PURPOSE: Life expectancy for black Americans is five to eight years less th
an for Whites. The socioeconomic status (SES) of Blacks is also less than f
or Whites, and SES is associated with early mortality. This paper estimates
the proportion of the racial difference in mortality attributable to SES b
y specific causes of death.
METHODS: Data on 453,384 individuals in the National Longitudinal Mortality
Study were used to estimate the hazard ratio associated with black race, w
ith and without adjustment for income and education (measures of SES), in 3
8 strata defined by cause of death and age.
RESULTS: For women, SES accounted for much (37-67%) of the black excess mor
tality for accidents, ischemic heart disease (ages 35-54), diabetes, and ho
micide; but not for hypertension, infections, and stomach cancers (11-17%).
For men, SES accounted fur much of the excess risk (30-55%) for accidents,
lune cancer, stomach cancer, stroke, and homicide; but not for prostate ca
ncer, pulmonary diseases, hypertension, and cardiomyopathy (0-17%).
CONCLUSIONS: These results confirm those specific causes of death likely to
underlie the overall excess mortality of Blacks, and identify those causes
where SES may play a large role. (C) 2000 Elsevier Science Inc. All rights
reserved.