Purpose-To evaluate the efficacy of two, theory-based, multi-media, middle school sexual education programs in delaying sexual initiation.Methods-Three-armed, randomized controlled trial comprising fifteen urban middle schools; 1,258 predominantly African-American and Hispanic 7 th grade students followed into 9 th grade. Both programs included group and individualized, computer-based activities addressing psychosocial variables. The risk avoidance (RA) program met federal abstinence education guidelines; the risk reduction (RR) program emphasized abstinence and included computer-based condom skills-training. The primary outcome assessed program impact on delayed sexual initiation; secondary outcomes assessed other sexual behaviors and psychosocial outcomes.Results-Participants were 59.8% female, mean age 12.6 years. Relative to controls, the RR program delayed any type of sexual initiation (oral, vaginal or anal sex) in the overall sample (adjusted odds ratio [AOR]: 0.65, 95% CI: 0.54-0.77), among females (AOR: 0.43, 95% CI: 0.31-0.60) and African-Americans (AOR: 0.38, 95% CI: 0.18-0.79). RR students also reduced unprotected sex at last intercourse (AOR: 0.67, 95% CI: 0.47-0.96), past three months' frequency of anal sex (AOR: 0.53, 95% CI: 0.33-0.84) and unprotected vaginal sex (AOR: 0.59, 95% CI: 0.36-0.95). The RA program delayed any sexual initiation among Hispanics (AOR: 0.40, 95% CI: 0.19-0.86), reduced unprotected sex at last intercourse (AOR: 0.70, 95% CI: 0.52-0.93) but increased the number of recent vaginal sex partners (AOR: 1.69, 95% CI: 1.01-2.82). Both programs positively impacted psychosocial outcomes. program, comparable in duration, delivery, and theoretical framework, to further examine how both approaches impact middle school students. The primary hypothesis tested whether students, who received either the RR or the RA intervention, would delay sexual initiation into 9 th grade relative to controls. Secondary hypotheses tested intervention effects on other sexual behaviors and psychosocial factors.
Conclusions-The
METHODS Study Design and ParticipantsA randomized controlled trial was conducted from 2006-2010 in fifteen urban middle schools in a large south-central US school district. Seventy-eight percent of students were classified as economically disadvantaged. Assignment of schools to one of three conditions (5 to the RA condition, 5 to the RR condition, and 5 to a control condition) was conducted prior to baseline assessment using a multi-attribute randomization protocol, accounting for school size, racial/ethnic composition (percent of African-American and Hispanic students), and geographic location.[18]Assuming 15% of controls would initiate any type of sex by 9 th grade, 25% attrition, intraschool correlations=0.005, and α=0.05 (two-tailed), an initial sample size of 1,500 7 th grade students was estimated to provide 80% power to detect 10% pairwise differences in sexual
Author Manuscript Author ManuscriptAuthor ManuscriptAuthor Manuscript initiation between intervention and co...