Development of the Dundee Caries Risk Assessment Model (DCRAM)--risk model development using a novel application of...

Area of Science:

• Pediatric Dentistry
• Dental Public Health
• Biostatistics

Background:

• Early childhood caries (ECC) poses a significant public health challenge.
• Accurate caries risk assessment is crucial for timely preventive interventions in preschool children.
• Existing risk models may require refinement for specific community settings.

Purpose of the Study:

• To develop and validate a novel caries risk assessment model for preschool children.
• To utilize a specific statistical analysis, chi-squared automated interaction detector (CHAID), for model development.
• To create a model applicable to a particular community setting, predicting caries at age 4 from early data.

Main Methods:

• Longitudinal data collection from approximately 1500 children in Dundee, Scotland.
• Data included dental examinations, salivary microbiological samples, and parental/health visitor questionnaires at ages 1, 2, 3, and 4 years.
• Chi-squared automated interaction detector (CHAID) analysis applied to 1-year data to predict caries at 4 years.

Main Results:

• Four caries risk models were developed using CHAID analysis for predicting caries at age 4.
• Models included 'any' caries (dmft >0) and 'high' caries (dmft ≥3) risk categories.
• The optimal model, the CHAID high caries-risk model at the d(3) level (d(3)mft ≥3), demonstrated 65% sensitivity and 69% specificity.

Conclusions:

• An appropriate caries risk assessment model, the Dundee Caries Risk Assessment Model, was successfully developed.
• This model predicts caries risk at age 4 years using data collected at age 1 year.
• The model is suitable for use in a specific community setting, facilitating targeted preventive strategies.