Subjects were divided into groups based on whether they had an externalizing disorder or an internalizing disorder at the time of the baseline interview. The groups were: Externalizing, Internalizing, Both, or Neither. Alcohol use disorder diagnosis was then assessed at each interview period, using the DSM 5 distinctions for Mild AUD (2–3 symptoms), Moderate AUD (4–5 symptoms), and Severe AUD (6 or more symptoms. The age of onset was defined as the first age when the required number of symptoms occurred during the same year. If two interviews performed on the same subject at different times provided divergent estimates of age of onset, the earliest age was taken as correct. Subjects were stratified into mutually exclusive categories based on the most severe AUD diagnosis they received during any part of the follow-up period (i.e., subjects with severe AUD were not counted in the Mild or Moderate AUD categories, though they may have aged through a time when they would have qualified for one or both of those diagnoses). Subjects with age of onset of AUD prior to age of onset of internalizing/externalizing disorders were excluded from analysis.
