Impact of Smoking and Smoking Cessation on Lung Cancer Mortality in the Asia-Pacific Region

Transcription

1 American Journal of Epidemiology Advance Access published March 16, 2007 American Journal of Epidemiology Copyright ª 2007 by the Johns Hopkins Bloomberg School of Public Health All rights reserved; printed in U.S.A. DOI: /aje/kwm002 Original Contribution Impact of Smoking and Smoking Cessation on Lung Cancer Mortality in the Asia-Pacific Region R. Huxley 1, K. Jamrozik 2, T. H. Lam 3, F. Barzi 1, A. Ansary-Moghaddam 1, C. Q. Jiang 4, I. Suh 5, and M. Woodward 1 on behalf of the Asia Pacific Cohort Studies Collaboration 1 Nutrition and Lifestyle Division, The George Institute for International Health, University of Sydney, Sydney, Australia. 2 Division of Health Systems, Policy and Practice, University of Queensland, Brisbane, Australia. 3 Department of Community Medicine, University of Hong Kong, Hong Kong, People s Republic of China. 4 Guangzhou Occupational Diseases Prevention and Treatment Centre, Guangzhou, People s Republic of China. 5 Department of Preventive Medicine and Public Health, Yonsei University College of Medicine, Seoul, Korea. Received for publication April 24, 2006; accepted for publication November 17, Cigarette smoking is becoming increasingly common in Asia while quitting remains rare, in part because of a lack of knowledge about the risks of smoking. This study compared the risk of death from lung cancer associated with smoking habits in Australia and New Zealand and in Asia by using data from the Asia Pacific Cohort Studies Collaboration: 31 studies involving 480,125 individuals. Cox regression models were used. The hazard ratios for lung cancer mortality associated with current smoking were, for men, 2.48 (95% confidence interval (CI): 1.99, 3.11) in Asia versus 9.87 (95% CI: 6.04, 16.12) in Australia and New Zealand; p for homogeneity < For women, the corresponding estimates were 2.35 (95% CI: 1.29, 4.28) in Asia versus (95% CI: 10.0, 37.3) in Australia and New Zealand; p for homogeneity < Quitting was beneficial in both regions; the hazard ratios for former compared with current smokers were 0.69 (95% CI: 0.53, 0.92) in Asia and 0.30 (95% CI: 0.22, 0.41) in Australia and New Zealand. The lesser effect in Asia was partly explained by the fewer number of cigarettes smoked and the shorter duration of follow-up in Asian studies. These results suggest that tobacco control policies in Asia should not solely concentrate on preventing the uptake of smoking but also attend to cessation. Asia; lung neoplasms; mortality; smoking; smoking cessation Abbreviations: ANZ, Australia and New Zealand; APCSC, Asia Pacific Cohort Studies Collaboration; CI, confidence interval. Across several countries in Asia, the pattern of cigarette smoking among the male population has evolved in parallel to that of the US adult population, but with a time delay of years (1, 2). In Japan (1) and South Korea (2), where the prevalence of cigarette smoking among men is estimated at about 60 percent, the habit did not become popular until the mid-1950s. Furthermore, although a similarly high prevalence of smoking has been reported among Chinese males (about 70 percent in some regions), the popularity of cigarette smoking in China has been a relatively recent phenomenon, reaching its peak in only the mid-1990s (3). The comparatively early stage of the smoking epidemic in large parts of Asia will mean that the hazards of cigarette smoking will not be fully realized for at least two more decades (3, 4). Discouraging young people from starting to smoke is the optimal strategy for reducing mortality from lung cancer in the long term (5), but, for the 1.25 billion current smokers in the world, this approach is too late (6). Correspondence to Dr. Rachel Huxley, The George Institute for International Health, P.O. Box M201, Missenden Road, Sydney, New South Wales 2050, Australia ( rhuxley@george.org.au). 1

2 2 Huxley et al. Fortunately, there is now good evidence that stopping smoking in middle age can avoid much of the excess risk associated with smoking (5). However, most of the data concerning the benefits associated with cessation of smoking have come from European (5) or US studies (7), with comparatively little information from Asian countries (8 11). The quit rates (former smokers as a proportion of ever smokers) (12) in several Asian countries are significantly lower than in the West (6, 11, 13), in part because of the relatively recent popularity of the habit but also because the hazards of smoking and the benefits associated with stopping are not as widely known in this region. For example, data from the China National Prevalence Survey showed that nearly 70 percent of smokers believed that the risk of