Tobacco Documents Online

 ' I I I I I I I I I I I I I I I I INDOOR BURNING COAL AIR POLLUTION AND LUNG CANCER - A CASE-CONTROL STUDY IN FUZHOU, CHINA Luo Ren-xia*, Wu Bin**, Yi Ying-nan** and Lin Ru-tao** * Fuzhou Senior Medical School, Fuzhou, China ** Fujian Medical College, Fuzhou, China Abstract A case-control study on risk factors for lung cancer was carried out in Fuzhou, China. One hundred and two newly-diagnosed primary lung cancer cases in urban Fuzhou (78 male and 24 female cases) were matched with 306 population-based controls. The primary histological types were adenocarcinomas (57 cases, 55.9%) and squamous cell carcinomas (39 cases, 38.2%). Controls were obtained from the general population by random, stratified sampling and consisted of noncancer cases matched for sex, ethnicity and age. Cases and controls were interviewed by trained professionals using a standardized questionnaire. Information was obtained on: smoking habit, living conditions, history of respiratory diseases, air pollution, and forty other variables. The data were evaluated by conditional logistic regression analysis. The major risk factors for lung adenocarcinoma were: indoor air pollution from burning coal, chronic bronchitis, and high economic income. These conclusions were based on the following model, which was derived from unconditional logistic regression analysis: Ln Px/Q, = 1.7923 X, + 1.4122 X6 + 0.9263 X5 X,:indoor burning coal air pollution X6:chronic bronchitis XS:high economic income When the same analysis was applied to lung squamous cell carcinoma, the following results were obtained. Ln P,/Q, = 2.6486 X, + 1.1160 X2 + 1.9647 X3 + 1.5104 X4 + 1.5678 X5 X,:indoor burning coal air pollution X2:number of cigarettes per day X3:deep smoke inhalation X,:passive smoking history before 20 years of age X,:high economic income Thus, risk factors for lung squamous cell carcinomas were: amount of cigarettes smoked per day, deep smoke inhalation, passive smoking history before 20 years of age, burning coal indoors, and high economic income. I

I The results showed that the major risk factors for lung cancer in Fuzhou were: burning coal indoors, smoking, passive smoking history before 20 years of age, chronic bronchitis, and high economic income. Cigarette smoking significantly increased the risk of lung squamous cell carcinoma, but it had no significant association with the risk of lung adenocarcinoma. In summary, our research supports the hypothesis that smoking and indoor air pollution are the major risk factors for lung cancer in Fuzhou. Burning coal indoors and smoking were associated with lung cancer mortality in a major city in southern China. Introduction Cancer is the leading cause of mortality in Fuzhou, China. Among the various types of cancer that were followed in Fuzhou over the years, a downward trend has been observed in the death rates associated with stomach, esophageal, and cervical cancers. In direct contrast, there has been a marked increase in the mortality of lung cancer, making it now among the leading causes of cancer deaths. To investigate risk factors which may be associated with the elevated mortality of lung cancer, a case-control study was performed. In this report, we describe fuu3ings from such a study. Methods Lung cancer cases involved Fuzhou residents who were identified from a special reporting system designed to cover all incidence of lung cancer in hospitals located in urban Fuzhou. All were newly diagnosed, histologically confirmed primary lung cancer cases (according to code 162 described in the International Classification of Diseases, Ninth Revision) and were collected over a 1.5-year period during 1990-91. Controls were randomly selected by sex and age from the general population of urban Fuzhou, using a 2-stage sampling procedure. In the first stage, 20,000 people were randomly sampled from the 760,000 general population in urban Fuzhou, by means of a neighborhood census roster kept by a committee in each neighborhood. They were further grouped by sex and age (35-75 years). For every case, four controls were chosen randomly from the same sex and age group, and one of the four controls was used as a match for the case. Cases and controls were interviewed by trained personnel using a standardized questionnaire. The interviews were conducted usually in the subject's home. Information was obtained on smoking habits (active and passive), demographic variables, occupation, living conditions, air pollution, history of respiratory diseases, and forty other variables. The odds ratio (OR), a measure used to evaluate the relationship between various risk factors and lung cancer, was estimated by conditional logistic regression analysis (1). Results A total of 124 lung cancer cases were identified over the study period, of which 102 (82.3 %) completed the interview. Excluded were 12 who had died, 8 who were too ill to participate in the -2- I 1 I I I I I I I I I I I I I I I I

1 I I I I I I I I I I I I I I I I I interview, and 2 who refused an interview. Diagnosis of all cases were confirmed by histological review of tissue specimens obtained from surgery or bronchoscopy. Among the 102 who completed the interview, 78 were males and 24 were females. Interviews were also performed with 306 controls. Table I shows that the distribution of cases and controls was similar with respect to sex, age, education, and term of residency in Fuzhou. Table 1. Distribution of cases and controls by sex, age, education, and term of residency in Fuzhou Sex 78 76.5 234 76.5 Male Female 24 23.5 72 23.5 Age (yr) 14 13.7 42 13.7 35-49 50-59 27 26.5 81 26.5 60-- 61 59.8 183 59.8 Education 18 17.6 58 19.0 No school Primary school 31 30.4 102 33.3 Secondary school 26 25.5 76 24.8 Beyond high school 27 26.5 70 22.9 Years of residency in Fuzhou 24 23.5 67 21.9 30-39 40-49 25 24.5 56 18.3 50-75 53 52.0 183 59.8 -3- I

