Minimal Clinically Important Changes in HOOS-12 and KOOS-12 Scores Following Joint Replacement.

<bold>Background: </bold>For patient-reported outcome measures (PROMs) to provide meaningful information to support clinical care, we need to understand the magnitude of change that matters to patients. The aim of this study was to estimate minimal clinically important changes (MCICs) for the 12-item Hip disability and Osteoarthritis Outcome Score (HOOS-12) and Knee injury and Osteoarthritis Outcome Score (KOOS-12) among people undergoing joint replacement for osteoarthritis.<bold>Methods: </bold>Individual-level data from the Australian Orthopaedic Association National Joint Replacement Registry's pilot PROMs program were used for this analysis. Preoperative and 6-month postoperative HOOS-12 and KOOS-12 domain and summary impact scores plus a rating of patient-perceived change after surgery (on a 5-point scale ranging from "much worse" to "much better") were available. Three anchor-based approaches-mean change, receiver operating characteristics (ROC) based on Youden's J statistic, and predictive modeling using a binary logistic regression model-were used to calculate MCICs based on patient-perceived change.<bold>Results: </bold>Data were available for 1,490 patients treated with total hip replacement (THR) (mean age, 66 years; 54% female) and 1,931 patients treated with total knee replacement (TKR) (mean age, 66 years; 55% female). Using the mean change method, the MCIC ranged from 24.0 to 27.5 points for the HOOS-12 and 17.5 to 21.8 points for the KOOS-12. The ROC analyses generated comparable MCIC values (28.1 for HOOS-12 and a range of 15.6 to 21.9 for KOOS-12) with high sensitivity and specificity. Lower estimates were derived from predictive modeling following adjustment for the proportion of improved patients (range, 15.7 to 19.2 for HOOS-12 and 14.2 to 16.5 for KOOS-12).<bold>Conclusions: </bold>We report MCIC values for the HOOS-12 and KOOS-12 instruments that we derived using 3 different methods. As estimates obtained using predictive modeling can be adjusted for the proportion of improved patients, these may be the most clinically applicable. These MCIC values can be used to interpret important changes in pain, function, and quality of life from the patient's perspective.<bold>Levels Of Evidence: </bold>Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence. [ABSTRACT FROM AUTHOR]