smoking was negligible, and only 4 percent recognized the association of smoking with coronary heart disease (11). In addition, aggressive marketing strategies by tobacco companies, combined with a lack of comprehensive tobacco control measures, have helped to sustain the popularity of smoking across large parts of Asia (14). The primary aims of this study were to quantify the risks of lung cancer mortality associated with current smoking and the benefits associated with smoking cessation in populations of the Asia-Pacific region. We used data collected by the Asia Pacific Cohort Studies Collaboration (APCSC). MATERIALS AND METHODS Details of study identification, data collection, and event verification for studies in the APCSC are described elsewhere (15, 16). Briefly, studies were included if they had continued follow-up for at least 5,000 person-years and had recorded vital status at the end of follow-up. Studies were excluded if entry was dependent on a particular condition or risk factor. Mortality was classified according to the International Classification of Diseases, Ninth Revision: lung cancer was selected as code 162. Because most studies identified events by using record linkage, verification of death from lung cancer was not routinely reported. All data provided to the secretariat were checked for completeness and consistency and were recoded, when necessary, to maximize comparability across cohorts. Summary reports were referred to principal investigators of each collaborating study for review and confirmation. Studies were classified as Asian if their participants were recruited from mainland China, Hong Kong, Japan, Korea, Singapore, Taiwan, or Thailand and as ANZ if from Australia or New Zealand. The associations between smoking and lung cancer are likely to vary across populations because of differences in the accumulated hazards of smoking and differences in background lung cancer mortality caused by other risk factors (17). Because of between-country variations in the maturity of the smoking epidemic, we examined the risks associated with cigarette smoking separately for each country. However, to avoid issues concerned with statistical analyses based on small numbers of events, we restricted countryspecific analyses to countries in which studies in the APCSC recorded more than 10 deaths from lung cancer. All data on cigarette smoking were based on self-report at the time of entry into one of the included studies, which recorded whether individuals were current, former, or never smokers. Participants were classified as current if they smoked currently, never smokers if they had never smoked, and former smokers if they had smoked but reported having already quit at study baseline. Note that participants classified as current smokers at study baseline would also have included those who quit smoking during follow-up, for which we had no information. Of the 31 studies, 18 had recorded the mean number of cigarettes smoked per day. Current smokers were further classified by the average number of cigarettes they smoked per day. Groups smoking fewer than 20, 20, and more than 20 were chosen to provide an approximately equal three-way partition; 20 cigarettes corresponds to one standard pack. Confidence limits for dose-response hazard ratios were calculated with the method of floating absolute risks by using these three ordinal groups of cigarette smoking and treating never smokers as the reference group (18). Floating absolute risks is an alternative method of presenting relative risk estimates for polychotomous risk factors. Instead of choosing one level of the risk factor as a reference category, each level is assigned a floated variance that describes the uncertainty in risk without reference to another level. The advantage of the floating absolute risks method is that the parameter estimates may, assuming a Normal distribution, be considered approximately independent, which means that it is possible to compare parameter estimates not only with the baseline estimate but also with each other. Analyses used individual participant data and were restricted to persons between the ages of 35 and 69 years at enrollment (19). Cox proportional hazards models, stratified by study and adjusted for age, were used to estimate hazard ratios associated with smoking for both current and former smokers. A sex-interaction term was added to assess heterogeneity in the strength of the association for men and women. A random-effects meta-regression was used to relate loghazard ratios from Cox models to follow-up durations using individual study data (18). RESULTS Mortality from lung cancer in current smokers Data from 480,125 individuals were available for analysis (83 percent Asian, 34 percent female; table 