1513. Clinically Relevant Percent Reduction (CRPR): A Novel Definition of Clinically Significant Change for Lumbar Spine Surgery
Authors: Anthony Michael Asher; Jacquelyn Pennings, PhD; Inamullah Khan, MD; Ahilan Sivaganesan, MD; Richard Call; Clinton Devin, MD; Anthony Asher, MD; Kristin Archer (Charlotte, NC)
Patient-reported outcomes (PROs) are integral to assessing efficacy of surgical procedures from a patient perspective. Minimum Clinically Important Difference (MCID), a measure of absolute changes from baseline for different PROs, is commonly used to determine whether a medical intervention improves perceived clinical outcomes. The sensitivity of MCID for achieving that aim, however, particularly at extremes of baseline PROs, has been questioned. The objective of this study was to compare achievement of MCID 12 months after surgery to a novel efficacy measure—Clinically Relevant Percent Reduction: CRPR (a 30% reduction from baseline PRO scores)—and their correlation to satisfaction with lumbar surgery.
The NPA QOD Lumbar Registry was queried for lumbar surgeries including discectomy, fusion, and laminectomy without fusion. Preoperative and 12 month postoperative PROs were recorded for disability (ODI) and back and leg pain (NRS-BP/LP). Achievement of MCID for lumbar surgery, defined in the literature as score reductions of 12.8, 1.2, and 1.6 for ODI, NRS-BP, and NRS-LP, respectively, and 30% reduction from baseline PRO values were compared with patient satisfaction (based on a dichotomized NASS lumbar spine satisfaction score) with ROC curves.
A total of 20281 patients were identified for the study. The area under the ROC Curves (AUROC) for the different PROs for CRPR and MCID, respectively, were as follows: ODI 0.768, 0.733; NRS-BP 0.737, 0.693; NRS-LP 0.712, 0.683. Additionally, the AUROCs differed substantially when patients had either high or low disability.
These analyses suggest CRPR may be a superior method to assess perceived surgical effectiveness than MCID. In particular, CRPR better captures the experiences of patients with very high or low baseline disabilities. These authors recommend using a 30% PRO reduction from baseline measure instead of absolute change MCID values for determining clinically important change based on ODI, NRS-LP, and NRS-BP after lumbar surgery.