Role of Smoking in Global and Regional Cardiovascular Mortality
http://www.100md.com
《循环学杂志》
Harvard School of Public Health, Boston, Mass (M.E.); American Cancer Society, Atlanta, Ga (S.J.H., M.J.T.)
School of Population Health, University of Queensland, Brisbane, Australia (A.D.L.).
Abstract
Background— Smoking is a major cause of cardiovascular disease mortality. There is little information on how it contributes to global and regional cause-specific mortality from cardiovascular diseases for which background risk varies because of other risks.
Method and Results— We used data from the American Cancer Society’s Cancer Prevention Study II (CPS II) and the World Health Organization Global Burden of Disease mortality database to estimate smoking-attributable deaths from ischemic heart disease, cerebrovascular disease, and a cluster of other cardiovascular diseases for 14 epidemiological subregions of the world by age and sex. We used lung cancer mortality as an indirect marker for accumulated smoking hazard. CPS-II hazards were adjusted for important covariates. In the year 2000, an estimated 1.62 (95% CI, 1.27 to 2.04) million cardiovascular deaths in the world, 11% of total global cardiovascular deaths, were due to smoking. Of these, 1.17 million deaths were among men and 450 000 among women. There were 670 000 (95% CI, 440 000 to 920 000) smoking-attributable cardiovascular deaths in the developing world and 960 000 (95% CI, 770 000 to 1 200 000) in industrialized regions. Ischemic heart disease accounted for 54% of smoking-attributable cardiovascular mortality, followed by cerebrovascular disease (25%). There was variability across regions in the role of smoking as a cause of various cardiovascular diseases.
Conclusions— More than 1 in every 10 cardiovascular deaths in the world in the year 2000 were attributable to smoking, demonstrating that it is an important preventable cause of cardiovascular mortality.
Key Words: cardiovascular diseases ; health ; mortality ; risk factors ; smoking
Introduction
Worldwide, the number of deaths from cardiovascular diseases increased from 14 million in 19901 to >16 million in 2000.2 Aging of the world’s population is a major factor contributing to this increase because cardiovascular mortality rates tend to increase with age. Although age-specific mortality rates for all cardiovascular diseases combined have declined in both developing and developed countries,3,4 the rates for specific causes (eg, ischemic heart disease [IHD]) have increased in some populations.5 The basis for this epidemiological change is a population-wide rise in some cardiovascular disease risk factors,6 including decreasing physical activity, changes in diet, tobacco smoking, air pollution,7 and possibly fetal and childhood exposures that may increase the risk of cardiovascular diseases in adulthood.8 Smoking and obesity are 2 cardiovascular risk factors with unequivocal global increase over recent decades. Tobacco use is currently widespread in most populations in industrialized and developing countries and has been causally associated with mortality from multiple cardiovascular diseases.9–12
See p 456
Recent estimates of cardiovascular mortality attributable to smoking were based on all cardiovascular diseases combined.13 Such estimates are important for motivating global tobacco control efforts such as the Framework Convention on Tobacco Control. Combined estimates, however, do not address regional differences in the background risks of specific cardiovascular diseases (eg, IHD versus stroke). Background risks vary because of regional differences in nutritional, behavioral, and environmental factors and medical care, all with important population-specific dynamics. For example, background mortality from cardiovascular diseases is substantially lower in China than in India or Eastern Europe.2 As a result, cardiovascular diseases make up a smaller fraction of total smoking-attributable deaths in China compared with India and Eastern Europe; in contrast, chronic obstructive pulmonary disease and lung cancer make up a much larger proportion of total smoking-attributable deaths in China.14,15 Economic development and urbanization, however, are changing diet and exercise in China, with implications for obesity and other cardiovascular risk factors.16 Changes in cardiovascular risks and access to pharmacological interventions are also occurring in other developing and developed countries.6,17 Parallel to changes in these risk factors, smoking increased substantially in most developing countries over the last quarter of the twentieth century, with an estimated 930 million of the world’s 1.1 billion smokers currently living in the developing world.18,19
In this report, we use 2 unique data sources, the American Cancer Society Cancer Prevention Study II (CPS-II) and the World Health Organization (WHO) Global Burden of Disease (GBD) mortality database, to estimate cause-specific cardiovascular mortality attributable to smoking for IHD, cerebrovascular disease, and a cluster of other cardiovascular diseases. A second contribution of this article is using hazard estimates that are systematically and consistently adjusted for major covariates such as diet, body mass index, and alcohol. Previous global estimates of smoking-attributable cardiovascular mortality12,13 used hazard estimates with no or little adjustment for major covariates. Rather, these studies used "hazard correction factors" to ensure that the mortality effects of tobacco were not overestimated.12,13 These somewhat-arbitrary hazard correction factors resulted in great uncertainty in the estimates of mortality attributable to smoking. Because a comparable measure of exposure and systematically adjusted relative risks are used, the estimates presented here provide a consistent baseline for assessing the cardiovascular mortality consequences of smoking in different regions of the world.
Methods
Relative Risks of Cause-Specific Cardiovascular Mortality
Measuring Exposure to Accumulated Smoking Hazards
The accumulated hazards of smoking depend on factors such as the age at which smoking began or stopped, duration of smoking, number of cigarettes smoked per day, whether the smoked tobacco product was in the form of cigarettes or in other forms such as bidis or cigars, cigarette characteristics, and smoking behavior such as degree of inhalation.22 Many of these factors vary over time because of changes in the socioeconomic determinants of smoking, including income, marketing by the tobacco industry, the effects of tobacco control programs, prices, taxes, and sociocultural norms. Therefore, current smoking prevalence or tobacco consumption alone would be insufficient indicators of the accumulated hazards of smoking, even if detailed data were available in all countries. This is particularly important in low- and middle-income countries, where smoking has increased substantially over the past few decades.18,19
To capture the accumulated hazard of smoking, we used the smoking impact ratio (SIR) method of Peto et al12 adapted to the specific conditions of developing countries.23 The method uses lung cancer mortality, data on which are available or can be estimated from validated methods in most populations, as an indirect indicator of the accumulated hazards of smoking. A background-adjusted SIR is defined as population lung cancer mortality in excess of that for never smokers relative to excess lung cancer mortality for a known reference group of smokers, adjusted to account for differences in never-smoker lung cancer mortality rates across populations: SIR=[(CLC–NLC)/(SLC–NLC)]x(NLC/NLC),23 where CLC is age-sex specific lung cancer mortality rate for 2000 in the study population (eg, country of analysis) from the WHO GBD database, NLC is age- and sex-specific lung cancer mortality rate of never smokers in the same population, and SLC and NLC are age- and sex-specific lung cancer mortality rates for smokers and never smokers in a reference population. SIR was calculated for each age group, each sex, individual countries and then averaged (population weighted) in each of 14 epidemiological subregions of the world (Table 2). The age groups used in the analysis were 30 to 44, 45 to 59, 60 to 69, 70 to 79, and 80 years. No deaths before 30 years of age were attributed to smoking because there are few cardiovascular deaths in individuals <30 years of age and relative risks are unstable.
Conceptually, by using excess lung cancer mortality as the indicator of the accumulated hazards of smoking in both study and reference populations, SIR converts the smokers in the study population, who may have different smoking histories, into equivalent smokers in the reference population in which hazards for other diseases (eg, different cardiovascular diseases) have been measured.12 We used CPS-II as the reference population (SLC and NLC). CPS-II is 1 of the few studies of smoking and cause-specific mortality that capture the full effects of the smoking epidemic because most of the CPS-II current smokers were lifelong cigarette smokers (with a mean consumption of 20 cigarettes per day).12
In applying the method to developing countries, we accounted for the important contribution of coal, a common fuel used in parts of the developing world, as a determinant of lung cancer among never smokers. For those populations that use coal in poorly vented stoves (eg, China and parts of Southeast Asia), never-smoker lung cancer mortality rates (NLC) were estimated from data on lung cancer mortality among Chinese nonsmokers and prevalence of coal use in poorly vented stoves.15,23,24 The remaining risk factors for lung cancer mortality (ambient air pollution, occupational hazards, indoor air pollution from radon or biomass smoke, etc) in various combinations affect a subset of the population of each country. It is unclear what the net impact of these risks on lung cancer rates among never smokers in specific countries may be. Therefore, the impacts of risk factors other than coal were considered as sources of uncertainty for never-smoker lung cancer mortality (Table 3).
