Statistical methods in genetic research on smoking.

A growing body of evidence suggests that genetic factors have an important influence on the onset and course of smoking. Here we review some of the statistical methods that have been used to test for genetic influences on smoking behaviour, with a particular focus on studies of large national twin samples. We show how many of the hypotheses that have been tested using a genetic model-fitting approach have also been reformulated using logistic regression models that will be more familiar to epidemiologists. Such an approach is more easily extended to allow for sociocultural, as well as genetic, influences on smoking behaviour. Using either approach, data are consistent in indicating that certainly in men, and possibly in women, genetic factors play an important role in predicting which individuals who become cigarette smokers progress to being long-term persistent smokers.