WestCare uses AWS to give clinicians more time to focus on healing
WestCare launched its SPIRIT initiative to reduce clinical documentation burden and modernize its EHR platform. The project uses Amazon Bedrock with Cohere to generate draft clinical documentation from EHR data and AWS infrastructure for scalable, reliable operations.
- Organization
- WestCare
- Industry
- Healthcare
- Location
- United States
- Published
- August 2025
Reported outcomes
99.9%
timeTime & speed
Strategic outcomes
Catalog median for time & speed deployments: +80% across 203 reported metrics. Compare benchmarks →
Primary read
Use case focus
Showing 3 of 4
- 1Clinical documentation automation
- 2Workforce productivity
- 3EHR modernization
- Clinicians spent over 230,000 hours in 2023 on clinical documentation.
- Documentation burdens contributed to burnout, turnover, and less time for direct patient care.
- WestCare built a generative AI workflow on Amazon Bedrock with Cohere to draft assessments, treatment plans, and case notes from EHR data.
- It also re-architected its EHR platform and automated CI/CD using Amazon EC2 and data workflows on Amazon S3, with governance and clinician training.
- Estimated 50 percent reduction in documentation time, freeing over 115,000 hours annually for direct patient care.
- 20 percent increase in patient throughput.
- 35 percent reduction in provider burnout.
- 50 percent faster release cycles.
- 99.9 percent uptime.
Architecture
WestCare’s SPIRIT initiative combines a re-architected EHR platform with a generative AI documentation workflow. Amazon Bedrock with Cohere generates draft clinical documentation from EHR data. Amazon EC2 hosts an automated Jenkins CI/CD pipeline for the EHR platform. Amazon S3 supports ETL jobs to transform and store PostgreSQL data in FHIR format for API-based interoperability. The deployment includes clinician training and AI governance controls such as acceptable-use acknowledgment and client opt-out consent.
Sources & evidence1
AI-generated summary. Verify important details with the linked sources before relying on this case.
Explore related AI use cases
Was this useful?