kras99 - stock.adobe.com
OpenAI launches Epic integration with ChatGPT for Healthcare
Users will be able to use ChatGPT to query the Epic Systems medical record, helping streamline their care management workflows.
A new partnership between Epic Systems and OpenAI will let healthcare professionals use ChatGPT to query the EHR. The integration, unveiled yesterday, is designed to make it easier for providers to find and synthesize data for better care management.
OpenAI also released its Healthcare Public Data plugin, which will let users get ChatGPT answers based on nine official healthcare datasets such as PubMed, DailyMed and the CMS Coverage Database.
"Healthcare organizations need AI that works across the systems and information central to care and operations," OpenAI said in the announcement. "Patient context, medical evidence, public healthcare data, and organizational knowledge often live in different places. Connecting these sources in a governed workspace helps teams find the right information, understand it in context, and put it to work across the business."
The Epic integration is big news for healthcare professionals bogged down trying to parse information from the EHR ahead of patient visits. OpenAI said organizations that deploy the integration can now query ChatGPT with context from the Epic record, such as asking which screenings a patient is due for or which labs the clinician should review before the visit.
This means users can get EHR data context when they use ChatGPT, which could let them review patient history or prepare for appointments. In some deployments, ChatGPT will be integrated directly into the EHR workflow, allowing provider users to leverage the AI without leaving the patient chart.
This functionality has been a time-saver for providers at UCSF Health, which was a pilot partner on the project.
"By bringing relevant information together more quickly and comprehensively, the technology has the potential to reduce time spent synthesizing data and give clinicians more time with patients," Suresh Gunasekaran, president and CEO, UCSF Health, said in the press release. "We're also engaging frontline teams to validate these capabilities in practice and help shape where they can add the most value."
Verified health resources have entered the chat
OpenAI also outlined its new Healthcare Public Data plugin, which directs ChatGPT to pull from a list of nine leading health information sources for its answers. Those sources include the following:
- ClinicalTrials.gov
- CMS Coverage Database
- CMS Open Data
- DailyMed
- openFDA
- Medicare Care Compare
- National Plan and Provider Enumeration System
- PubMed
- RxNorm
These plugins will build on what ChatGPT already does -- help "teams answer clinical questions and synthesize medical research with trusted clinical search across thousands of medical sources," OpenAI said.
But the Healthcare Public Data plugin will be more dedicated, drawing in nine official public healthcare resources. OpenAI said healthcare teams can work within ChatGPT using these specific sources, which should make it easier for them to verify key information.
For example, a research team might use ClinicalTrials.gov to identify trials in active recruitment or a pharmacist could use DailyMed to look at the most recent warnings for a medication. Teams can also look at all nine Healthcare Public Data plugin resources at once, which could be effective for those doing multidisciplinary work, like population health teams.
OpenAI touted the safety and efficacy of these integrations, saying it worked with its physician partners to evaluate the connected EHR context across 27 clinical use cases. Across 4,363 ratings, physicians rated 99.1% of responses as safe across all use cases. They also rated 93% of tested ChatGPT responses as having "good" or better accuracy.
Sara Heath is an executive editor at Xtelligent Healthcare Media, where she covers patient engagement, healthcare policy and health IT.
