To address the second aim and evaluate Hypothesis 3 (alcohol involvement at follow-up by age), subjects who completed the interview were divided into four age groups (50 – 54, 55 – 59, 60 – 64, and 65+). These age divisions were imposed in order to obtain a more fine-grained (primarily five-year) characterization of outcome, because the total age span of this sample encompasses different life stages and circumstances (e.g., employment vs. retirement, children in the home vs. out of the home, etc.).
