Comparison of validity, and responsiveness between general and disease-specific quality of life instruments (Thai version) in knee osteoarthritis.

Short-form 36 (SF-36) and Western Ontario and McMaster Universities Osteoarthritis index (WOMAC) are common instruments for measuring quality of life (QoL) in patients with knee osteoarthritis. The goal of the present study was to compare the performance of both instruments in evaluating QoL in patients with knee osteoarthritis as diagnosed by the American College of Rheumatology (ACR) criteria. Treatment included nonsteroidal anti-inflammatory drugs and patient education for 6 weeks. Face-to-face interview by an orthopaedist was done at baseline and after treatment, including collection of demographic data and use of both SF-36 and WOMAC questionnaires. Evaluation of instrument performance included reliability, validity, and responsiveness measures. Reliability was tested by analysis of internal consistency using Cronbach's alpha at baseline and after treatment. Construct validity was computed by determining the correlation between each domain of SF-36 and WOMAC (Pearson's test). Responsiveness was compared between baseline and after treatment of both SF-36 and WOMAC in each domain using the paired t test. Fifty-two patients (8 men, 48 women) with a mean age of 58.4 years were included in the present study. About 75% of subjects had less than secondary education levels and most were from agricultural communities. Sixty-four percent had mild grade knee osteoarthritis. The internal consistency of WOMAC revealed good levels of reliability, both at baseline and after treatment, in all dimensions. The reliability of SF-36 was relatively low, especially in the role physical and bodily pain dimensions (Cronbach's alpha < 0.700). Construct validity between each dimension in SF-36 and WOMAC demonstrated coefficients ranging from -0.05 to -0.409. Both WOMAC and SF-36 showed good responsiveness when comparing scores before and after treatment in all domains. In conclusion, both the Thai version WOMAC and SF-36 were valid, reliable, and sensitive to change in evaluating QoL in Thai patients with knee osteoarthritis.