The influence of school enrollment and alcohol outlets on sexual risk among rural South African young women Public Deposited

Downloadable Content

Download PDF
Last Modified
  • March 19, 2019
  • Rosenberg, Molly
    • Affiliation: Gillings School of Global Public Health, Department of Epidemiology
  • Young women in South Africa are at extremely high risk for sexually transmitted infections and teen pregnancy. The identification of new intervention targets is critical to reduce the burden of these outcomes, yet place-based structural determinants of sexual risk have not been previously explored in this population. In Aim 1, we estimated the effect of school enrollment on teen pregnancy using longitudinal census data for 15,457 South African young women aged 12-18 years. A Cox proportional hazard model was constructed to compare the hazard of pregnancy between school enrollees and dropouts. Our findings suggested that young women who remained in school were at lower risk for teen pregnancy [aHR (95% CI): 0.57 (0.50, 0.65)]. In Aim 2, we estimated the association between visits to alcohol outlets and sexual risk using cross-sectional data from a sample of 2,533 South African young women. We also tested for interaction by alcohol consumption. Visiting alcohol outlets was associated with having more sex partners [aOR, one versus zero partners (95% CI): 1.51 (1.21, 1.88)], more unprotected sex acts [aOR, one versus zero acts (95% CI): 2.28 (1.52, 3.42)], higher levels of transactional sex [aOR (95% CI): 1.63 (1.03, 2.59)], and HSV-2 infection [aOR (95% CI): 1.30 (0.88, 1.91)]. Generally, the dual combination of exposure to alcohol outlets and alcohol consumption yielded stronger associations with the sexual risk outcomes than anticipated given the associations observed with each risk factor alone. In Aim 3, we estimated the association between number of alcohol outlets per village and prevalent HSV-2 infection using cross-sectional data from a sample of 2,174 young women living across 24 villages in rural South Africa. We used generalized estimating equations with log links to account for the clustered nature of the data. Young women who lived in villages with more alcohol outlets were more likely to be infected with HSV-2 [PR (95% CI): 1.08 (1.01, 1.15)]. Overall, the findings from all three aims suggest that place-based exposures may be important determinants of sexual risk among young women in South Africa.
Date of publication
Resource type
Rights statement
  • In Copyright
  • Miller, William
  • Thirumurthy, Harsha
  • Emch, Michael
  • Pettifor, Audrey
  • Van Rie, Annelies
  • Doctor of Philosophy
Degree granting institution
  • University of North Carolina at Chapel Hill Graduate School
Graduation year
  • 2014
Place of publication
  • Chapel Hill, NC
  • There are no restrictions to this item.

This work has no parents.