The post-war public health effects of civil conflict
Introduction
Civil wars kill people directly and immediately. They also destroy property, disrupt economic activity, and divert resources from health care. Huge refugee flows put people into crowded conditions without access to clean water and food; refugees become trans-border vectors of infection. Crime and homicide rates rise in wars and may remain high afterwards in a culture accustomed to violence. Many of these effects last for years after the fighting. A recent article in this journal reviewed the public health literature on both immediate and long-term effects (Pedersen, 2002; also see Krug, Dahlberg, Mercy, Zwi, & Lozano, 2002, Chapter 8). The health effects during specific civil wars are relatively well known1, but the general and longer-term impact is not. These new deaths (and disabilities) are overwhelmingly concentrated in the civilian population. For example, Davis and Kuritsky (2002) report that severe military conflict in sub-Saharan Africa cut life expectancy by more than 2 years and raised infant mortality by 12 per thousand.
Most reports on the consequences of civil wars for civilians are case studies. This is an important limitation, since the severity of war for civilians both in the immediate and longer-term varies considerably across wars, as do the specific diseases and conditions by which the impact occurs. Here we offer a comprehensive worldwide analysis of the post-conflict consequences for healthy life expectancy. We modify a previously developed statistical model (Ghobarah, Huth, & Russett (2003), Ghobarah, Huth, & Russett (2004)) to explain World Health Organization data which take into account both years of life lost to disease and injury and years of healthy life lost to long-term disability. They allow us to assess the effect of civil wars on the subsequent impact of 23 major diseases and conditions, on segments of the population distinguished by gender and five age groups.
The model here improves on our previous analyses in several important respects discussed below: (1) deletion of two independent variables—ethnic heterogeneity and type of government—that have a strong impact on health spending but little further direct impact on health outcomes; (2) improved data on civil war casualties; (3) a more sensitive term for the effect of civil wars in neighboring countries, measuring the interaction of the severity of such wars with the permeability of the borders; (4) substituting the rate of female educational achievement for that of both sexes combined. These improvements produce much clearer and stronger effects.
Overall, WHO (2000, pp. 168, 174) estimates that 269,000 people died and 8.44 million years of healthy life were lost to death and disabilities in 1999 as direct and immediate effects of all wars—civil and international. Based on the results we report here, perhaps another 15 million lives were lost in 1999 to death and disability indirectly, from various diseases in war-torn countries and their neighbors, from the lingering effects of civil wars during the years 1991–1997. Those outcomes in return affect politics (e.g., Price-Smith, 2002, Chapter 2).
Section snippets
New measures of public health
WHO's measure of overall health achievement, Health Adjusted Life Expectancy (HALE), discounts the total life expectancy at birth in each country by the number of years the average individual spends with a major disability as the burden of disease or injury—the gap between total life expectation and expected years without disability. It is estimated from three kinds of information: the fraction of the population surviving to each age level (calculated from birth and death rates),
Theoretical framework
In developing hypotheses about the longer-term effects of civil war on public health we draw on a general theoretical framework for studying the factors affecting public health. We identify four major influences on public health in societies, and contend that political conditions and processes are important causes of each of these major influences on health.
(1) The extent to which populations are exposed to conditions that increase the risk of death, disease, and disability. At the most basic
Hypotheses on civil war and public health
This outline of important general causal connections between politics and public health allows us to focus specifically on the linkages between civil war and long-term health. Civil wars directly affect all the major contributors to health: exposure to disease, medical care, public health interventions, and overall socio-economic conditions. We posit two related hypotheses for how civil wars produce health consequences that extend well into the post-war period: H1 More DALYs are lost with the
Controlling for other causes of public health
While our primary focus is on the impact of civil wars on public health, we must control for other factors that public health scholars and health economists argue are important causes of cross-national variation. H3 The higher the level of total health expenditures, the fewer DALYs lost.
Higher income improves health through public and private decisions to spend money on hospitals, preventive and curative health care, sanitation, and nutrition. Earlier work by economists (e.g., Pritchett & Summers,
A multivariate analysis of all deaths and disabilities
We test these hypotheses using cross-sectional least squares regression analysis. Missing data for some countries on some variables reduces the number of observations to 165, with most of the missing being mini-states. Table 2 shows ten equations for deaths and disabilities from all causes combined, with breakdowns by age and gender. The explanatory variables are listed across the top, and each column gives the estimated coefficient and the t-ratio. Coefficients and t-ratios which reach the
The who and how of civil war effects
We can now look at the impact of civil wars through specific diseases and conditions. The WHO data on impacts of various diseases by age and gender allow us to compute 210 equations8
Contiguous civil wars
Table 4 shows the effect of civil war in a contiguous country, above any effect of civil war at home. Our initial analysis found that having a civil war in an adjacent country was itself a major contributor to loss of healthy life expectancy overall. In the disease-specific analysis we find about as many (66) disease-age-gender groups for which a contiguous civil war significantly increased death and disability. This too is far above the 10 or 11 categories we would expect by chance to cross
Conclusion
We developed the central hypothesis that civil wars produce long-term damage to public health and medical systems that extend well beyond the period of active warfare, and tested it in the context of a more general political-economic model of conditions affecting death and disability cross-nationally. Using newly available data on disability-adjusted life years lost from various diseases and conditions by age and gender groups, we found that, controlling for the other influences, civil wars
Acknowledgements
We thank the Weatherhead Initiative on Military Conflict as a Public Health Problem, the Ford Foundation, and the World Health Organization for financial support, Gary King, Thomas Gariepy, Melvin Hinich, Kosuke Imai, Jennifer Leaning, Roy Licklider, Lisa Martin, Christopher Murray, Joshua Salomon, Nicholas Sambanis, and William D. White for comments, and Sarah Croco for research assistance.
References (68)
- et al.
The impact of public spending on healthDoes money matter?
Social Science & Medicine
(1999) Political violence, ethnic conflict, and contemporary warsBroad implications for health and well-being
Social Science & Medicine
(2002)Military expenditures, health, and education
Armed Forces and Society
(1996)- et al.
Violent acts and violent timesA comparative approach to postwar homicide rates
American Sociological Review
(1976) Security and economy in the third world
(1988)- Bates, R. H. (1999). Ethnicity, capital formation, and conflict (Working Paper No. 27). Cambridge, MA: Harvard...
- et al.
Rethinking the trauma of war
(1998) - et al.
Public health and war in Central America
- et al.
An institutional explanation of the democratic peace
American Political Science Review
(1999) - Centers for Disease Control and Prevention. (1992). Famine-affected, refugee, and displaced populations:...