During the final decades of the 20th century, armed conflicts occurred in 6 of the 8 countries of the Central American isthmus: El Salvador, Guatemala, Honduras, Nicaragua, Panama, and the state of Chiapas, Mexico.
These conflicts, which could be both prolonged (13 years in El Salvador) and bloody (at least 200,000 deaths and disappearances in Guatemala1), arose in settings of great poverty. The principal preconflict epidemiologic enterprise of the region was the description of poverty-related health issues, particularly childhood malnutrition and its associations with landlessness and infectious disease.2–10
Preconflict poverty and hunger were compounded by inequality. In the 1980s, the ratio of per capita GDP between the extreme quintiles was estimated to be 13 in Nicaragua and 30 in Guatemala.11 The postwar Guatemalan Truth Commission specifically cited inequities in land distribution as a cause of war.1
With the exception of the still-unresolved situation in Chiapas, the various conflicts have ended either militarily or diplomatically. Yet poverty, inequality, and violence continue to plague Central America. The Pan American Health Organization currently identifies improvement in health equity as a main objective; a consortium of Central American human rights organizations has once again identified worsening inequality and violence as threats to realization of economic and social rights.12,13
We discuss some important efforts to describe associations between health, violence, and poverty during and immediately after the Central American conflicts. The lessons learned may be of use to others who must confront, and describe, similar situations elsewhere. Because of our greater knowledge of these settings, we will emphasize work conducted in El Salvador, Guatemala, and Chiapas.
Description of politically-motivated killing replaced the study of poverty-related health issues as the chief epidemiologic enterprise of the wartime era. The de facto epidemiologists included human rights advocates, mothers of the disappeared, and journalists; the de facto study instruments were autopsy reports and witnesses’ testimonies. It fell to postwar Truth Commissions to merge and interpret data from many disparate sources to describe conflict-related mortality and to determine which reported episodes of violence had, indeed, occurred.
The Salvadoran Truth Commission evaluated 27,000 reports of human rights abuses but chose not to estimate the total number of dead other than “thousands upon thousands.” The Commission focused on exhaustive documentation of 34 episodes of violence (some, such as the Rio Sumpul massacre, resulting in hundreds of deaths), while remarking that 300 to 500 persons may have been assassinated every single month in 1982 alone.14 The Guatemalan Truth Commission, assisted by the American Association for the Advancement of Science (AAAS), concluded that approximately 160,000 persons were assassinated and 40,000 more disappeared (presumably murdered) during the conflict, and that there had been 626 massacres.1 Both commissions determined that political violence had been purposefully committed on a massive scale, primarily (though not exclusively) by military forces on the government side.
The quantitative, population-level characteristics of conflict-related mortality were described most thoroughly by the Reconstrucción de la Memoria Histórica (REMHI) project, coordinated by the Catholic Church’s Archdiocese of Guatemala. Based on 6494 survivors’ testimonies, the REMHI project reported, for example, that of 165 massacres, 29% killed more than 20 people; 64% occurred in predominantly indigenous Quiché province; 38% were preceded by kidnappings or assassinations; 40% provoked the mass exodus of survivors; and 3% included victims who were forced to participate in the killing of other victims.15,16 Thus the analyses of REMHI and AAAS expanded human rights reporting beyond its historic, case-based methodology to encompass population-level description of the correlates of political violence.
The Truth Commissions also used forensic evidence to substantiate case reports. For example, at El Mozote, El Salvador, at least 500 civilians were slain in December 1981. Contemporary reports were published in The New York Times and The Washington Post, and a single survivor courageously provided testimony. However, the very existence of the massacre was disputed until 11 years later, when exhumations conducted by forensic anthropologists provided confirmatory physical evidence, and the Salvadoran Truth Commission summarized the case.14,17
In general, reporting of war-related mortality respected the standard categories of war crimes defined by Common Article 3 of the Geneva Conventions (“violence to life and person, in particular murder of all kinds” and “the carrying out of executions without previous judgment”).18 However, REMHI and other project coordinators developed their own expanded taxonomies of violence because the Geneva case definitions were inadequate to describe the full horror and complexity of local situations.15,16,19
The reporting itself led to further mortality. Bishop Juan Gerardi, the organizer of Guatemala’s REMHI report, was assassinated within days of the report’s release; Jesuit priests Segundo Montes and Ignacio Martín-Baró, who reported on the socioeconomic and mental-health impact of the Salvadoran civil war, were also assassinated, as was Herbert Anaya of the Salvadoran nongovernmental Human Rights Commission.20–23
MORBIDITY AND POVERTY
During the height of conflict, descriptions of morbidity and poverty were focused largely on political violence and its direct consequences.
