Of the probands who were interviewed at baseline (n=1247), 793 (63.6%) were interviewed at the 5-year follow-up (Supplemental Figure 1 provides a flow chart of proband selection). Of these, 9 had incomplete data to inform AUD remission status, leaving 784 probands with baseline and follow-up interviews and AUD remission status. Of these, 178 were excluded from the current study because they did not have a first-degree relative with lifetime AUD, necessary to the purpose of this study, leaving 606 probands for the current analysis. These probands comprised the index group with whom relatives were paired. Half of probands (n=303) were randomly selected and paired with a biologically-related first-degree relative; the remaining probands were paired with a randomly selected, unrelated individual of the same race from the remaining group of first-degree relatives. The pool of relatives available for matching to probands comprised 2305 first-degree relatives with lifetime AUD who participated in the baseline or the 5-year follow-up interview.