Estimating Tobacco-Attributable Mortality
Analysis of Uncertainty
Despite the recent improvements in the data sources required for global estimates of smoking-attributable mortality (eg, mortality registration and adjusted relative risk estimates), substantial uncertainty remains about the estimates of smoking-attributable cardiovascular mortality, especially in developing countries where data are more sparse. Some sources of uncertainty were addressed in a quantitative uncertainty analysis. These include (1) uncertainty in the parameters used to calculate the SIR, including lung cancer mortality in each country based on its mortality reporting system (CLC), background lung cancer mortality among never smokers as a result of exposure to other risk factors such as occupational carcinogens and air pollution (NLC), and lung cancer mortality in CPS-II smokers and nonsmokers (SLC and NLC), and (2) statistical uncertainty in the estimated CPS-II relative risks for each disease (Table 1).
Uncertainty for each of the inputs was estimated separately, as described elsewhere.13 Uncertainties of individual parameters were combined in a stratified sampling simulation25 to estimate an uncertainty distribution for the fraction of cause-specific mortality attributable to smoking. The uncertainty reported here is the 95% range of the combined distribution.
The above quantitative analysis does not include a number of other sources of uncertainty. First, there is uncertainty in the total number of cardiovascular deaths (from the GBD database), especially in developing countries where vital registration is absent or incomplete.26 Second, using SIR, which is based on excess lung cancer mortality, as the marker for accumulated smoking hazard for cardiovascular diseases introduces additional uncertainty. Because lung cancer has a longer lag than cardiovascular diseases, using SIR as the metric of accumulated hazard could result in an overestimation of risk when there has been a sharp decline in smoking (eg, North America and those countries in Western Europe where smoking has declined) and an underestimation of risk when there have been large increases in smoking. On the other hand, the cardiovascular impacts of smoking may be underestimated in developing countries, where smoking has been rising. A third source of uncertainty is the extrapolation of relative risks from CPS-II to other populations, which may have different exposure to other cardiovascular risk factors and background disease levels. The nature of this source of uncertainty and its implications for research and surveillance are described in detail in the Discussion.
Results
The estimated numbers of cardiovascular deaths attributable to smoking are summarized in Table 3, disaggregated by region and sex in Table 4, and displayed by region and disease in the Figure. Globally, an estimated 1.62 (95% CI, 1.27 to 2.04) million cardiovascular deaths were attributable to smoking in 2000, 11% of total global adult (30 years of age) cardiovascular deaths. The proportion of adult cardiovascular deaths attributable to smoking was 17% for men and 5% for women and 7% and 16% for developing and industrialized countries, respectively. An additional 90 000 deaths from hypertensive disease were excluded from these estimates, consistent with the analysis of the 2004 Report of the Surgeon General.9
Adult (age, 30 years) cardiovascular mortality resulting from smoking in GBD epidemiological subregions by disease in 2000. Pattern for hypertensive disease is different because it was excluded from main estimates (see Tables 1 and 5). See Table 2 for description of regions.
Regional Patterns
The largest numbers of cardiovascular deaths attributable to smoking were seen in the industrialized regions (340 000 deaths, 17% of total adult cardiovascular deaths in Eastern Europe and the former Soviet republics [EUR-C]; 220 000 deaths, 22% of total adult cardiovascular deaths in North America [AMR-A]; 200 000 deaths, 13% of total adult cardiovascular deaths in Western Europe [EUR-A]) and the high-mortality developing region of Southeast Asia (SEAR-D; 300 000 deaths, 10% of total adult cardiovascular deaths, dominated by India in terms of population). The large numbers of cardiovascular deaths attributable to smoking in EUR-C and SEAR-D are due to a combination of 2 factors: (1) the large number of cardiovascular deaths resulting from large population size and high background rates of cardiovascular deaths caused by other cardiovascular risk factors such as blood pressure, cholesterol, and body mass index27–29 and (2) a relatively high proportion of cardiovascular deaths caused by smoking. For example, SEAR-D has substantially less adult population than the developing regions of the Western Pacific (WPR-B; dominated by China in terms of population) (480 million versus 740 million) but a similar number of adult cardiovascular deaths (2.9 million versus 3.1 million). Together with a higher smoking-attributable fraction, smoking caused a considerably larger number of cardiovascular deaths in SEAR-D (300 000; 10%) than in WPR-B (120 000; 4%). This is in contrast to similar estimates for cancers, where WPR-B had a much higher total number of deaths (1.8 million cancer deaths in WPR-B versus 800 000 in SEAR-D) and hence higher smoking-attributable cancer deaths (240 000 in WPR-B versus 190 000 in SEAR-D) even with a lower smoking-attributable fraction (14% in WPR-B versus 24% in SEAR-D). Similarly, the adult population of EUR-C (140 million) is substantially smaller than that of EUR-A (260 million) and AMR-A (190 million), yet the total number of cardiovascular deaths in this region (2.0 million) is considerably larger than those in EUR-A (1.6 million) and AMR-A (980 000 million) because of the background risk of cardiovascular mortality resulting from other risks such as elevated blood pressure and overweight and obesity.30 Therefore, although the fraction of all cardiovascular deaths attributable to smoking in EUR-C (17%) is lower than in AMR-A (22%) and only slightly higher than in EUR-A (13%), the number of smoking-attributable deaths in this region is the largest of all industrialized regions.
The proportions of total adult cardiovascular deaths attributable to smoking were generally highest in industrialized regions (12% to 22%), where smoking has been entrenched for several generations (Table 4). At the same time, the increase in smoking over the last quarter of the twentieth century in a number of developing regions, including parts of Latin America, the Eastern Mediterranean, and Southeast Asia, has resulted in 10% of all current adult cardiovascular deaths being attributable to smoking; the number is higher among men (Table 4).
The number of cardiovascular deaths attributable to smoking among men was higher than among women (Tables 3 and 4), with men accounting for 68% of all smoking-attributable cardiovascular deaths in industrialized regions and 78% in developing regions. The exception to this pattern was AMR-A, where female deaths accounted for 58% of total smoking-attributable cardiovascular deaths. Even in North America the fraction of smoking-attributable cardiovascular deaths was slightly lower among women (21% for women versus 23% for men). Because total cardiovascular deaths were higher among women, however, population-attributable fractions were applied to larger baselines for women than for men. Eighty-one percent of smoking-attributable cardiovascular deaths in developing regions were in individuals between 30 and 69 years of age compared with 61% in industrialized regions.
Disease-Specific Patterns
Globally, IHD accounted for the largest number of cardiovascular deaths attributable to smoking (870 000 deaths; 54% of all smoking-attributable cardiovascular deaths), more than twice as many cerebrovascular disease deaths attributable to smoking (400 000 deaths; 25%) (Table 5). The relative contributions of different diseases to smoking-attributable cardiovascular deaths varied across developing and industrialized regions because of variations in both background death rates and fractions attributable to smoking (the Figure). In developing regions, 360 000 IHD and 200 000 cerebrovascular deaths made up 54% and 30% of all smoking-attributable cardiovascular deaths, respectively. In industrialized regions, cerebrovascular disease had a smaller role relative to developing regions, with 510 000 IHD and 210 000 cerebrovascular deaths making up 54% and 22% of all smoking-attributable cardiovascular deaths (see also the article by Yusuf at al5). The relatively smaller role of cerebrovascular disease in industrialized countries was countered by a larger role for the cluster of "other cardiovascular diseases" (25% of all smoking-attributable cardiovascular deaths in industrialized regions versus 16% in developing regions). The contribution of IHD to total smoking-attributable cardiovascular deaths varied from a low of 30% to 40% in the Western Pacific (WPR-A and WPR-B) and parts of sub-Saharan Africa (AFR-E) to a high of 60% to 70% in the Eastern Mediterranean (EMR-B and EMR-D), SEAR-D, and EUR-C (the Figure); the contribution of cerebrovascular disease ranged from 10% to 17% in AMR-A, EUR-A, and parts of EMR-B to >40% in WPR-B, dominated by China.
Discussion
Smoking is currently a very important determinant of cardiovascular mortality among men in all regions of the world and among women in industrialized countries. The proportion of cardiovascular deaths caused by smoking exhibited large variations among different regions of the world (men, from 10% of total cardiovascular mortality in sub-Saharan Africa and parts of Latin America and the Western Pacific to 23% in industrialized regions of Europe and North America; women, from 5% of total cardiovascular mortality in the developing world to >20% in North America). The interregional variation was even more apparent for deaths caused by smoking from specific cardiovascular conditions. In particular, the contribution of IHD to smoking-attributable mortality varied by a factor of 2. Smoking also caused large numbers of deaths from cerebrovascular disease in several developing regions. This interregional variation occurs because the shape and maturity of the smoking epidemic is highly affected by region-specific socioeconomic and cultural determinants of smoking, and because background mortality varies across populations.5
Clinical, experimental, and epidemiological evidence has established the role of environmental tobacco smoke as a risk factor for cardiovascular disease in otherwise healthy nonsmokers.31–35 Because environmental tobacco smoke is also a risk factor for lung cancer,36,37 its effects are captured partially by the SIR, as described in detail elsewhere.38 Separate estimates of cardiovascular mortality as a result of exposure to environmental tobacco smoke would have to consider the physical proximity of smokers and nonsmokers and building ventilation. These parameters, in turn, depend on cultural and geographical factors and on regulation of smoking in public areas.
