The purpose of this study was to assess the efficacy of brief group interventions, the positive choices intervention (PCI) and a standard intervention (SI), to increase condom use and intention to use condoms and to change condom use attitudes and beliefs. The design of the study was a randomized comparative trial. Participants were 347 heterosexual African American crack cocaine users living with HIV infection. Data were collected at intake and at three and nine months after intake. Behavioral and sociocognitive data were collected. Although both brief interventions achieved positive results, there were significant differences in outcomes between the interventions groups. The mean number of sex partners was significantly lower in the PCI group at three months. The proportion of those assigned to the PCI reporting sex with a paid partner significantly decreased, while the proportion disclosing their serostatus to their partners increased. There were no significant differences on these measures in the SI group. Significant time effects were found on measures of condom use, condom use attitudes, and self-efficacy beliefs. These measures significantly increased from intake to one month for both groups. One significant time-by-group effect was found. The measure of situational self-efficacy significantly increased in the PCI group, but not the SI group. Results also showed significant time-by-time effects. Mean condom use, intention to use condoms, attitudes, and condom use self-efficacy beliefs showed significant difference between three and nine months. However, there was no clear pattern of change. Findings suggest that brief group interventions designed to reduce HIV can help heterosexual drug users living with HIV infection increase condom use and intention to use condoms and change condom use attitudes and beliefs. A significant time-by-group effect was observed only for situational self-efficacy, suggesting limited additional efficacy of the PCI intervention. Given similar positive findings between groups, more research is needed to determine which components of brief interventions produce changes in motivations and risk behaviors.