No abstract

OBJECTIVES. To assess long-term trends in cigarette smoking according to the combined influence of sex and education, this study examined smoking prevalence in successive US birth cohorts. METHODS. Data from nationally representative surveys were examined to assess smoking prevalence for six successive 10-year birth cohorts stratified by race or ethnicity, sex, and educational attainment. RESULTS. Substantial declines in smoking prevalence were found among men who had a high school education or more, regardless of race or ethnicity, and slight declines among women of the same educational background were revealed. However, little change was found in smoking prevalence among men of all race/ethnic groups with less than a high school education, and large increases were found among women with the same years of schooling, especially if they were White or African American. CONCLUSIONS. These data suggest that persons of low educational attainment have yet to benefit from policies and education about the health consequences of cigarette smoking.

Knowledge of the epidemiology of tobacco use and dependence can be used to guide research initiatives, intervention programs, and policy decisions. Both the reduction in the prevalence of smoking among US adults and black adolescents and the decline in per capita consumption are encouraging. These changes have probably been influenced by factors operating at the individual (e.g., school-based prevention programs and cessation programs) and environmental (e.g., mass media educational strategies, the presence of smoke-free laws and policies, and the price of tobacco products) levels (for a discussion of these factors, see, e.g., refs. 2, 48, 52, 183, and 184). The lack of progress among adolescents, especially whites and males, and the high risk for experimenters of developing tobacco dependence present cause for great concern (48, 183-186). In addition to those discussed above, several areas of research can be recommended. 1. Better understanding of the clustering of tobacco use with the use of other drugs, other risk behaviors, and other psychiatric disorders could better illuminate the causal processes involved, as well as the special features of the interventions needed to prevent and treat tobacco dependence. 2. To better understand population needs, trend analyses of prevalence, initiation, and cessation should, whenever possible, incorporate standardized measures of these other risk factors. Future research should compare the effect of socioeconomic status variables on measures of smoking behavior among racial/ethnic groups in the United States. 3. For reasons that may be genetic, environmental, or both, some persons do not progress beyond initial experimentation with tobacco use (2, 48, 183, 187-192), but about one-third to one-half of those who experiment with cigarettes become regular users (48, 193, 194). Factors, both individual and environmental, that can influence the susceptibility of individuals to tobacco dependence need further attention. 4. To estimate their sensitivity and specificity, comparisons of the National Household Survey on Drug Abuse indicators of dependence with DSM-based criteria are needed. Public health action continues to be warranted to reduce the substantial morbidity and mortality caused by tobacco use (195). A paradigm for such action has been recommended and involves preventing the onset of use, treating tobacco dependence, protecting non-smokers from exposure to secondhand smoke, promoting nonsmoking messages while limiting the effect of tobacco advertising and promotion on young people, increasing the real (inflation-adjusted) price of tobacco products, and regulating tobacco products (186).

The associations between depressive symptoms and regular smoking appears to be established by adolescence.

These data suggest that smoking initiation at a young age can increase the risk of nicotine addiction during adolescence and that sports participation may influence smoking behavior. Interventions to prevent smoking should be available before age 12 years to help combat the smoking epidemic among youth.

No abstract

The authors explored two methodological issues in the estimation of smoking-attributable mortality for the United States. First, age-specific and age-adjusted relative risk, attributable fraction, and smoking-attributable mortality estimates obtained using data from the American Cancer Society's second Cancer Prevention Study (CPS II), a cohort study of 1.2 million participants (1982-1988), were compared with those obtained using a combination of data from the National Mortality Follow-back Survey (NMFS), a representative sample of US decedents in which information was collected from informants (1986), and the National Health Interview Survey (NHIS), a nationally representative household survey (1987). Second, the potential for residual confounding of the disease-specific age-adjusted smoking-attributable mortality estimates was addressed with a model-based approach. The estimated smoking-attributable mortality based on the CPS II for the four most common smoking-related diseases-lung cancer, chronic obstructive pulmonary disease, coronary heart disease, and cerebrovascular disease-was 19% larger than the estimated smoking-attributable mortality based on the NMFS/NHIS, yet the two data sources yielded essentially the same smoking-attributable mortality estimate for lung cancer alone. Further adjustment of smoking-attributable mortality for disease-appropriate confounding factors (education, alcohol intake, hypertension status, and diabetes status) indicated little residual confounding once age was taken into account.

No abstract