Human immunodeficiency virus seroprevalence and risk assessment of a homeless population in Denver

TO determine the prevalence of human immunodeficiency virus (HIV) infection among the homeless of Denver and to describe behaviors in the homeless that may be associated with HIV infection.

A cross-sectional cohort study.

From July 1990 through June 1994, the authors conducted an unlinked survey collecting demographic and risk exposure data, and from August 1990 through June 1992, a more detailed risk behavior survey was completed on persons who attended the largest homeless clinic in Denver.

For the combined survey years, the overall seroprevalence rate in the unlinked survey was 0.9%. Men were more likely to be seropositive than women (1.3% versus 0.1%) (P < 0.001). Black and Hispanic men had higher seroprevalence rates than white men (3.1% and 2.2% versus 0.5%) (P < 0.001). Gay and bisexual men, men who were injection drug users, and men with partners at risk had a nearly fivefold higher seroprevalence rate compared to other risk groups (3.1% versus 0.7%, P < 0.001). During the 4 study years, 14% of homeless persons tested positive for tuberculosis. In the risk behavior survey, 41% of the clients reported previous injection drug use (since 1978), and 22% reported recent use (past 12 months); of this 22%, 16% reported sharing their works (needles and paraphernalia). Seventy percent of the study participants stated that they changed their sexual behavior, and 39% reported using condoms in the past 12 months.

In this homeless Denver population, there is a low seroprevalence of HIV but a high rate of HIV risk behavior. Certain groups of homeless persons are at high risk for HIV infection, and targeted interventions are necessary.