A comparison between different outcome measures based on meaningful important differences in patients with lumbar spinal stenosis

Patient-reported outcome measures (PROM) are used to measure treatment efficacy in clinical trials. The impact of the choice of a PROM and the cut-off values for 'meaningful important differences' (MID) on the study results in patients with lumbar spinal stenosis (LSS) is unclear. The objective is to study the consequences of applying different PROMs and values for MID for pain and disability on the proportions of patients with improvement. Prospective multi-center cohort study. Proportions of patients with improvement using established MID cut-off values were calculated and compared for PROMs for pain and disability. 466 patients with LSS completed a baseline and 6-month follow-up assessment and were analyzed. Treatment modalities included surgery (65 %), epidural steroid injections (15 %), or conservative care (20 %). The prevalence of patients fulfilling the criteria for MID ranged from 40 to 70 % across all outcome measures and cut-offs. The agreement of the spinal stenosis outcome measure (SSM) symptom subscale with other pain scales, and the SSM function subscale with other function scales was fair to moderate (Cohen's kappa value between 0.24 and 0.5). Disagreement in the assessment of MID (MID reported by patients in one scale but not the other) was found in at least one-third of the patients. The MID in outcome scores for this population varied from 40 to 70 %, depending on the measure or cut-off score used. Further, the disagreement between domain specific measures indicates that differences between studies may be also related to the choice of an outcome measures. An international consensus on the use and reporting of outcome measures in studies on lumbar spinal stenosis is needed.