BACKGROUND: Knee arthroplasty does not always require extensive long-term follow-up. If knee range of motion (ROM) could be assessed reliably by patients, some follow-up visits might be replaced by patient-reported outcome measures, and this additional information could be reported directly to registers. We developed and tested the validity and reliability of a simple scale for patients to self-report their passive knee ROM.

METHODS: Through an iterative process, we created a 2-item scale with 11 illustrations of knee motion in 15° increments. The validity and reliability was tested in knee osteoarthritis and arthroplasty patients at different treatment stages, many with poor ROM. Patient estimates were compared to passive goniometer measurements performed blindly by a physiotherapist and a junior orthopedic surgeon.

RESULTS: The mean difference between 100 patients' (70.9 years) estimates and goniometer measurements was -0.7° (standard deviation, 12.3°) for flexion and 1.1° (standard deviation, 11.6°) for extension, both not significant. Correlation was 0.79 and 0.63, and kappa values at retest were 0.84 and 0.66. For flexion < 110°, sensitivity of patient estimates was 88% and specificity was 88%. For a limit of 100°, values were 95% and 81%. For extension deficits >10°, sensitivity was 78% and specificity 70%. Values were 100% and 66% for a 15° limit.

CONCLUSION: The Copenhagen Knee ROM Scale is a patient-friendly and feasible alternative to passive ROM measurement for registers, research, and selected clinical use. This scale appears reliable and valid compared to reports of similar tools, and patient estimates are better correlated to goniometer measurements.