Intraobserver and Interobserver Reliability of the Oberg-Manske-Tonkin (OMT) Classification: Establishing a Registry on Congenital Upper Limb Differences. Issue 1 (January 2018)

• Record Type: Journal Article
• Authors: Bae, Donald S.; Canizares, Maria F.; Miller, Patricia E.; Roberts, Summer; Vuillermin, Carley; Wall, Lindley B.; Waters, Peter M.; Goldfarb, Charles A.
• Is Part Of: Journal of pediatric orthopaedics. Volume 38:Issue 1(2018)
• Publication Date: 2018-01
• Subjects: intraobserver -- interobserver -- reliability -- classification of congenital differences -- upper limb -- registry; Pediatric orthopedics -- Periodicals; 618.927
• DOI: 10.1097/BPO.0000000000000732
• Languages: English
• ISSNs: 0271-6798

Abstract

Abstract : Background: The purpose of this investigation was to determine the reliability of the Oberg-Manske-Tonkin (OMT) classification system applied to patients enrolled in a prospective, multicenter cohort study of congenital upper limb differences. Our hypothesis was that the OMT classification would exhibit high intraobserver and interobserver reliability and thus would be an appropriate tool for the new Congenital Upper Limb Differences registry. Methods: Four pediatric hand surgeons independently reviewed the medical records, clinical photographs, and radiographs of 60 randomly selected patients previously enrolled in the Congenital Upper Limb Differences registry. Patient's information was summarized and shared electronically (PowerPoint). Each rater classified each congenital anomaly according to the OMT classification system. Responses were recorded using a web-based data capture tool (REDCap). Three weeks later, the surgeons repeated the process. Intraobserver reliability for each rater was assessed using Fleiss' κ. Interobserver reliability was assessed using percent exact agreement (when all 4 raters were in agreement) as well as Fleiss' κ. Results: Using the OMT classification, intraobserver reliability for the 4 pediatric hand surgeons showed almost perfect agreement, with κ values in the range of 0.89 to 0.93. Interobserver reliability demonstrated substantial agreement, with κ value of 0.79 (95% confidence interval, 0.77-0.82) in the first reading and 0.80 (95% confidence interval, 0.77-0.83) in the second reading. The highest possible agreement (κ=1) was seen for the following diagnoses: congenital dislocation of radial head (OMT I.A.2.v), Madelung deformity (I.A.2.vii), radial polydactyly (I.B.2.iii), triphalangeal thumb (I.B.2.iv), Kirner deformity (I.B.4.vi), and osteochondromatosis (3.B.4.i). Conclusions: The OMT classification of congenital upper limb differences exhibits substantial to almost perfect intraobserver and interobserver reliability among pediatric hand surgeons at different institutions. The OMT is therefore well-suited for classification of congenital upper limb anomalies in patients enrolled in a multicenter registry. Level of Evidence: Level II—diagnostic.