Estimating Childhood Mortality in Ghana: Does Using Direct or Indirect Methods Matter?

Abstract

ABSTRACT There are essentially two main methods for estimating demographic indices, namely direct and indirect methods. The direct approach for estimation works well with reliable, timely, and adequate data. The lack of dependable data as a result of the poor functioning of vital registration systems and poor reporting of events in surveys in developing countries had spawned indirect approaches. Using data from the 2014 Ghana Demographic and Health Survey (GDHS), the study sought to compare childhood mortality estimates using direct and indirect methods. The MORTPAK software version 4.3 and STATA version 14 were used to analyse the data. The direct estimations were done using STATA version 14 and the indirect estimations were carried out using the QFIVE application based on MORTPAK software version 4.3. There were variations in estimations from direct and indirect methods. The results showed that under-five and infant mortality were higher, using both indirect and direct techniques for males compared to females. Estimations from indirect methods were higher than those from direct estimations. The Palloni-Heligman version of the Brass method should be used to estimate infant and under-five mortalities in Ghana. This can help fill the gap of under-reporting of infant and under-five mortality.