Objective: To (a) determine the inter-observer variability vis-à-vis mRSS assessment by members of the Scleroderma Research Group before and after mRSS-assessment training by an experienced rheumatologist and (b) determine intra-observer variability.

Material and Method: Between June and August 2013, we conducted a descriptive study of Thai adult SSc patients and all rheumatologists in the Scleroderma Research Group at Srinagarind Hospital, Khon Kaen University, Northeast Thailand. Eleven rheumatologists assessed the mRSS of 22 SSc patients three times (i.e., before and after training, and eight weeks after training). The intra-class correlation coefficient (ICC) and its 95% CI were estimated at week 8 after training.

Results: The mean and standard deviation (SD) of mRSS for inter-observer variability analysis was slightly decreased from before training, after training (by an experienced rheumatologist), and at week 8 after training (17.3±11.9, 16.5±11.1, and 16.2±10.3, respectively). Intra-observer variability had moderate agreement before training (ICC 0.59; 95% CI 0.38-0.78), which increased to good agreement after training and at week 8 after training (ICC 0.60; 95% CI 0.42-0.76 vs. 0.68; 95% CI 0.53-0.82, respectively).

Conclusion: Inter-observer variability for mRSS assessment decreased after training and the reduction persisted for eight weeks after training. The ICC rose from moderate agreement at baseline to good agreement at the end of the study. The mRSS assessment by members of the Scleroderma Research Group was reliable.

Keywords: Systemic sclerosis, Scleroderma, Modified Rodnan skin score, Validation, Inter-observer variability, Intra-observer variability