Three Ways AI Can Improve Physicians’ Workflow

Posted on November 26, 2018 I Written By

Anne Zieger is veteran healthcare consultant and analyst with 20 years of industry experience. Zieger formerly served as editor-in-chief of FierceHealthcare.com and her commentaries have appeared in dozens of international business publications, including Forbes, Business Week and Information Week. She has also contributed content to hundreds of healthcare and health IT organizations, including several Fortune 500 companies. Contact her at @ziegerhealth on Twitter or visit her site at Zieger Healthcare.

For far too many physicians, EHRs and other important health IT seem to get in the way of getting the job done. But according to one pair of physician-authors, emerging AI technology has the potential to improve physician workflow instead.

“We see opportunities for AI to be a solution for—rather than a contributor to—burnout among physicians and achieving the Quadruple Aim of improving health, enhancing the experience of care, reducing cost and attaining joy in work for health professionals,” wrote AMA chief medical information officer Michael Hodgkins, MD, MPH and Shantanu Nundy, MD, director of the Human Diagnosis Project.

In an article for the journal Health Affairs, Drs. Hodgkins and Nundy outlined three ways in which AI could be used to make physicians’ work easier and more satisfying. They include:

  • Delivering educational information to the point of care: At present, most educational efforts targeting physicians don’t do a good job of keeping physicians up to date, as they aren’t targeted enough, the article asserts. However, by using AI, healthcare organizations can offer personalized content to physicians by reviewing their existing research habits. By analyzing practice data, online search queries and assessments, AI can provide a streamlined infostream offering only what they need.
  • Producing clinical documentation: The authors argue that AI will someday be able to complete clinical documentation tasks on the physicians’ behalf. In their view, these AI applications will analyze a given physician’s free-text narrative, extract relevant information and insert the information into the right data fields in their EHR. (Researchers are testing out some concrete approaches for doing this.)
  • Collecting information needed for quality-measurement reporting: Hodgkins and Nundy envision a scenario in which AI tools spare doctors the need to perform hours of redundant quality reporting duties. As in the documentation example, such tools would review clinical documents and extract needed information, though this time in search of meeting external requirements. They would then populate data fields in need of completion on submission forms.

These are comparatively straightforward applications of AI. In addition to the trio of possibilities suggested above, AI could eventually deliver clinical decision support on the fly, speed and improve the accuracy of medical image interpretation and more.

In the meantime, however, it’s hard to disagree with these authors that physicians could benefit a great deal from AI tools that make basic clinical workflow faster and less draining.