Patient acceptable symptom state for patient-reported outcomes in people with non-specific chronic low back pain

Background: The patient acceptable symptom state (PASS) is a treatment-response criterion developed to determine the clinical relevance of a treatment effect. Its estimates for some patient-reported outcomes (PROs) in non-specific chronic low back pain (cLBP) are lacking and the stability of PRO estimates between independent cLBP populations is unknown. We hypothesized that these PRO estimates will be stable. Objectives: To estimate and compare the PASS for PROs between 2 independent cLBP populations. Methods: We conducted a secondary analysis of a randomized controlled trial (PREDID) and a cohort of outpatients with non-specific cLBP. Using an anchoring question, participants who self-rated their health as "excellent", "very good" or "good" at 1 month were considered to have an acceptable symptom state. PASS estimates for 5 PROs were calculated by using the 75th percentile method. Estimates were compared between the 2 populations with bootstrap resampling. Results: A total of 256 participants with non-specific cLBP were included: 135 patients with cLBP and active discopathy from the PREDID trial and 121 outpatients with cLBP without active discopathy followed up in an independent cohort. Overall, 137/256 (54%) participants had an acceptable symptom state at 1 month. PASS estimates were 47.5 (95% confidence interval [CI] 40.0 to 50.0)/100 for lumbar pain (0, no pain and 100, maximal pain), 30.5 (30.0 to 40.0)/100 for radicular pain, 39.3 (33.6 to 45.3)/100 for Quebec Back Pain Disability score (0, no disability and 100, maximal disability), 10.0 (9.2 to 10.0)/21 for the Hospital Anxiety Depression anxiety subscale (0, no anxiety, and 21, maximal anxiety) and 6.7 (6.0 to 8.0)/21 for the depression subscale (0, no depression, and 21, maximal depression). PASS estimates did not differ between the 2 populations. Conclusions: Our study provides PASS estimates for 5 PROs commonly used in cLBP. Our estimates were stable between 2 independent populations of people with cLBP. The stability of our PASS estimates suggests that they are relevant for interpreting PRO values in clinical trials and practice. (C) 2020 Elsevier Masson SAS. All rights reserved.