Strategies are discussed for locating and interviewing a community-based sample of 492 adults, approximately half of whom were diagnosed in 1974 with ADHD. Completion rates, reasons for attrition, and amount of time and intensity of effort associated with subject location are reported. We found that 81% of the original sample had been located after 28 months, and that the reconstituted sample closely resembled the original larger childhood sample in background characteristics. Findings confirm that staff persistence was an important factor in containing attrition, and that longer contact intervals are likely to be a necessary feature of follow-up with socially and behaviorally disordered samples.