Whittle J, Conigliaro J, Good CB, Lofgren RP.
Racial differences in the use of invasive cardiovascular procedures in the Department of Veterans Affairs medical system.
N Engl J Med 1993;329(9):621-7. (Comment in N Engl J Med 1993;329(9):656-8.)

This study assessed the use of cardiac catheterization, coronary artery angioplasty (PTCA), and coronary artery bypass grafting (CABG) among white and black male veterans treated at Veterans Affairs (VA) hospitals. Records of hospitalizations from the Patient Treatment File for 1987 through 1991 (fiscal years) were used. Patients over 30 years of age with a primary discharge diagnosis of cardiovascular disease or chest pain were included. The first such discharge during the study period was considered the index hospitalization for this study.

Cardiac catheterization was performed during or within 60 days of the index admission for 18% of this group; the rate was 19.3% for whites and 11.8% for blacks (RR of catheterization for whites versus blacks=1.64, 95% confidence interval=1.61 to 1.68). For PTCA, 1.8% of whites and 0.8% of blacks underwent this procedure (RR of PTCA for whites versus blacks=2.18, 95% confidence interval=2.01 to 2/37). For CABG, 5% of whites and 1.6 % of blacks underwent this procedure (RR of CABG for whites versus blacks=3.17, 95% confidence interval = 2.99 to 3.37).

Adjusting for demographic characteristics, region, year of discharge, coexisting conditions, diagnosis, and whether the facility was equipped to perform CABG, white race remained a statistically significant predictor of the use of each of the procedures. The adjusted OR for white versus black race was 1.38 for catheterization, 1.50 for PCTA, and 2.22 for CABG. When the analyses were controlled for site of index hospitalization (CABG facility availability versus no facility), excluded nonspecific diagnoses (such as other cardiovascular diseases and chest pain), and were restricted to veterans with limited incomes, the white-black ratios remained statistically significant.

The authors conclude “our data, coupled with the results of previous studies, suggest the existence of race-related inequities in our health care system.”

The authors were able to argue against two explanations previously offered for racial differences in rates of procedures based on these data. First, the racial differences cannot be explained by differences in patient incomes because the current study was conducted in a VA system where differential access to care and provider incentives to use expensive procedures do not exist. Also, the sub-analysis of those with more homogenous incomes in this study was similar to the main analysis. Second, a lower rate of procedure use in the hospitals that treat more blacks was ruled out because the racial pattern was consistent across hospitals in this study, even though the rate of procedure use varied widely.

Alternative explanations offered include: differences in severity of coronary artery disease, unmeasured differences in the presence or severity of coexisting conditions, cultural differences in attitudes toward procedures or medical care in general, differences in access to care, and systematic racial bias. “More than one, and perhaps all, of these factors may be at work.”

For example, the authors note that racial differences were lowest for catheterization and greatest for CABG. They state that “if blacks have less severe coronary artery disease than whites, lower rates of PTCA and CABF among black patient would be appropriate.” They also state that “if blacks are less likely than whites to accept recommendations that they undergo invasive procedures, this reluctance might be greater for CABG and PTCA than for cardiac catheterization.” These suppositions are supported by previous data as well as findings from the current analysis. Finally, the authors argue that “any subtle bias against the use of invasive procedures in the treatment of black patients might be most evident when decision making is less clearly dictated by the clinical situation. Expert physicians participating in the RAND study of the appropriateness of care were more likely to disagree on the appropriateness of CABG than on the appropriateness of cardiac catheterization. Thus there may be more room for clinical judgment in the use of CABG.”

The authors conclude that “the extent to which subtle or overt racism underlies racial differences in the use of cardiac procedures is unclear. We believe that inadequate health education, differences in patient’s preferences for invasive management, delivery systems that are unfriendly to members of certain cultures, and overt racism may all play a part.”