• Paul W Stratford, Deborah M Kennedy, and Daniel L Riddle.
• School of Rehabilitation Science, Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Ontario, Canada. stratfor@mcmaster.ca
• J Clin Epidemiol. 2009 Mar 1; 62 (3): 347-52.

ObjectiveWe propose a study design to evaluate the ability of measures to assess physical function relatively independent of pain. We illustrate the design using data from patients scheduled for total hip (THA) or knee (TKA) arthroplasty.Study Design And SettingA repeated-measure study design was developed and applied to patients undergoing THA or TKA. A performance battery (PB) (walking, stair, and timed-up-and-go) was used as the gold standard measure for "physical function." Recent evidence indicates physical performance deteriorates 1 month after surgery compared to presurgical measures and patients with THA deteriorate to a greater extent than patients with TKA. Our design assessed whether the results from self-report measures (Western Ontario and McMaster Universities Osteoarthritis Index [WOMAC] and the Lower Extremity Functional Scale [LEFS]) of physical function were consistent with changes found using the PB.ResultsThe performance measures demonstrated significant Occasion (pre- to 1 month postarthroplasty) and Group (hip vs. knee)-by-Occasion effects; the LEFS showed a significant Occasion effect; and the WOMAC PF showed neither.ConclusionOur findings support the proposed design in that the PB was more sensitive to change than the self-report measures.

22 keywords...

hide…