A Computational Framework for Tailored Preventive Care Recommendations Using Electronic Health Records.

Area of Science:

• Health Informatics
• Artificial Intelligence in Healthcare
• Preventive Medicine

Background:

• Healthcare systems are largely reactive, with 90% of expenditures linked to chronic and mental health conditions.
• Current preventive care clinical decision support (CDS) in electronic health records (EHRs) uses a "one-size-fits-all" approach, neglecting patient-specific risk factors.
• Social and environmental determinants of health significantly impact outcomes, yet are underutilized in preventive care strategies.

Purpose of the Study:

• To develop a computational framework for personalized preventive care recommendations.
• To integrate U.S. Preventive Services Task Force (USPSTF) guidelines with patient-specific EHR data.
• To leverage artificial intelligence (AI) for analyzing complex health data and generating tailored advice.

Main Methods:

• Developing a computational framework integrating preventive care guidelines and EHR data.
• Extracting patient-specific risk factors (family history, social history, ethnicity, chronic conditions) from EHRs.
• Utilizing AI to analyze USPSTF guidelines and patient data for personalized recommendations.

Main Results:

• A novel framework for personalized preventive care recommendations has been developed.
• The system integrates diverse patient data, including social determinants of health, for tailored advice.
• Justifications for recommendations are provided, grounded in both EHR data and established guidelines.

Conclusions:

• Personalized preventive care, informed by AI and comprehensive patient data, is crucial for improving public health.
• This AI-driven approach enhances preventive strategies beyond traditional, generic methods.
• The framework offers a pathway to more effective and individualized disease prevention in healthcare systems.