Patterns of gendered risk factors and associations with intimate partner violence and low educational attainment among adolescent girls and young women in Lesotho: A latent class analysis.

Background: Globally, adolescent girls and young women (AGYW) are disproportionately impacted by economic, demographic, and social factors associated with a wide range of negative outcomes.

Objective: The objective of this study was to use latent class analysis (LCA) to identify groupings of AGYW in Lesotho based on patterns of gendered risk factors, and to assess the association between the identified groupings and intimate partner violence (IPV) and low educational attainment.

Participants And Setting: Data were from the 2018 Lesotho Violence Against Children and Youth Survey. AGYW reported gendered risk factors: teen pregnancy, child marriage, intergenerational sex, early sexual debut, being HIV positive, transactional sex, endorsement of one or more negative traditional gender norms, and one or more norms supportive of violence against women.

Methods: LCA identified latent classes of eight gendered risk factors. Multivariable logistic regression assessed associations between latent classes and IPV victimization and low educational attainment.

Results: A three-class solution was selected, and classes were named as: Low Risk class, Behavioral Risk class, and Attitudinal Risk class. Odds of low educational attainment and IPV were higher in the Attitudinal Risk class than the Low Risk class. Odds of low educational attainment and IPV were higher in the Behavioral Risk class than the Low Risk class and the Attitudinal Risk class.

Conclusions: In Lesotho, gendered risk factors form distinct classes that have variable associations with low educational attainment and IPV. LCA can be an important approach to better understand the complicated relationship gendered risk factors have with each other and with certain outcomes, to further elucidate the influence that gender has on the health of AGYW and to provide more targeted prevention programming.