not in others (Hosmer and Lemeshow, 2000). Variables that were not associated with the outcome category and that did not contribute to model fit based on comparison of fit statistics were eliminated from the model if they were not necessary for interpretation. Once final models were identified, interactions of gender with marital status were tested, based on evidence that marriage is differentially associated with gender in individuals with AUDs (Schneider et al., 1995), and that women who are regular drinkers are more likely than their male counterparts to select partners with alcohol dependence (Agrawal et al., 2006). All statistical tests were adjusted for familial clustering using the Huber-White robust variance estimator in the Stata statistical program (StataCorp, 2009).
