Abstract
We performed a cross-sectional, community-based survey, supplemented by interviews with community leaders in Chiapas, Mexico, to examine the prevalence and predictors of child malnutrition in regions affected by the Zapatista conflict.
The prevalence rates of stunting, wasting, and underweight were 54.1%, 2.9%, and 20.3%, respectively, in 2666 children aged younger than 5 years. Stunting was associated with indigenous ethnicity, poverty, region of residence, and intracommunity division. The results indicate that malnutrition is a serious public health problem in the studied regions.
In Chiapas, Mexico, long-standing conflicts related to land tenure, religion, and other issues have been further complicated by an armed conflict between the Mexican government and the Ejército Zapatista de Liberación Nacional (“the Zapatistas”), which began in 1994 over Zapatista demands that the Mexican government address the alarming poverty conditions among the indigenous population. Social polarization and intolerance have led to forced displacement of more than 16000 Chiapanecan citizens, politically motivated violence (including murder), and intracommunity divisions so complete as to have produced villages within which separate governments, clinics, schools, justice systems, and other services for adherents of the separate factions operate.2–5 We postulated that chronic interparty and intracommunity conflict in Chiapas might be associated with malnutrition, particularly stunting, in children. Because no published studies have addressed this question in this setting, we sought to describe the prevalence of malnutrition in children aged younger than 5 years in the 3 Chiapanecan regions most adversely affected by the Zapatista conflict and its association with various socioeconomic and conflict-related factors.
METHODS
In the regions most adversely affected by the armed conflict in Chiapas,6 we conducted a cross-sectional, population-based household survey, supplemented by semistructured interviews with community-based governing councils and health teams. We investigated the health conditions of and access to care for children in the conflict zone, including rates of childhood malnutrition, pulmonary tuberculosis, and maternal mortality. The methodology of the study has been described previously.6
Weight was measured with baby and hanging scales.7 Height was measured with infantometers (for children aged younger than 2 years) and stadiometers (for children aged 2 to 4 years).7 Age was obtained by asking the parent for each child’s birthdate, confirming whenever possible (65% of children) with birth records or vaccination cards. Anthropometric data were analyzed with the Epi Info 2000 Epinut package (Centers for Disease Control and Prevention, Atlanta, Ga), which classified children as stunted, wasted, or underweight; standardized z scores (obtained using Epi info) of less than 2.0 for these classifications were used as the cutoff point, and participants with impulsive results (those for whom the obtained measures of height and age gave an impossible nutritional status) were excluded.8 We estimated the association between stunting and other variables using logistic regression (consistent confidence intervals9 were used to adjust for clustering within communities), weighted to reflect probability of selection. Given the transversal design of the study, and the conditions of the region during the fieldwork that impacted the nonresponse rates, the estimated odds ratios were adequate to measure the magnitude of the association between the nutritional status and factors analyzed, but did not necessarily approximate the prevalence ratio, because of the high prevalence of stunting.
RESULTS
The final sample included 21 government-aligned communities, 6 opposition-aligned communities, and 19 communities divided by political-party affiliation (government vs Zapatista-aligned).6 Four of the government-aligned communities were internally divided by religious or intraparty differences. We identified 2838 children aged younger than 5 years from 1779 households (20 households declined to participate).
Table 1 ▶ presents the individual, household, and community characteristics of the children in the study. All communities were predominantly indigenous (89.3% of the children in the study). The overall prevalence of stunting was 54.1% (n=2666 with analyzable data) and was significantly higher in divided communities (P<.001). The overall prevalence rates of wasting and underweight were 2.9% and 20.3%, respectively. Nearly all factors associated with either poverty or intracommunity division were significantly associated with stunting in bivariate analyses (data not shown). In multivariate analyses, child’s age, dirt-floored house (a proxy for low socioeconomic status), maternal education, indigenous ethnicity, region of residence, and intracommunity division retained their significant associations with stunting (Table 2 ▶).