1). Overall, 876 deaths from lung cancer occurred among men and 174 among women, with more than three quarters of these in Asian populations. The percentage of current smokers varied by sex and region: 22 percent of men and 14 percent of women from ANZ reported smoking at study baseline, while the corresponding proportions in Asia were 59 percent and 3 percent. On average, male and female smokers from ANZ smoked one third to one half more cigarettes per day compared with smokers from Asia (21 vs. 16 cigarettes per day for men and women from ANZ compared with 16 vs. 15 cigarettes per day in Asia, respectively). Among men who reported smoking habits at study baseline, the hazard ratios for fatal lung cancer, smoking versus never smoking, were fourfold greater in ANZ compared with Asia: 9.87 (95 percent confidence interval (CI): 6.04,

3 TABLE 1. Baseline summary statistics and details of studies included in the analyses of the impact of smoking and smoking cessation on lung cancer mortality in the Asia- Pacific region* Country Study Baseline No. of subjects Mean age (years) Female (%) Median follow-up (years) No. of lung cancer deaths Current smoker (%) Former smoker (%) Mean CPDy Males Females Males Females Males Females Males Females Australia Busselton , Longitudinal Study of Aging Melbourne , National Heart Foundation , Newcastle , Perth , AAAy , New Zealand Fletcher Challenge , Australia and New Zealand Subtotal , China Anzhen , CISCHy , Fangshan , Guangzhou Occupational , Seven Cities Cohorts , Xi an , Yunnan , Hong Kong Hong Kong Japan Akabane , Civil Service workers , Hisayama , Konan Miyama Ohasama , Saitama , Shibata , Shigaraki Town , Shirakawa , Singapore Singapore Heart , Singapore NHS , South Korea KMICy , Taiwan CVDFACTS , Kinmen , Asia Subtotal , Total , * Blank cells denote that no data were available. y CPD, number of cigarettes smoked per day by current smokers; AAA, Western Australian Abdominal Aortic Aneurysm Screening Program; CISCH, Capital Iron and Steel Company Hospital Cohort; KMIC, Korean Medical Insurance Company. Smoking, Quitting, and Lung Cancer Mortality 3 Downloaded from by guest on May 31, 2013

4 4 Huxley et al ) versus 2.48 (95 percent CI: 1.99, 3.11); p for homogeneity < Within those Asian studies that reported more than 10 deaths due to lung cancer, the hazard ratios for men varied by country: 2.16 (95 percent CI: 1.69, 2.76) in China, 3.64 (95 percent CI: 1.32, 10.1) in Japan, and 4.64 (95 percent CI: 2.35, 9.16) in Korea; p for homogeneity ¼ Among females, the regional differential was much larger: the hazard ratios for fatal lung cancer, current smokers versus never smokers, were eightfold higher in ANZ compared with Asia: (95 percent CI: 10.0, 37.3) versus 2.35 (95 percent CI: 1.29, 4.28); p for homogeneity < Because of the small number of deaths from lung cancer among women, country-specific hazard ratios were not derived. There was no evidence of a sex difference in the strength of the association between smoking and fatal lung cancer in either Asia (p ¼ 0.68) or ANZ (p ¼ 0.11). Hazard ratio Asia ANZ p for homogeneity = Dose-response relation between cigarette smoking and mortality from lung cancer Dose-response comparisons were restricted to Asia and ANZ and to men only because of the limited number of deaths from lung cancer in other subgroups. The following analyses were based on a subsample of 169,143 men (85 percent Asian) and 569 events (76 percent in Asians) that excluded former smokers. In both regions, there was a clear dose-response association, although the strength of the relation differed significantly (p for homogeneity across all cohorts ¼ 0.001): in Asia, men who smoked more than 20 cigarettes a day, compared with never smokers, had a fourfold greater risk of dying from lung cancer, whereas, in ANZ, men who smoked more than 20 cigarettes a day, compared with never smokers, had a ninefold greater risk of dying from lung cancer (figure 1). Mortality rates for lung cancer in never, former, and current smokers As with current smokers, the percentage of former smokers also varied by sex and region; in ANZ, 40 percent of men and 21 percent of women reported being former smokers at study baseline compared with 10 percent and 1 percent, respectively, in Asia. The total number of deaths from lung cancer among former male smokers was 134, of which nearly half were in Asia (n ¼ 63). In women, there were 23 deaths among former cigarette smokers, and all but two 0 <20 20 >20 Number of cigarettes smoked per day FIGURE 1. Hazard ratios for lung cancer mortality by number of cigarettes smoked per day (base ¼ nonsmoker) by men in Asia and in Australia and New Zealand (ANZ). Bars show 95% confidence intervals (CIs) calculated by using the floating absolute risk