Development of a Risk Prediction Model for Assessing Dental Readiness in the Canadian Armed Forces.

Area of Science:

• Military Health
• Dental Public Health
• Predictive Analytics

Background:

• The Royal Canadian Dental Corps (RCDC) mission is to maintain high dental readiness in the Canadian Armed Forces (CAF).
• Assessing and predicting dental readiness is crucial for operational effectiveness.

Purpose of the Study:

• To develop a risk prediction tool for estimating dental readiness in active CAF personnel.
• Predicting non-deployable status within 12 and 18 months.

Main Methods:

• Developed elastic net logistic regression models using dental history and demographic data.
• Utilized two cohorts: recruits (2016-2017) and longer-serving members (LSM) (2014-2018).
• Evaluated model performance using area under the curve (AUC), F1 score, and Brier score.

Main Results:

• Non-deployable classification occurred in 5.1% (12 months) and 9.6% (18 months) of the study population.
• Models achieved an AUC of 0.77 for recruits and 0.70 for LSMs.
• The prediction tool demonstrates potential for identifying individuals at risk of becoming non-deployable.

Conclusions:

• The developed prediction model shows promise for enhancing military dental readiness.
• Further improvements require consistent, high-quality epidemiological data collection with standardized terminology.
• A recalibrated, automated model could aid decision-making and resource allocation for the RCDC.