The Knee Injury and Osteoarthritis Outcome Score (KOOS): A Review of Its Psychometric Properties in People Undergoing Total Knee Arthroplasty

Abstract

STUDY DESIGN: Systematic literature review, clinical measurement. OBJECTIVE: To review and summarize the evidence regarding the psychometric properties of the Knee injury and Osteoarthritis Outcome Score (KOOS) in patients undergoing total knee arthroplasty (TKA). METHODS: A comprehensive review of the existing literature was undertaken using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) systematic review guidelines. Studies were included if they reported the psychometric properties of the KOOS or the KOOS-Physical Function Shortform (KOOS-PS). Papers written in both English and German were analyzed. Studies of patients undergoing primary total TKA or TKA revision, or those with severe osteoarthritis and awaiting TKA, were considered. The methodological quality of the included articles was assessed using the Consensus-Based Standards for the Selection of Health Measurement Instruments checklist. RESULTS: Six studies fulfilled the inclusion criteria. Of these, 5 evaluated psychometric properties of the KOOS and 1 evaluated the KOOS-PS. Both the KOOS and KOOS-PS have moderate-to-high construct validity with existing validated self-report measures. However, construct validity of the KOOS function in sport and recreation subscale was weak. Further, these instruments demonstrated a high level of responsiveness, with effect sizes and standard response means of greater than 0.80. Overall, both questionnaires demonstrated clinically acceptable reliability (intraclass correlation coefficient of 0.70 or greater). However, somewhat lower reliability was observed for the KOOS function in sport and recreation subscale (intraclass correlation coefficients of 0.45 and 0.65, respectively) and the other symptoms subscale (internal consistency, = .56). CONCLUSION: The KOOS and KOOS-PS exhibit clinically acceptable psychometric properties. Their strength is in large effect sizes to measure outcomes over time and their weakness is in weak-to-moderate reliability and weak construct validity in some subscales of the KOOS