The estimated number of global cardiovascular deaths attributable to smoking in this analysis (1.62 million) was practically identical to the number obtained by applying a constant "correction factor" of 30% to unadjusted hazards for multiple diseases combined (1.69 million).13 The differences in estimates from the 2 methods across regions and sexes ranged from >25% underestimation among women in sub-Saharan Africa and parts of Southeast Asia and the Western Pacific to >20% overestimation among men in North America, Western Europe, and the Western Pacific when covariate-adjusted site-specific hazards were not used. This variation in underestimation or overestimation arises because all cardiovascular diseases, which in reality have different distributions in different regions and different hazards, were grouped together in previous estimates. Using relative risks with additional adjustment for hypertension and diabetes mellitus resulted in practically identical global estimates (1.62 million versus 1.67 million deaths; difference not statistically significant) and for individual regions and diseases. This robustness to additional adjustment is consistent with the findings of Malarcher and colleagues39 (CPS-II) and Kawachi and colleagues40 (Nurse’s Health Study) who found that relative risks for smoking and cardiovascular diseases were not affected by residual confounding as a result of history of hypertension and diabetes.
The validity of smoking-attributable mortality estimates using the indirect SIR method has been confirmed against alternative methods that account for accumulated hazards of smoking.41 Uncertainty nonetheless remains, especially in developing countries where both complete mortality records and detailed studies of cardiovascular risk factors are less common. Some sources of uncertainty were described earlier. Two important additional sources of uncertainty in the current estimates with implications for future research are (1) the correlation between smoking and other cardiovascular disease risks within populations and (2) biological interactions between smoking and other risks that would result in uncertainty above and beyond that quantified in statistical hazard models when CPS-II relative risks are extrapolated to other populations. For many diseases, multiple risk factors "magnify" each another’s total risks as determined from standard proportional-hazard models. If multiple risks for the same disease are more common (positively correlated) in specific population subgroups, total (absolute) hazards would be larger than those estimated here, even if relative risks remain unchanged (smaller with negative correlation).42,43 Unhealthy diet among smokers is an example of positive correlation. In addition to correlation of risk factor exposures, multiple risks for the same disease may modify one another’s relative risk (eg, the relative risk of smoking may depend on diet or other factors that affect background cardiovascular mortality). The existing evidence indicates that extrapolation of hazards from CPS-II to other populations is a valid option. For example, the Chinese retrospective study of smoking and mortality demonstrated that background mortality rates, which varied by up to a factor of 4 for cardiovascular diseases, did not change the relative risk of mortality as a result of smoking (Figure 5 in Liu et al15). The robustness of relative risks for smoking to changes in other risk factors has also been confirmed in studies that stratified on serum cholesterol.44 More recently, a study in >50 countries illustrated that, when comparable methods were used, hazards of smoking for myocardial infarction did not vary significantly across regions of the world (Western Europe, Central and Eastern Europe, North America, South America and Mexico, Australia and New Zealand, the Middle East, Africa, South Asia, China and Hong Kong, Southeast Asia, and Japan).45,46 Despite these assurances, extrapolation of CPS-II relative risks is a source of additional uncertainty but necessary in the absence of recent local epidemiological studies in developing countries. (At any point in time, only recent epidemiological studies can be used for direct estimates because of changes in smoking patterns over time.)
The uncertainties in our analysis also identify 2 important research and surveillance needs to better understand how tobacco affects the global and regional cardiovascular epidemiology in combination with other risks. First, there is a need for epidemiological research on how smoking may interact with other risk factors. Parallel to this individual-level analysis, there is a need to improve population-level risk factor surveillance to collect data on correlation of multiple cardiovascular risks in population subgroups, differentiated according to factors such as income or education, race or ethnicity, or place of residence. Data on multiple risk factor interaction and correlation would allow better estimation of the total health effects of individual cardiovascular risk factors such as smoking. More important, data on correlated multiple risk exposure would be invaluable for targeting preventive interventions or treatment toward those subgroups at highest risk of cardiovascular disease.
Such research and surveillance needs are particularly important in developing countries. Currently, the number of cardiovascular deaths attributable to smoking is larger in industrialized regions than in the developing world (Table 3). However, available data on smoking trends suggest that smoking has increased in much of the developing world over the past few decades,18 as has population exposure to other cardiovascular risks such as overweight and obesity.4 Because the effects of smoking on cardiovascular disease emerge sooner than other diseases affected by smoking (eg, cancers and chronic obstructive pulmonary disease), this implies an expected increase in cardiovascular mortality in the developing world. At the same time, because the health benefits of smoking cessation occur faster for cardiovascular than other diseases,47 policies that prevent and reduce smoking will have immediate and large benefits for reducing cardiovascular mortality.48–52
Acknowledgments
This research was sponsored by the National Institute on Aging grant PO1–AG17625. The Cancer Prevention Study II is funded by the American Cancer Society.
References
Murray CJL, Lopez AD, eds. The Global Burden of Disease. Cambridge, Mass: Harvard School of Public Health, on behalf of the World Health Organization and the World Bank; 1996.
World Health Organization. World Health Report 2002: Reducing Risks, Promoting Healthy Life. Geneva, Switzerland: World Health Organization; 2002.
Murray CJ, Lopez AD. Alternative projections of mortality and disability by cause 1990–2020: Global Burden of Disease Study. Lancet. 1997; 349: 1498–1504.
Pelletier DL, Rahn M. Trends in body mass index in developing countries. Food Nutr Bull. 1998; 19: 223–239.
Yusuf S, Reddy S, Ounpuu S, Anand S. Global burden of cardiovascular diseases, part I: general considerations, the epidemiologic transition, risk factors, and impact of urbanization. Circulation. 2001; 104: 2746–2753.
Yusuf S, Reddy S, Ounpuu S, Anand S. Global burden of cardiovascular diseases, part II: variations in cardiovascular disease by specific ethnic groups and geographic regions and prevention strategies. Circulation. 2001; 104: 2855–2864.
Pope CA3rd, Burnett RT, Thurston GD, Thun MJ, Calle EE, Krewski D, Godleski JJ. Cardiovascular mortality and long-term exposure to particulate air pollution: epidemiological evidence of general pathophysiological pathways of disease. Circulation. 2004; 109: 71–77.
Singhal A, Lucas A. Early origins of cardiovascular disease: is there a unifying hypothesis; Lancet. 2004; 363: 1642–1645.
US Department of Health and Human Services. The Health Consequences of Smoking: A Report of the Surgeon General. Atlanta, Ga: USDHHS, Public Health Service, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2004.
Thun MJ, Apicella LF, Henley SJ. Smoking vs other risk factors as the cause of smoking-attributable mortality: confounding in the courtroom. JAMA. 2000; 284: 706–712.
Doll R, Peto R, Wheatly K, Gray R, Sutherland I. Mortality in relations to smoking: 40 years’ observation on male British doctors. BMJ. 1994; 309: 901–911.
Peto R, Lopez AD, Boreham J, Thun M, Heath C Jr. Mortality from tobacco in developed countries. Lancet. 1992; 339: 1268–1278.
Ezzati M, Lopez AD. Estimates of global mortality attributable to smoking in 2000. Lancet. 2003; 362: 847–852.
Gajalakshmi V, Peto R, Kanaka TS, Jha P. Smoking and mortality from tuberculosis and other diseases in India: retrospective study of 43 000 adult male deaths and 35 000 controls. Lancet. 2003; 362: 507–515.
Liu BQ, Peto R, Chen ZM, Boreham J, Wu YP, Li JY, Campbell TC, Chen JS. Emerging tobacco hazards in China, 1: retrospective proportional mortality study of one million deaths. BMJ. 1998; 317: 1411–1422.
Popkin BM. The shift in stages of the nutrition transition in the developing world differs from past experiences. Public Health Nutr. 2002; 5: 205–214.
Ezzati M, Vander Hoorn S, Lawes CMM, Leach R, James WTP, Lopez AD, Rodgers A, Murray CJL. Rethinking the "diseases of affluence" paradigm: global patterns of multiple nutritional risk factors in relation to economic development. PLoS Med. 2005; 2: e113.
World Health Organization. Tobacco or Health: A Global Status Report. Geneva, Switzerland: World Health Organization; 1997.
Jha P, Ranson MK, Nguyen SN, Yach D. Estimates of global and regional smoking prevalence in 1995, by age and sex. Am J Public Health. 2002; 92: 1002–1006.
