An analysis of differential factors affecting risk behaviors among out- of-treatment drug users in four cities

Although its explanatory value remains questionable, 'race/ethnicity' is often offered as an explanation of enhanced risk behavior or enhanced risk of acquiring HIV. The literature is often unclear about the underlying causes of racial differences in seroincidence. By examining the role of intervening factors in the association between race and HIV risk, it was thought that the role attributed to race could be clarified. This document describes the process by which the differential factors affecting HIV risk in a population of out-of-treatment drug users were analyzed. Condom use, number of partners, number of times injecting in the previous 30 days, number of times injecting with a used needle, and number of times injecting with a used needle without cleaning were examined. Besides race, the other factors considered are demographics (gender, age, mobility, job status, education, domestic status) and sexually transmitted disease history. Data for the study were derived from the Risk Behavior Assessment (RBA), the Risk Behavior Follow-up Assessment (RBFA), and the TCU Self-Rated Form. Data were collected from four Cooperative Agreement sites: Houston, New Orleans, San Juan, Puerto Rico, and Tucson. The data for several risk behaviors and for HIV serostatus were analyzed using a three-stage sequential logic regression model. Socio- demographic variables were analyzed first, followed by clinical variables. A standard software package (SPSS+) was employed.