An Integrated Framework for Reducing Hospital Readmissions using Risk Trajectories Characterization and Discharge...

Area of Science:

• Healthcare Management
• Patient Safety
• Health Informatics

Background:

• Patients face daily risks of adverse events after hospital discharge to lower care levels.
• Value-based purchasing initiatives increase focus on reducing post-discharge adverse events.
• Current prediction models primarily support post-discharge interventions, not discharge timing.

Purpose of the Study:

• To develop an integrated framework for risk prediction and discharge optimization.
• To support both pre-discharge (timing) and post-discharge (monitoring) interventions.
• To identify high-risk patients for adverse events during care transitions.

Main Methods:

• Combined kernel approach to capture non-linear relationships between length of stay and adverse event risk.
• Employed Principle Component Analysis (PCA) for tractable estimation.
• Developed a discharge timing optimization strategy based on the prediction model.

Main Results:

• The integrated framework effectively predicts patient risk for adverse events.
• Demonstrated the model's utility in optimizing discharge timing.
• Showcased a practical approach for implementing discharge and monitoring interventions.

Conclusions:

• The developed framework supports both discharge timing and post-discharge monitoring interventions.
• This integrated approach enhances patient safety during care transitions.
• The method offers a tractable and implementable solution for reducing adverse events.