A Poisson Regression Model to Examine Spatial Patterns in Antenatal Care Utilisation in Nigeria

ABSTRACT Despite the benefits of antenat]al care to the mother and the child, it is available less often in sub‐Saharan Africa than in other regions. Coverage of at least four visits is lower, with a slower rate of increase, than in other regions of the world. Using data from 2008 Nigeria Demographic and Health Survey, we applied a Poisson regression model with geo‐additive predictors to analyse individual‐level and community‐level determinants of antenatal care utilisation and quantify district specific spatial effect. The spatial structure was modelled using conditional autoregressive prior, whereas the metrical covariates were assumed to be non‐linear. Findings reveal substantial spatial variation with a distinct north–south divide in antenatal care utilisation in Nigeria. The effects of mother's age at birth, partner's age, and marital duration were non‐linear, and utilisation was found to differ significantly between never‐married and ever‐married women respondents. Results further show that older women, particularly those above 35 years old, utilise antenatal care services during pregnancy more than the younger ones. Policymakers and public health providers need to put in more efforts to address these differentials as a minimum of four visits for antenatal care during pregnancy is required by all women in all parts of the country. Allocations for antenatal care services should also take into account the differentials so that those states lagging behind can be brought up, at least to scale with others. Copyright © 2013 John Wiley & Sons, Ltd.