What is the best way of assessing outcome after total knee replacement?

Variable definitions of outcome have been used in the past to assess the results after total joint replacement surgery. These differ in their approach to the measurement of outcome but all must be valid (they measure what they are designed to measure), reliable (they consistently produce the same score), and responsive (able to detect changes that may occur during a period). Responsiveness is crucial to distinguish those patients who benefit from a procedure from those who do not, and a more responsive test will theoretically be able to identify more subtle changes in patient status. The responsiveness of 6 different scoring systems was compared. The results are based on a cohort of 71 patients undergoing total knee arthroplasty in a 7-month period. Responsiveness was determined by performing a paired t test among each patient's scores at 0, 3, and 6 months. The size of the resulting t value represented the comparative responsiveness of the 6 tests. The highest value achieved was with the Western Ontario and McMaster Universities osteoarthritis index and the Knee Society clinical rating scale. The worst scores were achieved by Short Form-36 and time trade off, a utility method of measurement. If small differences between groups of patients are to be shown, measures of outcome that are more responsive to patient change should be used.