Garfinkel L. Selection, follow-up, and analysis in the American Cancer Society Prospective Studies. In: Garfinkel L, Ochs O, Mushinkski M, eds. Selection, Follow-up, and Analysis in Prospective Studies: A Workshop. Bethesda, Md: National Cancer Institute; 1985. NCI monograph 67.
Burns DM, Garfinkel L, Samet JM, eds. Changes in Cigarette-Related Disease Risks and Their Implications for Prevention and Control: Smoking and Tobacco Control monograph 8. Bethesda, Md: National Cancer Institute; 1997.
Peto R. Influence of dose and duration of smoking on lung cancer rates. In: Zaridze DG, Peto R, eds. Tobacco: A Major International Health Hazard. Lyon, France: International Agency for Research on Cancer; 1986. IARC scientific publication 74.
Ezzati M, Lopez AD. Measuring the accumulated hazards of smoking: global and regional estimates for 2000. Tob Control. 2003; 12: 79–85.
Smith KR, Mehta S, Maeusezahl-Feuz M. Indoor air pollution from household solid fuel use. In: Ezzati M, Lopez AD, Rodgers A, Murray CJL, eds. Comparative Quantification of Health Risks: Global and Regional Burden of Disease Attributable to Selected Major Risk Factors. Geneva, Switzerland: World Health Organization; 2004: 1435–1493.
Morgan MG, Henrion M. Uncertainty: A Guide to Dealing with Uncertainty in Quantitative Risk and Policy Analysis. Cambridge, Mass: Cambridge University Press; 1990.
Mathers CD, Stein C, Ma Fat D, Rao C, Inoue M, Tomijima N, Bernard C, Lopez AD, Murray CJL. The Global Burden of Disease 2000: Version 2, Methods and Results (GPE Discussion Paper No. 50). Geneva, Switzerland: Global Programme on Evidence for Health Policy, World Health Organization; 2002. Available at: http://www.who.int/evidence.
Lawes CMM, Vander Hoorn S, Law MR, Elliott P, MacMahon S, Rodgers A. High blood pressure. In: Ezzati M, Lopez AD, Rodgers A, Murray CJL, eds. Comparative Quantification of Health Risks: Global and Regional Burden of Disease Attributable to Selected Major Risk Factors. Geneva, Switzerland: World Health Organization; 2004: 281–389.
Lawes CMM, Vander Hoorn S, Law MR, Rodgers A. High cholesterol. In: Ezzati M, Lopez AD, Rodgers A, Murray CJL, eds. Comparative Quantification of Health Risks: Global and Regional Burden of Disease Attributable to Selected Major Risk Factors. Geneva, Switzerland: World Health Organization; 2004: 391–496.
James WTP, Leach R, Mhurchu CN, Kalamara E, Shayeghi M, Rigby NJ, Nishida C, Rodgers A. Overweight and obesity (high body mass index). In: Ezzati M, Lopez AD, Rodgers A, Murray CJL, eds. Comparative Quantification of Health Risks: Global and Regional Burden of Disease Attributable to Selected Major Risk Factors. Geneva, Switzerland: World Health Organization; 2004: 497–596.
Ezzati M, Lopez AD, Rodgers A, Vander Hoorn S, Murray CJL, for the Comparative Risk Assessment Collaborative Group. Selected major risk factors and global and regional burden of disease. Lancet. 2002; 360: 1347–1360.
Law MR, Morris JK, Wald NJ. Environmental tobacco smoke exposure and ischaemic heart disease: an evaluation of the evidence. BMJ. 1997; 315: 973–980.
Steenland K. Passive smoking and the risk of heart disease. JAMA. 1992; 267: 94–99.
Glantz SA, Parmeley WW. Passive smoking and heart disease: mechanisms and risks. JAMA. 1995; 273: 1047–1053.
Pope CA III, Eatough DJ, Gold DR, Pang Y, Nielsen KR, Nath P, Verrier RL, Kanner RE. Acute exposure to environmental tobacco smoke and heart rate variability. Environ Health Perspect. 2001; 109: 711–716.
National Cancer Institute, ed. Health Effects of Exposure to Environmental Tobacco Smoke: Smoking and Tobacco Control Monograph 10. Bethesda, Md: National Cancer Institute; 1999.
Hackshaw AK, Law MR, Wald NJ. The accumulated evidence on lung cancer and environmental tobacco smoke. BMJ. 1997; 315: 980–988.
Environmental Protection Agency Office of Research and Development. Respiratory Health Effects of Passive Smoking: Lung Cancer and Other Disorders. Washington, DC: Environmental Protection Agency; 1992.
Ezzati M, Lopez AD. Smoking and oral tobacco use. In: Ezzati M, Lopez AD, Rodgers A, Murray CJL, eds. Comparative Quantification of Health Risks: Global and Regional Burden of Disease Attributable to Selected Major Risk Factors. Geneva:, Switzerland: World Health Organization; 2004: 883–956.
Malarcher AM, Schuman J, Epstein LA, Thun MJ, Mowery P, Pierce B, Escobedo L, Giovino GA. Methodological issues in estimating smoking-attributable mortality in the United States. Am J Epidemiol. 2000; 152: 573–584.
Kawachi I, Colditz GA, Stampfer MJ, Willet W, Manson JE, Rosner B, Hunter DJ, Hennekens CH, Speizer FE. Smoking cessation and decreased risks of total mortality, stroke, and coronary heart disease incidence among women: a prospective cohort study. In: Burns DM, Garfinkel L, Samet JM, eds. Changes in Cigarette-Related Disease Risks and Their Implications for Prevention and Control: Smoking and Tobacco Control Monograph No. 8. Bethesda, Md: National Cancer Institute; 1997.
Bronnum-Hansen H, Juel K. Estimating mortality due to cigarette smoking: two methods, same result. Epidemiology. 2000; 11: 422–426.
Greenland S. Bias in methods for deriving standardized morbidity ratio and attributable fraction estimates. Stat Med. 1984; 3: 131–141.
Ezzati M, Vander Hoorn S, Rodgers A, Lopez AD, Mathers CD, Murray CJL, for the Comparative Risk Assessment Collaborative Group. Estimates of global and regional potential health gains from reducing multiple major risk factors. Lancet. 2003; 362: 271–280.
Jee SH, Suh I, Kim IS, Appel LJ. Smoking and atherosclerotic cardiovascular disease in men with low levels of serum cholesterol: the Korea Medical Insurance Corporation Study. JAMA. 1999; 282: 2149–2155.
Yusuf S, Hawken S, Ounpuu S, Dans T, Avezum A, Lanas F, McQueen M, Budaj A, Pais P, Varigos J, Lisheng L. Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case-control study. Lancet. 2004; 364: 937–952.
Ezzati M. How can cross-country research on health risks strengthen interventions; Lessons from INTERHEART. Lancet. 2004; 364: 912–914.
US Department of Health and Human Services. The Health Benefits of Smoking Cessation: A Report of the Surgeon General. Bethesda, Md: USDHHS, Public Health Service, Centers for Disease Control, Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 1990.
Narkiewicz K, van de Borne PJ, Hausberg M, Cooley RL, Winniford MD, Davison DE, Somers VK. Cigarette smoking increases sympathetic outflow in humans. Circulation. 1998; 98: 528–534.
US Department of Health and Human Services. Women and Smoking: A Report of the Surgeon General. Bethesda, Md: USDHHS, Public Health Service, Centers for Disease Control Office on Smoking and Health; 2001.
Rehm J, Room R, Monteiro M, Gmel G, Graham K, Rehn N, Sempos CT, Frick U, Jernigan D. Alcohol use. In: Ezzati M, Lopez AD, Rodgers A, Murray CJL, eds. Comparative Quantification of Health Risks: Global and Regional Burden of Disease Attributable to Selected Major Risk Factors. Geneva, Switzerland: World Health Organization; 2004: 959–1108.
Cohen A, Anderson R, Ostro B, Pandey KD, Krzyzanowski M, Künzli N, Gutschmidt K, Pope A, Romieu I, Samet J, Smith KR. Urban ambient air pollution. In: Ezzati M, Lopez AD, Rodgers A, Murray CJL, eds. Comparative Quantification of Health Risks: Global and Regional Burden of Disease Attributable to Selected Major Risk Factors. Geneva, Switzerland: World Health Organization; 2004: 1353–1433.
Prüss-üstün A, Fewtrell L, Landrigan P, Ayuso-Mateos J. Lead exposure. In: Ezzati M, Lopez AD, Rodgers A, Murray CJL, eds. Comparative Quantification of Health Risks: Global and Regional Burden of Disease Attributable to Selected Major Risk Factors. Geneva, Switzerland: World Health Organization; 2004.(Majid Ezzati, PhD; S. Jan)
School of Population Health, University of Queensland, Brisbane, Australia (A.D.L.).
