Cost-effectiveness analysis of the Tu’Washindi na PrEP program in reducing HIV and IPV-related health burden among adolescent girls and young women in Siaya County, Kenya

Abstract

Background: Adolescent girls and young women (AGYW) in Siaya County, Kenya, face substantial risk of both HIV acquisition and intimate partner violence (IPV). IPV is associated with adverse mental health consequences, including depression, and reduced health-related quality of life. IPV may also increase HIV risk through sexual coercion, limited condom negotiation, and barriers to HIV prevention service use. Tu’Washindi na PrEP is a multilevel, community-based intervention developed to address these intersecting risks by strengthening communication, relationship power, and partner support for health decisions. Rigorous economic evaluation of integrated IPV and HIV prevention strategies is needed to inform intervention efficiency in high-burden settings.Methods: A Markov state-transition model with three-month cycles was developed to evaluate the costs and quality-adjusted life years associated with Tu’Washindi over a 20-year time horizon. The model included ten health states across three subgroups: HIV-negative and not on PrEP, HIV-negative and on PrEP, and HIV-positive. Each subgroup was further divided into three IPV severity states: No or Mild IPV, Moderate IPV, and Severe IPV. The model included state-specific utility weights, medical costs, and transition probabilities based on available literature. Tu’Washindi was assumed to increase PrEP uptake and adherence, reduce progression from the No or Mild IPV state to higher-severity states, and de-escalate IPV from higher- to lower-severity states. A payer perspective and three percent annual discount rate were applied to costs and quality-adjusted life years. Deterministic sensitivity analyses examined parameter uncertainty. Results: In the base case, Tu’Washindi produced an incremental cost of USD 61,001 and an incremental gain of 60 quality-adjusted life years (QALYs) over 20 years from the payer perspective. The resulting incremental cost-effectiveness ratio (ICER) was USD 1,018 per QALY gained. One-way sensitivity analyses indicated that relationship duration, intervention effect size, and program costs had the greatest influence on the ICER. Across sensitivity analyses, the ICER remained below the cost-effectiveness threshold of USD 1,508 per QALY, suggesting that the model conclusions were robust across plausible ranges of parameter uncertainty. Conclusion: Tu’Washindi may be cost-effective for reducing the health burden associated with HIV and IPV among AGYW in Siaya County. These findings support continued consideration of integrated HIV and violence prevention initiatives to strengthen the health, safety, and well-being of AGYW in settings with high HIV incidence.