Reliability and Validity of the Knee Injury and Osteoarthritis Outcome Score in Patients Undergoing Unicompartmental Knee Arthroplasty


School of Data Science and Analytics

Document Type


Publication Date



BACKGROUND: The Knee Injury and Osteoarthritis Outcome Score (KOOS) was developed to document outcomes from knee injury, including the impact of osteoarthritis on knee function. The purpose of this study is to determine the reliability and validity of the KOOS subscales for evaluating outcomes following unicompartmental knee arthroplasty (UKA). METHODS: KOOS Pain, Activities of Daily Living (ADL), Sport, Symptoms, and Quality of Life (QoL) scores collected from 172 patients who underwent UKA were used in the analysis. KOOS subscales were tested for reliability and validity of scores through a Rasch model analysis. RESULTS: KOOS Sport, KOOS ADL, and KOOS QoL had good evidence of reliability with acceptable person reliability, person separation, and item reliability. For overall scale functioning, KOOS Pain, Symptoms, and ADL all had 1 question that did not have an acceptable value for infit or outfit mean square value. Questions in KOOS Sport and QoL all had acceptable values. There was a positive, linear relationship between the Short-Form 12 Physical Component Summary and the KOOS subscales which indicated good evidence of convergent validity. These associations were also seen when the cohort was separated in medial and lateral UKA. CONCLUSION: Two of the 5 KOOS subscales (KOOS Sport and KOOS QoL) were considered adequate in measuring outcomes, as well as reliability. The KOOS ADL had borderline values; however, it had adequate infit and outfit values. The KOOS Pain and Symptom score performed poorly in this analysis. For documenting outcomes following UKA, this study supports the use of KOOS ADL, Sport, and QoL.

Journal Title

The Journal of Arthroplasty

Journal ISSN


Digital Object Identifier (DOI)