Title:
Malaria Incidence Modeling Based on Climatic and Vegetation Factors in the Commune of Zogbodomey, Southern Benin

Authors:
GBAI N. Innocent, ABDOULAYE Djafarou, HOUNZINME Sènadé Sylvie, TCHAOUSSI FOUSSENIN Ayouba, Benin

Abstract:

While it is true that climate affects human health, there are many reasons to believe that health also influences the epidemiology of malaria. Thus, in the current context of climate change, where we are witnessing an increase in the incidence of climate-dependent diseases, the implementation of operational tools such as early warning models appears to be of great use as a decision-making tool for controlling this disease.
 
The methodology adopted consisted of generating a linear regression model based on climatological and environmental parameters strongly correlated (at the 5% level) with the disease, after identifying the influence of these parameters on malaria. The model was then calibrated using backward elimination. Finally, validation was performed based on three statistical tests: the variance inflation factor test, the Durbin-Watson test, and the Goldef-Quandt test.
 
The results reveal that among the parameters most strongly correlated with malaria, only rainfall, vegetation index, maximum temperature, and minimum relative humidity proved to be the most relevant parameters of the model. The model itself proved quite reliable, with good predictive power and a fairly accurate reproduction of the disease's seasonality. Furthermore, it has a high correlation coefficient (r) of 87%, with a standardized standard deviation of 0.97, very close to the reference value of 1.

DOI: http://dx.doi.org/10.51505/ijaemr.2025.1528