Application of a Statistical Model to Forecast Drowning Deaths in Iran

• Main Authors: Mohammad Reza Omidi, Meysam Jafari Eskandari, Sadigh Raissi, Amir Abbas Shojaei
• Format: Article
• Language: English
• Published: Negah Institute for Scientific Communication 2019-07-01
• Series: Health in Emergencies & Disasters Quarterly
• Subjects: accidents; disasters; drowning
• Online Access: http://hdq.uswr.ac.ir/article-1-249-en.html
• ISSN: 2345-4210; 2345-4210

Background: One of the indicators for measuring the development of a country is its death rate caused by accidents and disasters. Every year, many people in Iran are drowned for various reasons. This study aimed to predict the trend of drowning mortality in Iran using statistical models. Materials and Methods: This research was a longitudinal study using time-series data of drowning deaths obtained from the Iranian Legal Medicine Organization during 2005-2017. The Autoregressive Integrated Moving Average (ARIMA) model was used for forecasting, which is based on the Box-Jenkins method consisting of the Autoregressive (AR) model, Moving Average (MA) model, and Autoregressive Moving Average (ARMA) model. The obtained data were analyzed in ITSM software. Results: A total of 14127 people have died due to drowning in Iran, during 2005-2017, with an average death toll of 1086 people per year. In 2017, the highest number of deaths caused by drowning was recorded in Khuzestan Province (n=161) and the lowest number in South Khorasan Province (n=1). Estimates of the drowning trend indicated that the number of drowning deaths in Iran would continue to decline in the coming years. Conclusion: The high accuracy of prediction using the Box-Jenkins method indicates its effectiveness for experts and managers to predict drowning death rates.