Physicians, nurses and other clinicians at Stanford Health Care are now able to interact directly with electronic health records, via a new software tool known as ChatEHR.
WHY IT MATTERS
The tool, built by data scientists at Stanford Medicine, is being piloted as a way to help clinicians engage more seamlessly with their EHRs, as reported this past week in a Stanford news item, and to help alleviate administrative burden by allowing them to ask questions about their patients’ medical histories.
The large language model (LLM)-based tool, similar to OpenAI’s GPT-4, can also help automate the summarization of patient charts, among other capabilities. ChatEHR uses information from an individual’s health records to provide its response.
The technology has been in the works for the past two years, when researchers first realized the potential for LLMs to make EHRs more user-friendly – and make workflows more efficient for clinicians.
ChatEHR is only available to a small group of clinicians at Stanford Hospital right now: about 30 or so physicians, nurses, PAs and NPs, who are tasked with monitoring its accuracy and helping refine its capabilities.
But the hope, Stanford IT leaders say, is that the software, while not able to give medical advice or decision support, will be able to gather data from the EHR about a patient’s longitudinal record and answer questions about their health history that will save time for clinicians at the point of care.
“ChatEHR opens up a new way for clinicians to interact with electronic health records in a more streamlined and efficient manner, whether that’s asking for a summary of the entire chart or retrieving specific data points relevant to the patient’s care,” said Dr. Michael Pfeffer, chief information and digital officer for Stanford Health Care and the School of Medicine, who helped develop and integrate the tool with the health system’s Epic EHR, in the Stanford news item.
[Watch our CIO Spotlight Q&A with Pfeffer from 2023, where he discusses the health system’s AI priorities.]
“This is a unique instance of integrating LLM capabilities directly into clinicians’ practice and workflow,” he added.
THE LARGER TREND
AI-enabled interactions with EHR systems are not new, of course, as natural language processing (NLP) and other tools have helped make documentation and charting easier for doctors and nurses.
But the hype around ChatGPT in 2022 helped spark widespread new interest in NLP and generative AI, and over the past three years it’s become apparent that LLMs could do big things for healthcare when deployed thoughtfully.
Stanford, a long-time leader in AI innovation, is well-positioned to innovate. And the ChatEHR tool looks to have some significant value for overworked clinicians.
At the same time, as shown by its slow rollout of the pilot project – and as demonstrated by two recent Healthcare IT News profiles of Stanford-based AI leaders – the health system is taking a careful and considerate approach to how it’s implementing these tools in clinical settings.
ON THE RECORD
“Making the electronic medical record more user friendly means physicians can spend less time scouring every nook and cranny of it for the information they need,” said Dr. Sneha Jain, a clinical assistant professor of medicine and a ChatEHR early adopter, in the Stanford news item.
“AI can augment the practice of physicians and other health care providers, but it’s not helpful unless it’s embedded in their workflow and the information the algorithm is using is in a medical context,” added Nigam Shah, chief data science officer at Stanford Health Care. “ChatEHR is secure; it’s pulling directly from relevant medical data; and it’s built into the electronic medical record system, making it easy and accurate for clinical use.”
Mike Miliard is executive editor of Healthcare IT News
Email the writer: mike.miliard@himssmedia.com
Healthcare IT News is a HIMSS publication.
Source link
