Underestimation of the prevalence of approximal caries and inadequate restorations in a clinical epidemiological study.

UNLABELLED: The prevalence of clinically diagnosed approximal caries and inadequate approximal restorations is usually underestimated, when compared with radiographic assessment as a 'gold standard'.
OBJECTIVE: The aim of this study was to determine the degree of this underestimation in a clinical epidemiological project.
METHODS: Clinical data (n = 951) of 17- and 23-year-olds were collected by calibrated examiners and bitewing radiographs were taken after obtaining the participant's consent. The radiographic examination of the approximal surfaces on the radiographs of 621 participants was based on criteria derived from the clinical protocol. A second examiner judged 20% of the radiographic material to determine the interobserver agreement, using Cohen's kappa. The agreement proved to be good (kappa = 0.66) for dentine caries diagnosis, and fair (kappa = 0.44) for restoration assessment.
RESULTS: Of the total number of recorded decayed or inadequately restored surfaces only 10.8% and 13.8% respectively were found clinically. Of the total number of clinically sound surfaces, 10.1% had dentine caries on the bitewing radiograph. The percentage of clinically adequate restorations needing replacement was 25.7%.
CONCLUSION: It was concluded that the prevalence of approximal caries and inadequate restorations in this clinical epidemiological study was highly underestimated. At least for these age groups the results contain a serious warning for epidemiologists and general dental practitioners with respect to the outcome of clinical caries diagnosis.