PREDICTORS OF THE DOUBLE BURDEN OF MALNUTRITION AMONG WOMEN OF REPRODUCTIVE AGE AND PRESCHOOL CHILDREN IN SUB-SAHARAN AFRICA: A CASE STUDY OF FOUR COUNTRIES

Jason Mulimba Were Mr, The University of Western Ontario

Abstract

There is only sparse research in Sub-Saharan Africa (SSA) that focuses on the double burden of malnutrition (DBM) among the most vulnerable population including women and infants. Therefore, the objectives of this thesis were (1) to describe the current state of knowledge regarding the DBM among preschool children and women of reproductive age in low- and middle-income countries, (2) to determine the patterns and predictors of the household level DBM in mother-child dyads in SSA, and (3) To determine the patterns and predictors of the individual level DBM among women of reproductive age in SSA. A scoping review and thematic analysis of the literature was done to examine the first objective. Theoretically- driven secondary data analyses from Cameroon, Nigeria, Rwanda, and Zimbabwe were performed to explore the second and third objectives. Data were obtained from the Demographic and Health Survey program. Our scoping review revealed existence of varied phenotypes operationally defining DBM, most of which used anthropometric measures as indicators for nutrition status. The following themes emerged as plausible mechanisms for DBM development: nutrition transition, breastfeeding, diet behaviour, biological mechanism, and statistical artifact. Data analyses for objective 2 indicated that the prevalence of household DBM ranged from 7.6–7.8%. Predictors for higher odds of DBM across all the countries were increasing child and mother’s ages, male child, maternal short stature, higher parity, short interpregnancy interval whereas those for lower odds were higher mother’s education and breastfeeding. Parity and age at first birth were found to mediate the relationship between mother’s education and DBM. Finally, objective 3 findings revealed that the prevalence of individual level DBM ranged from 3.1% (Rwanda) to 16.5% (Nigeria). Predictors for higher odds of DBM were age, higher parity, lower age at first birth, short stature, being married/formerly married, being a Christian or Muslim, higher education, higher household wealth, higher neighbourhood wealth. Parity and household wealth were found to be key mediators in the association between neighborhood variables and DBM. This thesis reveals that DBM may represent the current and possibly future nutrition burden in SSA and it requires creative ways to address the challenge it portends.