method. p values are shown for tests of homogeneity between regions (Asia vs. ANZ). Hazard ratios for the dose-response relation (<20, 20, and >20 cigarettes per day) are as follows: 1.35 (95% CI: 1.12, 1.62), 1.87 (95% CI: 1.57, 2.22), and 4.01 (95% CI: 3.12, 5.15) in Asia; 2.39 (95% CI: , 3.91), 5.56 (95% CI: 3.50, 8.81), and 9.14 (95% CI: 6.79, 12.3) in ANZ. occurred in cohorts from ANZ. Hence, because of the limited number of events, subsequent analyses were restricted to men and to regional (rather than country-specific) comparisons. Among men, the age-adjusted death rates from lung cancer among current, former, and never smokers varied substantially between Asia and ANZ (table 2). There was a more than 10-fold difference in these rates between smokers and never smokers in ANZ compared with only a twofold difference among male smokers and never smokers in Asia. Mortality rates among former smokers in both Asia and ANZ were intermediate to those for current and never smokers. Men who reported, at study baseline, that they had quit smoking were at substantially lower risk of lung cancer compared with current smokers (table 2). Among male former TABLE 2. Age-adjusted rates and hazard ratios of lung cancer among male current, former, and never smokers at study baseline in Asia and in Australia and New Zealand Country Rate per 1,000 (no. of events) Never smoker Former smoker Current smoker Current smoker (reference group) Never smoker Former smoker HR* 95% CI* HR 95% CI Asia 0.29 (94) 0.51 (61) 0.73 (502) , , 0.92 Australia and New Zealand 0.10 (19) 0.30 (71) 1.21 (117) , , 0.41 * HR, hazard ratio; CI, confidence interval.

5 Smoking, Quitting, and Lung Cancer Mortality 5 smokers in ANZ, the risk was reduced by 70 percent compared with 30 percent in male former smokers in Asia (p for heterogeneity <0.0001). We lacked the necessary information to examine the temporal relation between time since cessation of smoking and fatal lung cancer. However, the relative risk of death from lung cancer (former smokers vs. never smokers) was significantly related to duration of study follow-up (p ¼ 0.01), decreasing by 15 percent (95 percent CI: 4 percent, 25 percent) with every extra year. DISCUSSION Our findings, based on nearly half a million individuals from the Asia-Pacific region, confirm that cigarette smoking substantially increases the risk of dying from lung cancer among men and women in both Asia and the predominantly Caucasian ANZ. Among male smokers in China, Japan, and Korea, the hazard ratios for fatal lung cancer associated with smoking were approximately 2, 4, and 5, whereas, in ANZ, the hazard ratio was approximately 10. Among women, the corresponding hazard ratios were approximately 2 in Asia and 20 in ANZ. Although the relative risks of fatal lung cancer associated with cigarette smoking in Asia were smaller than those for ANZ, particularly for women, this finding is likely due in part to variation in the maturity of the smoking epidemic across Asian countries, especially in China, compared with ANZ. For example, in Japan and Korea, smoking became popular in the 1950s and 1960s, whereas it was not until the mid-1990s, more than three decades later, that the habit reached its peak, among men, in China. By comparison, the prevalence of smoking among Chinese women has remained low, about 4 percent. Moreover, cultural differences in smoking acceptability may have resulted in a greater likelihood of underreporting of cigarette smoking among women in Asia compared with those in ANZ, which would have resulted in an underestimation of the hazard ratios between smoking and fatal lung cancer. Our findings are consistent with those from previous studies of smoking and lung cancer in Asia. In a cohort of 1,268 elderly Chinese men, He et al. (20) reported a relative risk of fatal lung cancer of 2.31 (95 percent CI: 0.95, 5.61) for ever smokers compared with never smokers; in three large Japanese and Korean cohorts, the relative risks for fatal lung cancer were four- to fivefold higher for current smokers compared with never smokers (2, 8, 9). In studies from ANZ, there was evidence of a difference between the sexes in the strength of the association between smoking and lung cancer mortality, such that the risk was nearly doubled for female compared with male smokers. This increased hazard for women, compared with men, who smoke has been reported previously. For example, in a large US case-control study, Zang and Wynder (21) reported odds ratios for lung cancer associated with smoking that were between 1.2- and 1.7-fold higher for women than for men. We hypothesize that this finding may be due in part to the fact that women have taken longer to quit cigarette smoking (22). The effect would be that more male