Elsevier

Social Science & Medicine

Volume 62, Issue 2, January 2006, Pages 375-386
Social Science & Medicine

Social Epidemiology
Determinants of child morbidity in Latin America: A pooled analysis of interactions between parental education and economic status

https://doi.org/10.1016/j.socscimed.2005.06.007Get rights and content

Abstract

Diarrhea and respiratory infections account for more than two-fifths of all deaths among children under five. Parental education and economic status are well-known risk factors for child morbidity, but little is known about whether education and economic status operate synergistically or independently to influence children's health. Confirming the presence and direction of such interactions is important to better target education and development policies. Our objective is to test for interactions between parental education and economic status in predicting the risk of diarrhea and respiratory illness among children under five, before and after adjusting for key proximate risk factors.

We pool 12 Demographic and Health Surveys (DHS) and nine Living Standards Measurement Surveys (LSMS) from Latin America, creating two large databases. Quintiles of economic status are constructed from principal components asset indices. We use logistic regression to analyze episodes of diarrhea and respiratory illness, and interactions between economic quintile and maternal and paternal education are evaluated via likelihood ratio tests. We find that mother's education and quintile interact synergistically in the DHS data, while results are inconclusive in the LSMS data. The effect of increasing maternal education appears to be more protective for children in wealthy families than for children in poor families. Conversely, improvements in economic status reduce health risks more for children whose mothers are better educated. Father's education is protective and operates independently of economic status. Our findings imply that poverty alleviation efforts occurring in concert with programs to educate women and girls will be more effective for improving children's health than either approach alone.

Introduction

More than 10 million children under the age of five die every year (UNICEF, 2004). Diarrhea and respiratory infections, abetted by undernutrition, account for more than two-fifths of all deaths among children under five (Black, Morris, & Bryce, 2003). These diseases disproportionately affect poor children in low- and middle-income countries. Socioeconomic inequalities in the distribution of child mortality and morbidity are well documented, both among countries and regions of the world and within individual societies (Victora et al., 2003). For instance, under-five mortality rates in Nicaragua are almost three times higher among those in the poorest economic quintile than among those in the wealthiest quintile; in Brazil rates are six times higher among the poorest quintile (Wagstaff, 2000).

Many researchers have documented the paramount importance of socioeconomic factors for child health outcomes. An important question that has received less attention, however, is whether education and economic status operate synergistically or independently to influence children's risk of illness and death. Put differently, does a mother's education play the same protective role for children in poor households as it does for better-off children, or does the effect of education change with economic status? Confirming the presence and direction of an interaction between education and household wealth is important for effective targeting of education and development policies. A finding that there is a synergy between education and economic status, for instance, would imply that educational interventions alone would have limited effectiveness for improving child health in the absence of simultaneous economic development programs. In contrast, a significant negative interaction between economic status and education would imply that educating women in the lowest economic strata will have the greatest impact.

Among the few previous efforts to address this question, the evidence has been conflicting. In a cohort study of infant mortality in Nicaragua, Peña, Wall, and Persson (2000) found that the protective effect of maternal education was greater in poor households than nonpoor households. They hypothesize that education may empower poor women to better cope with harsh living conditions, while maternal education may provide little additional benefit to children in nonpoor households whose basic needs are already satisfied. Case studies from Kerala State in India provide evidence that significant investments in women's education can lead to very good child health outcomes even in resource-poor settings (Thankappan & Valiathan, 1998). In contrast, Dargent-Molina, James, Strogatz, and Savitz (1994) found that maternal education was protective only in economically and socially advantaged communities, but had no effect in the more disadvantaged communities. They posit that women's access to basic resources may be a necessary precondition for education to have the expected effect on child health. From a third perspective, Desai and Alva (1998) have argued that maternal education largely proxies for the family's socioeconomic status and area of residence, implying that education may not have an independent, causal effect. Others have also emphasized the importance of the contextual effects of local economic conditions for the association between maternal education and child health (Fotso & Kuate-Defo, 2005; Griffiths, Madise, Whitworth, & Matthews, 2004; Larrea & Kawachi, 2005).

