Global Estimation of Child Mortality Using a Bayesian B-Spline Bias-Reduction Method

Leontine Alkema, National University of Singapore
Jin Rou New, National University of Singapore

For the great majority of developing countries without well-functioning vital registration systems, estimating levels and trends in child mortality is challenging, not only because of limited data availability but also because of issues with data quality. We developed a Bayesian penalized B-spline regression model for assessing levels and trends in the under-five mortality rate (U5MR) for all countries in the world, whereby biases in data series are estimated through the inclusion of a multilevel model. This model was recently accepted by the United Nations Inter-agency Group for Child Mortality Estimation to measure countries' progress in reducing U5MR. In this paper, we present the model and the resulting estimates of the U5MR for selected countries.