Many reports catalogued military attacks on health units and health personnel, again categorized in war-crimes terms.24,25 Reports for a single year (1989–1990) in the Salvadoran civil war cited 179 episodes in which health workers were assassinated, disappeared, arrested, detained, harassed, threatened or attacked, or health units or training programs were attacked, threatened, or forcibly shut down.26 In Nicaragua, by late 1985, the contra war was said to have caused the deaths of 38 health workers.27 Although these episodes resulted in the suspension or interruption of health services, estimates of the magnitude of resulting civilian morbidity (and poverty) were not usually attempted. Epidemiologic description was inhibited by military restrictions on circulation of medical personnel, shortages of trained investigators, pervasive fear, and perceived futility. Mere reporting of mortality or morbidity data was unlikely to be effective,28 for example, where childhood immunization coverage was inadequate or maternal mortality ratios were high because vaccines and ambulances were consistently turned back at military roadblocks.
As security risks subsided, formal epidemiologic studies re-emerged, with the aim of describing associations between armed conflict and health indicators. The prevalence of stunting was found to be very high (54% in children <5 years old) in regions affected by the armed conflict in Chiapas, and was even higher (up to 63%) in communities divided by internal conflict.29 Childhood stunting prevalence declined nationwide (from 15% to 11%) after resolution of the civil war in El Salvador, but remained elevated (32%) in conflict-afflicted areas.30,31 Maternal and perinatal mortality were elevated in the conflict zones of Chiapas.32,33 In Nicaragua, a preconflict decline in malaria incidence was reversed (a 17% increase in 1983–1985 compared with a 38% decline from 1979–1984) in the region affected by the contra war.34 In El Salvador, the annual rate of infection with Mycobacterium tuberculosis apparently doubled during the civil war; tuberculosis control also faltered in Chiapas.35,36 In Guatemala, the mental-health consequences of conflict included persistent fear (26%), persistent sadness (15%), and prolonged or abnormal grieving (8%) in witnesses interviewed up to 14 years after political violence.15
Wartime epidemiologic reports focused primarily on measurement of mortality and violence, seldom on that of poverty, even though entire populations had lost their homes, their livelihoods, and their family members, and were sheltered only by standard-issue sheets of United Nations black plastic. Even so, graphic and moving descriptions of poverty were embedded within human rights reports and studies of morbidity. The REMHI testimonies, for example, described deliberate burning of food crops and the politically-motivated assassination of breadwinners (at least 42,047 children lost a parent to political violence) as mechanisms by which war left its victims in penury.15
Epidemiologic investigations of morbidity employed quantitative description of subjects’ socioeconomic status in environments in which income and wealth were not measurable in cash, the concept of the census tract was useless, and the effects of violence were pervasive. In response, investigators created variables denoting such concepts as “adherents of one faction are denied access to health services, housing, water systems, latrines, and/or electrical systems controlled by the other faction” and “armed incursion into a health facility for the purpose of frightening the health workers and patients away.”19,26,32 Related descriptors included history of forced displacement, inability to obtain food or health services because of roadblocks or minefields, destruction of crops and livestock by bombing, and orphanhood or widowhood caused by war.
The invention of study variables paralleled the invention of spoken language to describe the intolerable situations of war. Residents of El Salvador, for example, created a verb (“guindear”) that means, loosely translated, “to flee from the army in the middle of the night, carrying all of one’s children and possessions.”
The Salvadoran, Guatemalan, and draft Chiapanecan peace accords all viewed cessation of violence and redistribution of arable land as cornerstones of postwar antipoverty strategy. The Guatemalan accords stated “Land is central to the problems of rural development.”37 The Salvadoran accords mandated postwar land transfers benefiting small farmers and landless peasants and ex-combatants.38 In Chiapas, the Ejército Zapatista para la Liberación Nacional lobbied for a definitive “solution to the grave national agrarian problem.”39 Both the Salvadoran and Guatemalan accords defined specific timetables for land transfers, judicial reform, and demobilization of combatants, with implementation to be monitored by the United Nations.
