Asset 1
Digital reasoning talks AI needs in healthcare
How Healthcare can Navigate the AI Maze

Healthcare certainly loves its buzzwords. One of the favorites at this week’s World Health Care Congress, in Washington D.C., was one that carries a fair amount of baggage: Artificial Intelligence.

The promises that AI has made to the healthcare industry in the past few years have not always come to pass, but that didn’t prevent AI being appended to many technologies and use cases by vendors at this year’s summit. How can healthcare providers and health systems navigate the maze of generalities and broad descriptions that are all too common to this area of innovation?

My advice is to start at the end and work backwards. What is your (very) specific problem statement? Precisely what are you trying to solve? If you have a firm grasp of the intended outcomes and a clear level of specificity, it becomes much easier to filter out the AI noise and focus on vendors that have built an AI solution around the workflow or process within your scope.

Too many vendors offer a general value proposition of what their technology can do and, quite frankly, too many health systems meet them with a general idea of what they want.

One example that stood out at the WHCC was a presentation on the impact of AI on the oncology service line, by Crystal Dugger, Assistant VP of Clinical Operations at Sarah Cannon and Dr. Richard Geer, Chief Physician for HCA, Surgical Oncology. The takeaway from their session was that creating an impactful workflow and process comes before any clever technology. Describing a successful AI initiative, Crystal and Richard explained the clarity that comes from first designing a workflow that supports better outcomes, better engagement, higher satisfaction and higher revenues, and then exploring which technology can bring it to life and create value for all involved.

Too many vendors offer a general value proposition of what their technology can do and, quite frankly, too many health systems meet them with a general idea of what they want. It’s understandable. The perceived value of technology that can automate the reading of pathology reports, or identify triggers for sepsis, have an inherent appeal to anyone familiar with these clinical challenges. However, the value will only be measured with solutions that can integrate with a health system’s existing workflows or empower a better workflow that has been newly created.

A case study for the success of a more measured approach is the Cancer Patient ID project, created by Sarah Cannon HCA and Digital Reasoning. This is an AI solution aligned to a new nurse navigation workflow that was created by the experts at Sarah Cannon and designed to address a very specific problem with a very specific solution. Patient ID automated the identification of new cancer patients and prioritized them, resulting in numerous improvements in efficiency, timeliness, retention, and revenue opportunities. All of these outcomes were the intended results of a better workflow, which the AI technology enabled.

As a health system or IDN, this is the best way to think about which AI and vendor is best for your needs. Focusing on the problem statement and the ideal workflow to solve it allows you to zero in on the right AI vendor partner, and disregard the AI hype. The success of the relationship between Sarah Cannon and Digital Reasoning came from aligning technology to meet a workflow rather than matching a workflow to new technology. Start with this orientation and the maze will be much easier to navigate!

Written By
Chris Cashwell

Chris Cashwell is our Vice President of Healthcare Solutions. He describes himself as a “passionate pioneer” in healthcare IT and focuses on improving patient care and health systems’ bottom line. At Digital Reasoning, Chris oversees a dedicated team focused on deploying AI solutions that solve real-world healthcare challenges.

Find Chris on LinkedIn