Non-linear age dynamics of malaria infection and fine-scale environmental exposure in rural Uganda
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background
Age-specific patterns of malaria are well-established for children aged < 5 years. Less understood is the epidemiology of malaria in older children and adults, and the influence of granular environmental risk.
Methods
We analyzed data from SchistoTrack, a community-based cohort in rural Uganda. We studied 4308 participants aged 5 to 90 years from 52 villages across three lakeside districts of Mayuge, Buliisa, and Pakwach, with enrollment between January 2022 to February 2024. The primary outcome was malaria infection status by rapid diagnostic test (RDT). Secondary outcomes included microscopy-confirmed infection with parasite density quantification and self-reported fever within the past month. We fitted a generalized additive mixed model (GAMM) with adaptive age smoothing, adjusting for sociodemographic factors, household characteristics, healthcare access, and environmental exposures. Environmental exposure was quantified using the Normalized Difference Vegetation Index (NDVI) derived from Sentinel-2 satellite imagery (10 m resolution), processed through hexagonal aggregation with Gaussian neighborhood smoothing and validated against field malacology surveys and participatory community mapping.
Results
Overall RDT prevalence was 41.2% (1776/4308), with microscopy prevalence at 32.3% (1363/4219), which was predominantly Plasmodium falciparum (83.1%; 1133/1363). Most infections were low-density ( < 999 parasites/µL; 71.6%; 976/1363). Malaria prevalence showed non-linear age patterns, peaking at 10 to 11 years then declining through adolescence before stabilizing in adulthood. Among RDT-positive individuals, fever prevalence decreased with age from 30.8% in children (aged 5 to 10 years) to 11.2% in adults (aged ≥ 20 years). Dense vegetation (per unit NDVI increase: Odds Ratio (OR) 3.25, 95% Confidence Interval (CI) 1.33–7.96) and greater distance from government health centers (per log-km: OR 1.87, 95% CI 1.34–2.59) increased the odds of infection. Proximity to vegetated water bodies increased the odds of infection compared to beaches: ponds/swamps (OR 1.65, 95% CI 1.19–2.28), river/river marsh (OR 1.63, 95% CI 1.16–2.31), lake marsh (OR 1.40, 95% CI 1.07–1.83).
Conclusion
Malaria prevalence remains high in older children and adults, though with fewer febrile cases, and is influenced by the local environment. Our findings support age-specific interventions targeting school-aged children while maintaining adult surveillance, and using validated environmental indices to guide sub-district resource allocation in high-risk areas.
