.522' Test'fbr'trend P=0.390 <40 48 42.1 47 41.2 L00 3!1-69 40- 34 29:8 29 25.41 1.15 0.61-2:17' 0471 80- 32 28!I 38 33.3' 0.83 0.44-1.53' 0:543 Testfortrend' P'=0i410 age of l years (OR 3!46, 95% CI: 1.80-6'.65) and increased twofold for those'exposed to maternal smok- ing duringchildhood (OR 2:7, P = Oi003). There are several possible reasons'fbrthis:,('1) ezposvre,for children under the age of 7 may be longer and more, intense, because they usually remain in the home for long periods and have close contact with smokers, especialhy, their mothers;'(2) children'are moresusceptible'to carcinogens than adults due to imperfect functioning' of physieall metabolism, detoxification and immunity, (3)prenatall exposure, i.e. mother smoking during pregnancy. Thee children of parents who smoke have been shown to bee especially susceptiblb to respiratory'problems that'occur soon after exposure to environmental tobacco smoke;16' Correa et al."' in a,large case-control study,, foundlan increased lung cancer risk only for those exposed to maternal smoking bur not to paternal smoking: This finding complies with the result,of our study. Findings from other studies support the plausibility, of increased lung'cancer risk from early life exposures to environmental tobacco smoke.,An experimental'study in animals have demonstrated transplacental icarcinogene- sis with chemical compounds in, smoke.lx It has been shown that compounds in tobacco smoke can reach the fetus via the placenta and may'appear'in breast milk.19'0 Eversonz11 also suggested that exposure to maternal smoking dilring fetal life could increase the possibility of cancer in adulthoodl Therefore, health efl'ects. including cancer, are a major concern among children of smoking mothers as well as among',mothers exposed to environ- mental tobacco smoke. ForPS: cartinine." nicotine" and thiocyanate,2' the markers of PS'; have been frmund after exposure to,tobaceo smoke:,In addition.,elevated blood levels ofi carcinogens in passive smokers were also re- portedl'S'Recently,; an autopsy study26 found significaa epithelial lesions among deceasedlnon-smoking women who were married to smokers compared to rthose ntarried to non-smokers. Findings from epidemiological studies7''9'la'17"7 have provided,further evidence for the'effeet of PS in earln• life on cancer risk in adillthood. A'lllof these studies found1 an increase of cancer risk in adulthood!associated with, early life exposure rto'parents' smoking.,although the',risk from exposure to paternal ormaternal smoking was not of the same pattern. In a matched Icas'e-control study of 191 non-s'mokingpairsiJanerichenal.10'foundhousehold exposure to.. latedlby1mulf in each resic residence)du risk of lung c such increase stlreor'lifetin'ff studyb also fo smoking,wor of household cence (OR 2 these two stu With nespt PS on lung,c exposure as'tl risk of lung a years of expos the'cuml index (PSI). l increase in th, age of 7'(P < of Dalagcr4 ar 6fe exposure : It was sugi betlWeeni PS i exposure or c duming childh or real effects maybeunder attention in: ai lung cancer ri from ]anerich for this view, " other people' For Americai twenties! 1$e> childhood ma collection of aditlthood , an warranted. It should b, tendency in n 0.001')'in our i the'passive'st by smoking s controis can. matched case whether cancc tal or childh, smoke. Il viduals who ~ most, of the studied, thert

