Cost-effectiveness and cost-utility of COVID-19 vaccination in a South African privately insured population: a real-world claims-based analysis

Background: COVID-19 placed substantial strain on healthcare systems globally, including South Africa. While vaccine effectiveness is well established, context-specific evidence on the economic value of COVID-19 vaccination in African and real-world settings remains limited. This study evaluates the cost-effectiveness and cost-utility of COVID-19 vaccination within a South African privately insured population between 1 March 2020 and 30 December 2022 using health insurance claims data. Methods: A state-transition Markov model was developed in TreeAge Pro 2025 from a third-party payer perspective, comparing no vaccination, partial vaccination, and full vaccination. Disease progression was simulated over a 60-year lifetime horizon using two-week cycles to capture acute and long-term health outcomes. Model inputs, including COVID-19-related hospitalisations, mortality, healthcare utilisation, and costs, were derived from insurance claims data for 550,332 individuals enrolled in two large health insurance funds during the study period. Outcomes included COVID-19-related hospitalisations and deaths averted, life years (LYs) gained, and quality-adjusted life years (QALYs) gained. Costs and outcomes were discounted at 0% and 3% annually. Incremental cost-effectiveness and cost-utility ratios were calculated, and parameter uncertainty was assessed using probabilistic sensitivity analysis. Results: Both partial and full vaccination strategies were associated with fewer COVID-19-related hospitalisations and deaths and lower lifetime healthcare costs compared to non-vaccination. Discounted costs per individual were $51.57 for non-vaccination, $21.29 for partial vaccination, and $44.84 for full vaccination. Vaccination generated more modest gains in LYs and QALYs relative to non-vaccination reflecting the relatively young age profile of the insured population, yet even within this lower-risk cohort, vaccination generated measurable long-term health benefits alongside immediate cost savings. Both vaccination strategies were dominant, with negative incremental cost-effectiveness and cost-utility ratios. Probabilistic sensitivity analyses demonstrated high decision certainty, with nearly all simulations indicating cost savings alongside improved health outcomes. Conclusion: Using real-world insurance claims data, COVID-19 vaccination was cost-saving and associated with improved long-term health outcomes within a South African privately insured population. These findings indicate that funding vaccination during the pandemic represented an efficient allocation of healthcare resources within the insured sector. Caution is warranted in generalising these results beyond populations with different demographic and risk profiles.