Clinical Report: Utilizing Machine Learning to Identify Oral Frailty Determinants
Overview
This study identifies a high prevalence of oral frailty (OF) at 69.3% among long-term hospitalized patients with schizophrenia. Utilizing machine learning, key determinants of OF were identified, enhancing predictive model accuracy and providing insights for clinical interventions.
Background
Oral health issues are prevalent among patients with schizophrenia, significantly impacting their quality of life. Oral frailty represents a critical condition that can exacerbate these issues, leading to severe health complications. Understanding the determinants of OF is essential for developing targeted interventions in this vulnerable population.
Data Highlights
Metric
Value
Prevalence of OF
69.3%
Optimal Model AUC
0.779
Performance Improvement
6.57%
Key Findings
The prevalence of oral frailty in hospitalized patients with schizophrenia is 69.3%.
The random forest model achieved an AUC of 0.779 after two-stage feature selection.
Feature selection improved model performance by approximately 6.57%.
Core risk factors for oral frailty include Number of Teeth, Number of Psychiatric Hospitalizations, Self-discontinuation of Medication, Marital Status, and Age.
Oral frailty is associated with increased risks of malnutrition, cognitive impairment, and dysphagia.
Clinical Implications
The identification of key determinants of oral frailty can guide the development of individualized oral health interventions for patients with schizophrenia. Enhanced predictive modeling through machine learning can improve clinical outcomes by facilitating early identification and management of oral health issues.
Conclusion
The study highlights the significant prevalence of oral frailty in schizophrenia patients and the effectiveness of machine learning in identifying its determinants. These findings can inform clinical practices aimed at improving oral health in this population.
