Neonatal Mortality in the Cape Town Metro West Geographical Service Area 2014-2017

Abstract

Background Each neonatal death counts, as recognised by the Every Newborn Action Plan (ENAP). This is an important aspect in attaining the third Sustainable Development Goal by 2030. Accurate neonatal mortality data as well as an understanding of the causality and context is essential to plan interventions to reduce neonatal deaths and attain the third Sustainable Development Goals (SDG) of a neonatal mortality rate of less than 12 per 1000 livebirths by 2035. Objectives The objectives of this study were: (i) to determine neonatal mortality occurring in and out of health facilities in the Metro West GSA using the three audit programmes; Perinatal Problem Identification Programme (PPIP), Child Healthcare Problem Identification Programme (Child PIP) and Forensic Pathology Services (ii) to ascertain the cause of death specific neonatal mortality (iii) to describe the avoidable factors in each death as coded by the three audit programmes (iv) to make recommendations for the alignment of existing audit databases to obtain accurate neonatal statistics for the Metro West GSA. Methods This was a retrospective descriptive study of neonatal deaths undertaken in the public healthcare setting in the Cape Town Metro West GSA from January 2014 till December 2017. Existing data from PPIP, Child PIP and the CDR/FPS was used. Neonatal deaths were defined as in the first 28 days of life where there had been signs of life at delivery and a birthweight greater than 500g. Neonatal deaths were excluded where birth had occurred outside of the GSA or in the private health care setting. The audit data with regards to cause of death and avoidable or modifiable factors was obtained for each death. Results From a total of 134843 live deliveries, 1243 neonatal deaths were identified: 976(78%) from PPIP, 58(5%) from Child PIP and 209 (17%) from CDR/FPS. Sixteen per cent of the deaths occurred outside of healthcare facilities. The neonatal mortality rate (NMR) for PPIP was 7.2, Child PIP 0.43 and CDR 1,6 per 1000 livebirths. When the audit systems were combined, the annual NMR over the study period varied from 8.05 to 10.1 with a mean of 9.2 per 1000 livebirths over the entire period. Seventy-eight per cent of the deaths occurred in the early neonatal period with a mean early neonatal mortality rate of 7.2 per 1000 livebirths. The mean late NMR was 2 per 1000 livebirths. Where all neonatal deaths were considered for those more than 500g, the main cause of death was immaturity related, then infection related followed by congenital disorders and then hypoxia related. Seventy-four per cent of deaths occurred in those less than 2500g at birth and 41% were less than 1000g and defined as extremely low birthweight. In the group of neonates greater than 1000g, the main cause of death was infection related deaths, closely followed by congenital disorders and then hypoxia, followed by immaturity. Most of infection related deaths were collected by the CDR and Child PIP. A third of Child PIP and PPIP deaths and half of the CDR deaths were coded as avoidable. The prevalence of deaths due to abandonment either by passive or active neonaticide contributed towards the higher proportion of preventable deaths in the CDR group. Conclusions The burden of deaths due to immaturity is high and may be attributed to the finding that 41% of neonatal deaths were in the ELBW group. Current viability criteria that aim at optimum use of resources may improve survival amongst this group. Infection related deaths were shown by this study to have a greater burden than recorded from PPIP data; most of these deaths were derived from Child PIP and CDR data. Also, where 10% of neonatal deaths were sudden unexpected deaths (SUDIs), a better understanding and definition of this group is urgently required as many of these deaths were subsequently found to be secondary to lower respiratory infections. It is further relevant that where 20% of CDR deaths or 3% of all the study deaths were due to active and passive neonaticide, this entity should be monitored and investigated. The study showed that the GSA has achieved the SDG for NMR of less than 12 per 1000 livebirth. However, a mean NMR of 9.2 per 1000 livebirths is not comparable to other upper middle-income countries. As 38% of the deaths were coded as avoidable, appropriate programmes to address these factors could reduce the NMR to 5.7 per 1000 livebirths. A strong recommendation from this study would be to use all three audit systems to calculate the NMR, understand the causes of neonatal deaths and plan programmes to improve neonatal survival in this GSA.