Reliability and validity of the Western Ontario and McMaster Universities (WOMAC) Osteoarthritis Index in Italian patients with osteoarthritis of the knee

Abstract

Objective: The Western Ontario and McMaster Universities (WOMAC) Osteoarthritis (OA) Index is a tested questionnaire to assess symptoms and physical functional disability in patients with OA of the knee and the hip. We adapted the WOMAC for the Italian language and tested its metric properties in 304 patients with symptomatic OA of the knee. Methods: Three hundred and four consecutive patients, attending 29 rheumatologic outpatient clinic in northern, central, and southern Italy, were asked to answer two disease-specific questionnaires (WOMAC and Lequesne algofunctional index) and one generic instrument (Medical Outcomes Study SF-36 Health Survey—MOS SF-36). A sample of 258 patients was readministered the WOMAC 7–10 days after the first visit and the structured interview, which also assessed demographic and other characteristics. Internal consistency was assessed using Cronbach's alpha, reliability using intraclass correlation coefficients (ICCs), and construct and discriminant validity using Spearman's correlations, Wilcoxon rank sum test, and Kruskal–Wallis test. Results: All WOMAC subscales (pain, stiffness, and physical function) were internally consistent with Cronbach's coefficient alpha of 0.91, 0.81, and 0.84, respectively. Test–retest reliability was satisfactory with ICCs of 0.86, 0.68, and 0.89, respectively. In comparison with the SF-36, the expected correlations were found when comparing items measuring similar constructs, supporting the concepts of convergent construct validity. Very high correlations were also obtained between WOMAC scores and Lequesne OA algofunctional index. WOMAC physical function, but not WOMAC stiffness and pain subscales, was weakly associated with radiological OA severity ( P=0.03). Also, WOMAC pain score was inversely correlated ( P=0.01) with years of formal education. Examination of discriminant validity showed that the scores on the WOMAC and SF-36 followed hypothesized patterns: the WOMAC discriminated better among subjects with varying severity of knee problems, whereas the SF-36 discriminated better among subjects with varying levels of self-reported health status and comorbidity. Conclusion: The Italian version of WOMAC is a reliable and valid instrument for evaluating the severity of OA of the knee, with metric properties in agreement with the original, widely used version.