Philbin EF, DiSalvo TG.
Influence of race and gender on care process, resource use, and hospital-based outcomes in congestive heart failure.
Am J Cardiol 1998;82(1):76-81.

This study assessed administrative records for patients hospitalized for evaluation and treatment of congestive heart failure in order to examine the influence of race and gender on care processes, resource use, and clinical outcomes. Data included information on all patients discharged with a primary diagnosis of coronary heart failure (CHF) from New York State hospitals in 1995.

Among the 45,894 patients included in this analysis, 10.4% were black women, 7.3% were black men, 46.1% were white women, and 36.2% were white men. Black patients underwent noninvasive cardiac testing more often, but underwent coronary revascularization, any cardiac surgery, and pacemaker insertion slightly less often than did whites. When only the subset of patients with the diagnosis of coronary artery disease (CAD) was analyzed, blacks underwent more diagnostic cardiac catheterizations and exercise stress tests than whites. Other procedures, including coronary revascularization, occurred with equal frequency in blacks and whites.

Compared with men, women underwent fewer procedures and were treated less often by cardiologists. Most of these gender differences, with the exception of cardiology specialty care and cardiac catheterization, were more pronounced in the white group than in the black group.

Multivariate analysis was conducted to compare race and sex groups on resource use and clinical outcomes, adjusting for age, gender, race, hospital location (rural versus urban), hospital type (teaching versus nonteaching), Charlson Comorbidity Index, ischemic heart disease, ICU hospitalization, transfer, discharge to nursing home, and discharge against medical advise. Race was an important determinant of length of hospitalization (10.4 days for blacks and 9.3 days for whites), hospital charges ($13,711 for blacks and $11,074 for whites), mortality (OR for blacks versus whites=0.93; 95% confidence interval=0.74 to 0.94), and readmission rate (OR for blacks versus whites=1.30; 95% confidence interval=1.22 to 1.39). To explain these racial patterns, the authors tested the role of the following procedures: echocardiography, nuclear ventriculography, exercise stress testing, intravenous inotropic drug use, pacemaker implantation, and cardiology specialty care. Race remained a significant determinant of hospital charges after adjusting for the above procedures, suggesting that intergroup differences could not be attributed to differences in procedural rates. The inclusion of 'length of stay' in the models somewhat reduced the race effect on hospital charges. The inclusion of 'death' in the model also did not eliminate the race effect, suggesting that the charges of dying patients could not explain the race difference.

Race differences in length of stay and hospital readmission were statistically significant both in teaching and non-teaching hospitals. Blacks had significantly lower mortality than whites only in urban nonteaching hospitals and had significantly higher costs than whites only in urban teaching hospitals.

Gender was also an important predictor of length of stay, hospital charges, and mortality (women stayed in the hospital longer, were charged more, and were less likely to die).

Unlike previous studies, this study found that blacks had equal or higher rates of specialty and procedure use. The authors do not offer explanations for this unexpected finding, but do state that the racial differences in quality of care suggested by a difference in outcomes is not the result of inappropriately low levels of resource use in this study. The authors suggest that differences in the causes of CHF may partially account for the observed racial variation in readmission rates. For example, hypertensive heart disease is more common in blacks and the nature of this disease may result in recurrent hospitalization.

The authors conclude that the aggregate available evidence suggests a higher incidence of CHF among blacks but a potentially equal or lower case-fatality rate. Whether the latter is due to a lower prevalence of ischemic heart disease, less severe CHF, or other factors is unknown.