Spatial and Temporal Inequalities in Undernutrition Among Non-pregnant Women in Ghana: A Multilevel Analysis

Abstract

Nutritional status is an important determinant of women’s health and wellbeing. However, there appears to be a dearth of information on the spatial distribution of undernutrition among women in Ghana over time. We, therefore, examined the spatial and temporal variations in undernutrition among non-pregnant Ghanaian women. Drawing on data from the 2003, 2008, and 2014 Ghana demographic and health surveys, we used geospatial techniques to show spatial autocorrelation for undernutrition levels in the study sample. Multilevel logistic regression analysis was also applied to estimate sociodemographic and contextual factors underlying the variations in the country. The results show considerable local clusterings and regional variations in undernutrition levels in the sample coupled with steady improvements within the regions over time. There was also a notable north–south divide in the undernutrition levels among non-pregnant women in the country for the study period. The spatio-temporal patterns of undernutrition levels are explained by some individual-level sociodemographic factors such as age, marital status, educational attainment, household wealth, and water source as well as the percent of regional unmet need for family planning. Policies and interventions should focus on alleviating the observed regional disparities in undernutrition levels among women while targeting and tracking higher risk women in Ghana.