Tobacco Documents Online

Imernational Journal ofEpidemio/oqy Vol. 20,.No. 1 C IhternationaLEpidemiologiul Asaotiation 1991 Printed in Gnat Brnta Smoking and Other Risk Factors for Lung Cancer in Xuanwei, China ZHIYUAN UU•t, IKINGZHOU HEe AND ROBERT S CHAPMIAN•• Liu Z(Institute of,Environmental Health and Engineering, Chinese Academy of Preventive Medicine, 29 Nsn YVei,Road, Beijing 10050; China), He X and Chapman R S. Smoking and otherrisk factors for lung cancer in Xuanwei, China. lnter- national Journal ofEpieemiology 1991; 20: 26-31. In Xuanwei County, Yunnan Province, lung cancer mortality rates are among the highest in China in both males and females. Previous studies have shown a strong association of lung cancer mortality with indoor air pollution,from 'smoky' coal combustion. In the present case-control stud'y,,1t0 newly-diagnosed lung cancer patients and~426 con- trols were matched with respect to age, sex, occupation (all subjects were farmers), and village of residence (which provided matching with respect to fuel use). This design allowed assessment of known and suapected lung cancer risk factors other than those mentioned above. Data from males and females were analysed by conditional logistic regres- sion. In females who do not smoke, the presence of lung cancer was statistically significantly associeted with chronic bronchitis (odds ratio IpRI - 7:37,95%confidence interval (CI(: 2:40-22:66)'and family history oflunp cancer(OR 4.18, 95% Cl: 1.61-10.85). Females'resuhs also suggested an association ofduny cancerwith duration of cooking food (OR 1.00, 9.1'8 and 14.70), but not with passive smoking (OR 0.77, 95% Cl: 0.30-1.96). In males, lung cancerwas signifioa ntty associated with chronic bronchitis IOR 7.32, 95% Cl: 2.66-20:1'8); family, history oLlung wncer(OR 3.79, 95% Cl: 1.70- 8.42), and personal hittory of cookinp food (OR 3.36, 95% CI: 1.27-8.88).'h"Pmslei:Bo'se=i.sponse re0ationship of tunp~ cancei witFiarnokinq index (years of smoking'amount of smoking) was shown by risks of 1.00,:2.61; 2.17 and 4.70. " Examination of Chinese nationwide cancer mortality, statistics reveals that lung cancer mortality rates in Xuanwei' County, Yunnan Province, are among the highest in China.' From 1973 through 1975, annualized male lung cancer death rates, age-adjusted to the 1964 China population, were 27.7 and 6.8 per 100 000 in Xuanwei and China, respectively, Corresponding mortalities in females were 25.3 and 3.2 per 100 000. Marked geographical variation in lung cancer mor- tality exists within Xuanwei. The county t;an be divided into high-, medium- and low-mortality areas, in which ago-adjusted lung cancer mortalities in both sexes are 126.1, 20.9 and 6.0 per 100000, respectively. Xuanwei residents have traditionally burned three types of fuel, 'smoky' coal, 'smokeless' coal, and wood, for residential heating and cooking. Until the 1980s fuel was nearly always burned in an open, unven- tilated fire pit in the floor of the dwelling's main room;, 'Institute of Environmenta! Health and Engineering. Clrineae Audemy of Pn:ventive Medicine, 29 Nan Wei'Road, Beijing 100030, China. •'US F,aviroamental Protectioo Aaescy; Researeh Triangle Park. NC 27711 USA. tPresent addies: Epidemiotbfy Group,, Biotopal' and Medial Research Diviaion,,ArSonne National Laboratory, Bldg. 203-J152, 970() South Cas Arcnue. Argonne, IL l0439 USA. such fire pits are still widely used, though the use of ventilated stoves is increasing. Women1ave generally been responsible for starting and tending the domestic fire and cooking food, though men assume these res- ponsibilities in some families. ! 