The black/white mortality crossover: investigation in a community-based study

Abstract

The black/white mortality crossover at about age 75, a result of lower white mortality rates at younger ages and lower black rates at the oldest ages, has been observed in U.S. vital statistics since 1900. Though a persistant observation in such data, its validity has been challenged by questions about census enumeration and age reporting on death certificates. Analyses of 20 years experience of all-cause mortality in the community-based Evans County Study using a Weibull model of age specific mortality rates showed a statistically significant black/white mortality crossover for both men (at age 73) and women (at age 85). The finding of a crossover in this longitudinally followed population is significant because the age reporting for both survivors and age at death for nonsurvivors were obtained in the study protocol and did not rely on age reporting either in census data or on the death certificate. Differences in the age and sex patterns of mortality between two populations living in the same geographic region are relevant to questions about the etiology of the major age-related chronic diseases as well as to topics of current interest in health care policy.