Adolescent girls incarcerated in a state reformatory (N = 246) were recruited and assigned to an 18-session health education program or a time-equivalent HIV prevention program. Cohorts were assigned to conditions using a randomized block design separated by a washout period to reduce contamination. Post intervention, girls in the HIV risk reduction program demonstrated the acquisition of risk-reduction behavioral skills and improved condom application skill. At a follow-up assessment approximately 9 months after release from the correctional facility, girls in both conditions reported fewer unprotected sexual intercourse occasions and less sex while under the influence of alcohol or other drugs