011103 Tobacco smoking is very rare in Xuanwei women, yet women's lung cancer rates are comparable to men's. Also, a survey of past fuel use showed~ that in the high-, medium- and low-mortaliry areas of Xuan- wei, the percentages of families using smoky coal before 1958 were 87.6%, 60.1% and 6.1% respect- ively. Corresponding percentages of families using wood were 1.4%, 19.9% and 67.1%. Indoor eoncen- trations of betuo(a)pyrene (BAP) averaged 627 ug/100 metres' (m) in the high-mortality area, and 46 ug/100 m'' in the low-mortality area. In addition indoor pollution samples from the high-mottality, area exhibited higher Ames-test mutagenicity than those from the lo.v-mortality area." All of these obser- vations have served to suggest an a.csociation between indoor :moky coal buraina and Iung cancer in Xuanwei. The casetontrol study reported here was designed to supplement existing information by assesaing the influence of factors other than fuel type on the occvr rence of lung cancer in Xuan.vei': Such factors„includ- 26 2023513015

. Vol 20..No. 11 Prmled inGrcet Btitain for Van Wei Road, ~i, China./nt'er- oth males and aollution from s and 426 con- idence (which, ing cancerrisk ogistic regres- d with chronic !ncer (OR 4.18, >king food (OR es significantly 95% Cl: 1.70- onship of lung '.17 and 4.70. l: though the use of 'men have generally ending the domestic •n assume these res- n Xuanwei women„ re comparable to ~use showed that in, ality, areas of Xuan- using smoky coall and 6.1% respect- s of families using 1%, Indoor concen- (BAP) average& -mortality area. and' area. In addition highrmortality area: genicity than those 3ll of these obser- association between id lung cancer in There was designed on by assessing the -1 tvpe on the occur- such factors, inclUd- RICK.nACTORS FOR'LLNG CA,+CER IN CHINA ine tobacco smoking. family history of lung cancer. htctor,°ofchronic bronchitis.,and'rooking habits. have hten associated~with lune cancer in areas other than ~u~nv ei. but their importance as contributors to lung c~lt,•r in Xuanwei, has not vet been systematicallM1~ dcl,rmined. The present snudv also allowed compari- k,Il,,t the relative impact,of these factors in males and I'm.lles.. ~l JERIALS AtiD METHODS In;uanuei. 93.4% ofthe total population were farm- e,rs in 1y82. Because of this. and because lung cancer mortalita in Xuanwei farmers is highL' this study was oonfined to farmers. Concentrating the targetpopula- tion tende& to increase the validity of the findings.' $er,een November 1985 and December 1986, we iiientified 112 eases of newly-diagnosed lung cancer at ){uanwei hospitals and clinics. After exclusion of two patients with unknown addt-esses, 110 lung cancer p,ttienas(56 males and 54 females) were included in data analysis. Of these. 19 (17%a)~had been diagnosed nn~the basis of cvtologicaUpathologicall findings. and the remainder on the basis of clinical histories and .vravs'. Controls were selected to match lung cancer patients Nith respect to age (±2 years). sex. and village of resi- dence. Because fuel use habits and dwelling types are similar within individual Xuanwei, villages. this design was expected to provide effective matching with res- pect toindoor fuelitype and dwelling type. Such matching was sought because it.would increase the effectiveness with which factors other than fuell tvpe could be assessed. Cases and controls were matched on' village, with, as many eligible controls included as possible. Therefore- we selected more con- trols (or each case in a large village than in a small vil- la¢e. The numbers of controls per casevaried from one to five. After exclusion of 26 controls because of erro- neous questionnaire responses, 426 controls were included inidata analysis, an average of 3.87 controls pcr case. There were 9 cases with 1 control, 15 cases with 2 controls. 