Session number: 1084
Abstract title: Racial Variations in Health Service Utilization among Veterans with Osteoarthritis
Author(s):
KL Dominick - Durham VAMC
TK Dudley - Durham VAMC
HB Bosworth - Durham VAMC
Objectives: Evidence suggests that there are racial disparities in medical care for osteoarthritis (OA), but patterns of utilization have not been well studied within the VA population. This study compared physician visits, hip and knee arthroplasty, and prescription drug utilization between African American (AA) and white veterans with OA.
Methods: Subjects included 2,473 patients (35% AA, 65% white) with a diagnosis of OA coded on an outpatient visit to the Durham VAMC during FY 1999. Arthroplasties and OA-related physician visits were examined for one year following subjects’ first physician visit during the inception period. For a subset of patients who had a physician visit between October 1998 and March 1999 (N=1,437, 34% AA, 66% white), prescription drug utilization was examined for six months. Analyses were adjusted for age and service connection.
Results: During follow-up, similar proportions of AA and white subjects had at least one OA-related physician visit (AA=64%, white=63%, p=0.489) and underwent arthroplasty (AA=3.3%, white=3.0%, p=0.772). In a multivariate poisson regression model, AA race was not a significant predictor of physician visits (β=0.04, p=0.280). Multivariate logistic regression models revealed that AA race was also not significantly related to arthroplasty use (adjusted OR=1.13, 95%CI=0.70-1.84) or prescription analgesic use in general (adjusted OR=1.06, 95%CI=0.79-1.42). However, there were racial variations in the types of analgesics prescribed. AA subjects were significantly less likely to be prescribed opioid analgesics (adjusted OR=0.75, 95%CI=0.60-0.95) and COX-2 inhibitors (adjusted OR=0.11, 95%CI=0.03-0.47) but tended to have an increased risk of non-selective NSAID use (adjusted OR=1.16, 95%CI=0.92-1.48).
Conclusions: Studies have suggested that AA patients seek OA-related physician care more often but undergo arthroplasty less often than white patients. However, there were no racial differences in physician visits or arthroplasty use in this veteran sample. Pharmacological treatment did vary by race. Most notably, AA subjects were less likely to be prescribed COX-2 inhibitors, which have fewer gastrointestinal side effects than non-selective NSAIDs. Further research should examine reasons for racial differences in pharmacological treatment of OA.
Impact statement: Understanding racial variations in OA-related health care is an essential step toward ensuring appropriate treatment for the growing numbers of veterans with OA.