Estimating markov transition probabilities between health states in the social security Malaysia (SOCSO) dataset

Abstract

Occupational injury represents a considerable part of injury burden to the society as it may affect workers in their most productive years. The objective of this paper is to estimate the Markov transition probabilities of a worker’s health states over time using the Counting Method (CM) and the Proportional Odds Model (POM), focusing on disability among the Social Security Organization (SOCSO) contributors in Malaysia. Four health states namely active/work (A), temporary disability (T), permanent disability (P) and death (D) are considered, where the transition probabilities are estimated at yearly intervals based on age, gender, year and disability category.