Linkages among sibling survival and death clustering in Nigeria: a panel data analysis

Rahul Mishra, International Institute for Population Sciences (IIPS)
Mukesh Ranjan, International Institute for Population Sciences (IIPS)

This paper assess, firstly the extent of death clustering and secondly the effect of previous child survival on index child across the various regions of Nigeria. The NDHS data is used to create a panel data from retrospective child birth history. Further, dynamic mixed effect model was applied to capture the effect of previous sibling’s and mother level unobserved heterogeneity. Findings reveals that, merely 1% of the families with neonatal deaths and 3% of the families with infant deaths, contributes nearly 20% and 30% of deaths respectively. Probabilities of infant deaths clustering is highest in North-East region and lowest in South-West region. In South-South region, the model based calculated risk of neonatal death, shows that, if there was prior neonatal death in a family, then the probability of death for index neonate is seven times more likely, than neonates from families with no previous neonatal death.