Impact of Child Health Interventions on Under-Five Mortality in Ghana, Evidence from 2008 and 2014 Ghana Demographic and Health Surveys

Abstract

Background: Although under-five mortality (U5M) decreased worldwide and in Ghana in the era of the Millennium Development Goals (MDGs), the decline in Ghana has not been parallel with the level of interventions in maternal and child health. Between 1990 and 2015, there was a 53% decline in mortality among children underfive years globally. Ghana achieved a 50% decline in mortality (119/1,000 live births to 60/1,000 live births) between 1993 and 2014. However, about 6 million under-five deaths were recorded globally in 2015. Under-five mortality decline in Ghana has been considered slow compared to countries at the global level even though there has been an increase in coverage of some of the 26 listed interventions advocated for improving child survival in Ghana. What was not known was the contribution of these interventions to mortality reduction which this study sought to determine. Objective: The objective of the study was to assess the impact of child health interventions on under-five mortality in Ghana. Methods: Secondary data of the 2008 and 2014 Ghana Demographic and Health Surveys (GDHS) were analysed for this study. The main data sets comprised 2,992 and 5,884 observations of children under-five years from the 2008 and 2014 data sets respectively. Coarsened Exact matching (CEM) with logistic and Poisson regressions and Lives Saved Tool (LiST) were used for the impact assessment, while logistic and Poisson regressions were fitted to assess the association between these interventions and under-five mortality. The 2008 and 2014 data sets were pooled for the logistic regression and CEM analysis. Results: There were 6,098 children under-five years and 93 (1.5%) died. Among the children who died, 47 (47.0%) were less than one month old. Fifty-six (56), representing 65.7% of children who died were born to mothers below 35 years. Among the interventions, antenatal care visits coverage level was the highest (84.0%), while water connection in the home had the lowest coverage level (8.1%). About 58 (1.4%) of all children received all eight (8) interventions evaluated at the individual level, and none of those who received all the eight interventions died. Early initiation of breastfeeding reduced odds of death by 58% (aOR = 0.42, 95% CI: 0.2 - 0.8), while clean postnatal care caused a 59% reduction in the odds of death (aOR = 0.41, 95%CI: 0.2-0.9). Interventions that saved the most lives among children under-five years at the population level were malaria control interventions including insecticide treated net and or indoor residual spraying (ITN/IRS) (8,524 lives saved, 16% mortality rate reduction) and artemisinin-based combination therapy (ACTs) (5,702 lives saved, 10% mortality rate reduction), labour and delivery management (skilled delivery) (4,726 lives saved, 8% mortality rate reduction) and pneumococcal vaccine (2,406 lives saved, 8% mortality rate reduction). Reduction in the prevalence of wasting saved 11,918 lives and contributed to a 19% reduction of mortality rate while, reduction in the prevalence of stunting also saved 5,761 lives and contributed to an 11% reduction in under-five mortality rate. However, complementary feeding targeted at reducing mortality via reduction in stunting (-457 lives saved, -1% mortality rate reduction) and wasting (-62 lives saved, 0% mortality rate reduction) resulted in negative lives saved (excess deaths/additional deaths) between 2008 and 2014. Conclusion: Only two (2), interventions caused mortality reduction at the individual level. A further decline in under-five mortality in Ghana will require increase in the coverage levels of the few high impact interventions, especially those with low coverage levels. Neonatal level interventions should be prioritized, since neonatal mortality decline is slow and the proportion of neonatal deaths is on the increase in Ghana as at the global level. Additionally, attention should be paid to other children at higher risk of dying including multiple births and children from polygamous homes. Strategies to reduce the prevalence of stunting and wasting would also be beneficial.