15 with 3 controls, 13 with 4 controls, and 58 with 5 controls. A standardized questionnaire of the closed-question t%pe was developed. Study factors included', tobacco use history- family and personal medical history, d'om~ cstic fuel use history, indoor fuel use history;,personal history of cooking food, dwelling type, ethnic group Inatinnality), and socioeconomic and educational kvels. After strict interviewer training and field test- ing. thisquestionnai're was administered directly to all lung cancer patients and controls. No interviewer or study subjecrt knew the purpose of the study and hfRotheses. A summarv index of tobacco smoking wa5 d'e.el- oped for each subject. The smoking indlx was ealcu- lated by multiplving the duration of sm,rkine (in ~eart), bv the amounti of tobacco smoked (;in kilograms per month). A subject was eonsid'cred to have a positive familv'historv of lung cancer, if at least,one relati~e was reported to have had the disease. The relatives included; subjects' parents, siblings and childrem and parents"siblings. A subject was considered to have a positive history of chronic bronchitis if he or she had been,diagnosedbv a doctor to have this condition, or reported coughdonat least'three months per year for ar least two vears before the vear of interview: A female subject was considered to have been exposed to passive smoking if, there was at least one smoker (mainlc hus- band) who lived'in the same household. To assess the effects of individuallindependenr vari- ables, unmatched, unadjusted udds ratios (ORs) were calculated.` Confidenee intervals were calf:ulated using Miettinen's method.° Dose-response relationships were examined for variables related to smokin¢ and cooking. Trends within these relationships were tested by extension of the Mantel-Haenszel proced'ure. To develop adjusted; estimates of ORs associated with selected factors and interactions. conditional logistic regression models were also constructed for males and females." In these models, all variables were dichotomous- assuming valUes of0'or Il The selected, risk factors and interactions were treated as indepen- dent variables, and'the presence or absence of lung cancer was treated asthe dependent variable. These analyses were performed using the PECAN program.' 11) RESULTS. Distributions of characteristics in cases and controls are presented by gender in Tabla: I. Age. famih size. ethnic group. bik-thplace. educational level. and dwell- ing type were comparable in cases and controls, so these factors were not considered further in data analy- sis. The effect of active tobacco smoking was not eval- uated in females, since onlrv one female (a eontroll subject) had ever smoked tobacco. The village match- ing provided effective matching on fuel type because fuel-use habits (type and average amount) were similar in cases and controls. Crude and adjuste&ORs for smoking an&cooking habits are presented with 95% confidence intervals for males in Table 2. No relationship between lung'cancer and ever having smoked was observed. There was a suggestion of monotottic dose-response relationships of lung cancer with the age at which smoking began., duration of smoking and amount smoked by month. ~t"

28 tNTERA]ATIONAL JOURNAL OF ElIDFJiiQOLAC'Y Tws- I Canpoison of Lug cm.on caw .id cnnoolt. _ X,.w,-i, Factor Average age (yean). No. of people in family now No. of people in funily 20 yean ago Nan nationality (%) Boni jn Xuanwei (%) "[vo-cto.ey. d.-lling (%) Amount of .anoky ooal tMUnt (conilrear) Amount of waod burnt (tor"ryear) Femaies Caus Coetrol. Can Conunb z respoma. o~ t1~><mc '~trn S..r~s~v indei ivas o~aerved e adjtts'°te~'~Fa( in`men often cooked food (at least once a day) was 3.36 (95 CI: 1.27-8.88); The adjusted ORs were slightly 52 50 52 52 than the crude ORS. 5.6 5.4 5:6 5.4 5.8 5.5 5.9 5.5 94.6 96.9 98.2 97:0 100 100 100 98.0 9!8.2 99.1 100 100 Crude and adjusted ORs for cooking and passi smoking are presented for females in Table 3: No dose response relationship of lung cancer with age at whi the woman began to cook food was observed, but the OTt associated with the age at which the woman began to cook food (11-15 years old) was significant. Adjusted ORs associated with the