A multi-country examination of the associations between the anthropometric indicators of wasting and stunting
Description
For the most part, the empirical evidence suggests that wasting and stunting are different phenomena, but this issue remains controversial. Both are used as indicators of undernutrition. While previous studies have explored the relationship of wasting and stunting, there has been inadequate work in this area using standardized data from several countries with individual level measures that allow potential predictors to be analyzed and compared for each of the indices DHS (Demographic and Health Surveys) data from 27 countries from 3 world regions are used. DHS are cross-sectional and use a self-weighting, two-stage, probability sample. Interviews are conducted with one women between the ages of 15 and 49 within each household. The analysis was conducted on 62,059 children between 3 and 35 months of age Linear models using forward stepwise regression were developed for outcomes of height-for-age and weight-for-height stratified on age and urban/rural status. Linear models using correlations between height-for-age and weight-for-height for each country were developed using country-level predictors. Using the correlations, hypotheses of equality among the correlations between the countries and between the age groups were tested. Forward stepwise regression procedures were used to develop models for the outcomes of height-for-age and weight-for-height stratified on countries with low, medium, and high prevalences of wasting The correlations between weight-for-height and height-for-age are not strong. The strength of relationship between wasting and stunting for the age groups is stronger than for the country groupings. An increasing trend is seen over age Prevalence of stunting is higher than that of wasting. The pattern of wasting does not follow the same pattern as stunting. Several predictors were consistent with the UNICEF-based framework for child malnutrition. Age of child, educational level of the mother, child illness and child health are important predictors to both outcomes. Socioeconomic status, a household where one or more sons or daughters have died, prenatal care, number of children under five years in a household and knowledge of contraception are important predictors of stunting. A mother's work status, prevalence of wasting, occupation in agriculture, diarrheal history and duration a child was breastfed are important predictors of wasting