Postconflict planning for improvement in health indicators was less specific. The Salvadoran accords mainly proposed that a postwar forum be established for later consideration of “economic and social development.”38 The draft Chiapanecan agreement proposed “mechanisms to guarantee ... food, health, housing, and ... an adequate standard of living,” but did not specify timetables for or indicators of the realization of this right.39
Guatemala’s social and economic accords addressed health more directly. They obliged the postwar Guatemalan government “to step up public spending on health as a proportion of gross domestic product by at least 50% ... to allocate at least 50% of public health expenditure to preventive care and undertake to cut the 1995 infant and maternal mortality rate in half ... and to eradicate measles by the year 2000.”37 The United Nations human rights monitoring mission (MINUGUA) was tasked with documenting postconflict compliance with these agreements.
MINUGUA reported, for example, that clinics serving displaced persons had closed or lacked medications; that newly-established health services failed to deliver folic acid, ferrous sulfate, and tetanus vaccine to pregnant women; and that food production still had not moved beyond subsistence levels for displaced persons.40 MINUGUA’s efforts to document achievement of mortality-reduction goals were thwarted, though, by the absence of reliable baseline data for comparison.
None of the peace accords provided quantitative benchmarks for achievement of socioeconomic equity, or for reduction in indicators of absolute poverty. In the postconflict period, the worst violence—such as the death-squad assassinations and rural massacres of the 1970s and 1980s—was indeed curtailed. But the underlying problem of severe poverty was not resolved. At the turn of the millennium, the proportion of persons living on less than US$1 per day was 14% in Guatemala, 19% in El Salvador, and 45% in Nicaragua; the poorest quintiles of these populations were estimated to control only 3 to 6% of national income.41
RESEARCH LESSONS LEARNED FROM THE EXAMPLES OF CENTRAL AMERICA
1. Prioritize what is most likely to help the vulnerable. In Central America, documentation and denunciation of politically-motivated violence and its consequences took precedence over more conventional epidemiologic investigations of infectious and other diseases.
2. Measure both direct and indirect effects of armed conflict on health and poverty. The extrajudicial execution of a health worker or small farmer merits reporting as a war crime, but loss of that victim’s labor or protection may also result in reportable indirect increases in mortality and morbidity. Childhood malnutrition is an especially important indicator of war’s effects on the most vulnerable, because of its sensitivity to conflict-related orphanhood and to disruption in access to food and health services (in particular, infectious disease interventions).42
3. Respect security threats. Protracted negotiations with individuals, community leaders, and military commanders may be required to create the conditions necessary for informant and investigator safety, and the trust required to promote sharing of accurate information. The alternative, elected by many human rights investigators responding on an emergency basis to individual incidents, is to conduct investigations clandestinely; this strategy confers its own risks. Ethical treatment of victim/informants may also require provision of medical and mental-health attention (a role played by many Central American human rights monitoring organizations); the need for the latter may be increased by the retraumatizing effects of interviews.
4. Complex situations require interdisciplinary collaboration and local knowledge. Data collection and analysis in Central America were collective enterprises that involved humanitarian workers, attorneys, forensic examiners, members of the clergy, lay catechists, human rights activists, and the direct victims of violence, in addition to clinicians and epidemiologists. The setting of priorities, the identification of indirect effects of conflict, the development of appropriate study variables, and the effective dissemination of findings required first-hand observation and broad local participation.
5. Appropriate legal analyses may strengthen epidemiologic findings. Many early health-related human rights reports focused on crimes of war as defined by the Geneva Conventions.18,24,25,27 This approach effectively drew international attention to the violence. However, it was inadequate for the analysis of the indirect effects of war on health and poverty. The framework of the International Covenant on Economic Social and Cultural Rights (ICESCR) expanded the definition of health-related human rights concerns to include availability, accessibility, acceptability, quality, and equal treatment as components of the right to health, and required documentation of “progressive realization” of this right, including elimination of health disparities.43,44 Thus, public-health concerns other than mortality provoked by violence could be expressed within a legal framework.
6. Epidemiologic contributions to peace accords may support more effective postwar poverty eradication and health improvement. Reliable and understandable data on health and poverty, together with appropriate suggestions for postconflict benchmarks may be of practical use to those drafting and implementing peace agreements. Models of peacetime benchmarking processes have recently been under development in Mexico and Panama44–46 and are outlined in ICESCR. Of course, the negotiation and implementation of any peace agreement depend ultimately on firm political will, which is far more difficult to assemble than valid data.
7. In situations of crisis, data collection and interpretation are more effective if coordinated with political processes. During the Central American crises, the aggressive reporting of politically-motivated violence helped stakeholders to mobilize support for cessation of organized violence. Had a similar partnership been feasible (a difficult matter to judge retrospectively) between the description and reporting of poverty and an active political process to end poverty, vulnerable populations might have benefited more.