I I Table 2 shows the distribution of histological types of lung cancer among the 102 cases. Adenocarcinoma (55.9%) and squamous cell carcinoma (38.2%) were the most frequent types. Table 2. Distribution of pathological types of lung cancer cases I I I Pathological types Males Females Total.. ' Squamous cell carcinoma 36 46.2 3 12.5 39 38.2 Adenocarcinoma 38 48.7 19 79.2 57 55.9 Oat cell carcinoma 2 2.6 1 4.2 3 2.9 Unknown 2 2.6 1 4.2 3 2.9 Smoking. Smokers were identified as those who had ever smoked cigarettes on a regular basis for six months or longer. Cigarette smoking was reported in 79.5 % of the males, but only in 12.5 % of the females. In controls, the percentage of smokers in both males and females was significantly lower (61.5% and 2.8%). Overall, smokers had a 2.7-fold increase risk of lung cancer. Table 3 shows that smoking was significantly associated with an increased risk for squamous cell carcinoma (OR= 10.9; 95 % CI 2.5-47.9), but not for adenocarcinoma (OR=1.5; 95% CI 0.7-3.0). Table 4 shows that the OR of squamous carcinoma increased both with amount usually smoked per day and with the duration of smoking. Moreover, the later the age at which smoking started, the lower the associated risk with squamous cell carcinoma. Similarly, the risk of squamous carcinoma declined markedly with an increase in the number of years since smoking ceased (Table 5). Deep smoke inhalation was associated with an increased risk of squamous cell carcinoma (OR=10.4; 95% CI 3.6-30.6). Table 3. Odds ratios for specific types of lung cancer associated with smoking hTonsmoker ;' Stnoker. OR 959'o CI Controls 160 146 1.0 - Cases 37 65 2.7 1.5-5.0 Squamous cell 5 34 10.9 2.5-47.9 carcinoma Adenocarcinoma 29 28 1.5 0.7-3.0 -4- I I I I I I I I I I I

I I I I I I I I I I I I ' I I I Table 4. Odds ratios for squamous cell carcinoma associated with intensity and duration of cigarette smoking Smoking Cases Controls OR 95%-CI Average number of cigarettes/ day 0 5 51 1.0 1-19 3 39 1.2 0.1-10.0 20-29 22 23 24.6 4.2-145.7 30- 9 4 38.7 5.2-290.2 Duration/years 0 1-29 30+ 5 51 1.0 6 21 5.7 1.0-32.9 28 45 12.5 2.8-55.4 Table 5. Odds ratios for squamous cell carcinoma according to age when smoking began and years since cigarette smoking stopped Smoking Nonsmoker Age-started >_ 40 20-39 <_19 Years since stopped z10 Cases Controls 5 1 OR' 95% Cl 51 1.0 8 3.1 0.3-40.2 19 38 10.5 2.3-48.6 14 20 12.4 2.7-57.7 1 10 2.0 0.2-23.1 < 10 33 56 15.9 3.1-82.2 Passive smokine. We regard those who were exposed to smoking parents or other family members before 20 years of age as positive passive smokers. Passive smoking before 20 years old was associated with squamous cell carcinoma (OR=2.4, 95% CI 1.1-5.1). Indoor burning coal air oollution. In this investigation, we regard smoke generated in the living room while cooking with coal as a positive index of indoor burning coal air pollution. Indoor burning coal air pollution was associated with an increased risk of lung cancer (OR= 7.6, 95% CI 3.7-15.7). -5- I

I Lung diseases. Participants were asked about history of lung disease diagnosed at least three years prior to the time of the investigation. Table 6 shows ORs of 3.6 (95 % CI 1.8-7.2) and 2.5 (95 % CI 1.2-4.7) for chronic bronchitis and pulmonary tuberculosis, respectively, as significant risk factors being associated with lung cancer. Table 6. Odds ratios for lung cancer in relation to history of bronchitis and pulmonary tuberculosis Cases Controls OR 95%: CI : Chronic bronchitis No 83 286 Yes 19 20 1.0 3.6 1.8-7.2 Pulmonary tuberculosis No 86 283 1.0 Yes 16 23 2.4 1.2-4.7 Mental stress. Mental stress (such as being denounced at a public meeting or having relatives that died of an accident) was associated with lung cancer (OR=2.3, 95 % CI 1.3-3.9). Income. High economic income was associated with an increased risk of lung cancer (OR=2.9, 95 % CI 1.84.6). Dust. Persons who had a history of contact with dust or coal dust or other industrial dust in workplace had an elevated risk of lung cancer (OR=1.8, 95% CI 1.1-3.1). Conditional logistic resression analysis. Based on the analyses of individual factors, a conditional logistic regression model was constructed to evaluate simultaneously the effects of multiple variables on the risk of squamous cell carcinoma and adenocarcinoma, respectively. Included in the model were indicator variables for smoking, passive smoking, indoor burning coal air pollution, economic income, lung disease, alcoholic drink, dust and mental stress. There were 5 variables in the conditional logistic regression model for squamous cell carcinoma and 3 variables in the conditional regression model for adenocarcinoma. By the log-likelihood ratio of statistic (Crz=63.42 on 5 degrees of freedom and G2=23.75 on 3 degrees of freedom, respectively), we conclude that the fit of the models is good (P-value of both less than 0.01). -6- I I I I I I I I I I I 1 I I I I I I

