Differing postneonatal mortality rates of African-American and white infants in Chicago: an ecologic study.

OBJECTIVES: This study sought to determine whether neighborhood impoverishment explains the racial disparity in urban postneonatal mortality rates. METHODS: Stratified and multivariate logistic regression analyses were performed on the vital records of all African-Americans and whites born in Chicago by means of a linked 1992-1995 computerized birth-death file with appended 1990 U.S. census income and 1995 Chicago Department of Public Health data. Four community-level variables (low median family income, high rates of unemployment, homicide, and lead poisoning) were analyzed. Communities with one or more ecologic risk factors were classified as impoverished. RESULTS: The postneonatal mortality rate of African-Americans (N = 104,656) was 7.5/1000 compared to 2.7/1000 for whites (N = 52,954); relative risk (95% confidence interval) equaled 2.8 (2.3-3.3). Seventy-nine percent of African-American infants compared to 9% of white infants resided in impoverished neighborhoods; p < 0.01. In impoverished neighborhoods, the adjusted odds ratio (controlling for infant and maternal individual-level risk factors) of postneonatal mortality for African-American infants equaled 1.5 (0.5-4.2). In nonimpoverished neighborhoods, the adjusted odds ratio of postneonatal mortality for African-American infants equaled 1.8 (1.1-2.9). CONCLUSIONS: We conclude that urban African-American infants who reside in nonimpoverished neighborhoods are at high risk for postneonatal mortality.