Predictive validity of the reduced Cariogram model for caries increment in non-cavitated and cavitated lesions: cohort study

Abstract

Background: The aim of this study is to assess the caries prediction of the reduced Cariogram by comparing baseline caries risk profiles with non-cavitated and cavitated lesions over periods of six, twelve, and 18 months. Methods: From May 2016 to October 2017, seven schools in Bhakkar, Pakistan, participated in a cohort study. First base line examination was conducted followed by examinations at 6, 12 and 18 months. Children intraoral examinations were performed on portable dental chair with in school premises by a trained examiner. A modified ICDAS index was used to measure caries at baseline and at follow-up examinations after 6, 12, and 18-months. A receiver operating curve (ROC) analysis was performed to evaluate its effectiveness for predicting dental caries increment. Results: About 40% of children had a low-risk status, 30.5% medium risk, and 29.7% high risk, at baseline risk assessment. At 18 months, 73% of high-risk children, 59% of medium-risk children, and 41% of low-risk children showed a caries increment. For the reduced Cariogram model, the area under the curve on the 6, 12 and 18 months follow-up was 0.63, 0.65 and 0.70 respectively. Conclusions: Our findings indicates that a reduced Cariogram can predict the progression of caries in both cavitated and non-cavitated lesions and model exhibits a level of discriminatory ability. While it might not achieve a very high accuracy, it suggests that the model is able to predict caries increment effectively than random guessing.