Interobserver variability in pediatric radiographic quality assessment

Programs to monitor radiographic quality have been established in several Canadian provinces. These quality assurance programs depend in part on the evaluation of technical radiographic errors. Interobserver variability in the assessment of these errors reduces the reliability of such information. This study examined the interobserver variability of three dental specialists involved in documenting radiographic technical errors, and the need for retakes following the initial radiographic surveys of 200 children. Observers were trained and calibrated to standardize the evaluation of errors. The overall frequency at which errors were found in bitewing radiographs was consistent with the results of previous studies, but the number of periapical and panoramic technical errors was low. Interobserver variability occurred in the recording of radiographic technical errors, primarily in the recording of bitewing errors. There was also interobserver variability in the ordering of retakes. The interobserver variability in the assessment of radiographic quality could not be eliminated, despite prior observer training. This could indicate that the decision to retake a radiograph may be based on subjective criteria, especially in the absence of clinical information.