Background
“Let Us Protect Our Future” is a sexual risk-reduction intervention for sixth-grade adolescents in South Africa. Tested in a cluster-randomized controlled trial, the intervention significantly reduced self-reported intercourse and unprotected intercourse during a 12-month follow-up period.

Purpose
The present analyses were conducted to identify moderators of the intervention’s efficacy as well as, which theory-based variables mediated the intervention’s effects.

Methods
Intervention efficacy over the 3-, 6-, and 12-month follow-up was tested using generalized estimating equation models.

Results
Living with their father in the home, parental strictness, and religiosity moderated the efficacy of the intervention in reducing unprotected intercourse. Self-efficacy to avoid risky situations and expected parental disapproval of their having intercourse, derived from Social Cognitive Theory, significantly mediated the intervention’s effect on abstinence.

Conclusions
This is the first study to demonstrate that Social Cognitive variables mediate the efficacy of a sexual risk-reduction intervention among South African adolescents.