Who Receives Biologics For Treatment Of RA?

Purpose: To investigate disparities in prescription of biologics for treatment of RA by comparison of rates and predictors of use in an observational cohort.
Methods: We analyzed data from the cohort of RA patients in the ARAMIS (Arthritis, Rheumatism and Aging Medical Information System) databank from January 2000 to January 2006. Subjects were grouped by drug treatment for analysis. Groups were hierarchical, due to combination therapy, overlap with lower group(s) was allowed: group 1) NSAID only; group 2) Steroids (+/- 1); group 3) DMARD (+/- 1, 2); group 4) Biologics (+/- 1, 2, 3). Bivariable comparisons of biologic users versus each other treatment group were conducted. Multivariable logistic regression models were developed to investigate 1) predictors of biologic versus non-biologic DMARD use and 2) predictors of biologic use versus steroid predominant treatment.
Results: 1402 patients from 8 US centers were included in the analysis. 86.6% were white, 6.1% black, 4.4% Hispanic, and 2.6% Asian. Percent of the overall sample in each treatment group: NSAID 11.9%, Steroid 6.6%, DMARD 47.7%, Biologic 32.6%. In 2000, 13.3% of the sample used biologics, and by 2006 this was 41.4%. Compared to other groups, Biologic users were significantly (p<0.05): younger (mean age 54.4 yrs vs. 60.2 - 66.1 yrs), more likely female (83.3% vs. 72.1 -76.9%), more likely to be employed (40.6% vs. 17.5 - 31.2%), higher earners ($54.4K vs. $39.1 - 46.3K), and less likely to have Medicare insurance (34.4% vs. 44.0 - 61.7%). In multivariable models, significant predictors of biologic versus non-biologic DMARD use were age (per decade; OR 0.74, CI 0.61-0.89), income (per $10,000; OR 1.11, CI 1.04-1.19), HAQ-DI (per unit; OR 1.5, CI 1.10-2.04). Significant predictors of biologic versus steroid predominant treatment were age (per decade; OR 0.54, CI 0.35-0.82), pain (VAS 0-100, per 10 pt; OR 0.79, CI 0.66-0.94), disease duration (per year; OR, 0.95 CI 0.92-0.98), and comorbidity (OR 0.25, CI 0.007-0.87).
Conclusions: Factors other than disease severity influence who receives biologics. In this observational cohort of RA patients, biologic use was associated with younger age, and increased with income and HAQ-DI compared to DMARD use. In multivariable comparisons of biologic users versus a subgroup of patients predominantly managed with steroids, biologic use was associated with younger age, lower baseline pain level, shorter disease duration, and less co-morbidity.

 E. Morgan DeWitt, None; L. Lin, None; H. Glick, None; K. Schulman, None; B. Lingala, None; S. Reed, None.