Abstract
Background— Smoking is a major cause of cardiovascular disease mortality. There is little information on how it contributes to global and regional cause-specific mortality from cardiovascular diseases for which background risk varies because of other risks.
Method and Results— We used data from the American Cancer Society’s Cancer Prevention Study II (CPS II) and the World Health Organization Global Burden of Disease mortality database to estimate smoking-attributable deaths from ischemic heart disease, cerebrovascular disease, and a cluster of other cardiovascular diseases for 14 epidemiological subregions of the world by age and sex. We used lung cancer mortality as an indirect marker for accumulated smoking hazard. CPS-II hazards were adjusted for important covariates. In the year 2000, an estimated 1.62 (95% CI, 1.27 to 2.04) million cardiovascular deaths in the world, 11% of total global cardiovascular deaths, were due to smoking. Of these, 1.17 million deaths were among men and 450 000 among women. There were 670 000 (95% CI, 440 000 to 920 000) smoking-attributable cardiovascular deaths in the developing world and 960 000 (95% CI, 770 000 to 1 200 000) in industrialized regions. Ischemic heart disease accounted for 54% of smoking-attributable cardiovascular mortality, followed by cerebrovascular disease (25%). There was variability across regions in the role of smoking as a cause of various cardiovascular diseases.
Conclusions— More than 1 in every 10 cardiovascular deaths in the world in the year 2000 were attributable to smoking, demonstrating that it is an important preventable cause of cardiovascular mortality.
Key Words: cardiovascular diseases ; health ; mortality ; risk factors ; smoking
Introduction
Worldwide, the number of deaths from cardiovascular diseases increased from 14 million in 19901 to >16 million in 2000.2 Aging of the world’s population is a major factor contributing to this increase because cardiovascular mortality rates tend to increase with age. Although age-specific mortality rates for all cardiovascular diseases combined have declined in both developing and developed countries,3,4 the rates for specific causes (eg, ischemic heart disease [IHD]) have increased in some populations.5 The basis for this epidemiological change is a population-wide rise in some cardiovascular disease risk factors,6 including decreasing physical activity, changes in diet, tobacco smoking, air pollution,7 and possibly fetal and childhood exposures that may increase the risk of cardiovascular diseases in adulthood.8 Smoking and obesity are 2 cardiovascular risk factors with unequivocal global increase over recent decades. Tobacco use is currently widespread in most populations in industrialized and developing countries and has been causally associated with mortality from multiple cardiovascular diseases.9–12
See p 456
Recent estimates of cardiovascular mortality attributable to smoking were based on all cardiovascular diseases combined.13 Such estimates are important for motivating global tobacco control efforts such as the Framework Convention on Tobacco Control. Combined estimates, however, do not address regional differences in the background risks of specific cardiovascular diseases (eg, IHD versus stroke). Background risks vary because of regional differences in nutritional, behavioral, and environmental factors and medical care, all with important population-specific dynamics. For example, background mortality from cardiovascular diseases is substantially lower in China than in India or Eastern Europe.2 As a result, cardiovascular diseases make up a smaller fraction of total smoking-attributable deaths in China compared with India and Eastern Europe; in contrast, chronic obstructive pulmonary disease and lung cancer make up a much larger proportion of total smoking-attributable deaths in China.14,15 Economic development and urbanization, however, are changing diet and exercise in China, with implications for obesity and other cardiovascular risk factors.16 Changes in cardiovascular risks and access to pharmacological interventions are also occurring in other developing and developed countries.6,17 Parallel to changes in these risk factors, smoking increased substantially in most developing countries over the last quarter of the twentieth century, with an estimated 930 million of the world’s 1.1 billion smokers currently living in the developing world.18,19
In this report, we use 2 unique data sources, the American Cancer Society Cancer Prevention Study II (CPS-II) and the World Health Organization (WHO) Global Burden of Disease (GBD) mortality database, to estimate cause-specific cardiovascular mortality attributable to smoking for IHD, cerebrovascular disease, and a cluster of other cardiovascular diseases. A second contribution of this article is using hazard estimates that are systematically and consistently adjusted for major covariates such as diet, body mass index, and alcohol. Previous global estimates of smoking-attributable cardiovascular mortality12,13 used hazard estimates with no or little adjustment for major covariates. Rather, these studies used "hazard correction factors" to ensure that the mortality effects of tobacco were not overestimated.12,13 These somewhat-arbitrary hazard correction factors resulted in great uncertainty in the estimates of mortality attributable to smoking. Because a comparable measure of exposure and systematically adjusted relative risks are used, the estimates presented here provide a consistent baseline for assessing the cardiovascular mortality consequences of smoking in different regions of the world.
Methods
Relative Risks of Cause-Specific Cardiovascular Mortality
Measuring Exposure to Accumulated Smoking Hazards
The accumulated hazards of smoking depend on factors such as the age at which smoking began or stopped, duration of smoking, number of cigarettes smoked per day, whether the smoked tobacco product was in the form of cigarettes or in other forms such as bidis or cigars, cigarette characteristics, and smoking behavior such as degree of inhalation.22 Many of these factors vary over time because of changes in the socioeconomic determinants of smoking, including income, marketing by the tobacco industry, the effects of tobacco control programs, prices, taxes, and sociocultural norms. Therefore, current smoking prevalence or tobacco consumption alone would be insufficient indicators of the accumulated hazards of smoking, even if detailed data were available in all countries. This is particularly important in low- and middle-income countries, where smoking has increased substantially over the past few decades.18,19
To capture the accumulated hazard of smoking, we used the smoking impact ratio (SIR) method of Peto et al12 adapted to the specific conditions of developing countries.23 The method uses lung cancer mortality, data on which are available or can be estimated from validated methods in most populations, as an indirect indicator of the accumulated hazards of smoking. A background-adjusted SIR is defined as population lung cancer mortality in excess of that for never smokers relative to excess lung cancer mortality for a known reference group of smokers, adjusted to account for differences in never-smoker lung cancer mortality rates across populations: SIR=[(CLC–NLC)/(SLC–NLC)]x(NLC/NLC),23 where CLC is age-sex specific lung cancer mortality rate for 2000 in the study population (eg, country of analysis) from the WHO GBD database, NLC is age- and sex-specific lung cancer mortality rate of never smokers in the same population, and SLC and NLC are age- and sex-specific lung cancer mortality rates for smokers and never smokers in a reference population. SIR was calculated for each age group, each sex, individual countries and then averaged (population weighted) in each of 14 epidemiological subregions of the world (Table 2). The age groups used in the analysis were 30 to 44, 45 to 59, 60 to 69, 70 to 79, and 80 years. No deaths before 30 years of age were attributed to smoking because there are few cardiovascular deaths in individuals <30 years of age and relative risks are unstable.
Conceptually, by using excess lung cancer mortality as the indicator of the accumulated hazards of smoking in both study and reference populations, SIR converts the smokers in the study population, who may have different smoking histories, into equivalent smokers in the reference population in which hazards for other diseases (eg, different cardiovascular diseases) have been measured.12 We used CPS-II as the reference population (SLC and NLC). CPS-II is 1 of the few studies of smoking and cause-specific mortality that capture the full effects of the smoking epidemic because most of the CPS-II current smokers were lifelong cigarette smokers (with a mean consumption of 20 cigarettes per day).12
In applying the method to developing countries, we accounted for the important contribution of coal, a common fuel used in parts of the developing world, as a determinant of lung cancer among never smokers. For those populations that use coal in poorly vented stoves (eg, China and parts of Southeast Asia), never-smoker lung cancer mortality rates (NLC) were estimated from data on lung cancer mortality among Chinese nonsmokers and prevalence of coal use in poorly vented stoves.15,23,24 The remaining risk factors for lung cancer mortality (ambient air pollution, occupational hazards, indoor air pollution from radon or biomass smoke, etc) in various combinations affect a subset of the population of each country. It is unclear what the net impact of these risks on lung cancer rates among never smokers in specific countries may be. Therefore, the impacts of risk factors other than coal were considered as sources of uncertainty for never-smoker lung cancer mortality (Table 3).
Estimating Tobacco-Attributable Mortality
Analysis of Uncertainty
Despite the recent improvements in the data sources required for global estimates of smoking-attributable mortality (eg, mortality registration and adjusted relative risk estimates), substantial uncertainty remains about the estimates of smoking-attributable cardiovascular mortality, especially in developing countries where data are more sparse. Some sources of uncertainty were addressed in a quantitative uncertainty analysis. These include (1) uncertainty in the parameters used to calculate the SIR, including lung cancer mortality in each country based on its mortality reporting system (CLC), background lung cancer mortality among never smokers as a result of exposure to other risk factors such as occupational carcinogens and air pollution (NLC), and lung cancer mortality in CPS-II smokers and nonsmokers (SLC and NLC), and (2) statistical uncertainty in the estimated CPS-II relative risks for each disease (Table 1).
