Seasonality and spatial heterogeneity of confirmed malaria in Senegal (2009-2025): consistency with VECTRI simulations driven by ten bias-adjusted CMIP6 models and projected changes
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background
Malaria transmission in Senegal is seasonal and spatially heterogeneous, but the consistency of dynamical malaria models with routine surveillance has rarely been tested with monthly subnational data.
Methods
We analysed monthly confirmed cases from the National Malaria Control Programme (PNLP) regional database (14 regions, 2009-2025) and reports of the 79 health districts (2016-2025). Ten bias-adjusted NEX-GDDP-CMIP6 models (1995-2014; SSP2-4.5 and SSP5-8.5 to 2100) were evaluated against CHIRPS rainfall; VECTRI was driven by MRI-ESM2-0, the model closest to the ensemble median change.
Results
National incidence averaged 24.5 cases per 1,000 per year without trend over 2009-2024, but increased in Kédougou, Kolda and Tambacounda, which reported 58.7 % of cases for 11.3 % of the population. The test positivity rate (TPR) decreased from 22.5 % in 2016 to 16.2 % in 2024. Based on the seasonal centre of gravity, observed cases lagged the entomological inoculation rate (EIR) simulated by VECTRI by 1.06 months (95 % CI 0.88-1.23) and the lag was positive in 13 of 14 regions, but the regional differences in timing were not reproduced. Across 45 departments, the simulated EIR was correlated with the TPR (Spearman ρ = 0.67, 95 % CI 0.47-0.79). In 2025, confirmed cases decreased by 55.7 % in the 34 districts of the long-lasting insecticidal net (LLIN) mass campaign against 9.7 % elsewhere. Under SSP5-8.5 in 2081-2100, the models projected +4.6 °C and -32 % annual rainfall, and the simulated EIR decreased by 35 % (p < 0.001).
Conclusions
VECTRI driven by bias-adjusted CMIP6 climate reproduces the timing of the malaria season and the gradient of transmission intensity in Senegal, but not the regional differences in timing. The TPR is a population-free indicator for model evaluation. Seasonal malaria chemoprevention calendars should account for the regional timing documented here; the projected changes in timing are small.