Abstract
PURPOSE: To assess the relationship between school environment and health and behavior outcomes. METHODS: Data are from baseline and first follow-up of the HIV Prevention Trials Network (HPTN) 068 longitudinal trial established in 2012 of adolescent girls and young women in rural Mpumalanga Province, South Africa. Data from 2212 participants are included. We measured the association between four school environment domains: school resources, school safety, negative personal experiences, and school connectedness, and several health and behavior outcomes: depressive symptoms, low attendance, recent pregnancy, recent unprotected sex, transactional sex, and having an older romantic partner. We used a g-computation approach to estimate risk differences (RD) for the longitudinal relationship between the school environment (measured at the individual and school level) on individual health and behavior outcomes, controlling for baseline covariates. RESULTS: The mean age of participants at baseline was 15.4; mean age at first follow-up was 16.6. Individual baseline perceptions of an unsafe school environment (RD = 3.1 %, 95 % CI: 1.3-5.2 %) and more frequent negative experiences (RD = 4.0 %, 95 % CI: 2.0-5.9 %) were associated with higher absolute risk of depressive symptoms at follow-up. There was an overall trend toward higher risk of pregnancy, unprotected sex, and having an older partner among those who reported fewer school resources, lack of school safety, more negative personal experiences, and lack of school connectedness. CONCLUSIONS: Our findings provide evidence of an overall trend toward higher risk of depression, pregnancy, unprotected sex, and having an older partner among those reporting a worse school environment across four school environment domains.