Background/Purpose: Chronic back pain is the second most common reason for a physician’s visit and results in significant physical and psychosocial consequences in older adults. Identifying appropriate and reliable patient reported outcome measures is critical for research and clinical purposes. The NIH’s Patient Reported Outcomes Measurement Information System (PROMIS) instruments provide robust patient reported outcome (PRO) measures; however, these have not been evaluated alongside “legacy” instruments in older adults with chronic back pain to expand our understanding of treatment response. In this pilot trial, we used epidural steroid injections (ESI) as a vehicle intervention to better understand our PRO. This study aims to evaluate whether legacy and/or PROMIS biopsychosocial measures change according to those who responded or were non-responders to ESI.

Methods: We enrolled a convenience sample of older Veterans (age 60+) with chronic back pain with/without leg pain scheduled for lumbar ESI. Subjects completed “legacy” instruments and corresponding PROMIS computer adaptive test (CAT) item banks pre- and post-ESI in the following domains: pain interference, behavior and intensity; functional status; depression and anxiety; fatigue; sleep and social functioning. The effects of ESI on bio-psychosocial measures using legacy and PROMIS were assessed using a two-way analysis of variance (ANOVA) with one repeat factor (pre/post ESI) and one between factor (responder/non-responder). We defined responders vs non-responders to ESI based on the accepted minimally important significant difference for the Roland Morris Disability Questionnaire, used in back pain trials.

Results: Participants included 71 Veterans who were on average 67 years old, 94% men, 73% non-Hispanic white, 17% African American. Patients were obese with a mean BMI 32, 25% reported multi-site pain and 59% were diagnosed with depression, anxiety and/or PTSD. The majority (69%) reported pain duration ≥5 years with 93% reporting associated radiculopathy. The time between pre and post ESI assessments ranged between 3-8 weeks (based on routine follow-up appointment). Two-way ANOVA results showed that ESI responders (compared to non-responders) reported significant improvement in several domains from legacy (pain interference and behavior, depression and fatigue) and PROMIS (fatigue and social) measures. Table 1 includes legacy and PROMIS mean values (+SD) as well as ANOVA results for the interaction and factor effects.

Conclusion: These results suggest that responders to ESI showed improvement in several domains, more commonly among legacy than PROMIS measures. Fatigue was the only variable that improved using both instruments. Further research in a larger and gender diverse sample is warranted to gain a better understanding of PRO that may improve in older adults with chronic back pain receiving ESI.