Caries experience variables as indicators in caries risk assessment in 6–7-year-old Chinese children

Objectives The aim of the present study is to investigate variables of the past caries experience as indicators of future caries in a Chinese child population. Methods Caries was recorded at baseline and 2 years later in 433, 6–7-year-old Chinese children. Correlation coefficients between variables of the past caries experience and caries increment were calculated. Stepwise logistic regression analyses provided predictor variables. ROC curves presenting sensitivity as function of (100-specificity) were employed to summarize the obtained information. The area under the ROC curves was used as a measure of predictive accuracy. Results Twelve variables of baseline caries experience had a significant (p ≤ 0.05) correlation with caries increment in the permanent dentition. Among the variables, the number of primary molars with enamel or dentine caries, fillings or missings due to caries (d12mftmol), the number of fissures of the first permanent molar with enamel caries (D1fis) and the number of pits and fissures of the first permanent molar with dentine caries (D2pitfis) were the most powerful caries predictors, which resulted in a value of 0.74 under the ROC. Conclusions By including non-cavitated caries in the past caries experience of primary molars and permanent first molars, it was possible to construct a screening criterion (d12mftmol + D1fis + D2pitfis), which enabled the selection of true negative children and the selection of a high caries risk group in a 6–7-year-old Chinese child population.