Neonatal and postneonatal mortality: useful constructs or outdated concepts?

Infant deaths are traditionally divided into neonatal and postneonatal categories by age at death. In the past neonatal deaths were largely caused by perinatal and biologic conditions (endogenous causes) and postneonatal deaths resulted from nonperinatal conditions, injuries, and socioenvironmental causes (exogenous causes). This study explores the congruence between neonatal-endogenous, and postneonatal-exogenous causes of infant death, with underlying cause of death data for the Wisconsin resident birth cohorts from 1979 to 1986. Although birth weight-specific neonatal and postneonatal mortality rates within cause of death categories remained stable during the 1980s, differences in the distribution of endogenous and exogenous infant deaths by cause of death suggest that the neonatal-postneonatal classification is no longer a useful proxy. Although neonatal and endogenous infant mortality remain fairly synonymous, postneonatal deaths are no longer equivalent to deaths of exogenous causation. Alternatives to use of days of life for classifying infant deaths as neonatal or postneonatal include dating infant deaths by corrected gestational age, and mortality rate during the hospital stay. These approaches should be evaluated as alternatives to the neonatal-postneonatal classification.