Retention and attendance of women enrolled in a large prospective study of HIV-1 in the United States.

The objective was to assess study retention and attendance for two recruitment waves of participants in the Women's Interagency HIV Study (WIHS). The WIHS, a prospective study at six clinical centers in the United States, has experienced two phases of participant recruitment. In phase one, women were screened and enrolled at the same time, and in phase two, women were screened and enrolled at separate visits. Compliance with study follow-up was evaluated by examining semiannual study retention and visit attendance. After 10 study visits, the retention rate in the original recruits (enrolled in 1994-1995) was 83% for the HIV-infected women and 69% for the HIV-uninfected women compared with 86% and 86%, respectively, in the new recruits (enrolled in 2001-2002). In logistic regression analysis of the HIV-infected women, factors associated with early (visits 2 and 3) nonattendance were temporary housing, moderate alcohol consumption, use of crack/cocaine/heroin, having a primary care provider, WIHS site of enrollment, lower CD4 cell count, and higher viral load. Among HIV-uninfected women, the factors associated with early nonattendance were recruitment into the original cohort, household income >or=$12,000 per year, temporary housing, unemployment, use of crack/cocaine/heroin, and WIHS site of enrollment. Factors associated with nonattendance at later visits (7-10) among HIV-infected participants were younger age, white race, not having a primary care provider, not having health insurance, WIHS site of enrollment, higher viral load, and nonattendance at a previous visit. In HIV-uninfected participants, younger age, white race, WIHS site of enrollment, and nonattendance at a previous visit were significantly associated with nonattendance at later visits. Preventing early loss to follow-up resulted in better study retention early, but late loss to follow-up may require different retention strategies.