The Psychometric Properties of the Cone Evasion Walk Test in Patients With Knee Osteoarthritis.

Cross-sectional psychometric study at a single university orthopedic clinic. Thirty-six adults (28 female, 8 male) with a clinical diagnosis of knee OA, mean age 61.76 years, were recruited consecutively over 2021–2023. Each patient completed the CEW twice on the same day with a 1-hour seated rest between trials, after two practice attempts. The same experienced physiotherapist administered all tests. Patients also completed the TUG, the WOMAC questionnaire, and a pain rating (VAS) before and after testing. Reliability was assessed with ICC 2,1; concurrent validity with Spearman's correlation against the TUG.

Mean CEW completion time was 10.75 ± 3.30 seconds. Within-day reliability was excellent: ICC 0.97 (95% CI 0.87–0.99), SEM95 0.73 seconds, SDC95 2.02 seconds. Concurrent validity with the TUG was moderate (r = 0.58, p < 0.001). The CEW did not correlate with the WOMAC self-report score or with the number of falls in the previous 12 months. The second trial was 0.18 seconds faster on average but this difference was not statistically significant. Pain scores did not change after testing and no adverse effects were observed.

A prolonged CEW time or difficulty avoiding the cones gives the OT a performance-based signal that the patient is struggling with the turning and obstacle negotiation that underpin moving between rooms, navigating a kitchen or bathroom, and getting to community destinations. The SDC95 of 2.02 seconds gives a concrete threshold: if a patient's CEW time improves by less than that after an intervention, the change is likely measurement noise rather than real progress. The test can supplement self-report and direct observation when planning environmental modifications, assistive-device training, or fall-prevention strategies.

Both trials were done on the same day one hour apart, so this is within-day repeatability, not the day-to-day test-retest reliability a clinician would rely on in follow-up. The second trial was slightly faster (0.18 s), suggesting a small residual practice effect despite two warm-up attempts. Single clinic, single assessor, small sample of 36 patients; generalisability to other settings or populations is untested. Fall history was self-reported, which is vulnerable to memory gaps and recall bias. Symptom duration was not recorded, so it is unclear how long-standing the knee OA was in each patient. The CEW did not correlate with past falls or with the WOMAC self-report, so it cannot be read as a fall-risk predictor or a substitute for the patient's own account of function.

No funding was received. The authors declare no conflicts of interest.

Do not read the excellent reliability and moderate TUG correlation as evidence that the CEW is a validated standalone fall-risk screen. It showed no correlation with past falls or with self-reported function (WOMAC), its reliability was tested only within a single day at one clinic, and the sample was small. It is a promising addition to an assessment battery, not a replacement for the tools you already use.