Skin model for improving the reliability of the modified Rodnan skin score for systemic sclerosis

Abstract

BackgroundThe gold standard for skin thickness assessment in systemic sclerosis (SSc) is the modified Rodnan skin score (mRSS); however, inter- and intra-rater variation can arise due to subjective methods and inexperience. The study aimed to determine the inter- and intra-rater variability of mRSS assessment using a skin model.MethodsA comparative study was conducted between January and December 2020 at Srinagarind Hospital, Khon Kaen University, Thailand. Thirty-six skin sites of 8 SSc patients underwent mRSS assessment: 4 times the first day and 1 time over the next 4 weeks by the same 10 raters. No skin model for mRSS assessment was used for the first two assessments, while one was used for the remaining three rounds of assessments. The Latin square design and Kappa statistic were used to determine inter- and intra-rater variability.ResultsThe kappa agreement for inter-rater variability improved when the skin model was used (from 0.4 to 0.5; 25%). The improvement in inter-rater variability was seen in the non-expert group, for which the kappa agreement rose from 0.3 to 0.5 (a change of 66.7%). Intra-rater variability did not change (kappa remained at 0.9), and the long-term effect of using a skin model slightly decreased by week 4 (Δkappa 0.9–0.7).ConclusionsUsing a skin model could be used to improve inter-rater variation in mRSS assessment, especially in the non-expert group. The model should be considered a reference for mRSS assessment in clinical practice and health education.