Neighborhood effect, poverty and maternal nutritional status as determinants of child's weight at birth: A Bayesian approach

Background: Child's weight at birth has been shown to be associated with child's and maternal health; which in turn could be a determining factor of maternal and child mortality before, during and after birth. Influence of poverty, maternal status and geographical locations on child's weight at birth are explored.

Methodology: Child's weight at birth was measured on a five-level ordinal scale in Nigeria DHS. As a consequence, analysis based on traditional common regression techniques could not be employed. Statistical models that consider such ordering and adjusting for spatial and nonlinear effects are rarely used. This paper, however, explores possible influence of household poverty level and maternal nutritional status in the presence of geographical variations on child's weight at birth. In this paper, weight at birth is measured as small, average and large. All estimations are based on Bayesian approach using restricted maximum likelihood technique.

Results: Significant non-linear effect of mother's age at birth, positive association with mother's educational attainment, residing in urban areas; and being a male child are all positively associated with child's weight at birth. Poverty and maternal nutritional status are significantly related to child's weight at birth. Substantial geographical variations exist as low birth weight is associated with Sokoto, Zamfara, Katsina, Jigawa, Bauchi, Plateau and Nasarawa states in the North. Niger, Kaduna, Kogi, Anambra and Imo states are associated with high birth weight.

Conclusion: The approach provides policy makers with useful tools to designing appropriate health interventions in order to address the problem of child's birth weight.