Background/Purpose:   Use of disease-modifying anti-rheumatic drugs (DMARDs) is a nationally endorsed quality measure, yet recent studies suggest that only 60% of Medicare recipients with rheumatoid arthritis (RA) use DMARDs.   We investigated the prevalence and predictors of receiving glucocorticoids alone for the treatment of RA in a nationwide sample of Medicare beneficiaries.

Methods:  Data derive from a 5% random sample of U.S. Medicare fee-for-service beneficiaries.  We included individuals ≥65 years with at least two face-to-face clinical encounters for RA and Part D drug claims for either a DMARD anytime during the year or sustained glucocorticoid monotherapy, defined as an annual dispensed glucocorticoid supply of ≥180 days or an annual dispensed dosage of ≥900 mg of prednisone (or steroid equivalent). Using multivariate logistic regression, we examined predictors of sustained glucocorticoid monotherapy including sociodemographic characteristics, income (low-income defined as Medicare eligible for reduced cost sharing or state buy-in), health service utilization (number of inpatient and outpatient encounters and prescribing physician specialty) and medical co-morbidities.  In addition, we used the Area Resource File to examine area level predictors of socio-economic status, health care shortage areas, and Census geographic divisions. From the regression models, we calculated adjusted group proportions and 95% confidence intervals.

Results:    Of the 8,062 beneficiaries, 10% (n = 830) were classified as receiving glucocorticoid monotherapy.  In adjusted analyses, we found that glucocorticoid monotherapy was higher among those with advanced age (18% among those ≥85 years compared to 11% in those 74-79 years), Blacks (12% versus 10% in Whites), and among low-income beneficiaries (12% versus 10% in those with higher incomes).  Having a rheumatologist prescribe one or more medications during the measurement year was associated with significantly lower rates of glucocorticoid monotherapy (7% versus 16%).  More inpatient admissions and medical co-morbidities were also positively associated with glucocorticoid monotherapy.  There was little variation across the nation, with marginally higher rates of glucocorticoid monotherapy in the Middle Atlantic region (13%) compared to the Pacific region (8%).

Conclusion:    Approximately 10% of Medicare recipients with RA were treated with sustained courses of glucocorticoids alone in 2009.  Compared to DMARD users, glucocorticoid monotherapy users were older, more likely to be Black, had lower income, had more medical comorbidities and hospitalizations, and were less likely to have a rheumatologist prescribing their RA medication.  Although advanced age and accompanying medical co-morbidities may appropriately limit the use of DMARDs, differences by race, income and geographic region suggest disparities in quality of care.