Structural Modeling of School-Related Predictors of HIV/STD Testing Among Sexually Experienced High School Students

Background: In 2013, nearly half of U.S. high school students had ever had sexual intercourse. Among students who had sex in the previous three months, only 59.1% reported using a condom at last sex which puts many young people at risk for HIV/STDs. The purpose of this study is to elucidate critical individual and school-related predictors and pathways to HIV/STD testing among sexually experienced teens using a structural equation model.

Methods: Youth from seven urban high schools in a south Florida school district were surveyed (N=11681). Analyses were restricted to youth reporting ever having had vaginal, oral or anal sex in their lifetime and having complete data (n=3505). Data were collected using paper and pencil surveys administered in all regular classrooms during a preselected class period. Hypothesized relationships between school climate and sexual health service variables as predictors of HIV or STD testing were examined using structural equation modeling in Amos; standardized betas are reported.

Results: Direct effects on reported HIV/STD testing included having received a referral from a school staff member for sexual health services (β=.09) and higher levels of exposure to HIV/STD health posters or messages in and out of school (β=.09). Strong indirect effects on testing behavior were seen for bullying experience (β=.12) and exposure to HIV/STD health posters or messages in or out of school (β=.10) both operating through staff referral making for sexual health services. Fit indices supported the hypothesized model (NFI = .98; RMSEA = .02), although the χ² remained significant due to large sample size.

Conclusions: The structural model revealed direct and indirect paths to HIV/STD testing. School staff referrals for sexual health services predicted testing, and youth experience of bullying may indirectly facilitate these referrals. Exposure to testing messages also contributed to testing behavior both directly and indirectly through staff referral-making.