Estimating Prevalence of Post-war Health Disorders Using Capture-recapture Data

TL;DR

This paper introduces a novel statistical method using a trivariate Bernoulli model and Monte Carlo EM algorithm to accurately estimate the prevalence of health disorders among post-war populations, addressing under-reporting issues.

Contribution

It develops a new modeling approach for complex population estimation problems, improving accuracy and robustness over existing methods in post-war health surveillance.

Findings

Proposed method outperforms existing estimators in simulations.

Robustness demonstrated under model mis-specifications.

Applied to Gulf War and 9/11 data, providing valuable health insights.

Abstract

Effective surveillance on the long-term public health impact due to war and terrorist attacks remain limited. Such health issues are commonly under-reported, specifically for a large group of individuals. For this purpose, efficient estimation of the size of the population under the risk of physical and mental health hazards is of utmost necessity. In this context, multiple system estimation is a potential strategy that has recently been applied to quantify under-reported events allowing heterogeneity among the individuals and dependence between the sources of information. To model such complex phenomena, a novel trivariate Bernoulli model is developed, and an estimation methodology using Monte Carlo based EM algorithm is proposed. Simulation results show superiority of the performance of the proposed method over existing competitors and robustness under model mis-specifications. The method is applied to analyze real case studies on the Gulf War and 9/11 Terrorist Attack at World Trade Center, US. The results provide interesting insights that can assist in effective decision making and policy formulation for monitoring the health status of post-war survivors.