duration of cooking were much larger than crude ORs. There was a sugges- 4.2 4.2 4.0 4.1 tion of dose-response relationship of lung cancer with 0.8 0.9 O.g 1.0 the duration of cooking food for the adjusted ORs. No relationship of lung cancer with passive smoking was observed. Odd fo d il f l ung cancer an per• y hlstory o s Tw,ts 2 Oddr raoor (OR) oid 95% confWaar mkrvat, (CI) Jor twr rauos r fam ou.cer Nmain .cco.das w„rot;ng amd'cooAvia, Xan»n. Cliirw sonal history of chronic bronchitis were significantly 19as_1s86 associated with lung cancer in both sexes but duration Factor Ever-amoter No Yes Age of stacting to smoke (years). Never >20 ,K20 Trend (p value) Yean of smoking ~35 a35 7lend (p value) Amount of fmokin8 (kg/per month) Never W.5 0.6-1.0 >1.0 71end (p nhx) Smoking mdat Q 2- 20- 33- Thend (p value) Oare oooks tuod lio ... Ys . .. However, none of these relationships was statis China. 19d5-r9AS significant. T{j"CGIIt!'~i ftltLttUCally Siim[ Cases Controk OAc• ORa•` 95% Q crude ORs in Tabl e 4. 4 19 L00 1.00 52 205 1.20 1.26 0.30- 5.26 DISCUSSION 4 19 1.00 1.00 This study was inte ies which had sho smoky coal combu nded to wn a st stion wi suppl rong a th lun ement ssocia g canc previo tion of er in X us stud- indoor uanwei' 20 80' 1 19 1.10 0.25- 4 93 " 32 125 . 1.22 1.39 . 032- 6.06 County. Full und erstandi ng of l ung ca ncer a etiology 4 9 (p>0.05) 1.00 1.00 in Xtlanwei, and co effect of smoky coa mprehe l use, re nsive r quire s isk ass ystem essme atic ass nrof the essment "f 30 146 0.96 1.07 0.26- 4.59 Tww 3 Oldt Rsoot (OR) d 95% ao wJldosa r irvw h (CI) Jb. 22 59 1.77 1.71 0.36- 8.12 hoq anov n)G+welo .aco.di+q m aool c+w -d psrr swo"u8. (p>0.05) X+u riwe4 C.4 oK 191Bf -19q6 Faaor Csra Coatiola ORr' ORa•• 95% CI 4 19 1.00 1.00 25 93 1.28 1.41 0.33- 6.09 At d aurd 20 93 1.02 1.09 0.24- 4.82 0o eook 7 19 1.73 1.91 0.32-11.40 >15 13 73 1.00 1.00 (P>0.05) 11-15 30 69 2.44 2.37 1. 09- S.1! 4910 11 60 1.03 12S 0. 45- 3.49 4 30 1.00 1.00 71sad(p+d.m) (p>0A3) 24 - 99 1 82 2 61 69- 9 0 82 Yea daoof+ 16 74 . . 1.62 2.17 . . 0s3S- 8.64 .g t t30 7 53 1.00 1.00 12 21 4.28 4.70 1.03-21.40 31-44 28 85 2.49 S.li 1. 76- 47.49 (pt0.03) i45 19 64 2M 14.70 1.61-134.03 74ead (p..Ioe) (p>d.0S) 4t 200 1 00 1 00 ti P W 12 24 . . 2.27 3.36 1.27- 8.88 r. aeo eg . No 9 26 1.Rq 1.00 Y.. 45 176 0.74 0.77 0.30- 1.96 'ORe - Qnde odds ratio. "ORa - Odds r.rio afler .d#ustmiot by osaditioed lo jrdc • ORc - Crvde add, raio. regrssiao for ottic risk l.aon. . •• ORa - Odb esuo oNer adjo>reeat by ooadEtiooN logstie Seokin8 iodn - Ysr of Smob.a'Aaoo,t dwaka8f nQa.io. fnr wbe ri.k t•eho.L 2023513017 of using an unventilated fire pit was not (Table 4). AU conditional logistic regression ORs were larger than

.ships was stattsqco y cally signi.ficant dosc. ancer with smol~ ted OR men who i day),was 3.36 (95sye were slighdy 1ar$er cooking and passive -s in Table 3. No dose_ loer with age at wltich .vas observed, but the tich the woman begait >ld) waS significant e duration of coolung s. There was a sugges. Ip of lung cancer yl,ith the adjusted ORs. Na passive smoking .,,,, ,f lung cancer and per. itis were significantly )th sexes but dtuation Was not (Table 4). qll, )Rs were larger than, lement previous stud- sssociation of ind'oor. ig cancer in Xuaawep lung cancer aetiologyy risk assessment of the ystematic assessment 1.00 1.00 2.49 S.1g 1.76- 47.44 2.25 14.70 1. 