The authors aimed to confirm whether racial variation exists across a broad range of procedures and to assess whether racial variation is related to physician discretion, as defined by physician consensus. The authors obtained fiscal year 1988 hospital discharge data for all Massachusetts residents, identified 12 common procedures, and then calculated age- and sex- adjusted rate ratios (RR) for white and black patients.

“Whites had substantially higher rates for seven procedures (abdominal aortic aneurysm repair, appendectomy, carotid endarterectomy, cholecystectomy, lumbar disc procedures, open reduction/ internal fixation of the femur, and tonsillectomy), substantially lower rates for only one procedure (hysterectomy), and similar rates for the other four procedures. Of the seven procedures for which utilization was substantially higher among whites, four were ranked as moderate or high discretion (carotid endarterectomy, cholecystectomy, lumbar disc procedures, and tonsillectomy), and three were ranked as low discretion (abdominal aortic aneurysm (AAA) repair, open reduction/ internal fixation of the femur, and appendectomy). All four procedures for which utilization was similar between racial groups were ranked as moderate or low discretion. Hysterectomy, the only procedure for which utilization was substantially higher among blacks, was rated as high discretion.”

“Whites were 37% more likely to undergo ORIF, 13% more likely to undergo appendectomy, and nearly three and a half times more likely to undergo AAA repair than blacks. The higher prevalence of osteoporosis and hip fracture among whites may account for part or all of the variation in rates of ORIF…An explanation for the lower rates of AAA repair and appendectomy among blacks is less clear….While the limited data on the prevalence of unrecognized AAA suggests that white men may have a higher prevalence of the condition, the suggested difference in disease prevalence is not likely to explain the observed variation in the rate of AAA repair…Black patients may be less likely than whites to have their AAA’s diagnosed or perhaps more likely to refuse AAA repair if offered the procedure. Appendectomy was felt by our physician panelists to be a low discretion procedure…Although death from appendicitis is rare, a recent analysis has shown that death due to appendectomy is 3.17 times higher in blacks than in whites.”

“Variation among low discretion procedures suggests the possibility of differences in medical need or that there are differences in the quality of care received by race. Lower rates of moderate or high discretion procedures among black patients suggest that, when physicians are uncertain of the benefits of a procedure, whites are generally more likely than blacks to undergo those major surgical procedures. Possible mediators of this phenomenon include: differences in medical need, patient preferences, resource availability, physician bias, financial incentives or more subtle aspects of socioeconomic class. Because the optimal clinical strategy in these situations is often unknown, the impact of differential rates on outcome is unclear. Alternatively, inequality in rates of low discretion procedures is more problematic. Here, interracial variations more strongly suggest that there are either differences in the prevalence of disease, or disparities in the quality of care.”

“In summary…because the clinical indications for low discretion procedures are relatively clear, our findings highlight the possibility of suboptimal care by race.”