Estimating war deaths with population-based survey data

Ziad Obermeyer, University of Washington
Chang Park, University of Washington
Emmanuela Gakidou, University of Washington

Accurate estimates of war deaths are crucial for political, military, and public health planning, as well as for purposes of national history and reconciliation. Existing methods of measuring war deaths, however, can be challenging to implement and suffer from a number of known biases. Analysis of population-based data from large household surveys has emerged as an accurate and practical way of measuring war deaths for at least three reasons: retrospective data are readily available for a wide range of countries, random population sampling avoids many major sources of bias, and standard statistical methods can be used to represent uncertainty. Building on previous work using the World Health Surveys to estimate violent war deaths, we propose new methods for using Demographic and Health Survey data to estimate both direct and indirect war mortality, and present selected results from this approach.