61-13f.03 (p>o:os) RICK FACTORS FOR LUNG CANCER INCHINA ,9, ~y~ .onas (OR) and95:b confidence inren~als (Cl) Ior lung g cancer.rn.malesand /emalea, acrording,io Jamihal'hnrono/,lung d 1~ • Adl~ e eancer, hurory.oJnhron~c bronchuu and yearsusrngunvennlotedfr.e pit. Xuanwtr.China/9fvS-19lt6. f .tor Males Femalcs Case Control' ORa' (95% CI) Case Control ORa (95°'; Cl)i Case Control Total ORa195°o CI I 41 200 45 192 fl6 ?92 15 24 3:79 (1.70- 8:42) 9 10 4.18O.61-10.85)~ 24 34 3 75(2 05- 6 83') 39 209 38 184' 77 393, 17, IS 7L32 (2.6fr20.18): 16 18 7.37 (2 40-22.66) 33 33' 7:61 (3.6_=16 (p) 22 107 21 84 43 191, 34 117 1.78 (0,46- 6.93) 33 118 0,73 (0.20- 2.60) 67 235 1.12 (0'u,- 2.74) aher 4dlustment by condition.l logistic regression for other risk faaors. .pRa + ~ onlv of fud use „but of other known and suspected M4 f3cors as well. To achieve such an assessment, we choic a studr design which; provided effective case- ~ rol'matching with respect to fuel type and average ,mount used as well as to age and gender. Xii,nwei is a very tural area.In the present study:, ~nost all lung cancer patients were diagnosed by the counn, hospitals. Only 17% of the lung cancer cases weK yased on cytological/pathological findings, 83% were based on clinical histories and X-ray findings. Tbcrclore, misclassification of the cases may exist in tpc stslcfy. However„other, reasons may improve valid- ar of lung cancer diagnosis in the study: Because of pom medical care in Xuanwei; most lung cancer paieots had reached an advanced stage of the disease ,,yn diagnosed, and' local doctors had wide experi- eeceof this diagnosis because of the high lung cancer morbidity in Xuanwei. In order to assess validity,of the duplotis,•+e also followed up a number of lung cancer pu>'ents from the study; almost all of''whom died within = (eoaths of diagnosis. fe both males and females,, the current study dis- dacd cotssistent and statistically, significant associ- atws of lung cancer with chronic bronchitis and paitrvc family history of lung cancer. Lung cancer was a6o aooaued with the frequency of cooking food (in asks) and the duration of cooking food' (in'females), e.es atter the matching on fuel type inherent in the ody desiga. Not surprisingly, lung cancer was associ- Bed with aL-tire smoking in males. No association with p.rs ataoking was observed in females. !a %uanwei', women are generally responsible for moliq food, so the variable 'cooks or does not cook iwd'aould noobe assessed in females. However,,the OR+ mociated'' with the variable 'years of cooking' suggested that liing cancer risk increased with'increase 'tn duration of cooking (Table 3). There was no dose- response relationship between lung cancer and'age at which a woman began to cook food. This observation may be due to the possibility that women who reponed cooking food at less than ten'years old did not really cook food at that age. The OR in males who often cooked food was over three times greater than in those who did not'. It is likely that those who cook'food inhale more coal-smoke pollution than those who do not, Wu et a! reported that subjects exposed to burning coal used'' for heating or cooking in a stove or fireplace during the majority of childhood and the teenage years had a lung cancer risk 2.3 times higher than subjects who were not so exposed." Wang er a! have reported similar results from China." Gao et a! reported increased risk of lung cancer in Shanghai women who cook frequently with rapeseed oil." This observation raises the question of whether pollutants associated notonly with rthe cooking fuelbut also with the cooking method promote lung cancer. The observed association of lung cancer with posi- tive family history of the illness may be attributable in part to recall'bias, However, our results are consistent with other studies'~" which tends to reinforce the validity of the association. Our relative risk estimates for positive family history of lung cancer, 3.79 in males and 4.18 in females, were also similar to previous stud- ies. The association may be due partly to the fact that a subject's relatives lived in the same environment'as the subject for some time. MulvihilP has postulated that some abnormal types of gene might increase sensitivity to,environmental carcinogens." Further research will, be necessary to elucidate and'distinguish the roles of genetic and environmental'faetors in carcinogenesis.

