Case Study
Reducing Hospital Readmissions through Predictive Modeling
Using predictive analytics to identify high-risk patients and enable proactive care decisions.
The Challenge
- Difficulty identifying high readmission-risk patients
- Reactive vs proactive care
- Limited resource prioritization
Why It Matters
- Financial penalties
- Patient outcomes
- Operational strain
The Objective
- Readmission Risk
- Resource Efficiency
- Decision Visibility
Our Approach
Data Preparation
Developed relational database using a star schema model, enabling reliable analytics and scalable reporting.
Feature Engineering
Analyzed data to identify key patterns and engineered features to capture utilization and clinical risk drivers.
Predictive Modeling
Built machine learning models to predict readmission risk and support proactive care decisions.
Performance Tracking
Delivered insights through dashboards and reports to track key features and support data-driven decisions.
The Solution
- Uncovering risk factors and readmission risk classification
- EHR integration for early patient readmission risk detection
- Targeted transitional care planning and interventions
- BI reports for risk monitoring and operational insights
Improved patient readmission risk identification
More efficient hospital resource allocation
Increased visibility into risk driver patterns
Better alignment between care, analytics, and decision-making
The Impact
- A small group of drivers impact readmissions rates (elderly and Medicare patients, high severity and mortality cases, and major chronic conditions)
- Early risk factor detection enables efficient inpatient, chronic care, and transitional care management
- Adoption depends on operational usability and objectives (readmission reduction, resource allocation, and hospital utilization)
- The coupling of AI, ML, analytics, and healthcare significantly reduces readmission risk and improves resource allocation and value-based care
Key Takeaways
Project Delivery Team
Kwame Boateng Akomeah, Dev Arora, Tony Lordson, and Anil Kumar Swamy Bandaru
