A pilot comparison study of four clinician‐rated atopic dermatitis severity scales. (1st August 2015)

• Record Type: Journal Article
• Authors: Zhao, C.Y.; Tran, A.Q.T.; Lazo‐Dizon, J.P.; Kim, J.; Daniel, B.S.; Venugopal, S.S.; Rhodes, L.M.; Law, M.G.; Murrell, D.F.
• Is Part Of: British journal of dermatology. Volume 173:Number 2(2015:Aug.)
• Page Start: 488
• Page End: 497
• Publication Date: 2015-08-01
• Subjects: Dermatology -- Periodicals; Skin -- Diseases -- Periodicals; 616.5
• DOI: 10.1111/bjd.13846
• Languages: English
• ISSNs: 0007-0963

Abstract

Summary: Background: There are multiple severity outcome measures for atopic dermatitis (AD). There is a need to compare the reliability of these measures. Objectives: To compare the inter‐rater and intrarater reliability of the objective Scoring Atopic Dermatitis (oSCORAD), Eczema Area and Severity Index (EASI), Six Area, Six Sign Atopic Dermatitis (SASSAD) and Three Item Severity index (TIS); and to analyse the correlation between these outcome measures and the quality‐of‐life instruments Patient‐Orientated Eczema Measurement, Dermatology Life Quality Index and Skindex‐29. Methods: Twelve patients with AD attended a 1‐day scoring exercise by five trained dermatology clinicians. Inter‐rater and intrarater reliability were assessed using the intraclass correlation coefficient (ICC). Correlation between clinician‐rated and patient‐reported measures was analysed using Spearman's rho. Results: Regarding inter‐rater reliability, EASI and SASSAD showed good reliabilities, with ICCs of 0·730 [95% confidence interval (CI) 0·500–0·900] and 0·680 (95% CI 0·440–0·880), respectively. However, the ICCs were poor for TIS and oSCORAD, with 0·497 (95% CI 0·233–0·785) and 0·498 (95% CI 0·234–0·785), respectively. Separate body surface area (BSA) component analyses showed that the oSCORAD BSA component contributed to its inter‐rater variations. Regarding intrarater reliability, EASI and TIS showed excellent ICCs of 0·886 (95% CI 0·744–0·952) and 0·820 (0·614–0·923), respectively, while SASSAD showed a good reliability with an ICC of 0·720 (95% CI 0·424–0·878). However, the intrarater ICC was poor for oSCORAD, with 0·446 (95% CI 0·037–0·730). Regarding correlation with patient‐reported measures, only SASSAD demonstrated moderate correlation with Skindex‐29 (ρ = 0·611, P = 0·035). Conclusions: EASI demonstrated the highest inter‐rater and intrarater reliability, supporting it as the optimal AD severity outcome measure.