Relation of age and race with hospital death after acute myocardial infarction

Abstract

BACKGROUND: Prior studies have suggested that young blacks with acute myocardial infarction (AMI) may have higher hospital mortality rates than whites of similar age. However, the influence of age and race on short-term death has not been explored in detail. We examined the relation of age and race on short-term death in a large AMI population and ascertained the factors that may have contributed to differences in mortality rates. METHODS: We compared the crude and adjusted hospital mortality rates stratified by age among 40,903 blacks and 501,995 whites with AMI enrolled in the National Registry of Myocardial Infarction-2 in 1482 participating US hospitals from June 1994 through March 1998. RESULTS: Overall crude mortality was lower among blacks compared with whites (10.9% vs 12.0%, P <.0001). However, blacks had a significantly higher crude mortality rate compared with the whites in the age groups <65 years (<45 years, and 5-year age groups between 45 and 64 years). There was a statistically significant interaction between age and black race on hospital death (P value for interaction <.001). Each 5-year decrement in age from 85 years was associated with 7.2% higher odds of death in blacks compared with whites (95% CI, 5.7% to 7.6%). After adjusting for differences in the baseline, clinical presentation, early treatment, and hospital characteristics, 5-year decrements in age was still associated with increases in the odds for death in blacks compared with whites (5.4%; 95% CI, 3.6% to 7.2%). This interaction between age and black race was present in both sexes but was stronger among men. CONCLUSIONS: Blacks younger than 65 years had higher hospital mortality rates compared with whites hospitalized for AMI, and decreasing age was associated with progressively higher risk of hospital death for blacks. Differences in the clinical presentation, early treatment, and hospital characteristics could only partly explain this age-race interaction.