Determination of minimum clinically important difference (MCID) in visual analog scale for pain scores after hallux valgus correction

Abstract

Category: Bunion Introduction/Purpose: Surgical outcome studies rely on patient reported outcome measurements to assess the effectiveness of treatment. The concept of minimal clinically important difference (MCID) proposes a necessary threshold to achieve clinically significant treatment results, and refers to the smallest change in outcome measure important from the patient’s perspective. In the context of visual analog scale (VAS) questionnaires, MCID refers to a clinically significant change in pain score. Determination of MCID in patient-oriented outcome questionnaires is necessary to further evaluate the effectiveness of hallux valgus surgery. Further, MCID analysis of hallux valgus surgical outcomes could provide improved insight into post-operative patient satisfaction. The purpose of this study was to determine the MCID in pre- to post-operative VAS pain score in patients undergoing surgical treatment of hallux valgus. Methods: Adult patients undergoing surgical treatment of hallux valgus were retrospectively included. Pre- and post-operative VAS pain scores (0-10) and surveys inquiring about satisfaction with pain level after surgery were collected at a minimum of 1-year post-surgery. Patients were categorized as responders or non-responders based upon a completed 6-point pain satisfaction scale. Patients reporting satisfaction scores 0-3 were categorized as non-responders, and 4-6 as responders. Four MCID calculation methods were used that have been described in previous literature: the standard deviation (SD) approach, the average change approach, the minimally detectable change (MDC) approach, and the change difference approach. The total percentage of patients meeting the calculated VAS threshold score for each MCID method was determined. The likelihood of meeting the VAS threshold for each MCID method based on responder status, hallux valgus severity, and correction status of concomitant hammertoe deformity was also determined using bivariate analysis. Results: 170 patients were included with post-operative follow-up averaging 23.6 months. VAS MCID threshold scores were 1.77points (SD approach), 5.21points (average change approach), 1.98points (MDC approach), and 4.27points (change difference approach). The patient percentage meeting the VAS threshold score for each MCID approach was 73.5%, 40.6%, 73.5%, and 48.8%, respectively. Moderate deformity procedures (Ludloff) demonstrated greater likelihood than mild deformity procedures (Chevron, Modified McBride, Aikin, Silver) of meeting the average change, MDC, and change difference approach thresholds (p=0.036, 0.035, 0.034). Severe deformity procedures (Lapidus) demonstrated greater likelihood than mild deformity procedures of meeting the SD approach threshold (p=0.046). Hammertoe correction demonstrated greater likelihood than non-correction of meeting the average change approach threshold (p=0.038). Responders demonstrated greater likelihood than non-responders of meeting all MCID approach thresholds (p<0.001). Conclusion: This study demonstrated marked variability in determining VAS MCID for hallux valgus correction (range 1.77- 5.21 points). This study suggests an association between type of hallux valgus correction and likelihood of post-operative improvement, as there was greater chance of meeting MCID with correction of greater hallux valgus deformity or hammertoe deformity. MCID methods utilizing comparisons of responder status may not be appropriate for hallux valgus patients, as responders tended to improve with time and non-responders tended to decline. Additional investigation of the optimal MCID method for hallux valgus correction is necessary to narrow the range and determine surgical efficacy.