.png)
Over 70% of healthcare organizations report pursuing or having already implemented generative AI capabilities, according to a 2024 McKinsey survey. Artificial Intelligence (AI) isn’t new to healthcare, but recent advancements actively transform health systems, sparking questions from executives and providers nationwide. From streamlining operations and workforce management to enhancing patient experiences, knowing how to leverage AI effectively is critical to getting ahead as the industry moves forward.
In a recent Healthcare Dive webcast entitled “AI in Healthcare: Revolutionizing Operations and Patient Care,” Kathy Pham, VP of AI at Workday, and Mike Harmer, VP Workforce Intelligence & Planning at Intermountain Health, discuss how healthcare leaders can make the most of AI to benefit an organization’s mission as well as its bottom line.
“Healthcare is a perfect place for the use of this new technology because we’re sitting on a trove of data that can be used to solve challenges that really impact broader society,” remarks Pham. With this new technology, healthcare organizations can:
• Connect complex systems
• Enhance patient care and diagnoses
• Accelerate decision making
• Boost operational efficiency
• Get ahead of healthcare workforce shortages
• Improve financial projections, minimizing risk
• Strengthen supply chain resiliency
“The work we do at the end of the day saves lives,” notes Harmer, “and AI can help us do things better in every way…and that’s important.”
While the webcast addressed all of these benefits, we look at three in particular: clinician burnout, essential healthcare workers, and decision-making.
“Every day, when doctors, nurses, and everyone who supports the organization comes in, they have this deep mission to take care of people,” says Pham. “One of the challenges to think about when we build technology is, how do we supplement them?” Harmer agrees, noting, “It's important for us to realize that the more we can help clinicians, the more they can help patients.” One way Harmer and his team have used AI to avoid clinician burnout is by modeling upstream to ensure there will be enough people to help them deliver care. He explains how the medical group used AI to build a long-range planning solution that enabled them to look ahead at what specialties they’ll need—several years ahead—so physicians and advanced practice professionals (APPs) won't wind up short-staffed.
“When we see that we’ll have a gap in labor supply and start fixing that problem early, we don’t get in a situation down the road where our clinical teams experience burnout because there aren’t enough people, and we have to go to contingent labor, unnecessarily driving up the cost of care,” Harmer adds.
Healthcare is experiencing a talent shortage, and it’s projected to rise. But Pham sees potential in AI to help fill gaps by using skills data to match people with roles inside a healthcare setting in ways that might not have been thought of before. “These are things that the current state of machine learning and AI are quite good at,” says Pham, “taking all that information, connecting it, and giving us insights on which individuals can fit into what job.”
Harmer notes that Intermountain has long been interested in a skills-based talent strategy, but the administrative effort was an obstacle. The pandemic pushed the organization to break apart jobs into specific skills so they could reconfigure their workforce to fill urgent gaps at a time when it was simply not possible to “hire your way out of it.” This helped Intermountain identify skill adjacencies and empower caregivers to step into different roles, making their workforce much more agile. The success they experienced made Intermountain leadership realize that a skills-based approach (vs a job-based approach) is their best path to overcoming the projected workforce shortages that all health systems face. While there is still work to be done, they are making good progress and Harmer notes that “AI really makes possible these types of solutions” – even for organizations as large and complex as Intermountain.
Perhaps even more exciting are the ways in which people can take advantage of skills–both in being able to quickly reconfigure the workforce and seizing new opportunities. Harmer predicts “The future can look radically different from the past, in terms of career paths and succession for our caregivers.”
Harmer shared that one of the most important things healthcare organizations must do is shorten the gap between insight and action. “When we close that distance, we get better outcomes faster,” he says. For example, thanks to AI embedded in Workday, the analytics team at Intermountain discovered one facility was impacted by an especially high vacancy rate of 30% for nurses and clinicians. Seeing a 49-day time to fill, the talent acquisition (TA) team quickly focused on where the problems were. As a result, the organization reduced the time to fill to 26 days, bringing the vacancies at that location down to zero in just eight months, and those numbers have been sustained ever since.
Leveraging AI to surface insights more quickly than human analysis alone can provide a distinct advantage in a tight talent marketplace. Pham shares an example from another health system:
By using Workday AI solutions to analyze their talent pipeline, hiring managers at AdventHealth saw a 40% decrease in decision-making time. Pham explains how faster decision-making enabled hiring managers to double the number of job requisitions they closed in 90 days. As a result, AdventHealth had zero nursing openings—for the first time in 10 years–and saw an increase in patient safety scores. Of course, responsible AI is just part of the process. “AI can shorten the length of decision making, but, ultimately, decision-making lands on the individual,” notes Pham, emphasizing how AI can streamline a job but never take over the critical role of skilled workers across healthcare.
Know what’s in your data set. Given that large language models (LLMs) are trained on the data they’re provided, Pham suggests knowing exactly what’s in your data set and being conscious of what’s missing. This way, you can plan to compensate for gaps and avoid bias.
Understand internal risk factors. Consider the risks of implementing AI in certain systems. For example, Pham explains how payroll is a higher risk than expense reporting as someone’s livelihood is at stake. It doesn’t mean you don’t use AI here, but you should take extra care in higher-risk areas and insert humans in the loop.
Train staff on AI tools. Harmer shares how upskilling caregivers on TA teams in AI has helped them automate tedious tasks and, as a bonus, reduce TA team attrition from 20% to 6%.
Choose trusted partners. As a best practice, Pham suggests selecting good vendors who can help you with infrastructure and support behind the scenes so you can maximize AI’s capabilities.
“Healthcare is expensive…It’s more expensive than any of us would like,” says Harmer. “But now generative AI is creating new opportunities to think about work and the tools we have to help bring down costs.” But it requires you to do things differently.
Harmer shares a metaphor from a colleague who said to think of AI as a fitness tracker. The fitness tracker doesn’t do anything until you change how you work and what you do. “It’s not just about the technology; it’s about what you do with it,” says Harmer. If you do it right, you could revolutionize operations and patient care in your organization.
For more information and to watch the full webinar, access the recording here.
.png){kind=small}
