To clarify the current status of oral frailty (OF) in long-term hospitalized patients with schizophrenia and identify influencing factors using machine learning, thereby enhancing clinical understanding and intervention strategies.
Key Findings:
The prevalence of OF in this population was 69.3%, indicating a critical need for targeted oral health interventions.
The optimal predictive model was the random forest, achieving an Area Under the Curve of 0.779 after optimization, demonstrating significant predictive capability.
Key risk factors identified included Number of Teeth, Number of Psychiatric Hospitalizations, Self-discontinuation of Medication, Marital Status, and Age, which can inform tailored intervention strategies.
Interpretation:
The high prevalence of OF highlights the urgent need for targeted interventions, and the identified risk factors can guide the development of individualized oral health programs to improve patient outcomes.
Limitations:
The study is limited to a specific geographic region, which may affect generalizability to other populations.
The reliance on self-reported data may introduce bias, potentially skewing the identification of risk factors.
Conclusion:
The study demonstrates that machine learning can enhance the accuracy of predictive models for OF in schizophrenia patients, providing a foundation for clinical interventions.
