Identification of malaria hotspots in southwestern Benin through spatial joint modelling of malaria incidence and vector abundance

Monteiro, Gabriel Michel and Aïkpon, Rock Yves and Dandonougbo, Codjo and Sedda, Luigi and Djogbenou, Luc Salako (2025) Identification of malaria hotspots in southwestern Benin through spatial joint modelling of malaria incidence and vector abundance. Other. Verixiv.

Abstract

Background Identifying the spatial heterogeneity in malaria transmission is crucial for designing geographically targeted control interventions, especially in high-burden communities where hotspot identification and delineation can facilitate the decision-making process toward resource allocation to specific areas where they are most needed. This study is the first attempt to identify malaria hotspots by jointly modelling vector abundance and human malaria incidence, alongside key ecological drivers, providing new insights into entomological and epidemiological synergies for public health management. Methods We applied a Bayesian Framework for Joint Gaussian Spatial Processes to log-transformed Anopheles gambiae s.l. and Anopheles funestus counts, and malaria incidence in eight communes of southwest Benin. Entomological data came from mosquito surveillance activities and routine malaria incidence data from the District Health Information System 2. Malaria hotspots were delineated by computing a joint risk surface derived from interpolated predictive surfaces of malaria incidence and vectors abundance. Co-regionalization analysis explored local spatial correlations between malaria incidence and each mosquito vector suitability. Results Joint risk modelling identified contiguous malaria hotspots located mainly on Lake Ahémé shores, and around Hillacondji, Agoue and Grand-Popo. Four ecological factors emerged as consistent and key drivers for all three processes: wind speed, mid-infrared reflectance, leaf area index and land surface temperature. Contrary to common assumptions, An. funestus showed stronger spatial correlation with malaria incidence across 165.3 km2 compared to An. gambiae s.l. (96.7 km2), with 85.2 km2 showing synergistic effects of both species. Conclusion This study reveals high heterogeneity in the spatial association between malaria and its primary vector species, with An. funestus playing a potential prominent role than previously recognized. Our framework offers a useful insight of the distinct ecological preferences of each malaria vector species, highlights the need for species-agnostic, and spatially targeted interventions informed by entomological and epidemiological data until universal vaccines become widely available.