Temporal variations in United States infant death from perinatal related causes, by race, birth weight and time period of birth

Objectives. This population-based study was conducted to examine the effects of various technologies, that evolved in the late 1980s and early 1990s to improve infant survival, on time to death. The study examined the temporal variations in time to infant death from perinatal related causes, at varying birth weights and for two races, between 1985 and 1995, and considered the implications of the findings for future perinatal policy and interventions.;Methods. Non-experimental, retrospective, population based design, utilizing US national and state secondary vital records data linking births with infant deaths from 1985 to 1995. The outcome variable was time to infant death from perinatal related causes. Two birth weight cohorts (500--1499 grams and 500 to 6000 grams), and two race cohorts (infants born to Black mothers and infants born to White mothers) were utilized. Mortality frequencies, distributions and rates, percentages of time-to-death, and total days to death for all infants from perinatal related causes were calculated. Logistic regression analyses to estimate risk of infant death and survival analysis to estimate 'time to an event' (death), were conducted.;Results. The study found three major findings. The first concerned mortality changes. Mortality decreased for both races and both birth weight cohorts, especially between 1990 and 1995. Mortality changes differentially improved for infants born to White women, and the racial disparity gap increased for neonatal and postneonatal mortality in White favor. The second major finding concerned 'postponement' of death. There was postponement of potential neonatal deaths from the neonatal period to the postneonatal period between 1990 and 1995, especially for all birth weight infants and infants born to Black women. The third major finding concerned total number of days to death for all infants. Although very low birth weight infants and infants born to Black women experienced much lower total number of days to death, infants born to White women experienced a much larger reduction of the number of days to death. There was a White infant and very low birth weight advantage.;Conclusions. Between 1985 and 1995, a time of changing technologies, mortality and total number of days to death decreased, death was 'postponed' for some infants, and there was a continuing, but changing, racial disparity gap. This may have policy and fiscal implications for maternal and child health leadership, health care delivery organizations, resource utilization, outcomes research, and clinical care.