Exploratory analysis of the correlation between precipitation and storms and malaria cases from national surveillance in Mozambique.

BACKGROUND

Climate change is impacting the seasonal weather trends. Mozambique is one country that has been experiencing significant impacts of climate change. These impacts include changing weather patterns and more frequent and intense storms. Mozambique also has the 5th highest malaria prevalence globally. These storms have all impacted the efficiency of the Mozambique National Malaria Control Program to achieve their goals. The goal of our study was to determine the appropriate temporal lag between high precipitation events and malaria risk in Mozambique. This knowledge is imperative to understand when and how to best respond the impacts of these events.

METHODS

Malaria monthly case count data at the district level were provided by the Mozambique National Malaria Control Program. Precipitation data were obtained from NASA GIS DISC Earth Data and were spatially aggregated to the district and temporally aggregated to the month to match the malaria surveillance data. We investigated the correlation between precipitation and malaria-related values at the district level each month accounting for spatio-temporal autocorrelation. We used Pearson correlation coefficients to quantify the correlations. We used Poisson generalized linear mixed models to determine the relative risk of malaria associated with precipitation nationally and for each of the three regions in Mozambique.

RESULTS

We found evidence of strong spatial variation in malaria cases, malaria hospitalizations, and malaria deaths. There was evidence of a clear temporal relationship between precipitation and malaria cases throughout Mozambique. Nationally, there was an increased relative risk for malaria at an 8-week temporal lag. However, this varied when stratified by region. In the Northern Region there was an increased relative risk for malaria between 6-week and 8-week temporal lags; in the Central region there was an increased relative risk for malaria between 10-week and 12-week temporal lags; and in the Southern region there was an increased relative risk for malaria between 10-week and 13-week temporal lags.

CONCLUSION

With increasing storms and changing weather patterns due to climate change there is a need to understand the specifics of the association between elevated precipitation and malaria risk. We found important spatial heterogeneity in temporal lag times between precipitation events and malaria cases in Mozambique.