Gharibzadeh S¹ (MSc), Mahjub H ^{* 1,2} (PhD), Moghimbeigi A^1,2 (PhD), Sadri GH1,² (PhD)

  ¹ Department of Biostatistics , School of Public Health ,

  Hamedan University of Medical Sciences, Hamedan, Iran

  ² Health Science Research Center, Hamedan University

  of medical Sciences, Hamedan, Iran

  Received: 18 Aug 2009, Accepted: 30 Dec 2009

  Abstract

Introduction: The aim of this study is to map acute flaccid paralysis in the provinces of Iran, using mixture distributions.

Methods: In this study, information about the new cases of acute flaccid paralysis which were provided by Management Center of Ministry of Health and Medical Education in 1385 (2006) were analyzed. Poisson Mixture distribution was used to find out the number of at risk groups. Number of components and parameters of each component were found by nonparametric maximum likelihood method using the C.A.MAN software. Bayes’ theorem and programs which are written in the R2.8.0 software were used to assign each province to a parameter. Finally the results were shown on the map using ArcGis9.2.

Results: The results show that the incidence of acute flaccid paralysis can be classified into four categories, each having their own specific risk and probability.

Conclusion: This study showed that Mixture models can be used to accurately map acute flaccid paralysis in Iran.

Key words: Diseases Mapping, Poisson Distribution , Bayes Theorem, Iran

  Hakim Research Journal 2010 12(4): 11- 18.

  * Corresponding Author: Health Sciences Research Center, D epartment of Biostatistics & Epidemiology, School of Public Health, Hamedan University of Medical Sciences, P.O.Box: 65175- 4171. Hamedan, Iran. Tel: +98- 811- 8260661
Fax: +98- 811- 8255301, Email : mahjub@umsha.ac.ir