Development of a prognostic model for caries onset and progression from early childhood caries incidence in urban preschool children

Prediction Models of Early Childhood Caries Based on Machine Learning Algorithms

In this study, we developed machine learning-based prediction models for early childhood caries and compared their performances with the traditional regression model. We analyzed the data of 4195 children aged 1-5 years from the Korea National Health and Nutrition Examination Survey data (2007-2018). Moreover, we developed prediction models using the XGBoost (version 1.3.1), random forest, and LightGBM (version 3.1.1) algorithms in addition to logistic regression. Two different methods were applied for variable selection, including a regression-based backward elimination and a random forest-based permutation importance classifier. We compared the area under the receiver operating characteristic (AUROC) values and misclassification rates of the different models and observed that all four prediction models had AUROC values ranging between 0.774 and 0.785. Furthermore, no significant difference was observed between the AUROC values of the four models. Based on the results, we can confirm that both traditional logistic regression and ML-based models can show favorable performance and can be used to predict early childhood caries, identify ECC high-risk groups, and implement active preventive treatments. However, further research is essential to improving the performance of the prediction model using recent methods, such as deep learning.

Predicting dental caries increment using salivary biomarkers in a remote Indigenous Australian child population

BACKGROUND

The burden of childhood dental caries amongst Indigenous Australians is higher than in other Australians. Because of differences in lifestyle and the evolutionary history of the oral microbiota, associated risk indicators may differ. Here, we evaluate associations between caries increment, salivary biomarkers and baseline caries among children aged 5-17 years residing in a remote rural Indigenous community.

METHODS

This study was part of a trial assessing cost-effectiveness of an intervention to prevent dental caries among children. Baseline epidemiology and application of topical caries-preventive measures was conducted in 2015, followed-up in 2016 and 2017. Children who did not consent or failed to attend the prevention visits but did attend for follow-up epidemiology constituted a natural comparison group for evaluating the intervention. Saliva flow, pH, buffering and bacterial loads were measured at all visits. Caries was scored by the International Caries Detection and Assessment system. Outcome was caries increment. Explanatory variables were sex, being in experimental or comparison group, baseline caries, saliva flowrate and buffering, pH, and salivary loads of mutans streptococci (MS), Lactobacilli (LB), and yeast. Chi Square tests compared caries incidence in relation to explanatory variables and Generalised Linear Models explored associations between explanatory and outcome variables.

RESULTS

Of 408 participants at baseline, only 208 presented at 2-year follow-up. Of caries-free children at baseline, significantly fewer had incipient (p = 0.01) and advanced (p = 0.04) caries after two years. Children in the experimental group experienced fewer tooth surfaces with advanced caries (p = 0.02) than comparison children. Having caries at baseline (p = 0.02) and low salivary flow-rates (p < 0.001) saw a significant increase in advanced caries after two years. Children with high salivary loads of MS (p = 0.03) and LB (p = 0.004) experienced more advanced carious surfaces. Multivariable analysis revealed 58% reduction (p = 0.001) in advanced caries among children with high salivary flow rates. Caries increment was 61% (p = 0.03) more for incipient and 121% (p = 0.007) more for advanced caries among children who harboured higher loads of MS.

CONCLUSION

As with other ethnicities, children with low salivary flow and those with high MS had higher incipient and advanced caries increments after two years. Such risk assessments facilitate targeted preventive interventions for such communities.

TRIAL REGISTRATION

Australian New Zealand Clinical Trials Registry (ANZCTR), No: ACTRN12615000693527: 3 July 2015.

Pathways influencing dental caries increment among children: A cohort study

BACKGROUND

Dentists should assess pathways influencing the increment of dental caries among children to guide the prevention and treatment of the disease.

AIM

Evaluate the pathways that influence the increment of carious lesions in pre-school children.

DESIGN

This is a 2-year cohort study was conducted with a random sample of 639 pre-school children in southern Brazil. Caries experience, socioeconomic status (SES), social capital, and psychosocial characteristics were obtained at baseline. Increment of dental caries was assessed at 2 years follow-up in 467 children (cohort retention rate of 73.1%). Previously calibrated examiners assess the caries through the International Caries Detection and Assessment System (ICDAS). Structural equation modeling (SEM) was performed to test the pathways influencing dental caries increment.

RESULTS

Dental caries at baseline was heavily influenced by children's age (SC: 0.381, P < .01), tooth plaque (SC: 0.077, P = .02), parent's perception child oral health (SC: 0.295, P < .01), and household (SC: 0.148, P < .01). Increment of dental caries was directly affected by dental caries at baseline (Standardized Coefficients [SC]: 0.377, P < .01). Indirect paths were not significant.

CONCLUSIONS

Dental caries experience was the main factor of direct influence on the increment of caries, reinforcing the theory of risk accumulation over time.