Building caries risk assessment models for children

Area of Science:

• Pediatric Dentistry
• Dental Public Health
• Biostatistics

Background:

• Caries risk assessment is vital but lacks practical models.
• Biopsychosocial models are needed for diverse settings.
• Accurate prediction of caries increment and burden is essential for intervention.

Purpose of the Study:

• To develop and validate biopsychosocial models for caries risk assessment in children aged 3-6 years.
• To evaluate the predictive accuracy of models with and without biological tests.
• To identify cost-effective screening tools for caries control and treatment planning.

Main Methods:

• Prospective study of 1782 children (3-6 years old), followed for 12 months.
• Data collection included questionnaires, oral examinations, and biological tests (salivary, microbiological, plaque pH).
• Risk assessment models were developed using 50% of cases and validated on the remaining 50%.

Main Results:

• Screening models without biological tests predicted one-year caries increment with 82% sensitivity and 73% specificity.
• Full models with biological tests achieved 90% sensitivity and 90% specificity for one-year caries increment prediction.
• A community-screening model using only a questionnaire identified children with high caries burden (dmft > 2) with 82% sensitivity and 81% specificity.

Conclusions:

• Developed biopsychosocial models demonstrate high accuracy for caries risk assessment.
• Models incorporating biological tests offer superior prediction of caries increment.
• Questionnaire-based community screening models are effective for identifying high-caries-burden children, supporting cost-effective interventions.