Comparative validation of the WOMAC osteoarthritis and Lequesne algofunctional indices in Greek patients with hip or knee osteoarthritis

Abstract

To comparatively evaluate the reliability and validity of the Western Ontario and McMaster (WOMAC) and the Lequesne algofunctional indices in Greek patients with hip or knee osteoarthritis (OA). The Greek versions of WOMAC LK 3.1 and Lequesne indices were administered to 97 outpatients with OA. Internal consistency reliability was assessed by Cronbach's alpha and item-scale correlations. Test-retest reliability was examined with intraclass correlations. Patients were also asked to complete the Short Form 36 (SF-36) and a Visual Analog Scale capturing strength of pain, in order to assess construct validity. Additional demographic and clinical data were also recorded to evaluate further associations. Cronbach's alpha values of the WOMAC ranged between 0.92 and 0.98 for hip and 0.89-0.97 for knee OA. The respective values for Lequesne were 0.63-0.74 and 0.74-0.80. Item-scale correlations confirmed the superiority of WOMAC with respect to internal consistency reliability. Intraclass correlations were 0.79-0.97 and 0.57-0.98 for hip and 0.86-0.97 and 0.82-0.97 for knee OA, for WOMAC and Lequesne, respectively. The two indices showed high correlations with comparable subscales of SF-36 and the Visual Analog Scale. Significant relationships were identified for age, body mass index, duration of disease, duration of stiffness and radiographic classification. Our findings, in samples of knee and hip OA patients, indicate that the WOMAC index demonstrates better internal consistency reliability than the Lequesne counterpart, as well as equivalent test-retest reliability and construct validity.