There is a growing body of research that suggests that there are black/white differentials in prescription drug use. This study aimed to (1) replicate previous findings in this area using a full sample of Medicaid enrollees in Georgia and (2) determine to what extent racial differences in prescription drug use and spending would remain after controlling for demographic and eligibility differences. Data for this study were obtained from the Georgia Medicaid Statistical Information System, which contains information of Medicaid eligibility as well as use and payment for Medicaid-covered serviced. Only individuals receiving outpatient prescription drugs in 1992 were analyzed; nursing home patients were omitted.

There were notable demographic differences between blacks and whites: blacks had a larger proportion of children under 18 years of age and a smaller elderly population than whites; and more black enrollees lived in rural areas. There were also race differences in Medicaid eligibility criteria patterns. Nearly 76% of black and 84% of white enrollees received prescription drugs through Medicaid. This pattern was true for both males and females and the subgroups defined by Medicaid eligibility. The black/white differences were lower for older enrollees compared with children less than 18 years.

On average, black enrollees received 10.5 prescriptions (per full time enrollment equivalents), while white enrollees received 18.2 prescriptions. This race pattern was also true among the two subgroups that received more prescriptions: the elderly and women. Prescription drug expenditures by blacks and whites followed patterns similar to prescription use: Medicaid drug spending (per full time enrollment equivalents) was about 90% higher among whites compared with blacks ($383 versus $202). Multivariate analyses that controlled for all other demographic and eligibility variables revealed that, for children, adults, and the elderly, blacks used fewer prescriptions and had lower Medicaid prescription expenditures than whites. Additionally, urban enrollees used fewer prescriptions than rural enrollees, part-time enrollees used fewer prescriptions than full-time enrollees, and blind and disabled enrollees used the greatest number of prescriptions. The authors note that the percentage difference in prescription drug use between blacks and whites seemed to decline in the higher age groups but acknowledge that this decrease was partly due to a larger percentage of white enrollees not having drug coverage.

In order to demonstrate differences for particular drug classes, the most frequently used drug categories were examined: antibiotics, psychoactives, calcium antagonists, H2 inhibitors, and hypotensives. Collectively, these categories comprised 35.4% of the total drug utilization and 44.2% of the total drug spending in Georgia. Analysis using full time enrollment equivalents found that prescription drug use by whites was greater than that by blacks across all categories, with psychoactives showing the greatest difference and hypotensives the least. In further analyses of psychoactive drugs, blacks were shown to have significantly lower expenditures than whites in all therapeutic subcategories (antipsychotics, antidepressants, anxiolytics, sedative-hypnotics, and CNS stimulants).

The authors cite Horner et al. (Millbank Quarterly 1995; 73(3):443-462) in offering three possible explanations for the observed racial differences in this study. "First, there may be racial differences in the diagnosis and treatment of medical conditions, which may indicate difference in important clinical factors. Second there may be the inability to pay, regardless of race. Third, there may be a difference in patients' decisions to use services." Two other possible explanations are that the severity of illness may determine drug use and that access to prescription drugs may be determine, in part, by race. The authors note that "while unlikely, it is possible that if prescription drugs are considered preventive or first-line treatment modalities, it may be that blacks are using fewer prescription drugs because they present with more severe illness that requires more drastic interventions, such as hospitalization." They argue that it is more likely, however, that "if blacks are more severely ill than whites, it may be because they have less access to first-line drugs."

In interpreting the findings of this study, the authors caution that severity of illness might be a confounding factor in this analysis, that is, "if whites were sicker than blacks, the drug use differences might be overstated." They also note that spending per prescription was not as different as the number of prescriptions used or the spending per enrollee, indicating that "much of the difference between black and white spending rates was due to differences in levels of use." Finally, they suggest that, in order to examine why use rates are higher among whites, disease prevalence, treatment and severity differences, use of potential substitute non-drug services, and differences in products dispensed for black and white Medicaid patients should also be evaluated.

Note that issues related to potential physician bias were never mentioned in this article.