! I I I I I I I I I I I I I I Table 7. Conditional logistic regression model parameters Factor oe ctent STD error U-va ue -va ue Squamous ce carcinoma Indoor burning coal 2.6486 1.1906 2.2245 14.1 0.0261 air pollution (Xl) Amount of cigarettes 1.1160 0.4200 2.6572 3.1 0.0079 per day (X2) Deep smoke 1.9647 0.8028 2.4472 7.1 0.0144 inhalation (X3) Passive smoking 1.5104 0.7058 2.1400 4.5 0.0324 history before 20 years old (X4) High income (X5) 1.5678 0.6631 2.3645 4.8 0.0181 Adenocarcinoma Indoor burning coal 1.7923 0.5728 3.1288 6.0 0.0018 air pollution (Xl) High income (X5) 0.9263 0.3481 2.6612 2.5 0.0078 Chronic bronchitis 1.4122 0.5814 0.4290 4.1 0.0151 (X6) The main effects of lung squamous cell carcinoma was established as: Ln P,/Q, = 2.6486 X, + 1.1160 X2 + 1.9647 X3 + 1.5104 X, + 1.5678 X5 Xi:indoor burning coal air pollution X2:amount of cigarettes per day X3:deep smoke inhalation X4:passive smoking history before 20 years of age Xs:high economic income The main effects of lung adenocarcinoma was established as: Ln Px/Qx = 1.7923 Xi + 1.4122 X5 + 0.9263 X6 X,:indoor burning coal air pollution XS:high economic income Xb:chronic bronchitis -7-

I I Discussion I The results show that the major risk factors for lung cancer in Fuzhou were: amount of cigarettes smoked per day, deep smoke inhalation, passive smoking history before 20 years of age, burning coal indoors, and high economic income. I An indoor air contamination monitor in Fuzhou showed that the concentration of SO2 and NOz was higher. The pollution in the kitchen generated by coal combustion was more serious (2). According I to a study in Shanghai (3), the contact quantity of B(a)P per person in rooms where coal is used as fuel is equal to the amount of B(a)P in 20 cigarettes per day. Indoor air pollution was found to be one of the risk factors for lung cancer in Guangzhou (4). A great quantity of coal has been used as cooking fuel I since 1959 in Fuzhou, the mortality of lung cancer in Fuzhou has increased rapidly since 1983 and has coincided with the latency period for lung cancer (more than 30 years). I Cigarette smoking increased significantly the risk of lung squamous cell carcinoma, but there was no significant association with lung adenocarcinoma. The findings are generally in line with studies of lung cancer in other parts of the world (5-7). I After adjusting for smoking and other confounding factors, chronic bronchitis is still associated with lung cancer and included in main effect model. This result supports the hypothesis that chronic I bronchitis is one of the risk factors for lung cancer. The association between high income and lung cancer must be further investigated. I Conclusion Our research supports the hypothesis that smoking and indoor air pollution are the major risk I factors for lung cancer in Fuzhou. The increase in the use of burning coal indoors and smoking were associated with lung cancer mortality in a major city in southern China. I I I I N I O tp i ~ 00 W I ~ -8- ~ I I

I I I I I I I I I I I I I I I I I I References Breslow, N.E. and Day, N.E. (1980) Statistical methods in cancer research. Vol. 1- the analysis of case-control studies, p. 32. 2. Zheng, Z.Q. et al. (1989) Journal of Fuiian Medical College 23(1), 24. 3. Huang, Y. et al. (1985) Shan¢hai Environment Science 1, 26. 4. Liu, Q. (1987) Shanghai Tumor 7 (6), 256. 5. Damber, L.A. et al. (1986) Smoking and lung cancer with special regard to type of smoking and type of cancer. Br. J. Cancer 53 (5), 673. 6. Benharnou, E. et at. (1987) Lung cancer and women: Results of a French case-control study. Br. J. Cancer 55(1), 91. 7. Cha, Q. et al. (1992) Relationship between smoking and lung cancer cell type. Journal of Guangzhou Medical College 20 (1), 5. -9- I