Space–time mixture model of infant mortality in peninsular Malaysia from 1990 to 2000

Abstract

Disease mapping is a method used to display the geographical distribution of disease occurrence. Some traditional methods of classification for detection of high- or low-risk area such as traditional percentiles method and significant method have been used in disease mapping for map construction. However, as described by several authors, the classification based on these traditional methods has some disadvantages for describing the spatial distribution of the risk of the disease concerned. To overcome these limitations, an approach using space–time mixture model within an empirical Bayes framework is described in this chapter. The aim of this chapter is to investigate the geographical distribution of infant mortality in peninsular Malaysia from 1991 to 2000. The analysis showed that in the early 1990s the spatial heterogeneity effect was more prominent; however, toward the end of 1990s this pattern tends to disappear. Indirectly, this may indicate that the provisions of health services throughout peninsular Malaysia are uniformly distributed over the period of the study, particularly toward the year 2000.