ChatGPT Integrates Epic EHR, Public Health Data
On September 1, 2026, OpenAI announced a significant expansion for its ChatGPT for Healthcare offering, enabling healthcare organizations to integrate their Epic electronic health record (EHR) environments directly with the AI platform. This development is complemented by a new Healthcare Public Data plugin, designed to link the workspace to nine official healthcare datasets, according to an announcement published in OpenAI’s newsroom.
The company described these two new capabilities as complementary, aiming to enhance clinical decision-making and operational efficiency. The EHR integration facilitates the secure transfer of authorized patient context from Epic into ChatGPT, while the public data plugin provides structured access to vital sources such as PubMed, DailyMed, and CMS Coverage.
Streamlining Clinical Workflows with EHR Integration
By Decode Today News
The integration with Epic EHR environments is set to transform how clinicians access and synthesize patient information. Rather than sifting through extensive appointment notes, laboratory results, medication lists, and specialist documentation, clinicians can now pose direct questions to ChatGPT, grounded in a patient's authorized record. This read-only access ensures that the AI assists in information retrieval without writing back to the patient chart, preserving data integrity and existing EHR workflows, the company said.
OpenAI highlighted several key applications for this integration: clinicians can efficiently review patient history, quickly identify changes since previous visits, and prepare more thoroughly for upcoming appointments. This capability has the potential to significantly reduce the time spent on data synthesis, allowing healthcare professionals to allocate more time to direct patient care.
The integration supports two distinct experiences for healthcare teams. In the first model, authorized patient information from a supported EHR system is securely pulled into ChatGPT. The second experience, available in select deployments, embeds ChatGPT directly into the EHR layout. This allows clinicians to leverage AI-assisted workflows seamlessly, without needing to navigate away from the patient chart, thereby enhancing user experience and workflow continuity.
UCSF Health, a prominent healthcare provider, participated as a pilot partner in this initiative. Suresh Gunasekaran, president and CEO of UCSF Health, emphasized the potential impact. "As a pilot partner, we're exploring how the new EHR integration with ChatGPT for Healthcare can help clinical teams understand what has changed and what matters most across a complex patient record," he stated in the announcement. He added that by consolidating relevant information more rapidly and comprehensively, the technology could reduce data synthesis time, thereby granting clinicians more time with their patients.
Accessing Authoritative Public Health Data
The second core capability, the Healthcare Public Data plugin, consolidates connectors to nine official public healthcare sources. This plugin is designed to provide teams with structured, precise access to a wealth of authoritative information, crucial for research, pharmacy operations, and population health initiatives. These sources include key databases such as ClinicalTrials.gov, CMS Coverage, RxNorm, DailyMed, and PubMed, among others. The company highlighted that teams can work with specific records, fields, identifiers, and versions while directing queries exclusively to these vetted sources.
This plugin aims to simplify the comparison and verification of critical information. For example, users can effortlessly check trial eligibility criteria, confirm medication identifiers, review coverage policy versions, and examine provider records without the need to search each source individually. OpenAI provided several illustrative use cases:
- A research team could utilize ClinicalTrials.gov to identify actively recruiting trials and compare eligibility criteria with greater ease.
- A pharmacy team could leverage DailyMed to quickly confirm the latest labeling and warning information for specific medications.
- A population health team could aggregate research findings, active trial data, and Medicare coverage information to develop or enhance programs, such as a diabetes-prevention initiative.
Understanding the Mechanics of ChatGPT for Healthcare
ChatGPT for Healthcare is engineered to combine healthcare-specific functionalities with robust enterprise-grade controls, ensuring secure and compliant operations. These controls include critical features such as role-based access, allowing only authorized personnel to view sensitive data; single sign-on (SSO) for streamlined and secure access; and comprehensive audit logs, which provide an immutable record of all system interactions. This focus on compliance security is paramount in healthcare, where data privacy is strictly regulated.
For customers operating under an applicable Business Associate Agreement (BAA), the platform supports HIPAA-compliant workflows. This means that sensitive patient information handled within the system adheres to the stringent requirements of the Health Insurance Portability and Accountability Act. Furthermore, customers can integrate ChatGPT Work, Codex, various applications, and other plugins within the same workspace. This enterprise integration capability extends the available business context by supporting existing permissions and connecting to other widely used enterprise systems like Microsoft SharePoint, Google Drive, Salesforce, and Slack. This connectivity enhances the platform's utility across a broader spectrum of business operations within a healthcare setting, potentially improving operating margin through increased efficiency.
Rigorous Evaluation and Enterprise Controls
OpenAI has undertaken extensive efforts to define and measure the quality of health responses generated by ChatGPT. The company collaborates with hundreds of physicians across 60 countries, spanning 49 languages, and representing 26 medical specialties. To date, these physicians have reviewed more than 700,000 model responses, underscoring a significant commitment to accuracy and safety.
To specifically assess performance with connected EHR context, physicians evaluated responses across 27 clinical use cases. These included critical scenarios such as pre-visit reviews, clinical timelines, medication reviews, and handoff summaries. Across a total of 4,363 ratings, physicians rated 99.1% of responses as safe across all evaluated use cases, the company reported. In a separate, two-round evaluation, physicians reviewed hundreds of ChatGPT responses to clinical questions based on large U.S. healthcare datasets. For each of the five connected data sources tested, more than 93% of responses were rated as having "good" or better accuracy, according to OpenAI's self-reported evaluation results. These statistics highlight the AI infrastructure's capability to deliver reliable support in complex medical environments.
Availability and Partner Ecosystem
OpenAI has outlined clear pathways for customers to access these new capabilities. Existing ChatGPT for Healthcare customers are advised to contact their workspace administrator to enable both the EHR integration and the Healthcare Public Data plugin. For ChatGPT Enterprise customers, eligibility and configuration for a Regulated Workspace can be confirmed by contacting their respective OpenAI account team. It is important to note that while eligible U.S. ChatGPT for Clinicians users can install the Healthcare Public Data plugin, the EHR integration is not available for individual accounts.
The company has named several prominent healthcare organizations as launch partners for these products, signifying their early adoption and belief in the platform's potential. These partners include:
- AdventHealth
- Baylor Scott & White Health
- Boston Children’s Hospital
- Cedars-Sinai
- HCA Healthcare
- Memorial Sloan Kettering Cancer Center
- UCSF
This strategic integration of advanced AI with foundational healthcare systems like Epic EHR and a comprehensive public data plugin represents a significant step in AI's evolving role within the health sector. By enhancing access to structured data and patient context, OpenAI's ChatGPT for Healthcare aims to empower clinicians, streamline operations, and ultimately improve patient outcomes through advanced AI capabilities.
