Racial and ethnic differences in lung cancer incidence: how much is explained by differences in smoking patterns? (United States).

OBJECTIVES: Lung cancer rates vary considerably among U.S. racial/ethnic groups. We quantitatively analyzed the extent to which these differences can be attributed to differential patterns of smoking.
METHODS: We utilized survey data from the U.S. Census to estimate smoking patterns in the following racial/ethnic groups: non-Hispanic whites, non-Hispanic blacks, Hispanics, Asian/Pacific Islanders and American Indians. We used several dose-response models of smoking and lung cancer to predict relative lung cancer rates in these groups based on reported smoking patterns, specifically, on smoking status (current, former, never), cigarettes per day, age started, and age quit (for former smokers). Predicted rates were compared to observed population rates for these groups.
RESULTS: Black men had slightly lower predicted lung cancer rates than white men, but had 35-47% higher observed rates. Hispanic men had predicted rates about 25% lower than whites but observed rates 50% lower than whites; predicted rates for Hispanic women were 50% lower than whites compared to observed rates that were 60-70% lower. For Asian/Pacific Islanders, predicted and observed rates relative to whites were comparable. Predicted rates for American Indians were slightly higher than whites while observed rates were about 40% lower.
CONCLUSION: Differences in smoking largely explain lower lung cancer rates in Asian/Pacific Islanders relative to whites and partially explain lower rates in Hispanics compared to whites. Increased rates in black men and decreased rates in American Indians are not explained by differences in smoking.