Abstract
OBJECTIVE: Knee osteoarthritis (OA) can substantially limit function, which can be assessed both objectively and subjectively. We examined whether objective performance tests are associated with self-reported function. METHODS: We analyzed baseline data from the Osteoarthritis Registry of Biomarker and Imaging Trajectories (ORBIT) of participants ≥40 years old with symptomatic and radiographic knee OA. Subjects completed the Knee Injury and Osteoarthritis Outcome Score (KOOS) Pain and Activities of Daily Living (ADL) scales and other assessments of pain and comorbidity. Subjects performed the timed single leg balance test (SLB), 30-s sit-to-stand (30s STS), Timed-Up-and-Go (TUG), and 40-m fast paced walk (40 m Walk). We used Pearson correlation coefficients to examine associations between performance and KOOS subscales. We adjusted for potential confounders using partial correlations. RESULTS: We enrolled 101 subjects (mean age 63.7 (standard deviation (SD) 10.1), mean BMI 30.0 (SD 5.6), and 63% female). The mean (SD) values for the performance tests were: SLB 20.1 (18.9) seconds, 30s STS 11.7 (4.6) stands, TUG 9.4 (2.3) seconds, and 40 m Walk 27.6 (6.5) seconds. Correlations between performance tests and self-report measures did not exceed 0.39, with the absolute value of correlations between KOOS ADL and performance measures ranging from 0.24 to 0.39. Adjusted partial correlations were largely similar to the crude correlations. CONCLUSIONS: Self-reported function in persons with knee OA had weak to modest correlations with objective function. Objective performance tests capture elements of physical function that self-report data do not and point to the potential value of including objective measures of functional status in OA trials.