Variability in the assessment of acetabular index

Abstract

This study was undertaken to assess the inter- and intraobserver variability associated with the assessment of acetabular index. Five readers (three senior orthopaedic residents and two pediatric orthopaedic attendings) read each of 24 films on three separate occasions (360 total readings). An independent observer who was blinded to the results of the study assessed whether the radiographs would result in a reproducible acetabular index. Variability was increased for the radiographs that were deemed poorly reproducible by the independent observer. Interobserver variability exceeded intraobserver variability. Variability was comparable for senior orthopaedic residents and attending pediatric orthopaedic surgeons. Even in the group with the least variability (intraobserver variability for reproducible radiographs), the 95% tolerance interval was 8.35 degrees. The results of this study cast doubt on the reliability of the acetabular index based on a single reading.