Good Care Practices Can Mitigate the Negative Effects of Poverty and Low Maternal Schooling on Children's Nutritional Status: Evidence from Accra

Care is increasingly recognized as an important determinant of good health and nutrition among preschoolers, along with food security, availability of health services, and a healthy environment. Urban life presents special challenges to the provision of adequate child care (defined as household and community time, attention, and support given to meet the physical, mental, and social needs of the child). The trade-offs mothers face between their productive, reproductive, and caring roles are likely to be more acute in urban than in rural areas because their employment often requires them to be away from home for long hours. Alternative child care may be less available than in rural areas where extended families are more common.

The present study looks beyond the constraints to child care and examines how good care practices affect child nutrition in a sample of 475 households in Accra. The study focuses on the care of children 4-36 months old as provided by the main caregiver. The caregiving behaviors surveyed are child feeding practices and the use of health services for preventive care (immunization and growth monitoring). The main research objectives include (1) determining whether a meaningful care index (a measure of the extent of child care) can be derived from a simple recall questionnaire about child feeding and the use of preventive health services in a cross-sectional survey; (2) investigating whether care practices (as proxied by the care index) are associated with children's nutritional status; and (3) identifying subgroups of children that may benefit more than others from good maternal care.

Main Findings
This is perhaps the first study that has quantified care practices into an index using cross-sectional survey data. And it may be the first to use appropriate statistical modeling to control for factors at the child, maternal, and household levels when examining the link between care practices and child nutritional status. An additional strength of the study is the multi-disciplinary nature of the team involved in planning the survey and analyzing the data.

Although the care index constructed for this study reflects only two of a large number of aspects of the overall concept of care, the results show that it is possible to measure care based on information about child-feeding practices and use of preventive health care from a simple recall interview, a large sample, and a cross-sectional survey. Children 0-4 months old, however, had to be excluded from the sample because of difficulties with the data. The generalizability of the findings, therefore, should be restricted to households with children between 4 and 36 months old.

The study shows that care practices are strong determinants of children's nutritional status, particularly for children from poorer households and children with mothers having less than a secondary school education.

Care practices are strong determinants of children's nutritional status, particularly for children from poorer households and children with mothers having less than a secondary school education.

In Accra almost three-quarters of mothers had less than a secondary education. Among this group, better maternal care practices brought the height-for-age Z-scores (HAZ, a measure of nutritional well-being) of children to the same level as those for children from wealthier families or children with more educated mothers. The magnitude of the improvement—about half a Z-score—is biologically important and equals or exceeds the differences typically found between socioeconomic groups, maternal education groups, or groups before and after success- ful nutrition interventions. Surprisingly, good care practices provided no additional benefit to children from more educated mothers and wealthier households.

A correlate finding is that less maternal schooling and low income have a negative effect on children's HAZ only if mothers were poor caregivers. Greater maternal schooling and higher income made hardly any difference in HAZ for children whose mothers had average or good care practices. Thus, in this population, good care practices could compensate for low maternal education and insufficient income.

The literature showing the importance of maternal schooling for child health, nutrition, and well-being is extensive. These studies il-lustrate that maternal schooling is associated with improved child nutrition, but not among households with insufficient resources. The mechanisms by which maternal schooling affects child outcomes, however, are poorly understood. The present study is one of the few that sheds light on this issue. It shows that the beneficial effect of maternal schooling on children's HAZ does act through care practices, but only for mothers with less than secondary schooling.

A study in rural Lesotho points out that the positive effect of maternal schooling is mediated by increased knowledge of nutrition (and probably better care practices), but only among households that have access to a minimum level of resources. Poor mothers cannot translate knowledge about nutrition into optimal child-care practices. The opposite is found in Accra where care practices (resulting from nutrition knowledge) have a greater impact on children's nutrition among poorer households and make no difference among the upper income tercile. Differences in absolute levels of poverty between Accra and rural Lesotho probably account for these contrasting results. The contrast highlights the importance of conducting context-specific analyses of this type before designing program and policy recommendations.

Other Results
Household income and calorie availability are not statistically significant determinants of child nutritional status when maternal schooling and caring practices are included in the models. In this sample, household income and calorie availability only affect children's nutritional status through maternal schooling and caring practices. Other studies also show that the effect of maternal education on child outcomes often exceeds the income effect. Evidence suggests that maternal schooling may be the prime determinant of nutritional status among young children (less than 24 months), but income may become more important as children grow older and have greater daily requirements of nutrients and other basic needs.

Quality of housing (including household assets) is positively and significantly associated with children's HAZ (in most tests), even when controlling for income, calorie availability, maternal schooling, and caring practices. It is possible that because quality of housing represents the long- term wealth and socioeconomic status of the family, it is more strongly associated with HAZ, which is also a measure of cumulative long-term nutritional status.

Maternal work, either part-time or full-time, is not associated with children's nutritional status. Mothers appear to adapt their work patterns to fit the specific needs of their young children.

Policy Implications
The main findings have important policy implications for children's health. They suggest that if poor mothers with little formal education were trained in child feeding and use of preventive health services, the impact on the growth of children living in impoverished environments would be large. Although continued effort to improve formal education of mothers is crucial, a complementary short-term approach could target specific messages to mothers with low levels of education. This could markedly reduce differences in height-for-age attributable to poverty and poor education.

The importance of maternal schooling came out clearly in the analysis. There is no variable easier to target or measure than formal schooling. It is subject to few biases due to recall or intent to withhold information. The recommendation would therefore be to try to target specific education messages to mothers with little or no schooling and to do so during pregnancy. Waiting until the mother takes the young infant to the health clinic for immunization is already too late because good child feeding practices have to start at birth. Successful, exclusive breast-feeding takes hold during the first few weeks of a child's life. The specific messages should also be included in the formal education curriculum and a special effort should be made to reach teenage girls even before their first pregnancy. A successful education strategy would have a large payoff in reducing malnutrition in Accra.

To order a discussion paper, please fill out an online order form, email IFPRI-FCN@cgiar.org or send requests to Food Consumption and Nutrition Division, IFPRI, 2033 K Street, N.W., Washington, D.C. 20006, U.S.A.