Can AI help doctors communicate better with their patients?

That’s the question a team of researchers from The Dartmouth Institute for Health Policy and Clinical Practice, Trinity College Dublin, and the University of Edinburgh recently considered in an a BMJ article that considered the potential ways AI might enhance patient-provider exchanges.

“Many clinicians’ communications skills aren’t formally assessed–either during school or in early practice,” noted senior author Glyn Elwyn, MD, Dartmouth Institute Professor. “At the same time, there is a lot of evidence that clinicians often struggle when communicating with their patients. It’s hard to improve on something when you’re not being given any feedback and don’t know how you’re doing.”

Naturally, effective communication is critical to effective patient care, but it seems most doctors never get their communication skills formally checked out.

According to a discussion of the article at docwire, “Elwyn and colleagues claim that AI is an innovative solution that could completely change the communication process in healthcare. Providing practitioners with personalized, detailed assessments of their communication skills, AI may provide the first widely used tool for evaluating physician communication.”

In their article, Elwyn and his associates identify three main areas where AI and recordings of patient-physician interactions could be used to improve communication, including analyzing conversations, allowing patients to speak their mind, and a more general consideration of tone and style in voice.

Concerning analysis of conversations, the docwire piece says, AI could be used to analyze both words and phrases in patient-provider dialogues to ensure mutual understanding, as well as, potentially, analyze conversations with an eye toward suggesting diagnoses that haven’t been previously considered.

In addition, AI could compare the amount of time providers speak to the time left for patients, as the article says, among other things, there tends to be a correlation between the amount of time patients are allowed to talk and levels of medication adherence.

Finally, Elwyn and his colleagues suggest, just as algorithms have been used to analyze commercial pilots’ vocal pitch and energy, “(a)nalyzing these vocal features could help detect high-risk situations in which the doctor is under stress that inhibits their communication skills. In addition, this vocal analysis could also provide insight into the patient’s mental and physical health.”

There are, of course, issues such as patient privacy to be managed for these analyses to move forward, but as Elwyn sums up the idea, ““Five years ago, the idea of using AI to analyze medical communication wouldn’t have been on anyone’s radar. As the technology advances, it will be interesting to see whether healthcare systems can employ it effectively and whether providers will be open to using it as a tool for improving their communication skills.”