In addition to LeBaron, the team includes Drs. Tabor Flickinger and David Ling from the School of Medicine and associate professor Laura Barnes and research scientist Mehdi Boukhechba from the School of Engineering.
Setting their sights on difficult palliative and end-of-life conversations, the team’s yearlong pilot study, funded by UVA’s Engineering in Medicine Seed Pilot Program, set out to develop and test the ability of CommSense’s algorithms to accurately hear and score clinical conversations and to assess if clinicians would find the technology acceptable.
First, they identified the components of a “gold standard” palliative care conversation by searching literature and consulting communication experts. The best practices relied on open-ended questions, providing time for therapeutic silence, and sentence structures that reinforce connections such as, “I want to be sure I understand,” “It sounds like you are feeling …” and “Can you tell me more about ...?”
The team then established 10 communication metrics – what they called their “Ground Truth” – to set the watch’s listening parameters using Google’s speech-to-text application. The team programmed the watch to note “positive” words and phrases that convey empathy, warmth and openness. Conversely, the watch also marked “less positive” metrics such as talking too much, interrupting the patient or using medical jargon.