School attendance, sexual risk, and pregnancy among adolescent girls and young women enrolled in a conditional cash transfer feasibility study in Kenya

Abstract

Objective To evaluate feasibility of administering a Cash Transfer (CT) program conditional to school attendance and assess effect on school retention, self-reported sexual risk taking, HIV incidence and unintended pregnancy.

Methods After informed consent 50 HIV-negative schoolgirls were enrolled and followed over a 12-month period. Each participant and her guardian received 4 cash disbursements of USD 36 (total $72) conditional to the girl achieving 80% school attendance in the preceding quarter verified against the school register. Computer aided personal interview (CAPI) was used to monitor self-reported sexual risk taking at each scheduled study visit. HIV and pregnancy testing were used to ascertain eligibility and as biomarkers of sexual risk-taking at study exit.

Results: The mean age was 16 years, and all girls achieved 80% school-attendance. Highest attendance was in first quarter but progressively declined in subsequent visits (p<0.001) citing inability to pay school levies. There was a discrepancy between self-reports and school registers attendance records (kappa=0.22). At enrolment 30 girls had achieved sexual debut, over half (n= 16) at age < 15 years. Dating at point of recruitment was a marker of girls at increased risk of missing school (RR= 2.25; 95% CI: 1.13 – 4.47). During follow-up there was significant decline in current sexual activity (p < 0.0001), transactional sex (p=0.02) and engaging with older sexual partner (p < 0.003). There were seven incident pregnancies and one HIV seroconversion.

Conclusion: Cash transfer conditional to school attendance verified against school records is feasible and a potentially viable structural intervention against school-girl sexual-risk taking and keep girls in school.