In this article, we use readily available secondary data to clarify these associations and interactions at the household level. We pool household survey data from the Latin America and Caribbean region to analyze the socioeconomic determinants of diarrhea and respiratory infections among children under five, specifically testing for the presence of interactions between parental education and economic status. We hypothesize that the effects of both mothers’ and fathers’ education are likely to vary by level of economic status and proceed to identify the direction of this interaction.

Section snippets

Data

We analyzed 7 nationally representative Living Standards Measurement Surveys (LSMS), two national LSMS-type surveys, and 12 Demographic and Health Surveys (DHS) from countries across Latin America. After creating standardized variables, we grouped comparable national datasets into two pooled datasets, increasing our ability to detect underlying trends that are often obscured by noise in individual datasets.

The 21 household surveys included in our analysis were supported by national governments

Model

In their seminal conceptual framework, Mosley and Chen (1984) classified the determinants of child mortality into “proximate” factors (such as environmental contamination or nutrient deficiency), which contribute in an immediate and direct fashion to children's risk of mortality, and “distal” factors (economic status, education, cultural norms, and government policies), which influence exposure to the proximate factors. We base our analysis on a modified version of Mosley and Chen's framework,

Outcome variables

We investigate two binary child morbidity outcomes reported during the survey recall periods: (1) an episode of diarrhea and (2) a respiratory illness. For the respiratory outcome, the phrasing of survey questions forces inclusivity: the DHS questionnaires usually asked whether children had been “sick with a cough” during the past two weeks, while the LSMS questionnaires asked about a list of respiratory ailments including cough, cold, flu, bronchitis, pneumonia, and/or pertussis. We excluded

Univariate and bivariate results

Between one-fifth and one-third of children in the LSMS surveys reported an episode of diarrhea during the past month or two weeks (Table 1). In the DHS surveys, slightly lower rates were reported. Substantially higher rates of respiratory illness were reported in both types of surveys, ranging from one-third to more than one-half of children surveyed.

Table 2 compares the sample distributions of the key distal variables for the four pooled databases as well as bivariate results for both

Discussion

This analysis used logistic regression methods to identify the effect of economic status and parental education on children's risk of diarrhea and respiratory illness, and tested for the presence of interactions among these distal factors. We took advantage of a rare opportunity: a wealth of comparable, under-analyzed data from across Latin America and the Caribbean. When pooled, these datasets provided very large sample sizes and hence greater statistical power to reveal underlying trends that

Acknowledgments

This publication was made possible through support provided by the Office of Health, Infectious Diseases, and Nutrition, Global Health Bureau, US Agency for International Development, under the terms of Award No. HRN-A-00-96-90006-00 the Family Health and Child Survival Cooperative Agreement. The opinions expressed herein are those of the authors and do not necessarily reflect the view of the US Agency for International Development.

References (30)

  • C.G. Victora et al.

    Applying an equity lens to child health and mortality: More of the same is not enough

    Lancet

    (2003)
  • J.R. Cruz et al.

    Epidemiology of acute respiratory tract infections among Guatemalan ambulatory preschool children

    Reviews of Infectious Diseases

    (1990)
  • R. da Costa Lima et al.

    Do risk factors for childhood infections and malnutrition protect against asthma? A study of Brazilian male adolescents

    American Journal of Public Health

    (2003)
  • A. Deaton

    The analysis of household surveys: A microeconometric approach to development policy

    (1997)
  • S. Desai et al.

    Maternal education and child health: Is there a strong causal relationship?

    Demography

    (1998)
  • Cited by (88)

    • A machine learning logistic classifier approach for identifying the determinants of Under-5 child morbidity in Bangladesh

      2021, Clinical Epidemiology and Global Health
      Citation Excerpt :

      Stunting, wasting, and underweighted children have a severer risk of death due to superior forms of morbidity such as diarrhea, pneumonia, ARI, etc.11 Parental education and economic status remain well-known crucial critical factors of child morbidity.5,12,13 Higher parental academic attainment can contribute to healthier hygiene habits such as washing hands with soaps after toilet use, decreasing children's exposure to waterborne disease.12 There has been a significant connection between childhood morbidity status and better toilet facilities, water supply sources, and water treatment practices.5

    View all citing articles on Scopus
    View full text