Patient and Provider Characteristics Associated with Receipt of Image-guided Interventions for Low Back Pain.

Abstract

Background and purpose: Low back pain commonly causes disability, often managed with conservative image-guided interventions before surgery. Research has documented racial disparities with these and other non-pharmacologic treatments. We posited that individual chart reviews may provide insight into the disparity of care types through documented patient/provider discussions and their effect on treatment plans.

Materials and methods: This retrospective analysis involved adults newly diagnosed with low back pain within a large Utah healthcare system. The primary outcome was the association of provider and patient variables with the frequency of image-guided interventions received within one year of low back pain diagnosis between White/non-Hispanic and underrepresented minority cohorts. Secondary outcomes were receipt of additional treatment types (physical therapy and lumbar surgery), time to any treatment, time to image-guided intervention, and discussion/receipt of therapy between cohorts within one year of diagnosis.

Results: Among 812 subjects (59% White/non-Hispanic and 41% underrepresented minority), more White/non-Hispanic patients had at least one image-guided intervention within 12 months compared to underrepresented minority patients (7.2% vs. 12.5%, p = .001), despite underrepresented minorities having higher presenting pain scores (64.5% vs. 49.3%; pain intensity > 5, p = .001). Underrepresented minority patients more often saw generalists (71.7% vs. 52.6%, p < .001) and advanced practice clinician providers (33.6% vs. 25.6%, p < .02) compared to the White/non-Hispanic cohort. Both cohorts were referred to a specialist at the same rate (17.7% vs. 19.8%, p = .20); however, referral completion was noted less often (60.4% vs. 77.7%, p = .02) and took longer to complete in underrepresented minority patients (54 vs. 27.5; mean day, p = .003).

Conclusions: Underrepresented minority patients had more severe low back pain on presentation but received image-guided interventions less often than White/non-Hispanic patients. While there may be systematic provider barriers, such as absence of a decision-making discussion, data do not support provider bias as a contributing factor to differences in receipt of image-guided interventions. Non-medical barriers to referral completion should be further investigated to improve access to more specialized low back pain care.

Abbreviations: IGI = image-guided intervention; LBP = low back pain; URM = underrepresented minority; WNH = White/non-Hispanic; ICC = intraclass correlation coefficient.