CanadaHealthcare and life sciences2024 – ongoingFraser Health
AI prediction of hospital discharge readiness
Fraser Health deployed discharge predictions across 12 hospitals after reporting 86% test accuracy
Problem
Clinicians estimated discharge timing during rounds. Uncertainty made it harder to arrange home support and plan hospital capacity.
Approach
Fraser Health's analytics team built a model combining demand forecasting and generative AI. It analyzed clinical information and unstructured notes to predict readiness within 24 hours. Predictions appeared in the electronic health record and a dashboard used in daily work.
Results
Takeaway
The accuracy figure came from testing before deployment. Care teams retained the final discharge decision.
Categories
- Size
- Large enterprise
- Industry
- Healthcare and life sciences
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and MLGenerative AI
- Adoption
- Built in-house
- Evidence
- Company's own claim
Sources
Source summaries
Fraser Health described model development and integration into clinical workflows. It reported 86% accuracy before deployment across 12 hospitals and said clinicians retained final authority.
Full details are in the original sources.