Uncertainty for each of the inputs was estimated separately, as described elsewhere.13 Uncertainties of individual parameters were combined in a stratified sampling simulation25 to estimate an uncertainty distribution for the fraction of cause-specific mortality attributable to smoking. The uncertainty reported here is the 95% range of the combined distribution.
The above quantitative analysis does not include a number of other sources of uncertainty. First, there is uncertainty in the total number of cardiovascular deaths (from the GBD database), especially in developing countries where vital registration is absent or incomplete.26 Second, using SIR, which is based on excess lung cancer mortality, as the marker for accumulated smoking hazard for cardiovascular diseases introduces additional uncertainty. Because lung cancer has a longer lag than cardiovascular diseases, using SIR as the metric of accumulated hazard could result in an overestimation of risk when there has been a sharp decline in smoking (eg, North America and those countries in Western Europe where smoking has declined) and an underestimation of risk when there have been large increases in smoking. On the other hand, the cardiovascular impacts of smoking may be underestimated in developing countries, where smoking has been rising. A third source of uncertainty is the extrapolation of relative risks from CPS-II to other populations, which may have different exposure to other cardiovascular risk factors and background disease levels. The nature of this source of uncertainty and its implications for research and surveillance are described in detail in the Discussion.
Results
The estimated numbers of cardiovascular deaths attributable to smoking are summarized in Table 3, disaggregated by region and sex in Table 4, and displayed by region and disease in the Figure. Globally, an estimated 1.62 (95% CI, 1.27 to 2.04) million cardiovascular deaths were attributable to smoking in 2000, 11% of total global adult (30 years of age) cardiovascular deaths. The proportion of adult cardiovascular deaths attributable to smoking was 17% for men and 5% for women and 7% and 16% for developing and industrialized countries, respectively. An additional 90 000 deaths from hypertensive disease were excluded from these estimates, consistent with the analysis of the 2004 Report of the Surgeon General.9
Adult (age, 30 years) cardiovascular mortality resulting from smoking in GBD epidemiological subregions by disease in 2000. Pattern for hypertensive disease is different because it was excluded from main estimates (see Tables 1 and 5). See Table 2 for description of regions.
Regional Patterns
The largest numbers of cardiovascular deaths attributable to smoking were seen in the industrialized regions (340 000 deaths, 17% of total adult cardiovascular deaths in Eastern Europe and the former Soviet republics [EUR-C]; 220 000 deaths, 22% of total adult cardiovascular deaths in North America [AMR-A]; 200 000 deaths, 13% of total adult cardiovascular deaths in Western Europe [EUR-A]) and the high-mortality developing region of Southeast Asia (SEAR-D; 300 000 deaths, 10% of total adult cardiovascular deaths, dominated by India in terms of population). The large numbers of cardiovascular deaths attributable to smoking in EUR-C and SEAR-D are due to a combination of 2 factors: (1) the large number of cardiovascular deaths resulting from large population size and high background rates of cardiovascular deaths caused by other cardiovascular risk factors such as blood pressure, cholesterol, and body mass index27–29 and (2) a relatively high proportion of cardiovascular deaths caused by smoking. For example, SEAR-D has substantially less adult population than the developing regions of the Western Pacific (WPR-B; dominated by China in terms of population) (480 million versus 740 million) but a similar number of adult cardiovascular deaths (2.9 million versus 3.1 million). Together with a higher smoking-attributable fraction, smoking caused a considerably larger number of cardiovascular deaths in SEAR-D (300 000; 10%) than in WPR-B (120 000; 4%). This is in contrast to similar estimates for cancers, where WPR-B had a much higher total number of deaths (1.8 million cancer deaths in WPR-B versus 800 000 in SEAR-D) and hence higher smoking-attributable cancer deaths (240 000 in WPR-B versus 190 000 in SEAR-D) even with a lower smoking-attributable fraction (14% in WPR-B versus 24% in SEAR-D). Similarly, the adult population of EUR-C (140 million) is substantially smaller than that of EUR-A (260 million) and AMR-A (190 million), yet the total number of cardiovascular deaths in this region (2.0 million) is considerably larger than those in EUR-A (1.6 million) and AMR-A (980 000 million) because of the background risk of cardiovascular mortality resulting from other risks such as elevated blood pressure and overweight and obesity.30 Therefore, although the fraction of all cardiovascular deaths attributable to smoking in EUR-C (17%) is lower than in AMR-A (22%) and only slightly higher than in EUR-A (13%), the number of smoking-attributable deaths in this region is the largest of all industrialized regions.
The proportions of total adult cardiovascular deaths attributable to smoking were generally highest in industrialized regions (12% to 22%), where smoking has been entrenched for several generations (Table 4). At the same time, the increase in smoking over the last quarter of the twentieth century in a number of developing regions, including parts of Latin America, the Eastern Mediterranean, and Southeast Asia, has resulted in 10% of all current adult cardiovascular deaths being attributable to smoking; the number is higher among men (Table 4).
The number of cardiovascular deaths attributable to smoking among men was higher than among women (Tables 3 and 4), with men accounting for 68% of all smoking-attributable cardiovascular deaths in industrialized regions and 78% in developing regions. The exception to this pattern was AMR-A, where female deaths accounted for 58% of total smoking-attributable cardiovascular deaths. Even in North America the fraction of smoking-attributable cardiovascular deaths was slightly lower among women (21% for women versus 23% for men). Because total cardiovascular deaths were higher among women, however, population-attributable fractions were applied to larger baselines for women than for men. Eighty-one percent of smoking-attributable cardiovascular deaths in developing regions were in individuals between 30 and 69 years of age compared with 61% in industrialized regions.
Disease-Specific Patterns
Globally, IHD accounted for the largest number of cardiovascular deaths attributable to smoking (870 000 deaths; 54% of all smoking-attributable cardiovascular deaths), more than twice as many cerebrovascular disease deaths attributable to smoking (400 000 deaths; 25%) (Table 5). The relative contributions of different diseases to smoking-attributable cardiovascular deaths varied across developing and industrialized regions because of variations in both background death rates and fractions attributable to smoking (the Figure). In developing regions, 360 000 IHD and 200 000 cerebrovascular deaths made up 54% and 30% of all smoking-attributable cardiovascular deaths, respectively. In industrialized regions, cerebrovascular disease had a smaller role relative to developing regions, with 510 000 IHD and 210 000 cerebrovascular deaths making up 54% and 22% of all smoking-attributable cardiovascular deaths (see also the article by Yusuf at al5). The relatively smaller role of cerebrovascular disease in industrialized countries was countered by a larger role for the cluster of "other cardiovascular diseases" (25% of all smoking-attributable cardiovascular deaths in industrialized regions versus 16% in developing regions). The contribution of IHD to total smoking-attributable cardiovascular deaths varied from a low of 30% to 40% in the Western Pacific (WPR-A and WPR-B) and parts of sub-Saharan Africa (AFR-E) to a high of 60% to 70% in the Eastern Mediterranean (EMR-B and EMR-D), SEAR-D, and EUR-C (the Figure); the contribution of cerebrovascular disease ranged from 10% to 17% in AMR-A, EUR-A, and parts of EMR-B to >40% in WPR-B, dominated by China.
Discussion
Smoking is currently a very important determinant of cardiovascular mortality among men in all regions of the world and among women in industrialized countries. The proportion of cardiovascular deaths caused by smoking exhibited large variations among different regions of the world (men, from 10% of total cardiovascular mortality in sub-Saharan Africa and parts of Latin America and the Western Pacific to 23% in industrialized regions of Europe and North America; women, from 5% of total cardiovascular mortality in the developing world to >20% in North America). The interregional variation was even more apparent for deaths caused by smoking from specific cardiovascular conditions. In particular, the contribution of IHD to smoking-attributable mortality varied by a factor of 2. Smoking also caused large numbers of deaths from cerebrovascular disease in several developing regions. This interregional variation occurs because the shape and maturity of the smoking epidemic is highly affected by region-specific socioeconomic and cultural determinants of smoking, and because background mortality varies across populations.5
Clinical, experimental, and epidemiological evidence has established the role of environmental tobacco smoke as a risk factor for cardiovascular disease in otherwise healthy nonsmokers.31–35 Because environmental tobacco smoke is also a risk factor for lung cancer,36,37 its effects are captured partially by the SIR, as described in detail elsewhere.38 Separate estimates of cardiovascular mortality as a result of exposure to environmental tobacco smoke would have to consider the physical proximity of smokers and nonsmokers and building ventilation. These parameters, in turn, depend on cultural and geographical factors and on regulation of smoking in public areas.