current smokers at baseline in ANZ will have subsequently stopped smoking than will have female current smokers at baseline. Thus, the hazard ratio for fatal lung cancer among male current smokers reflects a more heterogeneous group containing a larger proportion of former smokers. Greater exposure to environmental carcinogens among nonsmokers in Asia may also have contributed to the lower relative risks observed for current smokers there. Liu et al. (17) reported that lung cancer mortality rates varied 10-fold among nonsmokers in different geographic regions of China, largely as a result of patterns of household energy use over the past few decades. Coal, a common household fuel in China traditionally burned in stoves and buildings with poor ventilation, has been associated with an increased risk of lung cancer (23). Differences in the types of cigarettes smoked, degree of inhalation, and types of smoking may have further contributed to variation in the strength of the association between smoking and lung cancer across the Asia-Pacific region. For example, data from the 1996 China National Prevalence Survey indicated that 20 percent of respondents smoked unfiltered cigarettes as opposed to filtered brands. The survey also highlighted ethnic and regional differences in smoking practices; for example, the Chinese pipe was more common in the south, whereas hand-rolled cigarettes were predominant in the northeast part of the country (12). Importantly, our findings indicate that stopping smoking is effective in lowering the risk of dying from lung cancer among men in both Asia and ANZ, similar to the findings from earlier studies. Although the risk reduction among male former smokers in Asia was only 30 percent (compared with 70 percent in ANZ), this difference may be due to a number of reasons. First, in APCSC, the average number of cigarettes smoked per day in Asia was much lower compared with ANZ; hence, the benefit associated with stopping smoking would be smaller. Second, the shorter duration of study follow-up among cohorts from Asia compared with ANZ (6.2 years vs. 7.9 years) may also have contributed to a smaller apparent reduction in risk of lung cancer associated with cessation of smoking. Third, the reasons for stopping smoking may have differed between regions. Data from the China National Prevalence Survey indicate that only about one quarter of smokers want to quit and that the primary reason for stopping smoking is ill health (11). Stopping smoking after its health consequences became evident to the person would do very little to alter the course of lung cancer, for example, and hence such patterns of smoking cessation are associated with only modest reductions in relative risk. By comparison, a recent survey in Australia indicated that more than 40 percent of current smokers were intending to quit within 3 months (24), possibly because of a greater awareness of the overall benefits associated with cigarette smoking cessation. Because information on smoking habits was collected at study baseline only, we lacked the necessary data to examine the impact of smoking habits several years prior to death. However, our findings indicate that, among former male smokers, the relative risk of fatal lung cancer decreased by 15 percent with every extra year of study follow-up, assuming that there was no relapse into prior smoking habits. Several important limitations of the APCSC analyses warrant attention. It is well established that duration of

6 6 Huxley et al. smoking, as well as age at which smoking starts, for which we had no information in APCSC, are more important determinants of subsequent lung cancer risk than the actual amount smoked (19). Hence, because Asians are known to have smoked for fewer years than Australians and New Zealanders, on average, our analyses underestimated the risk of lung cancer due to smoking among Asian populations compared with ANZ. We lacked data to examine whether the impact of smoking cessation on the risk of lung cancer varies with age, and, related to this, we were unable to explore the temporal relation between time since stopping smoking and lung cancer risk. Furthermore, APCSC analyses were based on self-reported cigarette smoking at baseline, and information on change in smoking habits during follow-up was available from too few studies to enable any meaningful analysis to be undertaken. Finally, the individual studies used different methods to verify cause of death. Related to this limitation, because studies lacked information on the proportion of lung cancer cases confirmed by histology, misclassification (either non lung cancer death being recorded as lung cancer or lung cancer death being recorded as death from another cause such as tuberculosis) would