Racial Disparity of the Relationship Between Cotinine Concentration and Age: The National Health and Nutrition Examination Survey 2005-2006 and 2007-2008

PURPOSE: Aging produces many physiological changes in the body. Thereby, effect of smoking on the body may vary with age, but not necessarily in a straight line. Previous studies assuming linearity identified a non-significant relationship between cotinine concentration and age. This study is aimed at determining whether there is a nonlinear association of serum cotinine concentration with age and whether there is any racial disparity in the association.

METHODS: A Generalized Additive Model (GAM) was used to examine graphically possible nonlinear associations. The analyses utilized a valid sample of 2170 smokers aged ≥20 years from the National Health and Nutrition Examination Survey 2005-2008.

RESULTS: The GAM demonstrated that the cotinine concentration went up steadily until age 67 at which cotinine was the highest and then dropped quickly (p>0.0001). Nonlinear pattern persisted even after controlling the daily cigarettes consumed (p=0.0088). The pattern differed little among racial groups. While it was peaked at the age of 51 among non-Hispanic whites, it was at 44 and 62 among non-Hispanic blacks. The initial and peak values of cotinine concentration were higher in African-Americans. The multivariate regression model suggested that nonlinear model fit the data better than a linear model did.

CONCLUSIONS: A significant nonlinear relationship existed between cotinine concentration and age. The finding suggested that the effect of smoking on the human body may be greater among those aged 50-67 and African-Americans than their counterparts. This finding may be a partial explanation for the association of lung cancer mortality with age and race.