Minimal Clinically Important Difference and Relative Change in Patient-Reported Outcomes After Surgery for Cervical Spondylotic Myelopathy: A Nationwide Study of 1,700 Patients.

Abstract

Healthcare policies have seen significant reforms, with a marked transition toward a patient-centered approach. This shift emphasizes the use of subjective patient-reported outcome measures as key metrics. However, these measures often face limitations, particularly in identifying clinically meaningful changes over time. To address this challenge, the concept of the minimal clinically important difference (MCID) was introduced. This study aims to evaluate the predictive effectiveness of relative change and its associated threshold, the minimal clinically important relative change (MCIRC), as a potential alternative to absolute differences and the MCID. Data prospectively collected between 2006 and 2021 from the Swedish Spine Registry (Swespine) were analyzed. Patient-reported outcome measures included the Numeric Rating Scale (NRS) for neck and arm pain, the EQ5D index, EQ VAS, and the Neck Disability Index (NDI). Anchor-based methods were used to calculate the MCID and MCIRC. The predictive performance of absolute differences and relative changes was compared using the area under the receiver operating characteristic curve. 1756 patients were included. The selected MCID values were -3 for NRS neck pain intensity, -2 for arm pain intensity, 0.09 for the EQ5D index, 7 for EQ VAS, and -12 for NDI scores. For MCIRC, the chosen values were -47% for NRS neck pain intensity, -40% for arm pain intensity, 386% for the EQ5D index, 52% for EQ VAS, and -32% for NDI scores. On area under the ROC curve, relative change was superior for NRS neck and arm pain scores and NDI, while absolute difference was superior for EQ5D and EQ VAS scores. Relative change, along with its associated minimal clinically important value (MCIRC), proved to be a more suitable indicator of subjective satisfaction for NRS and NDI scores. By contrast, absolute differences and the MCID were better suited for evaluating the EQ5D index and EQ VAS scores.