Observations from Mortality Trends at The Children's Hospital, Accra, 2003-2013.

objectiveFacility-based studies provide an unparalleled opportunity to assess interventions deployed in hospitals to reduce child mortality which is not easily captured in the national data. We examined mortality trends at the Princess Marie Louise Children's Hospital (PML) and related it to interventions deployed in the hospital and community to reduce child mortality and achieve the Millennium Development Goal 4 (MDG 4).

methodsThe study was a cross-sectional review of data on consecutive patients who died at the hospital over a period of 11 years, between 2003 and 2013. The total admissions for each year, the major hospital-based and population-based interventions, which took place within the period, were also obtained.

resultsOut of a total of 37,012 admissions, 1,314 (3.6%) deaths occurred and admissions tripled during the period. The average annual change in mortality was -7.12% overall, -7.38% in under-fives, and -1.47% in children ≥5 years. The majority of the deaths, 1,187 (90.3%), occurred in under-fives. The observed decrease in under-five (and overall) mortality rate occurred in a specific and peculiar pattern. Most of the decrease occurred during the period between 2003 and 2006. After that there was a noticeable increase from 2006 to 2008. Then, the rate slowly decreased until the end of the study period in 2013. There was a concomitant decline in malaria mortality following a pattern similar to the decline observed in other parts of the continent during this period. Several interventions might have contributed to the reduction in mortality including the change in malaria treatment policy, improved treatment of malnutrition and increasing paediatric input.

conclusionUnder-fives mortality at PML has declined considerably; however, the reduction in mortality in older children has been minimal and thus requires special attention. Data collection for mortality reviews should be planned and commissioned regularly in hospitals to assess the effects of interventions and understand the context in which they occur. This will provide benchmarks and an impetus for improving care, identify shortfalls and ensure that the gains in child survival are maintained.