A longitudinal randomized design was used to evaluate the impact of a theoretically based, stand-alone interactive video intervention on 300 urban adolescent girls' (a) knowledge about sexually transmitted diseases (STDs), (b) self-reported sexual risk behavior, and (c) STD acquisition. It was compared to two controls, representing high-quality informational interventions. One used the same content in book form; the other used commercially available brochures. Following randomization, the interventions were administered at baseline, with booster sessions at 1, 3, and 6 months. Self-reports revealed that those assigned to the interactive video were significantly more likely to be abstinent in the first 3 months following initial exposure to the intervention, and experienced fewer condom failures in the following 3 months, compared to controls. Six months after enrollment, participants in the video condition were significantly less likely to report having been diagnosed with an STD. A non-significant trend in data from a clinical PCR assay of Chlamydia trachomatis was consistent with that finding.