Monitoring osteoarthritis progression and therapy.

Monitoring the progression of osteoarthritis (OA) and the effects of therapy during clinical trials requires valid and reliable outcome measurements. This paper discusses the selection of outcome measures in Phase III clinical trials of OA, the importance of presenting these variables in terms of the response of individual patients, and the combination of information from several outcome measures into a composite index. Four domains-pain, physical function, patient global assessment, and joint imaging (for studies >;1 year in duration)-have been identified as core outcome measures for Phase III clinical trials of OA. Within the symptom severity domains, several measurement instruments may be considered, including visual analog scales (VAS), the Western Ontario McMasters Universities Osteoarthritis (WOMAC) Index, the Lequesne Functional Severity Index, and the Arthritis Impact Measurement Scales (AIMS). Imaging techniques consist of radiography, magnetic resonance imaging (MRI), or other techniques. Although evaluation of these variables is often based on the average improvement in the study population as a whole, evaluation in terms of individual patients is more relevant. Therefore, continuous data collected from individuals (e.g., pain VAS 0-100mm) require conversion to a dichotomous variable (e.g., improvement yes/no) so that the percentage of responders can be determined. Continuous data from individual patients' data may be converted to a dichotomous variable based on global assessments, statistical modeling, or predictive capacities. Such methods suggest that a change in joint space width of >/=0.5mm may be considered clinically relevant. A composite index combining symptomatic and structural efficacy variables, adverse events, costs, and the requirement for surgery into a single variable would be useful. The requirement for total hip arthroplasty has been evaluated as a composite outcome measure for trials of hip OA and found to be valid, simple, and clinically pertinent.