Reliability of coding depth of approximal carious lesions from non-independent interpretation of serial bitewing radiographs.

In order to help understand the effectiveness of preventive care and the source of variation in treatment decisions, it is important to determine intra- and interexaminer agreement on the presence and depth of radiolucencies on bitewing radiographs when serial radiographs on individuals are available for interpretation. Serial radiographs on 24 subjects were read in succession, and one of four depth codes was assigned to approximal lesions. The radiographs were then read a second time, both by the person who had initially read the radiographs and by a second reader. Intraexaminer agreement on presence of a lesion ranged from 60% to 90%; four of the five readers had an agreement of over 78%. Interexaminer agreement on presence of a lesion ranged from 71% to 84%. Intraexaminer agreement on depth of lesion ranged from 64% to 80%; four readers had an agreement of over 73%. Interexaminer agreement on depth of lesion ranged from 59% to 76%.