Longitudinal Invariance Testing Of The Knee Injury Osteoarthritis Outcome Score For Joint Replacement Scale (KOOS-JR)

Abstract

Background The Knee Osteoarthritis Outcome Score for Joint Replacement (KOOS-JR) is a seven-item patient reported outcome measure used to assess perceived knee health. Though commonly used, the longitudinal psychometric properties of the KOOS-JR have not been established and further characterization of its structural validity and multi-group invariance properties is warranted. Purpose The purpose of this study was to evaluate psychometric properties of the KOOS-JR in a large sample of patients who received care for knee pathology. Study Design Original research. Methods Longitudinal data extracted from the Surgical Outcome System (SOS) database of 13,470 knee pathology patients who completed the KOOS-JR at baseline, three-months, six- months, and one-year. Scale structure was assessed with confirmatory factor analysis (CFA), while multi-group and longitudinal invariance properties were assessed with CFA-based procedures. Latent group means were compared with statistical significance set at α ≤ .05 and Cohen’s d effect size as d = 0.2 (small), d = 0.5 (medium), and d = 0.8 (large). Results CFA results exceeded goodness-of-fit indices at all timepoints. Multi-group invariance properties passed test requirements. Longitudinal analysis identified a biased item resulting in removal of item #1; the retained six-item model (KOOS-JR-6) passed longitudinal invariance requirements. KOOS-JR-6 scores significantly changed over time (p ≤ .001, Mdiff = 1.08, Cohen’s d = 0.57): the highest scores were at baseline examination and the lowest at 12-month assessment. Conclusions The KOOS-JR can be used to assess baseline differences between males and females, middle and older aged adults, and patients receiving total knee arthroplasty or non-operative care. Caution is warranted if the KOOS-JR is used longitudinally due to potential measurement error associated with item #1. The KOOS-JR-6 may be a more viable option to assess change over time; however, more research is warranted. Level of Evidence 3 © The Author(s)