Perinatal and postneonatal mortality among Indigenous and non-indigenous infants born in Western Australia, 1980-1998

Objective: To describe cause-specific perinatal and postneonatal mortality for Indigenous and non-Indigenous infants using a new classification system. Design: Total population retrospective cohort study. Participants and setting: All registered births in Western Australia of birthweight greater than 399 g from 1980 to 1998, inclusive. Main outcome measures: Rates and time trends for all births.1980-1998, and cause-specific rates for births 1980-1993 of fetal, neonatal and postneonatal and childhood deaths. Results: For indigenous infants born 1980-1998, the mortality rate before the first birthday was 2.7 times (95% Cl, 2.5-2.9 times) that for non-Indigenous infants. Indigenous infants born 1980-1993 had a higher mortality rate in all cause-of-death categories. The highest relative risk was for death attributable to infection (8.1; 95% Cl, 6.5-10.0) which occured primarily in the postneonatal period; the source of the infection was less likely to be identified in Indigenpous deaths. From 1980-1998, the rate of neonatal deaths decreased at a greaterrate for Indigenous than for non-Indigenous infants. However, while stillbirth and sudden infant death syndrome rates for non-Indigenous births fell, they remained static for Indigenous births. Conclusions: The new classification system, which considers the underlying rather than immediate cause of death, enables investigation of the causes of all deaths, from stillbirths to childbirth. This system has highlighted the comparative importance of infection as a cause of death for Indigenous infants, particularly in the postneonatal period.