'alue 1.1079 012 699 522 671 543 in reach the ist milk,tg"0 o maternal ossibilitv of s: including of smoking to environ- otine'3andl 'ound after ate(' )od re au.,, re- significant ng women se,married, ).17427 have n early life lies found iarted with -zh the risk ig was nott d lstudy of lousehold PASSIVE SNtOFIING.4NDTFHE RISH:OF FEMALE LUNGCAItCER exposure to 25 or moresmoker-yearsdwhich wascalcu= lated byimultiplyingihe'numberofyea'rs thesubjecnlived in each residence by the number of smokers in thao nesidr:•nce)during!childhoodand adolescencedbubled the risk of lungcartcer (OR 2:07;,95'7o Cl: 1.16-3;68). but noo such increase in risk was observed for adulthood expo• sure orlifetime exposure. Recently, another case-control study° alsotbund an,increased lung cancer risk for non- smoking women~who reported 22'or more smoke-years of household exposure during childhood and adoles- cence (OR2:4, 95% CIi I.1,-5:4D: The findings tiromi these two studies correspond well with our results. Withi respect to the effect of cumulative exposure of PS,on lung,cancer Dalager et aL3'using pack-years off exposure as2he measure. found a significant trend in the risk of lirng cancer associated with the eumulative pack- yearsofexposure. In thisstudy, we;attempted to estimate . the cumulative effect of PS by using a, passive smoking, index (PSI). Ulsing,this index, we observed a significant increase in the OI2' with PSI for those exposed under the age of 7(P'<O.OOI ). This finding is comparable to that of Dalager„and it again indicates the importance of early life exposure for the risk of lung cancer. It was suggested in this:study thart„if the relationship between PS and lung,cancer is studied'only for adult exposure or overiifetime, the health effects of exposure dluing childhood maybe masked. As a result, the whole or reai effects of PS, especialhythat of earliy life exposure, may be,underestimated and may have,even escaped ourr attention in a,previous studi7:"eiFailure to find elevated lung cancerrisk duringadulthood andlifetiine exposure from Janerich et a!,'s report 1D provides further evidence for this view, The frequencyand,intensityofexposure to other people's tobacco smoke may change with age. For Americans; peak exposure occurs during a person's twenties.7 Besides. susceptibility'to carcinogens during childhood ma;• diGfer firom that in adlrlthDod. Therefore. collection of exposure information from early life to adulthood' and analysis of data by age of exposure is warranted.. It shouldlbenoted that the f ndingofa `dose-response'' tendency in the three youngest exposure groups (P < 0!009 ) in our study may be questioned„because some of the'passive'smokers were also activesmokers. Matching by smoking status over lifetime between the cases and. controlscan; to some extent, obviate this criticism. In aa matched case-control studyc Sandfer eral:y investigated whether cancer risk in adult life is related to transplacen- tal' or childhood exposure to their parents' cigarette smoke. Data were collected and analysed lfrom the indi- viduals who livedlwith bothinatura'l parents fior all or most of'the, first, 110 years of life. Of alll the subjects studiedj therewere 45% of the cases and 47P/d of the ~.)9 controlk who never smoked. One of their interesting findings was thart the relartiue,risk estimates for smoking- related cancers (cancer of the oral',eavity and pharynx: oesophagus; pancreas, respiratoryand intrathoracic,or- gans, urinary tract and icervix) iin relation to motheri's or father's smoking were similar for smokers and non- smokers (0.8 versus 0.8 and 1,.5'versus 1.7 respectiively): In another case-control study,'-7 it was observed that overall k:ancerriskincreased significantly with iincreasing nurttbersof household exposure to tobacco smokeand , thetrend wassimilarbetween.smokersand non-smokers. In our study; we included! 59 caneer patients who ever smoked. In an attempt to limit the effect of smoking on the lung cancer risk, individual matehing,by smoking status over lifetime (current- or ex-smoker versus never smoked over lifetime) ' was adopted. So, the eases and controls were comparable withirespect!to smoking status over lifetime (ever smoked or neversmoked): For those 59 smokingpairs; there were no significant differences between cases and controls forthe age,smoking began ( l!9;1 versus 20.7 years) and for years of smoking (32.6 versus 31.1 years). However; the number smoked!each day for the cases (mean 1L3,.4 gram/day; SD 7.72) was slightly higher (P = 0:0'13)', than that for the controls (mean 10.2, SD 6.41;): This difference might magnify,the observedlOR'in thirstudy but it,is unlikely to distorrthe true relationship between PS, especiallyPS in early life. and the risk offemale lungcancerobservedlin this studiy. Similarity oflthe OR infour out of five exposure periods amongst smokers.and non-smokers suggests no interac- tion between passive and active smoking. Of note, the ORinall five age'periods among non-smoking pairs were similartothose among all 114 pairs studied. These resu lts again suggest that active smoking is less likely tto have a strongeffect, ifany;onthe observedassociation between lung cancer and household exposure to tbbaeco smoke in ohe present stvdy. Information biasis'inheriteds incase-control studies, Atipresent, information on past exposure to environmen- tal tobacco smoke is often obtained by subject recalll through interview:It has beenfoundRhat patients usually attempt to explain theinillness:'9In this study, interview- ers were not' blinded to the subject's case/control status although they did not know the underliying hypothesis of the study. In an attempt to limit information bias, we, used structured questionnaire and standardized inter- view techniques,,which are oftenihelpful in minimizing both recall and interviewer bias. Also, it appeared that collecting information on lPS by each residence may help ~ subjects to recall pastexprnsure to environmentalsmoke, especially, for those'yes or no"questions. Coultas et a1.30' reported thaninformtation onwhethertherewasexposure to parental smoking dunng,childhood!was more reliable

230 INTERNATIIDNALJOURI*ALOFEPIDEMIOLOGY t trttemabonalJoul r ©'Jnternahonal'E; than that on how'muchiexposure there was in the same period. In this' studyl: however, we were not a'ble to validate the histories of past exposure and we did not have information onifetal exposure to tobaceo smoke. Also, our data were limited by' the small numbers of specific histologica'l types of.lung cancer, which limited some'interpretation of the resultsi I t should also be'noted i that the latencyperiod between h!ouseho'ldlexp+rnsuret'o cigarette smoke and diagnosis'of lung cancer may be very long. If so, the examination of the household PS from ages 311 to'69 in our study. therebyexcluding cases over 69 years, might underline the effect, of' PS' for those exposed only during'.that age period. Nevertheless, the findings firom this study provide further support'to the observation that PS may increase the risk of subsequent lung cancer, especially PS during childhood. It is alfr'o'suggested that assessment of PS should be dbne by dififerent,exposure periods. REFERENCES I Trichopoulos D. Kalandidi iA. Sparros L and MacMa:honB. Lung . cancer and ipassive:smoking. Jnr J Cancer 9981: 27: I 1411-48: , ' Hirayama T. Cancer.mortaGty.innonsmoking women with'smokingh husbands based bn a ilarge-scale cohort st ud yin Japan. Pler Med 1984'J3i680-90. 3 Dalager.NA. PickkL W~ Mason. Ti J.. Correa P. Fontham E er a7.'. Therelationiofpassive,.smokinetodungcancer. CancevRes 198fi:48o 4808-11. j Fonah'amE T. Correa P. Wu,W illiamsA. er aL Lungcancer in ~non- smokin¢women:: Amultieentercaserontrol studc.. Cancer Epidenriol'Biomarker.cPrer 1991 l 1: 35-43: ` Pershagen G- Hrubec.Z'and.Svensson C:.Passivt smok~ingandg lungeancer in Swedish women. Arn. J. Epideminl 1987: '.. 1'35:17-2a. 6'Siockwell H G. Goldman A.L. LvmaniGH rr, n! Em•ironmental tobacco smoke and Ilung ca ncer risk in nonsmoki ngwome~n. J_ NorJ Cancer /nst 199°_: 84: 14'17-2?. Sandler:DP: EversoniR'B. Wilcox AJ. Browder J:P! Cance~rriskin adulthood from early life exposuree too parents' smoking. Am.J' Puhlk Health, 1985: 75: 487-91. "'JanerichDT: Thompson WD: VardaL Revaf.Lunecanceri and exposure to tobacco:smoke in 2he household. ,h' ~En,¢l d'AYed 1990: 323:;632-36'. vChilmonczyk n A. Knigfit G J.,Palomaki iG E. Pulkkinen A 1.1Villiams. and Haddow.l E. Environmental'.tobaeco exposureduring in- fancv'. Am J.Puh'lir Healrh1990: R0l . t"JanvisM J. Strachan DP: Feverabend.C: Deternmnanlcoflpassive smoking in children in Edinburgh. Scot land. Aru JPuhlrcHealrh. 