Medical service use among individuals receiving HIV prevention services in Los Angeles County

Background: Timely provision of medical services among communities at increased risk for HIV infection is crucial to detect the infection and to further prevent the spread of the HIV epidemic. In the US, HIV has been disproportionately affecting certain populations: men who have sex with men (MSM), Blacks/ African Americans and Latinos/ Hispanics. A major problem is that too many individuals are being diagnosed with HIV in a late stage of infection. Objectives: To identify factors that affecting medical service utilization among communities at increased risk for HIV infection. Methods: The Gelberg-Andersen behavioral model includes predisposing characteristics specific to vulnerable populations for predicting medical service use. The candidate variables considered include social support (indicated by living arrangement and case management), attitudinal factors (including beliefs, knowledge, and attitudes), and behavioral variables (including alcohol and drug use, physical and sexual abuse, sex trading and risky sexual practices). The data were collected from 2,126 clients of HIV prevention agencies in Los Angeles County in 2004 with the Countywide Risk Assessment Survey (CRAS). Results: The factors that were positively associated with use of medical services in a logistic regression model included Black/African Americans compared to Whites; individuals who live in treatment center/ halfway house or mission/ shelter compared to individuals who live in a house or apartment; individuals who experience physical/ sexual abuse and individuals who have ever received HIV testing/ counseling. Factors inversely associated with medical service use were male gender, Latinos/ Hispanics, individuals with high school diploma or graduate school compared to those who did not complete High School; individuals who live in a car or other vehicle and individuals who consume alcohol. Analysis was conducted using SAS 9.3. Conclusions: Most of the findings are consistent with the Gelberg-Anderson model. The exceptions being that victims of physical or sexual abuse were more likely to use services instead of less likely as predicted by the model and living arrangement was the strongest predictor of medical service use.