TABLE 1—
Characteristics of Children Aged Younger Than 5 Years in Conflict Zones of Chiapas, Mexico, by Presence of Intracommunity Division
| Intracommunity Division | ||||
| Characteristics of Children and Their Households or Villagesa | All Children (N = 2838), no. (%) | No Community Division (n = 1295), no. (%) | Community Divided, Other Than by Political Partyb (n = 301), no. (%) | Community Divided by Political Party (n = 1242), no. (%) |
| Stunting | 1442 (54.1)c | 592 (48.5) | 165 (62.7) | 685 (58.0) |
| Wasting | 79 (2.9)d | 34 (2.8) | 6 (2.2) | 39 (3.3) |
| Underweight | 549 (20.3)e | 230 (18.5) | 73 (26.7) | 246 (20.6) |
| Maternal education, mean years completed ±SD | 2.9 ±2.6 | 3.1 ±2.7 | 2.2 ±2.7 | 2.8 ±2.4 |
| Whole family slept in same room | 1764 (62.2) | 729 (56.4) | 200 (66.7) | 835 (67.3) |
| Household has dirt floor | 2186 (77.1) | 919 (71.1) | 220 (73.1) | 1047 (84.4) |
| Household cooked with wood fuel | 2738 (96.7)f | 1228 (95.0) | 289 (96.3) | 1221 (98.6) |
| Maternal language fluency | ||||
| Mother speaks indigenous language only | 1486 (52.6)f | 654 (50.9) | 167 (55.7) | 665 (53.8) |
| Mother bilingual | 1040 (36.8)g | 479 (37.3) | 59 (19.7) | 502 (40.6) |
| Mother speaks Spanish only | 297 (10.5) | 153 (11.9) | 74 (24.7) | 70 (5.7) |
| Household has piped water | 1918 (67.6) | 947 (73.1) | 126 (41.9) | 845 (68.0) |
| Household has electricity from local grid | 2476 (87.2) | 1167 (90.1) | 264 (87.7) | 1045 (84.1) |
| Household has toilet or latrine | 2084 (73.4) | 926 (71.5) | 165 (54.8) | 993 (80.0) |
| Household has refrigerator | 152 (5.4) | 99 (7.6) | 19 (6.3) | 34 (2.7) |
| Travel time to nearest clinic, min, mean ±SD | 25.5 ±31.3 | 25.5 ±27.0 | 3.0 ±9.0 | 31.0 ±36.1 |
| Travel time to nearest hospital, min, mean ±SD | 147.5 ±120.3 | 170.9 ±157.5 | 106.1 ±58.7 | 133.0 ±71.9 |
| Region of residence | ||||
| Norte region | 1130 (39.8) | 464 (35.8) | 113 (37.5) | 553 (44.5) |
| Altos region | 880 (31.0) | 354 (27.3) | 188 (62.5) | 338 (27.2) |
| Selva region | 828 (29.2) | 477 (36.8) | 0 | 351 (28.3) |
| History of displacement of residents from village | 1197 (42.2) | 308 (23.8) | 163 (54.2) | 726 (58.5) |
| Survey only allowed to include 1 of 2 antagonistic factionsh | 604 (21.3) | NA | 30 (10.0) | 574 (46.2) |
| Alleged bias in provision of public services (other than health) | 342 (12.1) | 0 | 158 (52.5) | 184 (14.8) |
| Alleged bias in provision of community-level health services | 905 (31.9) | 35 (2.7) | 301 (100) | 569 (45.8) |
| Political-party affiliation of household | ||||
| Aligned with government party | 1910 (67.3) | 1034 (79.9) | 301 (100) | 575 (46.3) |
| Political party affiliation not stated | 482 (17.0) | 0 | 0 | 482 (38.8) |
| Aligned with opposition | 446 (15.7) | 261 (20.2) | 0 | 185 (14.9) |
| From systematically selected households | 1755 (61.8) | 870 (67.2) | 209 (69.4) | 676 (54.4) |
Note. Results given as percentage of number of children with complete data unless otherwise indicated.
a There were no differences among the following characteristics: maternal age (mean years ±SD = 27.8 ±7.6), total number of persons living in household (mean ±SD= 7.1 ±2.7), total number of children younger than 5 years living in household (mean ±SD = 1.9 ±0.8); child had up-to-date immunizations (for age) = 72.6%.
bDivision by religion, in the absence of political-party division, or division within a single political party.
cNot included in the analysis: 94 children (36 without community division; 23 community divided, other than by political party; 35 community divided by political party).
dNot included in the analysis: 4 children (2 without community division; 2 community divided by political party).
eNot included in the analysis: 27 children (10 without community division; 7 community divided, other than by political party; 10 community divided by political party).
fNot included in the analysis: 6 children (2 without community division; 1 community divided, other than by political party; 3 community divided by political party).
gNot included in the analysis: 4 children (all of them without community division).
hIn some divided communities, the survey was only permitted if those in the opposing faction were not surveyed.
