Psychometric properties of loaded and unloaded timed stair tests in assessing people with knee osteoarthritis

OBJECTIVE: To compare timed stair test performance between individuals with knee osteoarthritis and healthy older adults, examine reliability under loaded and unloaded conditions, explore correlations with functional measures, and identify optimal cut-off times. DESIGN: Cross-sectional study. SUBJECTS/PATIENTS: Sixty individuals with knee osteoarthritis and 30 healthy older adults. METHODS: Participants completed unloaded and loaded timed stair test trials 7 days apart. Knee strength, balance, mobility, and self-reported outcomes were assessed. RESULTS: Mean completion times for osteoarthritis participants were 23.91 s (unloaded) and 25.40 s (loaded). Intra-rater (ICC3,1 = 0.951-0.990 across raters and assessment days) and inter-rater (ICC3,2 = 0.999 [unloaded] to 1.000 [loaded]) reliabilities were excellent. Test-retest reliability was good for loaded (0.835) and excellent for unloaded (0.918) conditions. Times correlated negatively with affected side knee strength, balance, and confidence, and positively with pain severity, mobility tests, and daily activity difficulty. Optimal cut-offs were 18.67 s (unloaded) and 19.94 s (loaded), demonstrating sensitivities of 70.0% (unloaded) and 68.3% (loaded), a specificity of 86.7%, and area under the curve values of 0.78 (unloaded) and 0.77 (loaded). CONCLUSION: The timed stair test is a reliable, valid, and practical measure of functional performance in knee osteoarthritis, effectively distinguishing movement capability under loaded and unloaded conditions.