Predictors of increased condom use following HIV intervention with heterosexually active drug users.

Abstract

Research with injection drug users (IDUs), at risk for acquiring and transmitting HIV, has focused primarily on their risky drug practices, with far less attention paid to their risky sex behaviors. The purpose of this study was to determine what variables were associated with an increase in condom use following an HIV intervention with 3357 IDUs in nine cities. Participants reported using condoms during 15% of their sexual encounters prior to the HIV intervention, and during 22% of their sexual encounters six months later. A logistic regression analysis indicated that individuals who increased their condom use were likely to be HIV seropositive (odds ratio OR = 2.49), to have received AIDS information prior to the intervention (OR = 1.28), to have multiple sex partners (OR = 2.14), to be single with multiple sex partners (OR = 1.34), or to have exchanged drugs or money for sex (OR = 1.33). Discussion focuses on the generally low incidence of condom use and the need for increased intervention, particularly among drug users in monogamous relationships and sex workers.