TABLE 2—
Association of Stunting With Child, Family, and Community Characteristics in Conflict Zones of Chiapas, Mexico: Multivariate Analysis
| Odds Ratio (95% Confidence Interval) | P | |
| Age of child, mo | ||
| ≤5 | Reference | — |
| 6–11 | 3.0 (1.8, 5.1) | < .001 |
| 12–23 | 9.1 (4.8, 17.3) | < .001 |
| 24–35 | 8.2 (4.2, 16.1) | < .001 |
| 36–47 | 9.0 (4.8, 17.0) | < .001 |
| 48–59 | 12.1 (5.6, 26.1) | < .001 |
| Maternal education (at least 1 year) | 0.9 (0.9, 1.0) | .006 |
| Household has dirt floor | 1.6 (1.2, 2.2) | .002 |
| Maternal language fluency | ||
| Monolingual Spanish speaker | Reference | — |
| Bilingual indigenous: Spanish | 1.8 (1.1, 3.1) | .018 |
| Monolingual indigenous language | 1.9 (1.3, 2.8) | .001 |
| Region of residence | ||
| Norte region | Reference | — |
| Altos region | 2.5 (1.3, 4.7) | .006 |
| Selva region | 1.5 (1.0, 2.3) | .055 |
| Intracommunity division | ||
| Community not divided | Reference | — |
| Community divided, not by political party | 2.7 (1.4, 5.3) | .005 |
| Community divided by political party | 1.8 (1.1, 2.8) | .013 |
DISCUSSION
The overall prevalence of stunting observed was substantially greater than that reported by the Mexican National Nutritional Survey (17.8% nationally; 29.2% in Chiapas State)10 and was consistent with observations from other surveys performed in Chiapas during the Zapatista conflict.11,12 In fact, the prevalence of stunting present in the children we examined resembled that in child residents of conflict zones in Afghanistan and Angola (63.7% and 57.3%, respectively)13,14 more than it did that in children in northern Mexico (7.1%). Such high levels of malnutrition place these children at higher risk for diminished school and work performance, as well as mortality.15–21 These effects may worsen disparities between the rural indigenous residents of Chiapas and others, thus predisposing them to continued cycles of violence and polarization. Furthermore, our findings may have underestimated the true prevalence of malnutrition, because we suspect that communities with higher levels of conflict and poverty were more likely to refuse participation in the study.
A key finding of our study was the strong association between factors related to conflict—both the Zapatista conflict and other sources of intracommunity division—and poor nutritional status. Although causality could not be firmly established by our cross-sectional methods, intracommunity divisions and intercommunity conflict may decrease access to food and increase vulnerability to infectious diseases in several ways: violence and social tensions may disrupt traditional mechanisms of intracommunity cooperation, thus interfering with cultivation and marketing of crops; maternal stress may diminish breastmilk production22,23; and religious or political discrimination may impede access to health services. However, because the study sample was not representative of the entire population of the study regions, the generalizability of our observations is limited. Nevertheless, given the environment of conflict,6 our data are unique. If circumstances permit, further confirmatory studies should be conducted.
Childhood malnutrition is a serious public health problem in the conflict-affected regions of Chiapas. Conflict-related divisions may serve to increase disparities between this and other Mexican populations. The observed levels of malnutrition compromise the health of Chiapanecan children.24,25
Acknowledgments
This investigation was funded by the El Colegio de la Frontera Sur, El Centro de Capacitación en Ecología y Salud para Campesinos-Defensoría del Derecho a la Salud, Universitat Autónoma de Barcelona, and the Grand Service Foundation.
We are indebted to the leaders and residents of the studied communities and to participating health workers from both the governmental and the Zapatista health sectors, for their permission, participation, and trust. We gratefully acknowledge the collaboration and participation of Physicians for Human Rights in the project on which this study was based. We would also like to thank Guadalupe Vargas, Roberto Solís Hernández, Alejandro Flores Hernández, Herlinda Méndez Santiz, Juan Carlos Nájera, Julio César Arias, Blanca Coello (all at the Colegio de la Frontera Sur), Juan Manuel Canales, Jonathan Kirsch, Kerri Sherlock, and Dave McFarlane for their invaluable assistance with interviews and database management.
