Fifty years of violent war deaths from Vietnam to Bosnia: analysis of data from the world health survey programme

Abstract

War causes more deaths than previously estimated, according by researchers at IHME and Harvard Medical School. The study, Fifty years of violent war deaths from Vietnam to Bosnia: analysis of data from the world health survey programme, proposed a new approach to estimating violent war deaths using household survey data and passive reports.

Research findings

The researchers found that from 1955 to 2002, an estimated 5.4 million war deaths occurred in 13 countries, of which 3.8 million occurred in Vietnam. From 1995 to 2002, survey data indicate 36,000 war deaths annually in these 13 countries.  This finding contradicts passive surveillance data, such as media reports, which report only one-third of this total, suggesting that nationally representative surveys yield estimates of war deaths that are less optimistic than those based on passive surveillance. The researchers used a new method based on survey data and passive surveillance data to more realistically estimate 378,000 global war deaths annually from 1985 to 1994.

Analytical approach

Researchers conducted an analysis of 2002 to 2003 nationally-representative world health surveys that included sibling reports of mortality and whether such deaths resulted from war injuries. They adjusted for sampling bias and censoring and corrected for known problems with mortality surveys to produce estimates of deaths in 13 countries over the past 50 years. Researchers then compared their estimates with estimates in an existing database of passive reports.

Research objective

There exist a number of challenges to estimating war mortality, including the fact that many wars occur in places without pre-existing surveillance infrastructure and that civil registration systems that do exist during peacetime fall apart during war. The study researchers thus utilized a novel methodology to provide an accurate estimate of violent war deaths. IHME is dedicated to developing new methodological tools for estimating population-level health when existing methods are inadequate.

Recommendations for future work

The results of this study highlight the need to re-evaluate claims made on the basis of current data from media reports, which state that the number of deaths related to war has declined consistently since the mid-20^th century. Although the methods presented in this paper help present more realistic estimates of war mortality, more resources are urgently needed for empirical measurement of war deaths. Additionally, future surveys can reduce uncertainty around estimates from sibling histories by including larger sample sizes, gathering more structured sibling histories, and using multiple nonsibling respondents per household. Accurate estimates of war mortality are crucial for political, military, and public health planning, as well as for purposes of national history and reconciliation.