- Healthcare providers still facing frustrations years after switching to electronic health records (EHRs) may soon find some relief from burnout as artificial intelligence moves closer to reality, suggests a viewpoint article published in JMIR Medical Informatics.
Artificial intelligence could help providers assemble and repackage the necessary components of clinical documentation – just like individual, standardized Lego pieces combine to produce an infinite variety of objects – to build notes that accurately reflect a patient encounter or diagnosis.
More intuitive interfaces, coupled with the emerging ability of AI to quickly identify important features of patients and their care, could help providers overcome some of their biggest ongoing gripes about digitalized data.
“In contrast to the historic paper-based documentation workflow, the EHR user must painfully search through the bins of items buried in the software to extract the correct ‘pieces’ of information necessary to complete the entry, requiring click after click after click in that process,” writes a team of researchers from MIT, Beth Israel Deaconess Medical Center, the University of Virginia, and Hospital Israelita Albert Einstein in Brazil.
EHR documentation is one of the most time-consuming tasks in the modern care environment, and one that users seem to dread. A recent AMA study found that clinicians spend twice as much time tapping on the keyboard as they do sitting face-to-face with their patients, which has contributed to widespread dissatisfaction with electronic tools and contributes to physician burnout.
Yet EHRs are here to stay whether providers like them or not, point out Rodrigo Octávio Deliberato, MD, PhD, Leo Anthony Celi, MD, MPH, MS, and David J Stone, MD.
“While the freedom involved in creating paper notes might represent a positive, nostalgic memory, the healthcare system is not going to abandon EHRs with all the manifold advantages that they represent and provide,” the article says.
Instead of longing for the days of manila folders and punchline handwriting, developers and physicians should look to budding competencies in machine learning and natural language processing to supplement EHR notetaking.
At the moment, the process of searching for the right “pieces” of the clinical note is mostly a manual one, the authors continue, that accounts for much of the time, stress, and cognitive drain involved in working with EHRs.
Users must dig through their software to access the right data – or take ill-advised copy-and-paste shortcuts that often result in errors, outdated information, or unnecessary duplications that complicate patient care.
“Instead of rummaging around in a variety of bins for the right pieces…we propose that a carefully engineered implementation of AI into the note creation software elements of the EHR would not only reduce the required rummaging through bins of pieces, but could assist in the assembly of those pieces into the desired output (i.e., a useful, readable, and cogent note that meets all the necessary requirements for clinical documentation).”
Providers and vendors are already laying the groundwork for this approach by integrating multiple sources of data into the EHR.
For example, data streams from the Internet of Things, including home monitors, wearables, and bedside medical devices, can feed predictive analytics algorithms or help to auto-populate notes with vital signs or records of changes in health metrics.
These building blocks, when coupled with additional data sources and wrapped in an artificial intelligence package, “provide a potential means by which to ‘de-bin’ the process of data element selection and assist in the assembly of the data pieces with the goal of improved and more efficient electronic note creation,” the team states.
“AI has the potential to assist users in extracting the right information from the different information systems (i.e., previous electronic notes and bedside monitors, and imaging, laboratory and pharmacy systems), assembling this information into the proper places in the note to assist in the formulation of the assessment with some bounds of certainty, and to analyze that assessment to develop a data-driven plan of action.”
The result would be active, adaptive decision support tools that suggest what the final note should include and what data is best to use in order to paint a clear, accurate, and comprehensive portrait of the individual patient.
Ideally, artificial intelligence could solve some of the user interface problems that make it difficult for providers to visualize and consume historical data, as well. AI tools could pull out the most important features of a medical history – such as a diabetes diagnosis added in the past six months or a recent hospitalization – to help providers make the most of their limited time with patients.
Before any of these improvements can occur, however, EHR developers must address an even more fundamental problem: reducing the amount of time clinicians sit at the keyboard while maintaining an extremely high degree of data integrity and accuracy.
“AI cannot analyze and repackage data until the latter has been incorporated into the system,” the team notes. “The current history and physical examination, whether taken at the bedside or the office examining room, cannot be leveraged for note writing until they are so entered.”
“Better, easier means for this must be devised: this might involve free text entry by voice recognition or keyboard, natural language processing of free text to enter structured data into the system, or new AI modalities as this exploding field develops.”
Stakeholders must also address the core interoperability issues that have made it so difficult for providers to access and share information with the current generation of health IT tools. While the industry has been working diligently to overcome the challenges of proprietary standards, data siloes, privacy concerns, and the lingering competitive disadvantages of sharing data too freely.
Meanwhile, providers and vendors searching for solutions to their daily EHR use woes should view artificial intelligence with the cautious optimism of a technology in its very early infancy.
“AI is not the panacea to every problem in healthcare, but for a relatively repetitive and clearly defined task such as clinical note creation, it seems to provide a fairly ideal solution,” the editorial says.
“It also bestows an opportunity to support an interdisciplinary care environment by learning from inter-specialty communication specifics and facilitating shared decision making by mining patient input and feedback. The final note would be the product of the user, but a user who is not exhausted by painful de-binning and endless clicking to insert the right data in the right places.”
If successfully deployed in the clinical environment, an artificial intelligence tool that can reliably construct meaningful and comprehensible clinical notes could provide significant workflow improvements for EHR users on the edge of burnout.
“An AI-enhanced system would boost the clinical workflow element of documentation, and maybe even inject some fun into the process of note writing,” the article concludes. “We certainly hope that an EHR company or some budding entrepreneur will take notice of this article and consider our idea in creating the next generation of EHRs.”