1992:82a.1225-29. ~~ FriedmanG D:.Pttilti D& Bawol RD...Prevaknoe.andicorrclates of passive smoking;.Ahr JPuhlir Health 198:3i 73: 4111-05 . t~ PronG E.Burch JID. HoweG R. MillerA B1 The,reliabilitp• of passive smoking histories reported in a case-control stude• ofllunecancer: Mn J'Eprdenunl 1988: 127: 267-73. 'Qian:Y P. Ad.vances.in eptdemiology. Be•ijrng : Pu•sra/'rhr People'.r Healrh1985:3:197-207. " US Departmenrof Health and Human~Serntces: TGe Health Cbnse- quenrc.cinfSmoking.-Canrnr: ArryxrrrafrheS'ur,Qran~Gcm•ral.. DH HS publication ino- (PHS)8?-50179. 198'_. Spitzer WO.Lawrcnce V. Dales'R evul. Linksbeta•aenipassive smoking , anddisease:A best+evidence synthesis, Cli'nlnrecrrAfe•d.1990::13: . 17-4^_. ~° US:.Department.ofHealth and Human Services. Thi• Health Can.re- qwncarujlnrolunrma•Srnoking: ARePorrnlirheSurgeonGeneral. Washington. D'C: GovernmentiPnnting Office. 1986. DHHS publication no. (CDC) 87-8398.1 ' Correa.P: Pickle L W. Foniham E. Lin Y. Haenszel W, Passivesmoking and lungcartcer: Lancer1983: ii: 295-97.. t"Drew R.T. Boorman'.G A. Haseman J K. McConnell E.E:Busey Wtvt i Moore 1 A. Theeffect of •rge and exposure duration on cancer , induction by a known carcinogen in irats:.mice. and .h'amsters. Toxir App! Pharnracol 1983: 68:120-30. ° Luck M iand Nau H: Nieotine'and cotinine concentrations in serum i and lurine of infants exposed via passive smoking orimilk from smoking mothers. J'Prdlarr 1985i 107: 81 r>-820. '° "Woodward A.. Grgurinovi¢h N and Ryan P. Breast feeding and smoking hygiene: major influeneeson~.cotinine.in unne of smoker s infants: JEprdenrini'Cinnmuniny Health I986: 40:309- I 5. '' Everson RB. Individuals transplacentally exposed tmmaternal smok- ingmay.b'eat increased eancerriskin adult life. Lanl 1980::u: 123-27: == Wald N 1. Boreh'am:J. Bailey A: Ritchie C: Haddow J E. Knigh't:G. Urinary cotinine'as marker of breathing other people s tobacco smoke. l:ancer 1984: i: 290-31. Greenberg R A. Haley N J. Etzel R A. Loda F A'. Measunn¢th'e exposure of linfantss to tobaeco smokee Nicotine and cot inme in urine:and saliva. N Eng! J Afed'1984: 310i 107:L78. -'9ottoml Kuh'nert BIR:,Kufinert P'Nt. ReeseA-L. Maternalpa~ssivel smokane.andg fetal serum thiocyanatelevels: Am JOhsrev G rnerrel '198?: 144: 78 7=91. -' \9aclure M: Kataz R B A..Brvant M S. Skipper P L. Tannenbaum.S R. Elevated blood levels of ca rci nogens in passive:smkers Aur JPuh6n Healdr 1989179i 13SII-84; -'Tnehopoulos D1 Mallo F. Tomatis L rr a!'Actn e and passive.smo.kmg and pathologia:al.indicatorsoflung.cancer rtsk.in.an autopsy stl J,4AYA 19921 268: I G97'-701. - Sandkr D~P: Wull 'J.. E.erson: R.Bi Cumulancerffactsof luetimee passtve smoking, on l risk. Cuneer 19S5i a: ? 1:2-14. ~Nang,F L. Dui ~xD. Liu N. Ltu C.B: Analysis.of risk factorsfor female lung adenocarcinomatn Harbin: Indool -pollution: Crir.;i lPren Alcrl 1989:: 2_1: ? 7t~- ?". -' Sartwell P E.. Retrospective.studres; a revte.r far the clinieian. Arrn /rrtern. Afrrl.1974: 81: 381-Fb C.o.ultasD B. Peake.G T and Sumet.J Nt. Questtonnairee assessment of litettmeand recent expos'.urc.tocnrironm:.ntal tobacco smoke. Arn J Epidvnuo1.1989: 1311:3 1,R'-47. ( Re ri.sed rersion rt?c eired OOafrdhrr l9y31. Brea Arnle Whit MILLICENT I JONATIHA'N Eidson M'.(IOffie TIM, Wiggins cC in New Mexico Yartiatitxt in'br( been assessec and non-Hispa mottalitydatac white' women extremelyJow ttwse for non+ occurrence we available data' rates increase' inadence and incidence and : in INew Mexico Patterns of b sively docum frequent solic'd surpassedib% related death incidence Ofi i litng cancer a breast cancer Moreover, d. Surveillance. Programiindi' 7 rising. Breast cani nic groups in Whltes. laU'e;' blacks. Asian data for An-r . OIFia of Epidct menLSanta Fc. ' Dopartments o1 . Regtstr), Cancci '. USA Rbprint request~ School0f Medir NE. Albuquerpu. ltrd:.