Linear and Nonlinear Associations between Meteorological Variables and the Incidence of Malaria in Zahedan District Southwest of Iran 2000-2019

Introduction: Iran is one of the malaria-prone countries. Malaria transmission is likely to be affected by many factors, including meteorological variables. This study aimed to evaluate the effect of climate variables on malaria incidence. Methods: A secondary analysis was conducted to examine the relation between malaria and meteorological variables in Zahedan district from 2000 to 2019. We built univariate and multivariate Seasonal Autoregressive Integrated Moving Average (SARIMA) models and Generalized Additive Models (GAM)/ Generalized Additive Mixed Models (GAMM) using R software. AIC, BIC and residual tests were used to test the goodness of fit of SARIMA models, and R 2 was used to select the best model in GAM/GAMM. Results: The SARIMA multivariate (1,0,1) (0,1,1) 12 model, including the mean temperature and minimum humidity variables without lag, was the best fit. In nonlinear analysis, the number of malaria cases positively correlated with the month from January and peaked in May (edf=6.29). There was a generally negative correlation between malaria and time in years (edf=8.41). The mean temperature, between 20 to 30°C had the highest and slightly positive relation with the incidence of malaria (edf=7.55). Rainfall showed a negative association with small fluctuations between 20 and 45 mm and a positive association over 50 mm (edf=7.52). Mean relative humidity from above 50% had a negative relation with the number of cases (edf=6.93). The hours of sunshine in a month, until 235 hours, had a negative correlation and above 340 hours had a positive correlation with the incidence of malaria (edf=7.4). Conclusion: Meteorological variables can affect malaria occurrence.