OpenAI is expanding ChatGPT's healthcare capabilities with a new integration connecting Epic electronic health records (EHRs) to ChatGPT for Healthcare. The integration allows clinicians to securely bring patient information into ChatGPT to prepare for appointments, review medical histories and identify changes in a patient's health over time. UCSF Health is currently piloting the technology.
ChatGPT can synthesize information from across a patient's record, including appointment notes, laboratory results, medications and specialist documentation. The system can also direct clinicians back to the underlying information in the patient's chart, allowing them to verify the source behind generated summaries and insights. Healthcare organizations can embed ChatGPT directly into supported Epic layouts, enabling clinicians to use the AI without leaving their existing EHR workflow.
Alongside the Epic integration, OpenAI launched a Healthcare Public Data plugin connecting ChatGPT to nine official healthcare data sources. These include ClinicalTrials.gov, CMS Coverage, RxNorm, DailyMed and PubMed. Clinicians and researchers can use the tool to search and compare structured healthcare information, identify recruiting clinical trials and evaluate eligibility criteria, while pharmacy teams can access current drug labeling and safety information.
The Epic integration is available to ChatGPT for Healthcare customers through their workspace administrators rather than individual accounts. The Healthcare Public Data plugin is also available to eligible U.S. ChatGPT for Clinicians users, expanding access to authoritative external healthcare information alongside the patient-specific context available through enterprise EHR connections.
The launch reflects OpenAI's broader push to embed AI directly into clinical workflows rather than requiring clinicians to use separate applications. By combining patient-specific EHR information with trusted external healthcare data, the company aims to help clinicians spend less time navigating fragmented information and more quickly access relevant context for clinical decision-making.
Click here to read the original news story.
