The main objective of the study was to assess how health service use differs by patient immune status (immunosuppression measured by CD4 strata), as well as how it differs across racial groups at comparable levels of immune dysfunction. The 571 study participants (219 were identified as non-white) received most of their medical care at public and nonprofit urban hospitals (in Atlanta, GA; Nassau County, NY; Newark and Jersey City, NJ; Miami, Fla.; New Orleans, La; Dallas, Tex; and Seattle, Wash). To measure the outcome variable, dates of all visits to HIV treatment clinics, inpatient admissions, and discharges were abstracted from medical records beginning with the date the patient learned about his/her HIV status and ending with either patient’s death or the data collection completion in December 1990.

The finding of racial difference in use – people of color had fewer HIV clinic visits within each of the CD4 strata – “was most pronounced for patients with a CD4 cell count of 50 or less per cubic millimeter.” In this stratum, whites had over six more clinic visits per person-year than non-whites (rate difference [RD] =6.2; 95% CI = 4.2, 8.3) and nearly nine fewer inpatient days (RD = 8.6; 95% CI = 6.5, 10.7). People of color had fewer HIV clinic visits across all three CD4 strata; however, only the ratio among those with CD4 counts greater than 200 was statistically significant. The analysis also revealed that “among participants with between 51 and 200 CD4 cells per cubic millimeter, whites had their counts monitored at the rate of 1.5 tests per year compared with 1.1 tests per year for blacks (CRR = 1.3; 95% CI = 1.2, 1.4).

The authors stated that these results are “consistent with the contention that whites are more likely to receive outpatient treatment while people of color are more likely to be admitted to the hospital.” Furthermore, private physician office visits were not counted in this study, and whites were 4.5 times more likely to report receiving care in such a setting. Thus, the rate ratios reported in this study may “underrepresent the true relationship between race and the likelihood of having an outpatient visit.”

The authors concluded, “there are differences in the use of medical care that are not attributable to disease progression and that may help explain the higher mortality rates for people of color.”