We investigated the association between AUD in young adulthood and parental EB using logistic regression. Parental EB in the three developmental periods can be combined in eight ways. Letting the first position represent parental EB between age 0–6, the second between 7–12, and the third between 13–18, the eight patterns can be denoted: (0,0,0), (1,0,0), (0,1,0), (0,0,1), (1,1,0), (1,0,1), (0,1,1), and (1,1,1). These patterns are included as a categorical variable in a single logistic regression model from which odds ratios were calculated, thus minimizing the number of tests. Absence of parental EB during all three time periods was the reference group. We report odds ratios (ORs) for comparisons of these parental EB patterns. As noted above, we adjust for parental education, sex, and cohort, and, based on preliminary analyses, we also allow the effect of cohort to be modified by sex. Besides ORs we also present the predicted probabilities, based on parameter estimates.
