Stochastic treatment regimes in climate-health research: Reassessing malaria risk under warming scenarios in Colombia.

Abstract

Malaria remains a significant public health challenge, mainly because climate change is altering transmission dynamics. This study investigates the relationship between rising temperatures and malaria cases across the 100 municipalities in Colombia with the highest incidence from 2007 to 2023. We employed causal machine learning techniques to analyze how incremental temperature impacts malaria incidence while controlling for valid confounding variables. Our findings reveal that with the currently observed temperature, malaria transmission intensifies with temperatures between 15 and approximately 23.5 °C but declines at higher temperatures, indicating an optimal range for transmission. Our results suggest an exposure-response relationship where higher temperature increases are associated with greater reductions in the probability of excess malaria cases. The Average Treatment Effect (ATE) on excess malaria cases for the evaluated temperature regimes showed a progressive decrease, from -0.007 when temperature increased by 0.5°C to -0.063 when temperatures were increased by 2.0°C, relative to current temperatures. These results suggest that further warming could constrain malaria transmission intensity in regions already experiencing high temperatures. This research underscores the importance of tailored public health strategies that consider local temperature profiles and socio-economic conditions in malaria control efforts.