Tobacco Documents Online

= 1988 EISevier Science Publishers 8,V (B omedical Divis on) Smoking and health 1987. M, Aoki ecal, edieors mjt materiai may, be protect'ed hy, aopyrigtit taw (Titte 17' U.S. Code}:. 79 pASS1VE SIfdKING I5 A RISK FACTOR F(JR LtJNG CANCER IN NEVER S?tiDKING NOMEN IN HONG ICDNG TAI HSNG LAM, KAR KEUNG CS£NG Department of Coatmunity Medicine, University of Hong Kong, Li Shu Fan Building, 5 Sassoon Road, Hong-Kong. INTEeODUCT ION In Hong Kong, lung cancer is the leading cause of death due to malignant neoplasms in both sexes. On a world scale, lung cancer death rates among men are not particularly high in Hong Kong. However, the rates in women are among the highest in the world. Four case control, studies have been carried out in Hong Kong to investigate the risk factors for lung cancer in women, particularly smoking and passive smoking. They are reviewed as follows: I, 1976-1977 STUDY The first major study on risk factors for lung cancer was a case control study on 208 male and 189 female patients. The controls were 204 male and 189 female hospital orthopaedic patients. 3ooking was found to be a major risk factor in males with a relative risk (RR) of 27.51. In females, the RR for smoking was only 3.48. 44.4% of the cases were non-smokers whose tumours were predomsnantly adenocarcinomas (45.2a).1 The role of passive smoking,was studied by simply asking the question of 'Are you exposed to the tobacco smoke of others at home or at work?' For non-smoking women* 40.5% of the cases and 47.5% of the controls had passive smoking. The RR for passive smoking was 0.75 (P-0.38).2 II. 1981-1983 STUDY In the second case control study, 200 female cases and 200 district female controls matched for age were interviewed in depth using a semi- structured questionnaire. The RR for ever smoking was 2.77. 44.9% of the cases had never smoked. Among the never-smoked'wives, 61.4% of the cases and 51.8% of the controls had smoking husband. The RR for passive smoking due to smoking husband was 1.48 (Pr@.16).3 III. 1981-1984 STUDY The third case control study included 163 female cases and 185 female controls from hospital orthopnedic patients. Unlike the previous two studies, only histologically and/or cytologically confirmed cases were included. A standardized questionnaire was used for interviewing. The RR for smoking was 4.12. The proportion of cases who were non-smokers was 46.0%. The role of passive smoking was studied only on the' adenocarcinoma cases. For non-smoking women, 61.71 of the adenocarcinoma cases and

280 44.4% of the controls had passive smoking,due to smoking husband. The -' RR for passive smoking was 2.01 (P<B.BS), Analysis was also carrie& out by the site of the tumour. For centrally site& tumour, the RR for passive smoking was 1.61: (P>0.05). For peripheral tumour, the RR was ' 2.64 (P<8.85):4 ~ IV. 1983-11986 STUDY This was the largest case control study on lung cancer in1women in Hong Rong. A standardized structured' questionnaire was designed for interviewing. All the cases were confirmed pathologicall~y. They were compared with 445 female healthy neighbourhood controls mntched for age. The RR of ever smoking was 3.81. 45.5% of the cases were never smokers. For never smoking women, 57.8% of the cases and 45.4% of the controls had passive smoking due to a smoking husband. The RR for passive smoking' was 1.65 (P<B.B1, 95% ., , C.I.-1.16, 2.35). ~ When broken down by cell type, the proportion of never smokers of 62'.4% was the highest in adenocarcinoma and it was only in this cell type that the RR for passive smoking was statistically significant (RR-1.87, P<8.01, 95% C.I.m1.23, 2.85). Significant trends for RR with amount smoked daily by husband were observed for all cell types combined and for adenocarcin oma only. TABLE I SUMMARY OF RESULTS ON PASSIVE SMC)RING A?SJt4; NONJ-SMURII+IG WOMEN IN 4 CASE COINTROS, ST(JDIES IN HONG RONG Cases/Controls Study* Passive smoking No passive smoking Total' no. of cases i~ controls Relative risk P value 1976-1977 34/66 50/73 223 0.75 0.38 Chan & Fung, 1983 1981-1983 54/71 34/66 225 1.48 0.16 Roo et al, 1985 1981-1984 37/64 23/80 204 2.01 0.03 Lam WR, 1985 1983-1986 115/152 84/183 534 1.65 8.897 Lam TH et al, 1987 Grand' Total 249/353 191/402 1,186 1.43** 0.004 •The study by Lam WK included only adenocarcinama whereas the other three studies included all oell types. ** Sucmaary relative risk by Mantel 8iaenazel's method

281 Si7MMARY OF RESULTS ON PASSIVE SMORING Table I shows the summary of results of the above four studies. Apart frcm the earliest study in wtiich~only one simple question was asked about passive smoking, they all showed a RR greater than unity. Statistical' significance was reached in the recent two. The Mantel-~ Baenszel's summary RR was 1.43 (P<0.81, 951 C.I.-1.12, 1.83).!' In a review of epidemiological and other evidence on passive smoking and lung cancer, Blot and Fraumeni estimated a 30% excess risk6 while wald et al calculated a relative risk of 1.35 by pooling the results of ten case control studies and three prospective studies.7 The suemary RR of the four case control studies in Hong Kong is close to these estimates. Because the local prevalence of smoking among women was low (4.11),8 the influence by misclassification bias would be much less than in western countries and could not account for the relatively high RR. The results in Bong Kong therefore strongly suggest that passive smoking -f r`is a risk factor for lung cancer in never smoking Chinese women. ' 3 A(_'RN CWLEDGEMENT We thank the International Developaent Research Centre and the University of Hong Rong (Comnittee on Research and Conference Grant and Medical Faculty Research Grant Fund), for financing the research project. Thanks are also due to Dr. D.W. Han and Dr. Reith Arnold for their support, to Roche Asian Research Foundation for sponsoring our presentation of the paper and to Mr. Richard Peto for his comments. AEfERFNCES 1. Chan WC, Colbourne MJ, Fung SC, Ho HC (1979) Br J Cancer 39:182-192 2. Chan WC, Fung SC (1982) In: Grund<nann E (ed) Cancer Campaign, Vol 6, Cancer Epidemiology, Fischer Verlag, Stuttgart and New York, pp 199- 3. 4. 5. 6. T. 8, 201 Koo LC, Ho JHC, Lee N (1985) Int J Cancer 35:149-155 Lam WK (1985) A clinical and epidemiological study of carcinoma of tl~ lung in Hong Kong. M.D. Thesis, University of Hong Kong, Hong Rong Lam TH, Rung ITM, Wong CM, Lam WK, Kleevens JWL, Saw D, Hsu C, ~ Seneviratne S, Lam SY, Lo laC, Chan WC (1987) Br J Cancer 56:673-678 Blot WJ, Fraumeni JF Jr (1986) J Natl Cancer Inst 77:993-1000 Wald NJ, Nanchahal K, Thompson 9G, Cuckle HS (1986) Br Med J 293s. 1217-1222 Hong Rong Census and Statistics Department (1985) Special Topics Report III, Social Data Colilected by the General Household Survey. Government Printer, Hong Kong