Bayesian spatial-temporal modelling to assess the impact of climate variability and control interventions on the burden of malaria in Kenya

Malaria, one of the oldest and most persistent infectious diseases, continues to pose a significant public health challenge, particularly in sub-Saharan Africa (SSA), where it disproportionately affects children under 5 years of age. The risk of malaria in Kenya is heterogeneous; with western Kenya experiencing a high burden. Contributing factors include a favourable climate for mosquitoes, weak health systems, and socioeconomic challenges.
This research aimed to evaluate the influence of climatic, environmental, and non-climatic factors, alongside control interventions, on malaria incidence and mortality in Kenya. Using data from Kisumu's Health and Demographic Surveillance System (HDSS) (2008–2022) and Kenya Malaria Indicator Surveys (2015 and 2020), the study employed advanced statistical and geostatistical models. It analysed trends in malaria incidence, parasitemia prevalence, and mortality, considering factors such as temperature, rainfall, bed net use, socioeconomic status, and proximity to health facilities. Additionally, the study applied empirical dynamic modelling to establish causal links and forecast malaria transmission.
The findings enhance our understanding of malaria epidemiology and highlight the significant and varied effects of climatic factors on malaria transmission. Results underscore the protective role of bed nets, the influence of socioeconomic disparities, and the spatial and temporal heterogeneity of malaria risk. The work provides critical tools for mapping and targeting malaria control efforts, with implications for the National Malaria Control Program (NMCP). Additionally, the forecasting model offers actionable insights for localized, short- and long-term malaria prediction, supporting more effective resource allocation and intervention strategies.