AB0840 Scleroderma classification criteria: Disease-specific methods for valid case comparisons. (23rd January 2014)

• Record Type: Journal Article
• Authors: Johnson, S.R.; Naden, R.L.; Fransen, J.; van den Hoogen, F.; Pope, J.E.; Baron, M.; Matucci-Cerinic, M.; Tyndall, A.; Distler, O.; Denton, C.; Gabrielli, A.; van Laar, J.; Mayes, M.; Steen, V.; Seibold, J.; Clements, P.; Medsger, T.; Carreira, P.; Riemekasten, G.; Khanna, D.
• Is Part Of: Annals of the rheumatic diseases. Volume 71(2012)Supplement 3
• Page Start: 686
• Page End: 686
• Publication Date: 2014-01-23
• Subjects: Rheumatism -- Periodicals; 616.723005
• DOI: 10.1136/annrheumdis-2012-eular.840
• Languages: English
• ISSNs: 0003-4967

Abstract

Abstract : Background: Classification criteria for systemic sclerosis (SSc) are being developed. Twenty-three candidate criteria have been identified and validated. Standardized cases need to be developed for criteria reduction and possible weighting. Objectives: To develop and test SSc-specific case forms for clarity and feasibility. We secondarily explored the agreement between SSc experts on the probability that cases were classified as SSc. Methods: Standardized forms were developed using: skin thickening; nailfold capillary pattern; finger tip ulcers/pitting scars; Raynaud's phenomenon; interstitial lung disease; renal crisis; reduced DLCO; reduced FVC; dysphagia, esophageal dilation; telangiectasia; flexion contractures; pulmonary hypertension; puffy fingers; pulp loss/acro-osteolysis; gastroesophageal reflux; calcinosis; tendon/bursal friction rubs; topoisomerase-I, centromere, RNA polymerase III, antinuclear or PM-ScL antibodies. Six experts tested 6 cases and the ranking procedure for clarity of the instructions, response option and time-to-completion. The forms were applied to purposively sampled SSc cases from international cohorts covering a wide range of probabilities that each case had SSc. Eight experts rank-ordered 20 cases from 1 (highest) to 20 (lowest). The consistency in rankings was evaluated using an intraclass correlation coefficient (ICC). Results: 83% (5/6) endorsed clarity of the form, 100% (6/6) clarity of the instructions, and 100% (6/6) clarity of the response option. The median time-to-completion was 10 minutes (range 10–20). The ICC for agreement across experts was 0.73 (95% CI 0.58, 0.86). Conclusions: Our SSc-specific forms have demonstrable clarity and feasibility. Using these cases, the expert panel had substantial agreement in rank order. Our methods reflect the rigors of modern psychometric science, and serves as a template for groups developing classification criteria. Disclosure of Interest: None Declared