SDoH Impact on Periodontal Disease Using Machine Learning and Dental Records.

Area of Science:

• Oral Health Research
• Public Health
• Data Science

Background:

• Social determinants of health (SDoH) critically influence periodontal disease (PD) outcomes.
• Understanding SDoH can inform policy and enhance holistic patient care.
• Machine learning (ML) offers advanced analytical capabilities for complex health data.

Purpose of the Study:

• To determine the association between PD and SDoH using big data analytics.
• To leverage linked dental electronic health records (DEHRs) and census data with ML.
• To identify key SDoH predictors of PD progression.

Main Methods:

• Utilized longitudinal DEHR data from 89,937 patients (2007-2023).
• Applied ML models (logistic regression, naive Bayes, random forest, XGBoost) for association analysis.
• Employed SHapley Additive exPlanations (SHAP) for model interpretability.

Main Results:

• XGBoost model achieved 94% accuracy in predicting PD outcomes.
• Key predictors for PD progression included social vulnerability index, poverty, population density, and environmental factors.
• Tobacco use and comorbidities also significantly correlated with PD progression.

Conclusions:

• SDoH play a critical role in PD progression and oral health inequities.
• Integrating SDoH into PD risk assessment and management is crucial.
• Big data analytics and ML provide valuable insights into oral health disparities.