Interrater reliability for the basic categories of the AO/ASIF's system as a frame of reference for classifying distal radial fractures.

This study is about agreement on the assignment into the three basic classes or categories (A, B, C) of the Arbeitsgemeinschaft für Osteosynthesefragen/Association for the Study of Internal Fixation's (AO/ASIF) classification system for distal radial fractures. A random sample of 124 distal radial fractures was classified by two experienced observers. Their agreement was calculated according to Cohen's kappa statistic. To investigate the possible bases for disagreement, all conflicting X-ray assessments were discussed in a consensus meeting. It appeared that the kappa value was .65 (good agreement) before the meeting; kappa rose to .86 (excellent agreement) after the consensus meeting. It appeared that the undisplaced fractures were a major source of disagreement. Further, the presence of articular involvement was an important issue. It was frequently noted that one observer classified the fracture as extraarticular (basic Class A), while the other observer chose classification as an intra-articular fracture (basic Class C) or vice versa. This phenomenon has been called the A/C reversal shift. It is concluded that radiological innovations might enhance agreement on articular involvement, and a separate category for undisplaced fractures should be defined in the Arbeitsgemeinschaft für Osteosynthesefragen (AO) system. However, agreement on relevant distinctive features and discussion of conflicting assessments may also be important in achieving excellent agreement.