30' RrrERNATIONAL fOt1RNAL OF PJtD@.ftOLAOY Tobacco smoling is generally accepted to be a major cause of lung cancer.'¢n We observed~an association of lung canccr with tobacco smoking in males, and this observation tended to enhance confidence in the results. However, the association was weaker than has been reported in many previous studies."•19 When con- sidered individually, duration of smoking, amount of smoking, and age at which~smoking began were only weakly associated with the illness. Only the smoking index derived by multiplying duration by amount of smoking was significantly associated'with lung cancer. These observations may be due partly to the fact that only 23 (8.2%) of 280males in this study were lifetime non-smokers. In such a small comparison group, even fairly'small changes in the distribution of non-smokers between~ cases and controls could have produced marked differences in observed ORs associated with smoking. More importantly many farmers in Xuanwei smoke tobacco through a long bamboo cylinder partly filled with water and the passage of smoke through the waler may filter out carcinogenic substances. Studies comparing the composition of water-filtered to un5l= tered tobacco smoke are currently in progress. It is also quite conceivable that the large amount of air pollutants inhaled' during indoor smoky coal burn- ing in Xuanwei partly overwhelm the qtciDOgeaiC-. effect of tobacco smolting' For example, as mentioned above, the average indoor concentration of BAP iathe Xuanwei region of high lung cancer mortality was 627 ug/100 m' in a recent survey. An individual inhal• ing 12 ma of air per day might therefore inhale 9154 ug of BAP in a year if he or she spent eight hours per day indoors. In contrast, an individual smoking 20 cigaret- tes per day might be ekpected' to inhale only about 70Q ug of BAP in one year n'111us, it is not especially surprising that the ORs associated with smoking in Xuanwei males were smaller than reported in other studies. Because unusual environmental conditions prevail in Xuanwei; it would not be adviaahle to generalize these ORs to other areas. Smoking is very rare in Xuanwei females. In addi- tion, we observed no association of lung cancer with passive smoking in females. Such an association has been reported in several previous investigatioas.' IIn tlon-smoki.ng women in Shanghai; Gao er at ob.erved a limited association of lung cancer with passive smok- ing; in that study the relative risk raaged hom 1.0 in womea ll'ving less than 20 years with a rmokfnj ho.- baad'to-1.7'in those living with a smoking husbaW for at lr.at 40 yean.°' However Koo et al have. not observed a consistent association of lung caocer with passive smoking in Chinese women:"7D'Ibese aathors also stated that correlatea of passive smoking such as diet and socioc~conomic status can act as ~ ~ confounders when the health risks of passive sm ok` are evaluated." The heavy indoor air pollutioo-~` Xuanwei may also overwhelm the carcinogenic etl,,, of passive'smoking. The effecYof passive smoking q lung cancer may depend'on local environmental fk tors and results obtained in a given region may the,, fore not be applicable to other regions. In summary, this study was undertaken to stl,, ptemenratisting evidence showing a strong associatiao of lung cancer with indoor use of smoky coal in~Xttan. wei. Our results disclose importanvassociations of lun, eancer, with factors other than fuel type and therefort indicate that these factors must be considere& in anr compreheasive, quantitative risk assessmet.: of litn, cancer in Xuanwei. Our results also confirm indirectly that smoky coal pollution is an important detetminant, of lung cancer in Xuanwei. A separate case-control study, which will aDow simultaneous direct analysis of the effects of indoor air pollution and other knownand suspected lung cancer risk factors in Xuanwei; : currently in progress. ACIQNOWLEDGEMENf'S The authors are grateful to Chaofu Huang and Deyi Shan, who helped with data collection. REFERENCES , ' Otfk,ie of neseare6 d eu+csr prevenrioa and treatment. Miestry of Headtti: D.eo of oorco nona/iry !n the Ptople't Reprblic oJ CAina, Beiiina.1980. ' Mumford J L. He X Z, C6aQman R S, a a1. Lan&wnxr anf io6aoc av polluooo in Xnaawei, China: Scrncr 1987; 2J5: 217-20. ' CLapman R S, Mumford J ~L, Harris D B a a!: The epdemioiogy of lung a.noer in X:uanvei, Qyna: Qurent proQea, esues..od r+aeaxc6 uratepes. Ard Enriron HeakA 1988; _V 186-5. 