Optimizing Age-Structured Sampling for Estimating the Seroconversion Rate in Malaria Seroepidemiology: A Simulation Study

Introduction Seroprevalence and the seroconversion rate (SCR) are valuable indicators of malaria burden, particularly in low-transmission settings. One potential strategy to improve the efficiency and precision of SCR estimation is to optimize age-structured sampling. For other infectious diseases such as measles, mumps, and rubella, optimal age allocation has been shown to vary depending on the epidemiological parameter of interest. However, such an approach has not yet been explored for malaria seroprevalence. Methods We employed Monte Carlo simulations to identify optimal age-based sampling strategies under varying assumptions by modifying 1) SCR, 2) seroreversion rate (SRR), and 3) whether SRR was known or unknown. Using the reverse catalytic model and its extension, we considered two transmission scenarios: a stable SCR and a reduction in SCR at a defined change point. Realistic ranges for SCR and SRR were selected based on previous reports, assuming serological responses to Plasmodium falciparum merozoite surface protein 1 (MSP1). Results In stable transmission settings, sampling older groups improved SCR estimates in low-transmission settings, whereas young children were more informative in high-transmission settings when SRR was known. When SRR was unknown, sampling a greater proportion of younger children yielded the highest precision. In particular, under conditions of low SRR and when SCR and SRR were jointly estimated, Q1 age allocations improved the relative precision compared to Q5 by 2.3-fold and 9.8-fold in low- and high-transmission settings, respectively. Under a changing transmission scenario with known SRR, SCR estimation was most precise when at least 40% of samples were drawn from individuals born after the change point. When SRR was unknown, a more balanced age distribution performed better, with 20–60% of samples from children born after the change point and increased sampling from older age groups. Conclusion Careful consideration of sampling design is critical for enhancing study efficiency, particularly in settings with limited personnel or financial resources. When estimating SCR from a single cross-sectional malaria serological survey, particular attention should be paid to the sampling strategy across age groups to ensure acceptable levels of precision.