Human Participant Protection At the time of study inception, there was no functioning institutional review board in Chiapas, and Chiapanecan researchers interpreted Mexican national regulations as meaning that minimal-risk community-survey protocols did not require formal review. However, oral informed consent was obtained at multiple levels in each community studied. Initially, consent was obtained from community authorities, generally common-lands commissioners (“comisariados ejidales”), and health and education committees. During their deliberations, community-level authorities were permitted to delete any questions they considered objectionable from the study protocol. Subsequently, an assembly of the entire adult population was called, to ask for collective consent. Finally, in each household, permission was requested of the head of the family. Communities were assured that the names of individual communities would remain confidential, as would each individual community’s survey results. The investigators also promised to present each community’s study findings to the community after data analysis. The study protocol was also approved by a panel assembled by Physicians for Human Rights, after deliberations conducted in accordance with the Declaration of Helsinki.
Peer Reviewed
Contributors H.J. Sánchez-Pérez originated and designed the study, performed the fieldwork, analyzed and interpreted the data, and wrote the brief. M.A. Hernán and D. Ford participated in the origination and design of the study, in the analysis and interpretation of data, and in revising critically the important intellectual content of the brief. A. Ríos-González participated in the fieldwork, in the analysis and interpretation of data, and in preparation of the brief. M. Arana-Cedeño participated in the origination and design of the study, in the fieldwork, and in the writing of the brief. A. Navarro and M.A. Micek participated in the analysis and interpretation of data and in revising critically the important intellectual content of the brief. P. Brentlinger participated in the origination and design of the study, in the analysis and interpretation of data, in revising critically the important intellectual content of the brief, and in drafting the final version.
References
- 1.Womack J. Chiapas, the bishop of San Cristóbal, and the Zapatista revolt. In: Womack J, ed. Rebellion in Chiapas: An Historical Reader. New York, NY: The New Press, 1999:3–59.
- 2.Burguete Cal y Mayor A. The de facto autonomous process: new jurisdictions and parallel governments in rebellion. In: Rus J, Hernández Castillo R, Mattiace S, eds. Mayan Lives, Mayan Utopias. Lanham, Md: Rowman & Littlefield; 2003: 185–186.
- 3.Martínez-Velasco G. Apreciaciones generales sobre la situación del desplazamiento en Chiapas. In: Seminario Regional Sobre Desplazamiento en las Américas. México City, México: Secretaría de Relaciones Exteriores, Oficina del Representante del Secretario General de las Naciones Unidas, Proyecto Brookings Institution—Universidad John Hopkins Sobre Desplazamiento Interno, February 18–20, 2004:9.
- 4.Centro de Derechos Humanos Fray Bartolomé de las Casas. La Situación Actual de los Derechos Humanos en Chiapas: Reporte Especial Para Mary Robinson, Alta Comisionada de Derechos Humanos de la ONU. San Cristóbal de las Casas, Chiapas, México: Centro de Derechos Humanos Fray Bartolomé de las Casas; 1999.
- 5.Oficina del Alto Comisionado de las Naciones Unidas Para los Derechos Humanos en México. Diagnóstico Sobre la Situación de los Derechos Humanos en México. Mexico City, México: Oficina del Alto Comisionado de las Naciones Unidas Para los Derechos Humanos en México; 2003.
- 6.Brentlinger P, 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. 2005;61:1001–1014. [DOI] [PubMed] [Google Scholar]
- 7.Habitch JP, Martorell R. Anthropometric field methods: criteria for selection in human nutrition. In: Jeliffe DG, Jeliffe EP, eds. Nutrition and Growth. New York, NY: Plenum Publishing Corp; 1979:365–387.
- 8.National Center for Health Statistics, World Health Organization. Growth Charts. Rockville, Md: National Center for Health Statistics, Dew Publications; 2000.
