HIV seroprevalence was 5 times greater for indigent adults than in San Francisco generally. Sexual behavior predicted HIV infection better than drug use, even among IDUs.

Maintaining study cohorts is a key element of longitudinal research. Participant attrition introduces the possibility of bias and limits the generalizability of a study's findings, but with appropriate planning it is possible to sustain contact with even the most transient participants. This paper reviews the essential elements of tracking and follow-up of marginalized populations, which are (1) collection of contact information, (2) thorough organization of tracking efforts, (3) attention to staff training and support, (4) use of phone and mail follow-up, (5) use of incentives, (6) establishing rapport with participants, (7) assurance of confidentiality, (8) use of agency tracking, (9) use of field tracking, and (10) attention to safety concerns. Diligent application of these tracking strategies allows researchers to achieve follow-up rates of 75 percent to 97 percent with vulnerable populations such as homeless, mentally ill adults, injection drug users, and runaway youth.