Carlisle DM, Leake BD, Shapiro MF.
Racial and ethnic differences in the use of invasive cardiac procedures among cardiac patients in Los Angeles County, 1986 through 1988.
Am J Public Health 1995;85(3):352-6.

While multiple studies have documented differences between African American and white patients with regard to the use of cardiac procedures, there is a paucity of data about use of such procedures among Latinos and Asians. The purpose of this study was to examine the rates at which African Americans, Latinos, and Asians receive invasive cardiac procedures once admitted to a hospital with a diagnosis of possible heart disease. Three questions were explored: (1) How does the rate of invasive cardiac procedures for these three racial/ethnic groups compare with those of whites in Los Angeles county?; (2) Do any racial discrepancies in these rates persist after adjustment for factors that might affect use of cardiac procedures?; and (3) What is the independent effect of hospital procedure volume on racial disparities after income and health insurance status are controlled?

Data for this study were derived from the California Hospital Discharge Data Set, which includes all patients discharged from California hospitals from 1986 to 1988. Patients included were those assigned principal discharge diagnoses of acute myocardial infarction, unstable angina, angina pectoris, chronic myocardial ischemia, or chest pain.

Although large numbers of each ethnic group received angiography, bypass graft surgery and angioplasty, African Americans and Latinos received fewer procedures and whites received more procedures than would be expected based on the racial distribution of the sample population. Whereas, the number of Asians receiving these procedures was reflective of their relative distribution in the sample population. Among patients discharged with heart disease diagnoses, the odds ratio (OR) for angiography compared with whites was 0.50 for African Americans, 0.78 for Asians, and 0.49 for Latinos; the OR for bypass grafts compared with whites was 0.35 for African Americans, 0.95 for Asians and 0.49 for Latinos; and the OR for angioplasty compared with whites was 0.38 for African Americans, 0.75 for Asians and 0.38 for Latinos.

There were a number of demographic and clinical characteristics that differentiated these ethnic/racial groups. Among these groups, African American and Latino groups were more likely to be female; white patients were older; and African American patients were less likely to have Medicare or private insurance as a primary payer, were more likely to have Medicaid or be enrolled in a HMO, and were more likely to have indigent or other services. Neighborhood income disparities were also substantial. Controlling for these variables, the OR for angiography compared with whites was 0.80 for African Americans, 0.85 for Asians, and 0.69 for Latinos; the OR for bypass grafts compared with whites was 0.58 for African Americans and 0.70 for Latinos; the OR for angioplasty compared with whites was 0.76 for African Americans and 0.75 for Latinos. Asians did not significantly differ from whites in the odds of bypass grafting and angioplasty after controlling for demographic and clinical variables.

With regard to the influence of the type of hospital, African American, Latino and Asian patients were significantly more likely to be admitted to hospitals that did not perform angioplasty, bypass graft surgery, and angioplasty. However, even after deleting discharges from hospitals that did not perform these procedures and controlling for demographic/ clinical variables, African Americans and Latinos still had significantly lower odds of all three procedures compared with whites (although the magnitude of the OR's were diminished). Asian-white differences in use of any of the procedures were no longer statistically significant. According to the authors, "these findings suggest that all disparities in the use of invasive cardiovascular procedures for Asians and an important component of such disparities for Latinos may be related to disproportionate representation of these populations at hospitals that do not perform such procedures."

With regard to the volume of procedures performed by the hospital (high, medium, low, and very low), African Americans and Latinos were significantly less likely than whites to be admitted to hospitals that performed large numbers of each procedure. Asians were as likely to be admitted to hospitals with high volumes of bypass graft surgery and angioplasty. After controlling for volume of hospital procedures (as well as deleting discharges from hospitals that did not perform these procedures and controlling for demographic/ clinical variables), Latinos had significantly lower odds of angiography (OR=0.90, 95% confidence interval=0.85 to 0.95) and bypass graft surgery (OR=0.87, 95% confidence interval=0.79 to 0.94), and African Americans had significantly lower odds of bypass graft surgery (OR=0.62, 95% confidence interval=0.56 to 0.69) and angioplasty (OR=0.80, 95% confidence interval=0.72 to 0.88). According to the authors, "the persistence of the relative magnitude of discrepancies in the odds of receiving either of the revascularization procedures (bypass graft surgery or angiography) for African Americans even after controlling for volume is troubling. Clearly the use of these revascularization procedures by African Americans is especially susceptible to other factors, such as patient preference and physician practice style that were not measured in this analysis."

The authors conclude that "it is far from certain that simply removing insurance-related barriers to care will eliminate the disparities noted in this analysis. This analysis does suggest that some as yet unidentified factors, which may include access barriers, health status, patient preference, and other issues, and which may vary among ethnic groups, affect the use of invasive cardiovascular procedures." They recommend an expansion of research efforts that incorporate primary data collection strategies that could include studies to validate administrative data reports of ethnicity, studies that abstract medical records for diagnosis at admission and for procedure use, studies that survey patients about their preferences for treatment and their understanding of the procedures, and studies that survey physicians about indications for procedure use.

The authors note several limitations, the most consequential of which is the paucity of clinical data available for these analyses.