Acceptability, reliability, validity and responsiveness of the Turkish version of WOMAC osteoarthritis index

Abstract

Objective: To evaluate the acceptability, reliability, validity and responsiveness of the Turkish version of Western Ontario and McMaster Universities (WOMAC) osteoarthritis (OA) index in physiotherapy outpatient practice in Turkey.

Method: Data were obtained from 72 patients with OA of the knee. They were asked to answer two disease-specific questionnaires (WOMAC LK 3.1 and Lequesne-Algofunctional Index of Severity for the knee) and one generic instrument (Medical Outcomes study SF-36 Survey-SF-36). Acceptability was assessed in terms of refusal rate, rates of missing responses, and administration time. Reliability was assessed using Cronbach's alpha. Content validity was assessed by examining the floor and ceiling effects, and skew of the distributions. Convergent and divergent validity was assessed by examining the Pearson's correlation coefficients. Responsiveness was determined by examining effect size (ES), standardized response means (SRM) and P values generated using Wilcoxon's test.

Results: The overall response rate was 100%. Alpha values for all WOMAC subscales exceeded the value of 0.70 at both baseline and follow-up assessments. Frequency distributions of scores were symmetrical. Subscales had negligible floor and ceiling effects. Both pain and physical function subscales were fairly correlated with the subscales measuring similar constructs of SF-36, whereas they were weakly correlated with other dimensions of SF-36. A good correlation was obtained between WOMAC total and Lequesne index. The pain and physical function subscales of WOMAC index were the most responsive subscales.

Conclusion: The Turkish WOMAC OA index is acceptable, valid, reliable and responsive for use in Turkish patients with knee OA.