Climatic Variables and Malaria Morbidity in Mutale Local Municipality, South Africa: A 19-Year Data Analysis.

The north-eastern parts of South Africa, comprising the Limpopo Province, have recorded a sudden rise in the rate of malaria morbidity and mortality in the 2017 malaria season. The epidemiological profiles of malaria, as well as other vector-borne diseases, are strongly associated with climate and environmental conditions. A retrospective understanding of the relationship between climate and the occurrence of malaria may provide insight into the dynamics of the disease's transmission and its persistence in the north-eastern region. In this paper, the association between climatic variables and the occurrence of malaria was studied in the Mutale local municipality in South Africa over a period of 19-year. Time series analysis was conducted on monthly climatic variables and monthly malaria cases in the Mutale municipality for the period of 1998-2017. Spearman correlation analysis was performed and the Seasonal Autoregressive Integrated Moving Average (SARIMA) model was developed. Microsoft Excel was used for data cleaning, and statistical software R was used to analyse the data and develop the model. Results show that both climatic variables' and malaria cases' time series exhibited seasonal patterns, showing a number of peaks and fluctuations. Spearman correlation analysis indicated that monthly total rainfall, mean minimum temperature, mean maximum temperature, mean average temperature, and mean relative humidity were significantly and positively correlated with monthly malaria cases in the study area. Regression analysis showed that monthly total rainfall and monthly mean minimum temperature ( R ² = 0.65), at a two-month lagged effect, are the most significant climatic predictors of malaria transmission in Mutale local municipality. A SARIMA (2,1,2) (1,1,1) model fitted with only malaria cases has a prediction performance of about 51%, and the SARIMAX (2,1,2) (1,1,1) model with climatic variables as exogenous factors has a prediction performance of about 72% in malaria cases. The model gives a close comparison between the predicted and observed number of malaria cases, hence indicating that the model provides an acceptable fit to predict the number of malaria cases in the municipality. To sum up, the association between the climatic variables and malaria cases provides clues to better understand the dynamics of malaria transmission. The lagged effect detected in this study can help in adequate planning for malaria intervention.