Impact of Climate Variability and Interventions on Malaria Incidence and Forecasting in Burkina Faso.

BACKGROUND

Malaria remains a climate-driven public health issue in Burkina Faso, yet the interactions between climatic factors and malaria interventions across different zones are not well understood. This study estimates time delays in the effects of climatic factors on malaria incidence, develops forecasting models, and assesses their short-term forecasting performance across three distinct climatic zones: the Sahelian zone (hot/arid), the Sudano-Sahelian zone (moderate temperatures/rainfall); and the Sudanian zone (cooler/wet).

METHODS

Monthly confirmed malaria cases of children under five during the period 2015-2021 were analyzed using Bayesian generalized autoregressive moving average negative binomial models. The predictors included land surface temperature (LST), rainfall, the coverage of insecticide-treated net (ITN) use, and the coverage of artemisinin-based combination therapies (ACTs). Bayesian variable selection was used to identify the time delays between climatic suitability and malaria incidence. Wavelet analysis was conducted to understand better how fluctuations in climatic factors across different time scales and climatic zones affect malaria transmission dynamics.

RESULTS

Malaria incidence averaged 9.92 cases per 1000 persons per month from 2015 to 2021, with peak incidences in July and October in the cooler/wet zone and October in the other zones. Periodicities at 6-month and 12-month intervals were identified in malaria incidence and LST and at 12 months for rainfall from 2015 to 2021 in all climatic zones. Varying lag times in the effects of climatic factors were identified across the zones. The highest predictive power was observed at lead times of 3 months in the cooler/wet zone, followed by 2 months in the hot/arid and moderate zones. Forecasting accuracy, measured by the mean absolute percentage error (MAPE), varied across the zones: 28% in the cooler/wet zone, 53% in the moderate zone, and 45% in the hot/arid zone. ITNs were not statistically important in the hot/arid zone, while ACTs were not in the cooler/wet and moderate zones.

CONCLUSIONS

The interaction between climatic factors and interventions varied across zones, with the best forecasting performance in the cooler/wet zone. Zone-specific intervention planning and model development adjustments are essential for more efficient early-warning systems.