have resulted in an underestimation of the risks associated with smoking. Moreover, these methods will have varied over time, and the lack of standardization could have had some unpredictable effect on the results. The importance of developing, and implementing, effective smoking cessation policies is highlighted by calculations showing that if interventions focus on prevention only, then 160 million current smokers will die before 2050, with the vast majority of deaths occurring in China (6). In the United States, more than 40 percent of persons who used to smoke have quit; in the United Kingdom, two thirds of current smokers want to give up the habit (6). However, in China in the late 1990s, nearly three quarters of smokers (205 of the 320 million) reported no intention of quitting, and only a small fraction of the number who did try to quit remained successful after 2 years (11). Inadequate knowledge of both the harmful effects of cigarette smoking and the benefits associated with quitting and a lack of comprehensive tobacco control programs (15), combined with a low uptake of nicotine replacement therapy and other pharmacologic aids for smoking cessation (11), may explain much of the continuing popularity of smoking among men in China as well as in other parts of Asia. As in the West, strategies designed to increase general awareness of the hazards associated with smoking, and the benefits derived from quitting, may help encourage the many smokers across Asia to quit the habit and, by doing so, prevent many of the large number of deaths predicted to occur if current smoking patterns persist (3). ACKNOWLEDGMENTS This project has received grants from the National Health and Medical Research Council of Australia, the Health Research Council of New Zealand, and The University of Sydney Cancer Research Fund, as well as an unrestricted educational grant from Pfizer Inc. (New York, New York). R. Huxley is supported by a University of Sydney SESQUI Postdoctoral Fellowship. The sponsors had no influence on the design, analysis, or interpretation of results and took no part in writing this paper. Executive committee of APCSC: M. Woodward, X. Fang, D. F. Gu, Y. Imai, T. H. Lam, W. H. Pan, A. Rodgers, I. Suh, H. J. Sun, and H. Ueshima; Statistical analyses: A. Ansary-Moghaddam, F. Barzi, and M. Woodward; Participating studies and principal collaborators (those in bold type provided data used in this paper): Aito Town: A. Okayama, H. Ueshima, and H. Maegawa; Akabane: N. Aoki, M. Nakamura, N. Kubo, and T. Yamada; Anzhen02: Z. S. Wu; Anzhen: C. H. Yao and Z. S. Wu; Australian Longitudinal Study of Aging: G. Andrews; Australian National Heart Foundation: T. A. Welborn; Beijing Aging: Z. Tang; Beijing Steelworkers: L. S. Liu and J. X. Xie; Blood Donors Health: R. Norton, S. Ameratunga, S. MacMahon, and G. Whitlock; Busselton: M. W. Knuiman; Canberra- Queanbeyan: H. Christensen; Capital Iron and Steel Company: X.G.Wu;CISCH (Capital Iron and Steel Company Hospital Cohort): J. Zhou and X. H. Yu; Civil Service Workers: A. Tamakoshi; CVDFACTS: W. H. Pan; East Beijing: Z. L. Wu, L. Q. Chen, and G. L. Shan; EGAT (Electricity Generating Authority of Thailand): P. Sritara; Fangshan: D. F. Gu and X. F. Duan; Fletcher Challenge:S. MacMahon, R. Norton, G. Whitlock, and R. Jackson; Guangzhou: Y. H. Li; Guangzhou Occupational: T. H. Lam and C. Q. Jiang; Hisayama: M. Fujishima, Y. Kiyohara, and H. Iwamoto; Hong Kong: J. Woo and S. C. Ho; Huashan: Z. Hong, M. S. Huang, and B. Zhou; Kinmen: J. L. Fuh; Konan: H. Ueshima, Y. Kita, and S. R. Choudhury; KMIC (Korean Medical Insurance Company): I. Suh, S. H. Jee, and I. S. Kim; Melbourne: G. Giles; Miyama: T. Hashimoto and K. Sakata; Newcastle: A. Dobson; Ohasama: Y. Imai, T. Ohkubo, and A. Hozawa; Perth: K. Jamrozik, M. Hobbs, and R. Broadhurst; Saitama: K. Nakachi; Seven Cities: X. H. Fang, S. C. Li, and Q. D. Yang; Shanghai Factory Workers: Z. M. Chen; Shibata: H. Tanaka; Shigaraki Town: Y. Kita, A. Nozaki, and H. Ueshima; Shirakawa: H. Horibe, Y. Matsutani, and M. Kagaya; Singapore Heart: K. Hughes and J. Lee; Singapore NHS92: D. Heng and S. K. Chew; Six Cohorts: B. F. Zhou and H. Y. Zhang; Tanno/Soubetsu: K. Shimamoto and S. Saitoh; Tianjin: Z. Z. Li and H. Y. Zhang; Western Australia Abdominal Aortic Aneurysm Screening Program: P. Norman and K. Jamrozik; Xi an: Y. He and T. H. Lam; and Yunnan: S. X. Yao. Conflict of interest: none declared. REFERENCES 1. Wynder E, Fujita Y, Harris RE, et al. Comparative epidemiology of cancer between the United States and Japan: a second look. Cancer 1991;67: Jee SH, Samet JM, Oh H, et al. Smoking and cancer risk in Korean men and women. Cancer Causes Control 2004;15: Peto R, Chen ZM, Boreham J. Tobacco the growing epidemic. Nat Med 1999;5:15 17.

