Lung carcinoma in African Americans and whites

BACKGROUND: African Americans are at higher risk for lung carcinoma than whites in the United States. This racial disparity is greater among younger people. The authors evaluated whether racial differences in lung carcinoma risk can be explained by differences in cigarette smoking behaviors. METHODS: For this study, 5588 population-based cases of African Americans and whites with pathologically confirmed lung carcinoma, diagnosed between 1984 and 1987, were identified through the Metropolitan Detroit Cancer Surveillance System. Also identified were 3692 population-based controls. Logistic regression methods were used to evaluate the risk of lung carcinoma associated with race both within cigarette smoking category and after adjustment for cigarette-smoking behaviors. RESULTS: The difference in lung carcinoma incidence between African Americans and whites was explained almost entirely by differences in smoking habits among study participants age 55-84 years. However, among males age 40-54 years, African Americans were 2-4 times more likely to develop lung carcinoma of any histologic type than whites even after adjustments were made for smoking habits. Similar excesses in risk among African American females age 40-54 years were demonstrated only for squamous cell and small cell carcinomas (odds ratios OR and 95% confidence intervals CI = 3.7 1.5-8.9 and 2.7 1-7.3, respectively). Also, in this younger age group, African American male nonsmokers and smokers of 1-40 pack-years had a significantly higher risk of lung carcinoma than white males belonging to the same age group with a similar cigarette-smoking history (ORs and 95% CIs = 8 2-32.8 and 3.1 1.9-5.4 respectively). CONCLUSIONS: Elevated risk for lung carcinoma among younger (but not among older) African Americans, particularly among males, exists both within cigarette smoking exposure level and beyond that associated with smoking habits. This may indicate a high risk group particularly susceptible to lung carcinogens, or it may indicate unique exposures not yet identified.