RESUMO
OBJECTIVE: To compare the risk for caries in children as determined by Cariogram® software (CS; Stockholm, Sweden) with and without its microbiological component and by a form based on Cariogram® (FBC). METHODS: Children (n = 28) aged 3-9 years were included. Data were collected clinically and from anamnesis. The salivary levels of Streptococcus mutans (SM) were evaluated. A linear regression model was used to determine which variables were predictive for each type of risk analysis. Caries risk was the dependent variable and the independent variables were caries experience, related disease, plaque amount, diet frequency, salivary levels of SM, fluoride sources and clinical judgment. A paired Student t-test was used for the following comparisons: (a) CS with and without SM; (b) CS without SM and FBC; (c) CS with SM and FBC. RESULTS: The mean dmft/DMFT was 5.56 ± 2.51. There was no difference between the methods (p < .05). Regardless of caries risk, the children presented the same levels of SM (p = .889). Caries experience, plaque amount, diet frequency and fluoride sources were predictors of caries risk in all assessment methods. Clinical judgment was a significant predictor in CS. CONCLUSIONS: Caries experience, plaque amount, diet frequency and fluoride sources are valuable predictors of caries risk; microbiological tests are not necessary for evaluating caries risk in children, which can be assessed similarly by CS without SM and FBC.