'Schlwehmao J J. Ca.reoROdDenln, 0owdac; .4w.fysi. New York. Oxford Univertiry Presa, 19ffi. ' L.iliedeld A M; lilienfcld D. ForndJaoiv oJ ep*demialoty. li Yort. Oxford Umvervty Prw, Secood. 1980:. • Mieatiacn O S. Escmability and aximaoo m ose-referem studfea. Am ! Epiderniol 1976; 143- 226-35. , Mantel N. Chi square usu witL one depee of troedom: eztemioo d the Mantmj-Haeaasel proeedvts. J Am Srr Auac 1963; !ac 690-70D: ' Bre.bw N E, Day N E: S-i+ca.l wrdLodi in cotcer ree+od~. Vo! 1, Ti.r .urr,ir ojann.cnwoer'aarfa. Iyaa. lARC soeaamc T.bticatiao No 32, 1960. ' Gn7 N H„LnDin J H. Rasin.ocin L V. WcSbood okuLtioes for -atcmodare-oaoad'rtn<iies and wmval oaAbes wfte oed deatE emq. dfowsrriks 1991; fa: 7D3-7: "Stor u B E. Wachoider S, Bredu. N E. Ma:dmm 6kehaood atmn{ dgeocral nst .odcis roo.madEed d.a. Ayp! S..CU 1993. ]T 17241. . . ° wn A H, Headaa.o. B E, Mke X( C, Yr b( C. Smoking and otVt esk fanen br Inaj nnaer io womea. !NG 1985, 74 7t7-5 t. ° w.na X Z. Lvag eaoeer wd ac pdtatioa uidoor med oadoor. . Fievfrorvwev wd H.aBM 1963; 1: 7-1Z • . - •(3w Y 7„Eict w'J, ar.n6 W, .r at.' isej oacm amoej eb:.re , .. .ras, he l Cr.c.19Q: A! 6UH. .P ,, 242351,3019 .M, wV

RICK FACTORS FOR LU14O CANCER IN'CHIKA HigginsunJ, Jcmmn O MI Epldemio6ugieal roview o(lung cancer in man. In: MohrE, Schwachl D. Tomatis L(ccd:): Arrpollurion and toncrr 1n nt.:n, Liyon. IARC. 1977. 'TukunutaC K. Lilicn(cld A!N. Familial uggrcgation uf1ung csncorr in human. JNC! I963: 30i 2B -312. (h+i W L. Elston R C, C'hcn V W. Baile.•W"ilson J'E. Rotlischild H. Incresccd'damdii I nsk for lung cancor. JNCI l!996t 76:l417-2'.: Guffmtan T Es Hu.in:en 0 D, Mulvlhlill 11 Familial respiratory tract cancer. JA:4A 1982:317i 1020-3. ' Mulvt hitl11). Hoft factors i n human lung tumotur- an example of eeo• gcnetlccin onco.ogy. JNCI 1976; 53: 3-7, Luwth L A. Ernstcr V L. Warncr K E. Abbottt 1,:Las:zlu J, Smoking andlun6 nncer: an overview. Cancer Rra 1984; (12: pr 1)! S96(>-5b. WvndcrE L. Graham A A. Tobacco.moking as a possib6o etiolog- ical factor in broncho:cnic caroinoma. lA^NA 1950: 143: 329-36. Lcvin M L. Guldttein H. Gcrtiardt.P R. Cancer at:d tobactro amok• inp; Preliminer,r repon. JAMA 193U: Iq; 35E-Fi: =Do1l R, Hill A 8. Smoking and carcinoma ofitie Iung; Preliminary repon. dr Mrd J 1950: 2: 73D-s8. WHQ: Ne,oLh hernndi of rhr Muwtan environment: C eneva. 1972, rrichopoulos D, Kat.ndid! A. Sparrot L. 41acMahon B. Lung cancer and paish, e smoktng.JnrJ'Cancrr 1961; 27t Ij. 31 ° Corrca P; Plcklcl W, Funtham E„Li,n Y: HaonsSCI W. Passtvc smok• inE and lung unccr. Th'r lancrt 1983. Il: SY5-7. a'Gamnkal!L. Pastive amukins and'runccr-Amcrncun ettpertenco. Prev Mcd 1944: 13: h91-7. '/BlotW1. Braumcni 1. F It. Pass:vcsmok'ine andlung cancar. JNCI' 1986: 77: 943- I utk1. ' Ktw L C. Hu J H-C. Sew D. Acttve snd'paune cmok'ing amnng fcmaleiung canccr patients and cuntrnh in HunY Kon;. J£ip C(in Cancer Rrt 1983: 4: ?67-75. r Kuo L C. Ho 1 H-C, Saw D, !s passivo smoktng sn adJled rtsk factor for lung canccr in Chtncat womtn'. J 6sp CZ'n Cancrr Ra 19&ic3: 27'JJ;3.. °KUo L C. Ho 1 H-C. Saw D, Ho C-Y. 1lcasurcmcnt of pustvc sunoking and cstimutcs odn.k (ur litng cancer amung non, smokiny Ch,inese females. Inr l Cuncrr 19g7t 39: 162-7. "' Kuo L C. Ho 1 H-C. RrhYndcr R. lifrhmorY correlatca o(cnviron- mcntai tobacco smoke: A studv un nunsmoking Hong Kung Chinese wivcs with smoking vlrscs nonsrnokinE husbands. Soc Sci Mrd I9tie: 26: 731r6). (Rrvued vrrsion received Seprsmber 1990) W I f I