Today, everyone in healthcare is acutely aware of the problem of physician burnout – and its consequences, which are deepening the staffing crisis.
Health informatics leaders at hospitals and health systems have already begun experimenting with artificial intelligence—now a huge boom in health care—to try to reduce physician burnout.
Could AI ever replace doctors? Some experts say yes—although only in confined, tiny ways. For example, Dr. Bruce Darrow, interim chief digital and information officer and chief medical information officer at Mount Sinai Health System in Up-to-date York, says that in some cases, when the clinical accuracy of doctors and AI is nearly identical, part of the clinical care of the future could indeed be transferred to artificial intelligence.
The idea is that AI can assist reduce physician workload and, by extension, clinician burnout. Even if AI only plays a supporting role, experts say the technology has the potential to absorb the massive workload clinicians have today.
Dr. Eve Cunningham is a jack of all trades. She is group vice president and head of virtual care and digital health at Providence and founder and CEO of MedPearl. The MedPearl platform is an AI-powered clinical intelligence engine designed by and for clinicians.
Cunningham will speak on the topic at the HIMSS AI in Healthcare forum case study session scheduled for Thursday, September 5.
We interviewed Cunningham to talk AI and burnout, and also saw a clip of her presentation.
Q. What exactly will your session on AI and clinician burnout cover?
AND. The theme of our session is to present three real-world examples of AI application in our healthcare system, with a focus on clinical workflows and physician burnout.
I intend to provide examples of: an in-house developed (built) AI-assisted technology, MedPearl; an example of working with a vendor (purchasing) to implement an AI-assisted solution; and an example of a desired solution that stalled due to technical challenges related to infrastructure, technical debt, and misaligned incentives.
While AI has the potential to revolutionize healthcare by improving decision-making and optimizing workflows, its implementation is not without challenges. A key aspect of our discussion is ensuring The applications and capabilities of artificial intelligence are useful tools, rather than an additional burden on physicians.
The relevance of this topic is underlined by ongoing discussions in healthcare settings about the balance between technological progress and the provision of human-centred care.
Q. What is one example of AI in action in your organization?
AND. An example of AI technology in our organization is the MedPearl platform, a clinical intelligence engine designed specifically by and for clinicians. This platform is an example of our approach to integrating AI into healthcare settings.
MedPearl improves clinical decision-making by consolidating clinical knowledge, patient data, and suggested point-of-care actions into a single interface, thereby optimizing physician workflow and reducing cognitive load.
The development of this platform was based on the collaboration of physicians and technologists, and the AI development plan was carefully designed taking into account not only technological aspects but also clinical safety and human factors issues.
By viewing AI as a tool and feature embedded within a larger product designed to support physicians rather than as a standalone entity, we have carefully planned the layered deployment of AI features that we enable within the platform.
This example is a microcosm of our broader approach to AI in healthcare – thoughtful integration tailored to empower the physician and build trust in the product from a quality and safety perspective.
Q. What two takeaways do you think session participants should take away and apply in their supplier organizations?
AND. The first lesson is that AI integration should be viewed as a deliberate, incremental process that aligns with specific clinical needs and strategic goals, rather than a quick, one-size-fits-all solution. Implementing AI in healthcare settings requires thoughtful consideration of the problems it is intended to solve and a clear understanding of the potential impacts on clinical workflows and provider well-being.
Second, we emphasize the importance of taking a strategic approach to selecting exploit cases. It is crucial to prioritize AI applications that align with the strategic priorities of the institution and address high-value clinical or operational issues.
We will share real-life examples at Providence, where AI exploit case selection and implementation are closely aligned with our strategic goals and are designed to support, not replace, human expertise.