The Effect of Climate Variables on the Incidence of Cutaneous Leishmaniasis in Isfahan, Central Iran

Abstract

BACKGROUND AND AIM: In recent years, there have been considerable changes in the distribution of diseases that are potentially tied to ongoing climate variability. The aim of this study was to investigate the association between incidence of cutaneous leishmaniasis (CL) and climatic factors in an Iranian city (Isfahan) which had the highest incidence of CL in the country. METHODS: CL incidence and climate data were inquired from April 2010 to March 2017 (108 months) for Isfahan city. Univariate and multivariate Seasonal Autoregressive Integrated Moving Average (SARIMA), Generalized Additive Models (GAM) and Generalized Additive Mixed Models (GAMM) were used to identify the association between CL cases and climate variables, and forecast CL incidence. AIC, BIC and residual tests were used to test the goodness of fit of SARIMA models; and R2 was used for GAM/GAMM. RESULTS:5297 CL cases were recorded during this time. The incidence had a seasonal pattern and the highest number of cases were recorded from August to November. In univariate SARIMA, (1,0,1) (0,1,1)12 was the best fit for predicting CL incidence (AIC=8.09, BIC=8.32). Multivariate time series regression (1,0,1)(0,1,1)12 showed that monthly mean humidity after 3 months lag was inversely related to CL incidence (β=-1.59, p=0.0072, AIC=8.52, BIC=8.66). GAMM results showed average temperature with 2-month lag, average relative humidity with 4-month lag, monthly cumulative rainfall with 1-month lag and monthly sunshine hours with 1-month lag were related to CL incidence (R2=0.94). CONCLUSIONS:The impact of climate variables on the incidence of Leishmania is not linear and GAM models that include non-linear structures are a better fit for prediction. In Isfahan, Iran, climate variables can greatly predict the incidence of CL and these variables can be used for predicting outbreaks. KEYWORDS: Time-series analysis; SARIMA; Leishmaniasis, Iran