VALIDATION DATA

For the validation study, the RADAI- F5 was collected at baseline, 1 week from baseline, and 6 months from baseline. Foot-related impairments and disability were evaluated using the FFI, and the FIS.
Global disease activity were measured using the DAS-28 ESR and Modified RADAI-F5.

High inter-item correlations (>0.8) were observed for questions 2 and 3, 2 and 4, and 3 and 4 of the questionnaire. Moderate inter-item correlations (>0.6) were observed for questions 1 and 2, and 2 and 5. All other inter-item correlations were between 0.4 and 0.6. Cronbach’s α = 0.90, indicating a high level of internal consistency.

The 1-week (test–retest) reliability was examined using a 2-way mixed intraclass correlation coefficient (ICC) with corre-sponding 95% CIs for baseline and 1- week scores. Mean ± SD RADAI- F5 scores for baseline and 1 week were 4.8 ± 2.58 and 4.91 ± 2.74, respectively. The mean ± SD difference between baseline and 1 week for the RADAI- F5 was 0.11 ± 1.37. The ICC for 1- week reproducibility was 0.87 (95% CI 0.80, 0.91; P ≤ 0.001), indicating very good reproducibility.

Absolute measurement error was evaluated using the standard error of measurement (SEm). The SEm value calculated for the RADAI-F5 from baseline and 1- week data was 0.97. The limits of agreement were – 2.57 and 2.80.

To evaluate the content validity and practical burden of the RADAI- F5, 3 items were evaluated: a 5- point Likert scale regarding questionnaire relevance to participants (ranging from extremely irrelevant to extremely relevant), a 5- point Likert scale regarding participants’ opinions on the readability/understanding of the new questionnaire (ranging from very difficult to very easy), and the time taken to complete the questionnaire. Participants largely considered the RADAI- F5 to be relevant for measuring foot disease activity in RA, with 53% and 29% indicating that it was rel-evant and extremely relevant, respectively. Participants largely considered the RADAI- F5 to be easy to read and understand, with 45.8% and 40% indicating that it was easy and very easy to read/understand, respectively.

The Kaiser-Meyer-Olkin value of 0.83 and highly significant Bartlett’s test (P < 0.001) indicated sampling adequacy and suitability for structure detection and factor analysis. Both Kaiser’s “eignenvalue >1” rule and scree plot examination suggested a 1- factor solution, and this solution explained 73.18% of the common variance. Item loadings on the factor were uniformly >0.4, indicating that all items contributed significantly to the aggregate score.