Scorpion stings pose a significant public health concern in Iran, resulting in approximately 45,000-50,000 cases and 19 deaths annually. The Khuzestan and Hormozgan provinces have the highest reported incidence rates, with an estimated 36,000 cases each year. This study focused on modeling the time series data of scorpion stings, specifically in Shoushtar City, from 2017 to 2022. Our objective was to investigate the presence of seasonality and long-term trends in the incidence of scorpion stings by utilizing advanced analytical techniques, such as the Autoregressive Integrated Moving Average (ARIMA) model. We applied the seasonal ARIMA model to fit a univariate time series of scorpion sting incidence. This study revealed a significant seasonal trend and an overall increase and decrease in scorpion sting cases during the study period. The best-fitting model for the available data was a seasonal ARIMA model in the form of ARIMA(0,0,1)(1,1,1)12. This model can forecast the frequency of scorpion sting cases in Southwestern Iran over the next two years. As a result, time series analysis can provide valuable insights into the patterns and trends of scorpion sting incidents, allowing for better planning and allocation of healthcare resources. By understanding the seasonal variations, proactive measures can be implemented to address the growing issue of scorpion stings in Iran effectively.
Keywords: ARIMA modeling; Box Jenkins model; Scorpion stings; Southwestern Iran; Time series analysis.