Bayesian model for work-related accidents in Iran: 2009

• Main Authors: M Salehi, M Imani, F Zayeri, N Vahabi, H Pirhosseini, M Arji
• Format: Article
• Language: fas
• Published: Iran University of Medical Sciences 2013-04-01
• Series: مدیریت سلامت
• Subjects: Work-related Accidents; Empirical Bayes; Relative Risk; Poisson-Gamma Modelin
• Online Access: http://jha.iums.ac.ir/article-1-1142-en.html

Introduction: It is of prime importance to consider the pattern and geographical changes of a disease, in each community independently, to determine high and low risk areas. Mapping diseases is a set of statistical methods which attempt to provide precise maps by which the geographical distribution of a disease is estimated. In this study, Bayesian methods were applied to estimate the relative death rate of work-related accidents in Iran.   Methods: For the purpose of this study, the data of work-related accidents of Iran in 2009 were analyzed. To estimate the parameters of the map, empirical Bayes method (Poisson-Gamma method) was applied using Winbugs 1.4 software. Moreover, the Arc GIs 9.2 software was used to set relative incidence of death and accident related maps.   Results: Regarding the estimates achieved by empirical Bayes method and applying Poisson-Gamma for the incidence of work related accidents in 2009, the maximum and minimum prevalence risk rate among men was 2.991 in Markazi province and 0.457 in Khorasan Razavi province, while they were 3.848 in Semnan province and 0.243 in Hormozgan province for women. Conclusion: Overall, the incidence of work-related accidents follows no specific geographical distribution pattern and in most provinces the pattern was different for men and women in Iran. By and large, the incidence of these events in the neighboring provinces of Tehran is more than the other parts of the country.