During the 1970s, 1980s, and early 1990s, armed conflict increased the burden of mortality, morbidity, and poverty throughout Central America. Academic epidemiology was nearly abandoned in the region during this period. However, qualitative and quantitative methods were used, formally and informally, to address the very nonacademic problem of violence. Lay and professionally trained investigators collaborated—often at great personal risk—to define methodologies that could address the intertwined problems of mortality, morbidity, hunger, crimes against humanity, and deliberate insults to human dignity. They adapted the intent and language of epidemiologic and human rights investigations not just to describe the magnitude and mechanisms of conflict-related human suffering, but to mobilize political and material assistance to impoverished victims. Their efforts did help achieve cessation of the worst violence although the problems of inequality and abject poverty remain unsolved.
Reductions in the linked burdens of poverty and violence are still of the gravest importance in Central America and elsewhere.47,48 The work described above offers lessons for epidemiologists, human rights advocates, and others who must find their own ways to describe the deadly collusion between poverty, war, and illness—and to take action based on their findings.
ABOUT THE AUTHORS
PAULA BRENTLINGER has participated in humanitarian relief efforts and research projects in Honduras, El Salvador, and Mexico. Her current work involves control of infectious diseases in postconflict settings in southern Africa. MIGUEL HERNAN has participated in research projects in postconflict settings in Iraq, El Salvador, and Mexico. His current work involves methods for the analysis of longitudinal studies.
We are grateful to Meredith Fort, Mary Maloney, Carlos Martín Beristain, Eileen Rosin, and Katherine Yih for helpful background information and thoughtful comments on earlier drafts of this paper.
1. Comisión de Esclarecimiento Histórico. Guatemala: Memoria del Silencio. Tz’inil Na’tab’al
. Guatemala City: Comisión de Esclarecimiento Histórico; 1999. Available at: http://shr.aaas.org/guatemala/ceh/mds/spanish/
. Accessed June 19, 2007.
2. Scrimshaw N. Historical concepts of interactions, synergism, and antagonisms between nutrition and infection. J Nutr
3. Fifty years working to improve food and nutrition in Central America. Pan Am J Public Health
4. Martorell R. Results and implications of the INCAP follow-up study. J Nutr
5. Martorell R. The role of nutrition in economic development. Nutr Rev
6. Durham W. Scarcity and Survival in Central America
. Stanford: Stanford University Press; 1979.
7. Rawson I, Valverde V. The etiology of malnutrition among preschool children in rural Costa Rica. Environ Child Health
8. Gomez F, Ramos Galvan R, Frenk S, et al. Mortality in second and third degree malnutrition. J Trop Ped
, Sept 1956; Reprinted in Bull World Health Organ
9. Valverde V, Martorell R, Mejia Pivaral V, et al. Relationship between family land availability and nutritional status. Ecol Food Nutr
10. Trowbridge F, Newton L. Seasonal changes in malnutrition and diarrheal disease among preschool children in El Salvador. Am J Trop Med Hyg
11. United Nations Development Program. Human Development Report 1999
. Geneva: United Nations; 1999.
12. Pan American Health Organization. Closing the Gaps in Health in the Least Protected Populations. Annual Report of the Director, 2006
. Washington, DC: Pan American Health Organization; 2006.
13. Lutheran World Federation. Central America 2005–2006, from a Human Rights Perspective
. Geneva: Lutheran World Federation; 2007.
15. Oficina de Derechos Humanos del Arzobispado de Guatemala. Informe Proyecto Interdiocesano de Recuperación de la Memoria Histórica. Guatemala. Nunca Más. I. Impactos de la Violencia
. Guatemala City: Oficina de Derechos Humanos del Arzobispado de Guatemala; 1998.
16. Martín Beristain C. Procesos de duelo en las comunidades mayas afectadas por la violencia política. Doctoral Thesis. Spain: University of Deusto; 2005.
17. Danner M. The truth of El Mozote. The New Yorker
. December 6, 1993;50–133.
18. International Committee of the Red Cross. Geneva Conventions of August 12, 1949
. Geneva: International Committee of the Red Cross; 1949.
19. Brentlinger P. Health sector response to security threats during the civil war in El Salvador. BMJ
20. Martín Beristain C. Health, memory, and human rights in Guatemala. Lancet
21. Martín-Baró I. El Salvador 1987. Estudios Centroamericanos
(San Salvador, El Salvador)1988;471–472.
