We ran a series of analyses to examine whether there were any cohort effects for the focal study variables and whether those who did not complete the follow-up assessment (n = 1710, 755 male, 951 female, 4 missing sex) differed from those who did complete the follow-up assessment. There was no evidence for cohort effects for any of the focal study variables (parental alcohol problems, conduct problems, positive urgency, dating several people, or alcohol use; all p-values > 0.09; full results available upon request from the first author). The attrition analyses indicated that males were more likely to miss the follow-up assessment than females (OR = 1.42, p < 0.01), that black participants were less likely to miss the follow-up compared to non-black participants (OR = 0.68, p < 0.01), and that Asian participants were less likely to miss the follow-up compared to non-Asian participants (OR = 0.75, p < 0.01). Those with higher conduct problems (OR = 1.22, p = 0.04) and baseline alcohol use (OR = 1.11, p < 0.01) were more likely to miss the follow-up assessment.
