Cross-cultural adaptation and psychometric validation of the first Arabic KOOS-12: a reliable tool for assessing knee outcomes in Arabic-speaking populations.

Abstract

Background: The Knee injury and Osteoarthritis Outcome Score-12 (KOOS-12) is a widely used patient-reported outcome measure assessing pain, function, and quality of life (QOL) in individuals with knee osteoarthritis or injury. However, the absence of validated short-form knee instruments in Arabic necessitates cross-cultural adaptation.

Methods: A prospective cross-sectional validation of the Arabic KOOS-12 (KOOS-12 AR) was conducted following established guidelines for translation and adaptation (independent forward translations, reconciliation, blinded back-translation, expert review, pilot testing, and proofreading). Adults > 40 years with knee osteoarthritis were consecutively recruited from two outpatient clinics and completed the KOOS-12 AR and the validated Arabic WOMAC (WOMAC-AR). A total of 201 patients were enrolled, and a clinically stable subset of 91 participants (the retest subgroup) completed a retest after 5-10 days. Psychometric evaluation included interpretability, internal consistency (Cronbach's α), test-retest reliability using intraclass correlation coefficients (ICC(2,1)), and measurement error indices such as standard error of measurement (SEM), minimal detectable change (MDC), and content, structural, and convergent validity.

Results: Internal consistency was excellent across domains (Cronbach's α = 0.93). Test-retest reliability was likewise excellent (ICC(2,1) = 0.990 for Pain, 0.992 for Function, 0.989 for QOL, and 0.987 for the Total score). SEM values were small (2.1-2.3 points), yielding MDC_individual ≈6 points across subscales. Bland-Altman analyses showed minimal bias (-0.36 to -0.02) and narrow limits of agreement (approximately ± 6-7 points). Structural validity was supported by principal component analysis (PCA): each subscale demonstrated a clear one-factor solution (Pain eigenvalue 2.50, 62.5% variance; Function 2.72, 67.9%; QOL 2.48, 62.1%). When all 12 items were analyzed together, the dominant first component explained 50.6% of total variance (Kaiser-Meyer-Olkin (KMO) = 0.91). Convergent validity with WOMAC-AR was strong, with Spearman's ρ = 0.879 for total scores (p < 0.0001).

Conclusion: The KOOS-12 AR demonstrated excellent reliability, validity, structural coherence, and measurement reproducibility. It represents a robust, culturally appropriate, and feasible instrument for evaluating knee pain, function, and QOL among Arabic-speaking patients.