22. Montes S. Los derechos económicos, sociales, y culturales en El Salvador. Estudios Centroamericanos
(San Salvador, El Salvador) 1988;476:515–537.
23. Comisión de Derechos Humanos de El Salvador (CDHES) No Gubernamental. Comisión de Derechos Humanos de El Salvador (CDHES), no gubernamental, se pronuncia en torno a los hechos sobre el asesinato de nuestro coordinador general. Estudios Centroamericanos
(San Salvador, El Salvador) 1988;471–472:144–146.
24. Eisenberg C, Halperin D, Hargreaves A, et al. Health and human rights in El Salvador. N Engl J Med
25. Gellhorn A. Medical mission report on El Salvador. N Engl J Med
26. International Commission on Medical Neutrality. Violations of Medical Neutrality: El Salvador
. Seattle: International Commission on Medical Neutrality; 1991.
27. Braveman P, Siegel D. Nicaragua: a health system developing under conditions of war. Int J Health Serv
28. Americas Watch. Nightmare Revisited. 1987–88. Tenth Supplement to the Report on Human Rights in El Salvador
. New York: Americas Watch; 1988.
29. Sánchez-Pérez HJ, Hernán MA, Rios-González A, et al. Malnutrition among children aged younger than 5 years in conflict zones of Chiapas, Mexico. Am J Pub Health
30. Grummer-Strawn L, Caceres JM, Herrera de Jaimes BP. Trends in the nutritional status of Salvadorian children: the post-war experience. Bull World Health Organ
31. Brentlinger P, Hernán M, Hernández-Díaz S, et al. Childhood malnutrition and post-war reconstruction in rural El Salvador six years after the cease-fire. A community-based survey. JAMA
32. Brentlinger PE, Sánchez-Pérez HJ, Arana Cedeño M, et al. Pregnancy outcomes, site of delivery, and community schisms in regions affected by the armed conflict in Chiapas, Mexico. A community-based survey. Soc Sci Med
34. Garfield R, Prado E, Gates JR, Vermund SH. Malaria in Nicaragua: community-based control efforts and the impact of war. Int J Epidemiol
35. Barr R, Menzies R. The effect of war on tuberculosis. Results of a tuberculin survey among displaced persons in El Salvador and a review of the literature. Tuber Lung Dis
36. Sánchez-Pérez HJ, Hernán MA, Hernández-Díaz S, et al. Detection of pulmonary tuberculosis in Chiapas, Mexico. Ann Epidemiol
37. Agreement on Social and Economic Aspects and Agrarian Situation concluded on 6 May 1996 between the Presidential Peace Commission of the Government of Guatemala and the Unidad Revolucionaria Nacional Guatemalteca. Available at: http://www.usip.org/library/pa/guatemala/guat_960506.html
. Accessed June 19, 2007.
40. MINUGUA. Suplemento al cuarto informe sobre la verificación de los acuerdos de paz de Guatemala (1 de Agosto 1998 – 31 de Octubre 1999). Situación de los compromisos sobre aspectos socioeconómicos y agrarios, reasentamiento e incorporación. Guatemala: MINUGUA, 1999.
41. The World Bank. World Development Report 2007; Development and the Next Generation
. Washington: The World Bank; 2006.
42. Toole M, Waldman R. The public health aspects of complex emergencies and refugee situations. Annu Rev Public Health
43. International Covenant on Economic, Social, and Cultural Rights. U.N. Doc A/6316(1966), 993 U.N.T.S.; 1966.
44. United Nations Committee on Economic, Social, and Cultural Rights. General Comment 14. The Right to the Highest Attainable Standard of Health: 11/08/2000. E/C12/2000/4
. Geneva: Office of the United Nations High Commissioner for Human Rights; 2000.
45. Lozano R, Soliz P, Gakidou E, et al. Benchmarking of performance of Mexican states with effective coverage. Lancet
46. Sandiford P, Salvetto M. Las desigualdades en salud en Panamá. Gac Sanit
47. Centro Internacional de Investigaciones en Derechos Humanos (CIIDH), Derechos Humanos Económicos, Sociales, y Culturales en Guatemala (DESCGUA), and Coordinación de ONG y Cooperativas (CONGCOOP). Situación de los Derechos Económicos, Sociales, y Culturales en Guatemala, 2006. A 10 años de la firma de los Acuerdos de Paz
. Guatemala City: CIIDH; 2006.
48. Roberts L, Lafta R, Garfield R, et al. Mortality before and after the 2003 invasion of Iraq: cluster sample survey. Lancet