7 Smoking, Quitting, and Lung Cancer Mortality 7 4. Lam TH, Ho SY, Heldey AJ, et al. Mortality and smoking in Hong Kong: case-control study of all adult deaths in BMJ 2001;323: Peto R, Darby S, Deo H, et al. Smoking, smoking cessation, and lung cancer in the UK since 1950: combination of national statistics with two case-control studies. BMJ 2000;321: Mackay J, Eriksen M. The tobacco atlas. Geneva, Switzerland: World Health Organization, Thun MJ, Day-Lally C, Myers DG, et al. Trends in tobacco smoking and mortality from cigarette use in Cancer Prevention Studies I (1959 through 1965) and II (1982 through 1988). In: National Cancer Institute, Smoking and tobacco control, monograph 8: Changes in cigarette-related disease risks and their implication for prevention and control. Washington, DC: National Institutes of Health, 1997: (NIH publication no ). 8. Ando M, Wakai K, Seki N, et al. Attributable and absolute risk of lung cancer death by smoking status: findings from the Japan Collaborative Cohort Study. Int J Cancer 2003;105: Marugame T, Sobue T, Satoh H, et al. Lung cancer death rates by smoking status: comparison of the Three-Prefecture Cohort study in Japan to the Cancer Prevention Study II in the USA. Cancer Sci 2005;96: Lam TH, He Y, Shi QL, et al. Smoking, quitting and mortality in a Chinese cohort of retired men. Ann Epidemiol 2002;12: Yang G, Ma J, Chen A, et al. Smoking cessation in China: findings from the 1996 national prevalence survey. Tob Control 2001;10: Health New Zealand. Surveillance and monitoring in New Zealand. ( stories/tfir3hrnzs.pdf). 13. World Health Organization. Tobacco or health: a global status report. Geneva, Switzerland: World Health Organization, Tong EK, Glantz SA. ARTIST (Asian regional tobacco industry scientist team): Philip Morris attempt to exert a scientific and regulatory agenda on Asia. Tob Control 2004; 13(suppl II): Asia Pacific Cohort Studies Collaboration. Determinants of cardiovascular disease in the Asia Pacific region: protocol for a collaborative overview of cohort studies. Cardiovasc Dis Prevent 1999;2: Asia Pacific Cohort Studies Collaboration. Smoking, quitting, and the risk of cardiovascular disease among women and men in the Asia-Pacific region. Int J Epidemiol 2005;34: Liu BQ, Peto R, Chen ZM, et al. Emerging tobacco hazards in China: 1. Retrospective proportional mortality study of one million deaths. BMJ 1998;317: Woodward M. Epidemiology: study design and data analysis. 2nd ed. Boca Raton, FL: Chapman and Hall/CRC, Doll R, Peto R. Mortality in relation to smoking: 20 years observations on male British doctors. BMJ 1976;2: He Y, Taihing L, Shi Q, et al. A prospective study of smoking, quitting and mortality in a cohort of elderly in Xi an, China. Zhonghua Liu Xing Bing Xue Za Zhi 2002; 23: Zang EA, Wynder EL. Differences in lung cancer risk between men and women: examination of the evidence. J Natl Cancer Inst 1996;88: Molarius A, Parsons RW, Dobson AJ, et al. Trends in cigarette smoking in 36 populations from the early 1980s to the mid- 1990s: findings from the WHO MONICA Project. Am J Public Health 2001;91: Xu ZY, Blot WJ, Xiao HP, et al. Smoking, air pollution, and the high rates of lung cancer in Shenyang, China. J Natl Cancer Inst 1989;81: Mitchell E, Sanders J. Current tobacco smoking by the New South Wales population and the consequences for health. N S W Public Health Bull 2004;15:87 91.