Validity and reliability of the Flare-OA scale for hip and knee osteoarthritis in a Turkish population: item reduction with Rasch analysis.

OBJECTIVES

The Flare-OA questionnaire is a self-reported instrument developed to assess flare in individuals with knee and/or hip osteoarthritis. This study aimed to translate and culturally adapt both the original and short versions of the Flare-OA into Turkish.

METHODS

The Turkish version of the questionnaire was obtained through a process of cross-cultural adaptation and translation. Patients aged 45 years or older with clinically and radiologically confirmed OA of the knee or hip were recruited. The Flare-OA scale, originally consisting of 33 items, was shortened to 19 items and then to a final 16-item version through Rasch analysis. The internal consistency of the Flare-OA was measured using Cronbach's alpha, and its stability over time was tested by evaluating test-retest reliability over a 15-day interval in patients with no clinical changes. The sensitivity to change was determined by calculating the standardized response mean (SRM) in those who reported symptom variation during the follow-up. Convergent validity was assessed by analyzing the correlations between the scale and previously validated measures, including the Hip Disability and Osteoarthritis Outcome Score (HOOS-PS), the Knee Injury and Osteoarthritis Outcome Score (KOOS), and the Mini-Osteoarthritis Knee and Hip Quality of Life Questionnaire (Mini-OAKHQOL).

RESULTS

The study included 185 participants, of whom 71.9% were women, with a mean age of 63.2 years (SD:9.1). Of these, 160 patients (86.5%) had knee OA and 25 (13.5%) had hip OA. In the past four weeks, 70 patients (37.8%) reported a worsening of symptoms in the affected joint. Cronbach's alpha coefficient was 0.987 (95% CI 0.984-0.990) for the 33-item and 0.972 (95% CI 0.966-0.978) for the 16-item. The intraclass correlation coefficient was 0.913 and 0.912 for the test-retest reliability (n=79) of the 33- and 16-item tests, respectively. Sensitivity to change was good in 9 patients with flare improvement [SRM 1.2 (95% CI 0.6-1.7), SRM 1 (95% CI 0.5-1.5), for 33- and 16-items, respectively] over the period. Discriminant validity was supported by statistically significant score differences between patients with and without flare for both the 33-item [36.2; 95% CI 29.9-42.6; SEM: 8; p<.0001] and the 16-item [36.7; 95% CI 30.3-43.0; SEM: 8.1; p<.0001] versions. There was a significant and negative correlation between the Flare-OA score and KOOS and mini-OAKHQOL (p<0.05). 16-item Rasch modeling allowed us to reduce the questionnaire to a 16-item version with good fit and a satisfactory interval scale.

CONCLUSION

The Turkish versions of the Flare-OA questionnaires (33- and 16-item) showed high reliability, validity, and clinical utility in evaluating flares in knee and hip OA. The 16-item version appears especially useful for routine use, although further validation is needed due to the limited sample size in the hip OA subgroup.