The estimated number of global cardiovascular deaths attributable to smoking in this analysis (1.62 million) was practically identical to the number obtained by applying a constant "correction factor" of 30% to unadjusted hazards for multiple diseases combined (1.69 million).13 The differences in estimates from the 2 methods across regions and sexes ranged from >25% underestimation among women in sub-Saharan Africa and parts of Southeast Asia and the Western Pacific to >20% overestimation among men in North America, Western Europe, and the Western Pacific when covariate-adjusted site-specific hazards were not used. This variation in underestimation or overestimation arises because all cardiovascular diseases, which in reality have different distributions in different regions and different hazards, were grouped together in previous estimates. Using relative risks with additional adjustment for hypertension and diabetes mellitus resulted in practically identical global estimates (1.62 million versus 1.67 million deaths; difference not statistically significant) and for individual regions and diseases. This robustness to additional adjustment is consistent with the findings of Malarcher and colleagues39 (CPS-II) and Kawachi and colleagues40 (Nurse’s Health Study) who found that relative risks for smoking and cardiovascular diseases were not affected by residual confounding as a result of history of hypertension and diabetes.
The validity of smoking-attributable mortality estimates using the indirect SIR method has been confirmed against alternative methods that account for accumulated hazards of smoking.41 Uncertainty nonetheless remains, especially in developing countries where both complete mortality records and detailed studies of cardiovascular risk factors are less common. Some sources of uncertainty were described earlier. Two important additional sources of uncertainty in the current estimates with implications for future research are (1) the correlation between smoking and other cardiovascular disease risks within populations and (2) biological interactions between smoking and other risks that would result in uncertainty above and beyond that quantified in statistical hazard models when CPS-II relative risks are extrapolated to other populations. For many diseases, multiple risk factors "magnify" each another’s total risks as determined from standard proportional-hazard models. If multiple risks for the same disease are more common (positively correlated) in specific population subgroups, total (absolute) hazards would be larger than those estimated here, even if relative risks remain unchanged (smaller with negative correlation).42,43 Unhealthy diet among smokers is an example of positive correlation. In addition to correlation of risk factor exposures, multiple risks for the same disease may modify one another’s relative risk (eg, the relative risk of smoking may depend on diet or other factors that affect background cardiovascular mortality). The existing evidence indicates that extrapolation of hazards from CPS-II to other populations is a valid option. For example, the Chinese retrospective study of smoking and mortality demonstrated that background mortality rates, which varied by up to a factor of 4 for cardiovascular diseases, did not change the relative risk of mortality as a result of smoking (Figure 5 in Liu et al15). The robustness of relative risks for smoking to changes in other risk factors has also been confirmed in studies that stratified on serum cholesterol.44 More recently, a study in >50 countries illustrated that, when comparable methods were used, hazards of smoking for myocardial infarction did not vary significantly across regions of the world (Western Europe, Central and Eastern Europe, North America, South America and Mexico, Australia and New Zealand, the Middle East, Africa, South Asia, China and Hong Kong, Southeast Asia, and Japan).45,46 Despite these assurances, extrapolation of CPS-II relative risks is a source of additional uncertainty but necessary in the absence of recent local epidemiological studies in developing countries. (At any point in time, only recent epidemiological studies can be used for direct estimates because of changes in smoking patterns over time.)
The uncertainties in our analysis also identify 2 important research and surveillance needs to better understand how tobacco affects the global and regional cardiovascular epidemiology in combination with other risks. First, there is a need for epidemiological research on how smoking may interact with other risk factors. Parallel to this individual-level analysis, there is a need to improve population-level risk factor surveillance to collect data on correlation of multiple cardiovascular risks in population subgroups, differentiated according to factors such as income or education, race or ethnicity, or place of residence. Data on multiple risk factor interaction and correlation would allow better estimation of the total health effects of individual cardiovascular risk factors such as smoking. More important, data on correlated multiple risk exposure would be invaluable for targeting preventive interventions or treatment toward those subgroups at highest risk of cardiovascular disease.
Such research and surveillance needs are particularly important in developing countries. Currently, the number of cardiovascular deaths attributable to smoking is larger in industrialized regions than in the developing world (Table 3). However, available data on smoking trends suggest that smoking has increased in much of the developing world over the past few decades,18 as has population exposure to other cardiovascular risks such as overweight and obesity.4 Because the effects of smoking on cardiovascular disease emerge sooner than other diseases affected by smoking (eg, cancers and chronic obstructive pulmonary disease), this implies an expected increase in cardiovascular mortality in the developing world. At the same time, because the health benefits of smoking cessation occur faster for cardiovascular than other diseases,47 policies that prevent and reduce smoking will have immediate and large benefits for reducing cardiovascular mortality.48–52
Acknowledgments
This research was sponsored by the National Institute on Aging grant PO1–AG17625. The Cancer Prevention Study II is funded by the American Cancer Society.
References
Murray CJL, Lopez AD, eds. The Global Burden of Disease. Cambridge, Mass: Harvard School of Public Health, on behalf of the World Health Organization and the World Bank; 1996.
World Health Organization. World Health Report 2002: Reducing Risks, Promoting Healthy Life. Geneva, Switzerland: World Health Organization; 2002.
Murray CJ, Lopez AD. Alternative projections of mortality and disability by cause 1990–2020: Global Burden of Disease Study. Lancet. 1997; 349: 1498–1504.
Pelletier DL, Rahn M. Trends in body mass index in developing countries. Food Nutr Bull. 1998; 19: 223–239.
Yusuf S, Reddy S, Ounpuu S, Anand S. Global burden of cardiovascular diseases, part I: general considerations, the epidemiologic transition, risk factors, and impact of urbanization. Circulation. 2001; 104: 2746–2753.
Yusuf S, Reddy S, Ounpuu S, Anand S. Global burden of cardiovascular diseases, part II: variations in cardiovascular disease by specific ethnic groups and geographic regions and prevention strategies. Circulation. 2001; 104: 2855–2864.
Pope CA3rd, Burnett RT, Thurston GD, Thun MJ, Calle EE, Krewski D, Godleski JJ. Cardiovascular mortality and long-term exposure to particulate air pollution: epidemiological evidence of general pathophysiological pathways of disease. Circulation. 2004; 109: 71–77.
Singhal A, Lucas A. Early origins of cardiovascular disease: is there a unifying hypothesis; Lancet. 2004; 363: 1642–1645.
US Department of Health and Human Services. The Health Consequences of Smoking: A Report of the Surgeon General. Atlanta, Ga: USDHHS, Public Health Service, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2004.
Thun MJ, Apicella LF, Henley SJ. Smoking vs other risk factors as the cause of smoking-attributable mortality: confounding in the courtroom. JAMA. 2000; 284: 706–712.
Doll R, Peto R, Wheatly K, Gray R, Sutherland I. Mortality in relations to smoking: 40 years’ observation on male British doctors. BMJ. 1994; 309: 901–911.
Peto R, Lopez AD, Boreham J, Thun M, Heath C Jr. Mortality from tobacco in developed countries. Lancet. 1992; 339: 1268–1278.
Ezzati M, Lopez AD. Estimates of global mortality attributable to smoking in 2000. Lancet. 2003; 362: 847–852.
Gajalakshmi V, Peto R, Kanaka TS, Jha P. Smoking and mortality from tuberculosis and other diseases in India: retrospective study of 43 000 adult male deaths and 35 000 controls. Lancet. 2003; 362: 507–515.
Liu BQ, Peto R, Chen ZM, Boreham J, Wu YP, Li JY, Campbell TC, Chen JS. Emerging tobacco hazards in China, 1: retrospective proportional mortality study of one million deaths. BMJ. 1998; 317: 1411–1422.
Popkin BM. The shift in stages of the nutrition transition in the developing world differs from past experiences. Public Health Nutr. 2002; 5: 205–214.
Ezzati M, Vander Hoorn S, Lawes CMM, Leach R, James WTP, Lopez AD, Rodgers A, Murray CJL. Rethinking the "diseases of affluence" paradigm: global patterns of multiple nutritional risk factors in relation to economic development. PLoS Med. 2005; 2: e113.
World Health Organization. Tobacco or Health: A Global Status Report. Geneva, Switzerland: World Health Organization; 1997.
Jha P, Ranson MK, Nguyen SN, Yach D. Estimates of global and regional smoking prevalence in 1995, by age and sex. Am J Public Health. 2002; 92: 1002–1006.
