You-Me-Us: Impact of a Relationships-Based Program to Reduce Sexual Risk Among Urban Middle School Students

Introduction: Romantic relationships are central in the lives of young people (Feiring, 1999). Indeed, by middle adolescence, most young people have been involved in at least one romantic relationship (Collins & Laursen, 2000), a context in which many sexual interactions occur (Manning, Giordano, and Longmore 2006). Nonetheless, most evidenced-based HIV/STI and pregnancy prevention programs used in school settings in the US were developed with HIV or pregnancy prevention funding, and most commonly address sexual risk taking as an individual behavior outside the context of romantic relationships. Numerous researchers have suggested the importance of attending to relationships in programs focused on sexual risk (Giordano, Manning, and Longmore 2005, 2010). The aim of this study was to develop and implement a relationships-based prevention program featuring two evidence based components (a prevention curriculum plus a school-wide social norm component) that center on relationship factors contributing to risk for unplanned pregnancy and/or sexually transmitted infections (STI) among urban middle school youth.

Methods:  The study featured a longitudinal group-randomized trial involving 9 schools randomized to intervention (5 schools, n=528) or control (4 schools, n=383). The curriculum was taught by outside health educators during 7^th grade. The school-wide component started late in 7^th grade and continued in 8^th grade. Participants completed 3 surveys: a baseline survey in the fall/winter of 7^th grade and two follow-up surveys 6- and 18-months after baseline. The sample, 911 7^th grade youth (mean age 12.4), represents 76% of the 7^th grade students enrolled in the participating schools at the time of the baseline survey. Six- and 18-month follow-up rates were 92% and 80%. Most youth described themselves as black (32%) or Hispanic (33%), and 45% said they speak a language other than English half or more of the time at home. Primary outcomes included intentions to have vaginal sex in next 3 months and by the end of 8^thgrade, and initiation of vaginal sex. Multilevel models were used to adjust for the correlation among students within the same school, and the correlation of repeated measurements taken on the same student over time. 

 Results: There were no statistically significant differences between intervention and control students on intention to have vaginal sex in the next 3 months or by the end of 8^thgrade (Cohen’s d = 0.04 and 0.03, respectively). The program reduced sexual initiation at the 6-month and 18-month follow-ups (OR=0.49 and 0.76, respectively), but these reductions did not reach statistical significance (p=.10 and .43 at 6- and 18-month follow-ups). Student and staff reaction data suggest the relationships-based approach was well received. Observations and log data highlight implementation challenges, particularly with the social norms component.

Conclusions: A relationship-based approach has promise for reducing sexual risk behaviors among urban youth as evidenced by the trends in sexual initiation in this small randomized trial. The content is fresh and well received, but an array of issues, such as classroom management, over-extended school staff, and general school dysfunction impacted implementation.