To test the efficacy of an innovative prevention approach for HIV-positive men who put their sexual partners at risk, we seek to adapt and expand a counseling intervention already proven effective in reducing risk among HIV-negative men. Investigators from UCSF, the San Francisco Dept. of Health, and the East Bay AIDS Research Institute propose a 5-year, randomized, controlled trial of a two-session counseling intervention. Rising HIV incidence as well as data showing that the use of HAART predicts a future diagnosis of a sexually transmitted disease suggests strongly that some HIV-positive people continue to engage in high-transmission risk sexual activity. The proposed intervention helps individuals identify and re-evaluate their "self- justifications," that is, their thoughts or beliefs when they decide to engage in high-transmission risk activity. In the first year, using qualitative methods, we will develop and pilot test a questionnaire to elicit such self-justifications. Over the next 3-1/2 years, we will compare the counseling intervention with typical CDC risk-reduction counseling at primary medical clinics in three Bay Area counties and at a large HIV mental health clinic in San Francisco. In year one, we will conduct qualitative interviews with 30 HIV-positive participants who have engaged in high risk unprotected sex with partners of discordant or unknown serostatus in the prior 12 months. In years 2-4, we will recruit 376 additional similar HIV-positive participants and randomize them to either the intervention or control group. Both groups will receive a booster session at 6 months. Follow-up data will be collected at 6 and 12 months. Primary outcomes include decreases in self-reported episodes of unprotected sexual intercourse with partners of discordant or unknown serostatus, decreases in the proportion of persons reporting high-risk behaviors, and reduced rates of laboratory-confirmed acquisition of other sexually transmitted diseases. This study will address a critical need: can a theory-based counseling intervention reduce future high risk sexual behavior among HIV-positive persons? If so, this intervention could be rapidly incorporated into primary care settings and slow the spread of HIV.