Garfinkel L. Selection, follow-up, and analysis in the American Cancer Society Prospective Studies. In: Garfinkel L, Ochs O, Mushinkski M, eds. Selection, Follow-up, and Analysis in Prospective Studies: A Workshop. Bethesda, Md: National Cancer Institute; 1985. NCI monograph 67.
Burns DM, Garfinkel L, Samet JM, eds. Changes in Cigarette-Related Disease Risks and Their Implications for Prevention and Control: Smoking and Tobacco Control monograph 8. Bethesda, Md: National Cancer Institute; 1997.
Peto R. Influence of dose and duration of smoking on lung cancer rates. In: Zaridze DG, Peto R, eds. Tobacco: A Major International Health Hazard. Lyon, France: International Agency for Research on Cancer; 1986. IARC scientific publication 74.
Ezzati M, Lopez AD. Measuring the accumulated hazards of smoking: global and regional estimates for 2000. Tob Control. 2003; 12: 79–85.
Smith KR, Mehta S, Maeusezahl-Feuz M. Indoor air pollution from household solid fuel use. In: Ezzati M, Lopez AD, Rodgers A, Murray CJL, eds. Comparative Quantification of Health Risks: Global and Regional Burden of Disease Attributable to Selected Major Risk Factors. Geneva, Switzerland: World Health Organization; 2004: 1435–1493.
Morgan MG, Henrion M. Uncertainty: A Guide to Dealing with Uncertainty in Quantitative Risk and Policy Analysis. Cambridge, Mass: Cambridge University Press; 1990.
Mathers CD, Stein C, Ma Fat D, Rao C, Inoue M, Tomijima N, Bernard C, Lopez AD, Murray CJL. The Global Burden of Disease 2000: Version 2, Methods and Results (GPE Discussion Paper No. 50). Geneva, Switzerland: Global Programme on Evidence for Health Policy, World Health Organization; 2002. Available at: http://www.who.int/evidence.
Lawes CMM, Vander Hoorn S, Law MR, Elliott P, MacMahon S, Rodgers A. High blood pressure. In: Ezzati M, Lopez AD, Rodgers A, Murray CJL, eds. Comparative Quantification of Health Risks: Global and Regional Burden of Disease Attributable to Selected Major Risk Factors. Geneva, Switzerland: World Health Organization; 2004: 281–389.
Lawes CMM, Vander Hoorn S, Law MR, Rodgers A. High cholesterol. In: Ezzati M, Lopez AD, Rodgers A, Murray CJL, eds. Comparative Quantification of Health Risks: Global and Regional Burden of Disease Attributable to Selected Major Risk Factors. Geneva, Switzerland: World Health Organization; 2004: 391–496.
James WTP, Leach R, Mhurchu CN, Kalamara E, Shayeghi M, Rigby NJ, Nishida C, Rodgers A. Overweight and obesity (high body mass index). In: Ezzati M, Lopez AD, Rodgers A, Murray CJL, eds. Comparative Quantification of Health Risks: Global and Regional Burden of Disease Attributable to Selected Major Risk Factors. Geneva, Switzerland: World Health Organization; 2004: 497–596.
Ezzati M, Lopez AD, Rodgers A, Vander Hoorn S, Murray CJL, for the Comparative Risk Assessment Collaborative Group. Selected major risk factors and global and regional burden of disease. Lancet. 2002; 360: 1347–1360.
Law MR, Morris JK, Wald NJ. Environmental tobacco smoke exposure and ischaemic heart disease: an evaluation of the evidence. BMJ. 1997; 315: 973–980.
Steenland K. Passive smoking and the risk of heart disease. JAMA. 1992; 267: 94–99.
Glantz SA, Parmeley WW. Passive smoking and heart disease: mechanisms and risks. JAMA. 1995; 273: 1047–1053.
Pope CA III, Eatough DJ, Gold DR, Pang Y, Nielsen KR, Nath P, Verrier RL, Kanner RE. Acute exposure to environmental tobacco smoke and heart rate variability. Environ Health Perspect. 2001; 109: 711–716.
National Cancer Institute, ed. Health Effects of Exposure to Environmental Tobacco Smoke: Smoking and Tobacco Control Monograph 10. Bethesda, Md: National Cancer Institute; 1999.
Hackshaw AK, Law MR, Wald NJ. The accumulated evidence on lung cancer and environmental tobacco smoke. BMJ. 1997; 315: 980–988.
Environmental Protection Agency Office of Research and Development. Respiratory Health Effects of Passive Smoking: Lung Cancer and Other Disorders. Washington, DC: Environmental Protection Agency; 1992.
Ezzati M, Lopez AD. Smoking and oral tobacco use. In: Ezzati M, Lopez AD, Rodgers A, Murray CJL, eds. Comparative Quantification of Health Risks: Global and Regional Burden of Disease Attributable to Selected Major Risk Factors. Geneva:, Switzerland: World Health Organization; 2004: 883–956.
Malarcher AM, Schuman J, Epstein LA, Thun MJ, Mowery P, Pierce B, Escobedo L, Giovino GA. Methodological issues in estimating smoking-attributable mortality in the United States. Am J Epidemiol. 2000; 152: 573–584.
Kawachi I, Colditz GA, Stampfer MJ, Willet W, Manson JE, Rosner B, Hunter DJ, Hennekens CH, Speizer FE. Smoking cessation and decreased risks of total mortality, stroke, and coronary heart disease incidence among women: a prospective cohort study. In: Burns DM, Garfinkel L, Samet JM, eds. Changes in Cigarette-Related Disease Risks and Their Implications for Prevention and Control: Smoking and Tobacco Control Monograph No. 8. Bethesda, Md: National Cancer Institute; 1997.
Bronnum-Hansen H, Juel K. Estimating mortality due to cigarette smoking: two methods, same result. Epidemiology. 2000; 11: 422–426.
Greenland S. Bias in methods for deriving standardized morbidity ratio and attributable fraction estimates. Stat Med. 1984; 3: 131–141.
Ezzati M, Vander Hoorn S, Rodgers A, Lopez AD, Mathers CD, Murray CJL, for the Comparative Risk Assessment Collaborative Group. Estimates of global and regional potential health gains from reducing multiple major risk factors. Lancet. 2003; 362: 271–280.
Jee SH, Suh I, Kim IS, Appel LJ. Smoking and atherosclerotic cardiovascular disease in men with low levels of serum cholesterol: the Korea Medical Insurance Corporation Study. JAMA. 1999; 282: 2149–2155.
Yusuf S, Hawken S, Ounpuu S, Dans T, Avezum A, Lanas F, McQueen M, Budaj A, Pais P, Varigos J, Lisheng L. Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case-control study. Lancet. 2004; 364: 937–952.
Ezzati M. How can cross-country research on health risks strengthen interventions; Lessons from INTERHEART. Lancet. 2004; 364: 912–914.
US Department of Health and Human Services. The Health Benefits of Smoking Cessation: A Report of the Surgeon General. Bethesda, Md: USDHHS, Public Health Service, Centers for Disease Control, Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 1990.
Narkiewicz K, van de Borne PJ, Hausberg M, Cooley RL, Winniford MD, Davison DE, Somers VK. Cigarette smoking increases sympathetic outflow in humans. Circulation. 1998; 98: 528–534.
US Department of Health and Human Services. Women and Smoking: A Report of the Surgeon General. Bethesda, Md: USDHHS, Public Health Service, Centers for Disease Control Office on Smoking and Health; 2001.
Rehm J, Room R, Monteiro M, Gmel G, Graham K, Rehn N, Sempos CT, Frick U, Jernigan D. Alcohol use. In: Ezzati M, Lopez AD, Rodgers A, Murray CJL, eds. Comparative Quantification of Health Risks: Global and Regional Burden of Disease Attributable to Selected Major Risk Factors. Geneva, Switzerland: World Health Organization; 2004: 959–1108.
Cohen A, Anderson R, Ostro B, Pandey KD, Krzyzanowski M, Künzli N, Gutschmidt K, Pope A, Romieu I, Samet J, Smith KR. Urban ambient air pollution. In: Ezzati M, Lopez AD, Rodgers A, Murray CJL, eds. Comparative Quantification of Health Risks: Global and Regional Burden of Disease Attributable to Selected Major Risk Factors. Geneva, Switzerland: World Health Organization; 2004: 1353–1433.
Prüss-üstün A, Fewtrell L, Landrigan P, Ayuso-Mateos J. Lead exposure. In: Ezzati M, Lopez AD, Rodgers A, Murray CJL, eds. Comparative Quantification of Health Risks: Global and Regional Burden of Disease Attributable to Selected Major Risk Factors. Geneva, Switzerland: World Health Organization; 2004.(Majid Ezzati, PhD; S. Jan)