Neonatal near-misses in Ghana: a prospective, observational, multi-center study
Date
2019-12-23
Journal Title
Journal ISSN
Volume Title
Publisher
BMC Pediatrics
Abstract
Background: For every newborn who dies within the first month, as many as eight more suffer life-threatening
complications but survive (termed ‘neonatal near-misses’ (NNM)). However, there is no universally agreed-upon
definition or assessment tool for NNM. This study sought to describe the development of the Neonatal Near-Miss
Assessment Tool (NNMAT) for low-resource settings, as well as findings when implemented in Ghana.
Methods: This prospective, observational study was conducted at two tertiary care hospitals in southern Ghana
from April – July 2015. Newborns with evidence of complications and those admitted to the NICUs were screened
for inclusion using the NNMAT. Incidence of suspected NNM at enrollment and confirmed near-miss (surviving to
28 days) was determined and compared against institutional neonatal mortality rates. Suspected NNM cases were
compared with newborns not classified as a suspected near-miss, and all were followed to 28 days to determine
odds of survival. Confirmed near-misses were those identified as suspected near-misses at enrollment who survived
to 28 days. The main outcome measures were incidence of NNM, NNM:mortality ratio, and factors associated with
NNM classification.
Results: Out of 394 newborns with complications, 341 (86.5%) were initially classified as suspected near-misses at
enrollment using the NNMAT, with 53 (13.4%) being classified as a non-near-miss. At 28-day follow-up, 68 (17%)
had died, 52 (13%) were classified as a non-near-miss, and 274 were considered confirmed near-misses. Those
newborns with complications who were classified as suspected near-misses using the NNMAT at enrollment had 12
times the odds of dying before 28 days than those classified as non-near-misses. While most confirmed near-misses
qualified as NNM via intervention-based criteria, nearly two-thirds qualified based on two or more of the four
NNMAT categories. When disaggregated, the most predictive elements of the NNMAT were gestational age < 33
weeks, neurologic dysfunction, respiratory dysfunction, and hemoglobin < 10 gd/dl. The ratio of near-misses to
deaths was 0.55: 1, yet this varied across the study sites.
Conclusions: This research suggests that the NNMAT is an effective tool for assessing neonatal near-misses in lowresource
settings. We believe this approach has significant systems-level, continuous quality improvement, clinical
and policy-level implications.
Description
Research Article
Keywords
Neonatal morbidity, Neonatal mortality, Neonatal near-miss indicators, Neonatal Near-Miss Assessment Tool (NNMAT)