Spatio-temporal modelling of severity of malnutrition and its associated risk factors among under five children in Nigeria between 2003 and 2018: Bayesian multilevel structured additive regressions

Malnutrition among under five children in Nigeria remains a major public health burden. It has been shown to be one of the leading causes of childhood death and has permanent consequences for cognitive development, school attendance, economic productivity in adulthood and maternal reproductive outcome. Understanding the associated risk factors of malnutrition is critical for developing appropriate interventions because its determinants interact at different degrees to affect these children. Thus, it is important, for programmatic purposes, to continue to identify the important risk factors and geographical variations especially in a country with massive variations in cultural belief and socioeconomic status of its populace. A multilevel structured additive regression, based on cumulative probit link function was adopted to explore the spatio-temporal variations on severity of three anthropometric indices among children in Nigeria using the Nigeria Demographic and Health Survey data from 2003 to 2018. Within a Bayesian context, appropriate priors were assigned on all functions and parameters. Evidently, strong and significant spatial variations was found to exist on severity of malnutrition among under five children in Nigeria showing a north–south divide. We found young age at first marriage, episodes of child illnesses, child birth weight, religion, mother’s education and household wealth to be associated with undernutrition. Findings will guide in developing effective strategies to combat the devastating effect of child malnutrition in Nigeria. Consequently, there can be hope of attaining the relevant sustainable development goals.