- 9.Liang K-Y, Zeeger SL. Longitudinal analysis using generalized linear models. Biometrika. 1986;73:13–22. [Google Scholar]
- 10.Encuesta Nacional de Nutrición, 1999: Estado Nutricio de Niños y Mujeres en México. Cuernavaca, Morelos, México: Secretaría de Salud, Instituto Nacional de Salud Pública, Instituto Nacional de Estadística, Geografía e Informática; 2001.
- 11.Ávila-Curiel A, Chávez-Villasana A, Shamah-Levy T. Encuesta Nacional de Alimentación y Nutrición en el Medio Rural 1996. México City, México: Instituto Nacional de la Nutrición Salvador Zubirán (INNSZ), Secretaría de Desarrollo Social (SEDESOL), Sistema Para el Desarrollo Integral de la Familia (DIF), Secretaría de Salud (SSA), Gobiernos de los Estados; 1997.
- 12.Ochoa-Díaz Lopez H, Sánchez-Pérez HJ, Ruíz-Flores M, Fuller M. Social inequalities and health in rural Chiapas, México: agricultural economy, nutrition, and child health in La Fraylesca region. Cad Saúde Pública. 1999;15:261–270. [DOI] [PubMed] [Google Scholar]
- 13.Assefa F, Jabarkhil M, Salama P, Spiegel P. Malnutrition and mortality in Kohistan District, Afghanistan, April 2001. JAMA. 2001;286:2723–2728. [DOI] [PubMed] [Google Scholar]
- 14.Agadjanian V, Prata N. Civil war and child health: regional and ethnic dimensions of child immunization and malnutrition in Angola. Soc Sci Med. 2003;56: 2515–2527. [DOI] [PubMed] [Google Scholar]
- 15.Martorell R. The role of nutrition in economic development. Nutr Rev. 1996;II:S66–S71. [DOI] [PubMed] [Google Scholar]
- 16.Haas JD, Murdoch S, Rivera J, Martorell R. Early nutrition and later physical work capacity. Nutr Rev. 1996;54(2 pt 2):S41–S48. [DOI] [PubMed] [Google Scholar]
- 17.Allen L. Malnutrition and human function: a comparison of conclusions from the INCAP and Nutrition CRSP studies. J Nutr. 1995;125:1119S–1126S. [DOI] [PubMed] [Google Scholar]
- 18.Pollitt E, Gorman K, Engle P, Rivera J, Martorell R. Nutrition in early life and the fulfillment of intellectual potential. J Nutr. 1995;125:1111S–1118S. [DOI] [PubMed] [Google Scholar]
- 19.Colombo M, de Andraca I, Lopez I. Mental development and stunting. In: Waterlow J, ed. Linear Growth Retardation in Less-Developed Countries. New York, NY: Vevey/Raven Press; 1988. Nestle Nutrition Workshop Series, Vol 14.
- 20.Murray C, Lopez A. Global mortality, disability, and the contribution of risk factors: Global Burden of Disease Study. Lancet. 1997;349:1436–1442. [DOI] [PubMed] [Google Scholar]
- 21.Schroeder DG, Brown KH. Nutritional status as a predictor of child survival: summarizing the association and quantifying its global impact. Bull World Health Organ. 1994;72:569–579. [PMC free article] [PubMed] [Google Scholar]
- 22.Chatterton R, Pamela D, Aldag C, Hodges K, Belknap S, Zinaman M. Relation of plasma oxytocin and prolactin concentrations to milk production in mothers of preterm infants: influence of stress. J Clin Endocrinol Metab. 2000;85:3661–3668. [DOI] [PubMed] [Google Scholar]
- 23.Grajeda R, Pérez-Escamilla R. Stress during labor and delivery is associated with delayed onset of lactation among urban Guatemalan women. J Nutr. 2002; 132:3055–3060. [DOI] [PubMed] [Google Scholar]
- 24.UNICEF, Alto Comisionado por los Derechos Humanos. Documento Integro de la Convención Sobre los Derechos del Niño Aprobada por la Ley 16.137: 28 de Septiembre de 1990. Geneva, Switzerland: UNICEF; 1990:8.
- 25.United Nations Committee on Economic, Social, and Cultural Rights. General Comment 14: The Right to the Highest Attainable Standard of Health: 11/08/2000. Geneva, Switzerland: UNCESCR (United Nations Committee on Economic, Social, and Cultural Rights); 2000. E/C12/2000.4.