A dramatic reduction in child mortality has been noted in Ghana in recent time and although the trend
has been projected to continue to decline, studies have shown that the decline in under-five mortality has stalled and
the effect more outstanding at different levels. Here we examine the issue of under five mortality using data from the
demographic and health survey by examining the trends in under-five mortality between 1988-2014 and develop a
predictive model to identify risk factors of under five mortality. Data from the Ghana Demographic and Health
survey between 1988-2014 was used to conduct a retrospective analysis of recorded children data (0-5 years). A
descriptive analysis of the explanatory and response variables was carried out followed by a subsequent trend
analysis of under-five mortalities from 1988-2014. A logistic regression was used to predict the association between
the explanatory variables and the dependent outcome variable. The study established that the decrease in under five
mortality that Ghana witnessed from 1988-2014 was not significant and that more measures need to be put in place
to reduce the phenomenon. Variables such as maternal age (R2
=0.0646, p<0.05), duration of breastfeeding
(R2
=0.3300, p<0.05) and the type of assistance at delivery (R2
=0.2898, p<0.05) were all strong predictors of under
five mortality relative to other explanatory variables. The study identified that issues such as young maternal age,
exclusive breast feeding beyond 6 months and unskilled or no assistance at birth were all contributors of under five
mortality in Ghana and as such it is necessary for all these predictors to be addressed and factored during
formulation and implementation of policies in an effort to curb under five mortality.