The author aimed to examine several hospital-based procedures for systemic utilization differences between males and females and among Asians, Blacks, Latinos, and Whites. Procedures for treating conditions commonly requiring hospitalization were addressed using the California Office of Statewide Health Planning and Development hospital discharge abstract data for 1989 and 1990.

For several of the procedures, ethnicity was associated with procedure utilization with substantial and often significant odds favoring Whites.

For Whites Compared to Blacks:
      Angioplasty                                    OR= 2.00; 95%    CI= 1.79-2.22
      Endarterectomy                            OR= 2.27; 95%    CI= 1.41-3.70
      Coronary bypass                          OR= 2.44; 95%    CI= 2.08-2.78
      Defibrillator implant                     OR= 2.86; 95%    CI= 1.28-6.25
      Kidney transplantation                OR= 3.04; 95%    CI= 2.27-4.17
      Compared to Latinos:
      Coronary bypass                          OR= 1.49; 95%    CI= 1.35-1.67
      Kidney transplantation                OR= 1.58; 95%    CI= 1.20-2.08
      Angioplasty                                    OR= 1.72; 95%    CI= 1.56-2.22
Compared to Asians:
      Angioplasty                                    OR= 1.30; 95%    CI= 1.15-1.47
      Endarterectomy                            OR= 2.08; 95%    CI= 1.18-3.85
(Additionally, Asians showed favorable odds to Whites for hip replacement:
OR= 2.13; 95%    CI= 1.30-3.45)

The authors note that the underlying causes for the differences in utilization may be “subtle and multifactorial.” “Several conventional explanations for these phenomena include patient preferences, barriers to access, and unmeasured clinical eligibility differences.” Additionally, one should consider “provider attitudes towards (women and) ethnic minorities as well as the development of technologies to meet any special needs of patients in these demographic populations.”

The authors emphasize the following points:
      • Inequality does not necessarily constitute inequity.
      • Differences in utilization are not necessarily quality-of-care differences (they were unable to assess
        if the procedures were overused in the high-odds groups or underused in low-odds groups).
      • Because ethnic minorities have been underrepresented in clinical trials, evidence supporting
        (existing) clinical guidelines do not necessarily apply to these populations.
      • Unmeasured clinical conditions might explain the differences in treatment seen.
      • Limitations of the research method may have affected these findings.

Policymakers must address not only financing but also the social and clinical conventions that determine the allocation of the technologies in order to equalize utilization of these procedures. The authors noted that it is unclear “what role, if any, patient preferences, referral patterns, or evidence of clinical effectiveness play in providers’ differential selection of procedure candidates by (gender and) ethnicity” as well as the “possible discrimination on the part of providers.”