A New Approach to Indirect Estimation of Child Mortality: Application to Malawi

Agbessi Amouzou, Johns Hopkins Bloomberg School of Public Health
Kenneth Hill, Harvard University

Information from women’s summary birth history consisting of only two questions - one on number of children ever borne alive and the other on children dead – is used traditionally for indirect child mortality estimation. However the estimation is based on several assumptions about fertility and mortality patterns and recent rates computed are biased. We propose and apply an innovative approach based on imputation of an existing full birth history onto summary birth history data using the summary birth history information. The resulting imputed full birth history is used to calculate child mortality rates using standard life table procedures. We applied the approach to the Malawi 2008 Population Census and the 2004 and 2010 demographic and health survey datasets. Preliminary results show very good promise of the approach with the imputed child mortality rates falling within 20% of less of the rates directly computed from the 2010 DHS survey.