Validity and reliability of the Rosacea Area and Severity Index: A novel scoring system for clinical assessment of rosacea severity
Nita Katarina Frifelt Wienholtz, Jacob P. Thyssen, Casper Emil Christensen, Simon Francis Thomsen, Katrine Karmisholt, Gregor B. E. Jemec, Hans B. Lomholt, Michael Heidenheim, Anne Birgitte Simonsen, Carsten Sand, Christian Vestergaard, Diljit Kaur‐Knudsen, Elisabeth Ammitzbøll, Erik Hillo Lørup, Anne Danielsen, Gitte Strauss, Lone Skov, Peter H. Andersen, Marianne Hald, Luise Winkel Idorn, Messoud Ashina, Claus Zachariae, Alexander Egeberg
Abstract
BACKGROUND: Rosacea is a common chronic inflammatory facial skin disorder. Standardized evaluation of the severity and extent of rosacea is important for baseline assessment and treatment effect. The currently used Investigator's Global Assessment (IGA) is unspecific and fails to consider subtypes/phenotypes of rosacea and area involvement. The Rosacea Area and Severity Index (RASI) was developed to give a more nuanced evaluation of rosacea features in four facial skin areas adjusted to the relative importance of each area of the face to obtain an overall severity score. OBJECTIVES: To validate RASI against the IGA and to assess the inter- and intraobserver reliability for RASI. METHODS: Sixteen dermatologists evaluated photographs of 60 adult patients with rosacea (3 photographs per patient, one from the front and one from each side). IGA and RASI scores were performed for interobserver reliability assessment. To determine intraobserver reliability, 14 dermatologists evaluated 10 other patients twice with at least 1 week interval. RESULTS: The IGA and RASI correlated well (Spearman correlation coefficient (SCC) = 0.75, 95% confidence interval (CI) = 0.72-0.78). Interobserver reliability was moderate for RASI and poor to moderate for IGA. Reliability was strongest for rhinophyma, followed by papules/pustules and erythema, and rather weak for telangiectasia. For area scores, interobserver reliability was strongest for cheeks, followed by nose, chin and forehead. We found a moderate-to-strong intraobserver agreement both for IGA and RASI. CONCLUSIONS: We have designed a new practical tool to examine clinical severity of rosacea. RASI proved simple and reliable in scoring clinical severity of rosacea with an agreement comparable to the currently used IGA although RASI will provide a more nuanced view of the current rosacea extent and severity. We suggest that RASI is used in the daily clinical setting as well as in clinical studies assessing the efficacy of rosacea therapies.