The EUSTAR model for teaching and implementing the modified Rodnan skin score in systemic sclerosis

Objective: To evaluate the ability to teach scleroderma experts and young rheumatologists to perform the modified Rodnan skin score test. Methods: Three international "teaching courses for teachers" were conducted with 6-9 experts who performed 3-9 skin score tests each. In addition, an international course for 90 young rheumatologists, in which 18 patients with systemic sclerosis (SSc) participated, was also organised. Finally, a local repeated training course for 5-9 rheumatologists was performed, in which 6-7 patients with SSc participated. Results: When 6-9 scleroderma specialists investigated the patients, the intraclass correlation coefficient (ICC) showed "good" to "excellent" values (0.865 and 0.710, respectively). When 90 young rheumatologists were involved in one teaching course, the coefficient of variation (CV) was relatively satisfactory (35%) owing to the high number of investigators, and with a considerable within-patient SD value of 5.4. Repeated teaching of 5-9 young rheumatologists in local courses clearly improved the consistency. The ICC increased from 0.496 to a "good" level of 0.722. The within-patient SD values for intraobserver variability ranged between 2.5 and 2.9. The intraobserver CV was about 20%. Conclusions: This study strongly supports the need for standardisation among different centres when using skin scoring for clinical trials. The intraobserver variability and within-patient SD values can be significantly reduced by repeated teaching. For inexperienced rheumatologists, at least one repeated teaching course is needed.