OTA 1998 Posters

Interobserver Reliability and General Applicability of Rank Order Classification of Fractures of the Tibial Plafond

Douglas R. Dirschl, MD; John L. Marsh, MD; Tom DeCoster; Shepard Hurwitz University of North Carolina School of Medicine, Chapel Hill, NC; University of Iowa School of Medicine, Iowa City, IA; University of New Mexico School of Medicine, Albuquerque, NM; University of Virginia School of Medicine, Charlottesville, VA

Purpose: Fractures of the tibial plafond are complex injuries, and previous studies have shown fair to poor interobserver reliability in the classification of these injuries. In an effort to explore methods of improving reliability in the classification of fractures, the authors wish to explore the rank order method of classification of tibial plafond fractures. The purpose of this study is to determine the interobserver reliability and the general applicability of the rank order method of classification of tibial plafond fractures by eliciting the participation of attendees at the OTA annual meeting.

Methods: Through the construction of a poster presentation for the 1998 OTA annual meeting, the authors will enlist the participation of meeting attendees in determining the reliability and general applicability of the rank order method of classification of tibial plafond fractures. The poster presentation will include high quality photographs of mortise and lateral radiographs of ten fractures of the tibial plafond. Fractures will be selected to represent a spectrum of injury severity. The ten fractures will be numbered and presented on a single poster to allow observers direct comparison of the ten cases. Observers will consist of all attendees at the OTA annual meeting who are willing to participate. Observers will receive an instruction and scoring sheet from a table in front of the poster, will rank the ten cases from least severe (#1) to most severe (#10) fracture, and will return their instruction and scoring sheets to the designated box on the table in front of the poster. Instructions to observers will be only that they rank the ten cases from least severe to most severe injury, taking into consideration all factors they believe important in performing such a ranking; observers will not be instructed to consider specific factors in making their rankings. Observers will be requested to participate in the ranking of the fractures only once during the OTA meeting. Observers will provide information on the scoring sheet as to their orthopaedic specialty training and experience, and their name (optional).

Results: All data obtained at the meeting will be analyzed to determine the interobserver reliability of the rank order method for classification of fractures of the tibial plafond. Interobserver reliability will be expressed using Cronbach's alpha. Results will be stratified according to orthopaedic specialty training (resident, fellow, traumatologist, and general orthopaedist) and amount of prior experience in the classification and treatment of fractures of the tibial plafond. Questionnaires will be sent following the OTA meeting to observers providing their names to assess general attitudes and responses to the applicability of classifying fractures according to the rank order method.

Discussion: The rank order method of classification has only rarely been used in clinical orthopaedic research, and its use has generally been limited to instances where objective measures to stratify data were not available, such as the quality of resident performance or the relative importance of various surgical procedures. The rank order method has been shown to have high inter-rater agreement for subjective outcome measures in other areas of medicine. The fair to poor interobserver reliability of accepted fracture classification methods seems to indicate these measures are not as objective as once thought, and it follows that the application of the rank order method to fracture classification is therefore logical. Orthopaedists may disagree about the classification of fractures, but the authors believe there will be a high level of agreement on relative fracture severity. It remains to future study to determine the wide spread applicability of rank order techniques for meaningful stratification of orthopaedic injuries. If the rank order method is found to be reliable and reproducible, then further study to determine the factors used to rank fractures may lead to the development of improved systems for fracture classification.