Cariogram validity in schoolchildren: a two-year follow-up study.

The validity of Cariogram in relation to caries increment over a 2-year period was evaluated. In 2007, the caries risk profile in a group of Sardinian schoolchildren (957) aged 7-9 years was assessed using the Cariogram software. A re-examination using the same criteria was performed 2 years later on 861 individuals from the original sample (drop-out 10.0%). The possible correlated variables were analyzed using the principal component analysis (PCA). The performances of Cariogram in predicting caries increment were evaluated by receiver operating characteristic (ROC) analysis. At follow-up examination, 54.4% of the sample had developed new carious lesions (mean DFS 1.6, 95% CI 1.5-1.8). The mean caries increment (ΔDFS) was 0.5, 95% CI 0.4-0.5. PCA showed that Cariogram, gingival status and dietary sugar frequency, both at baseline and at follow-up, tend to form a separate cluster (goodness of fit ≥0.75). Sensitivity and specificity measured by ROC analysis were 0.83 and 0.85, respectively, so the gain in certainty was 1.68, while the area under the ROC curve was 0.93. A strong correlation between caries risk profiles at baseline and caries incidence in the permanent teeth after 2 years was found. The validity of Cariogram was confirmed, the software fulfilling the criteria for a good risk assessment model: precision, accuracy and ease of use.