AI tool links HCW stress during pandemic to mental health issues

Editor's Note

An AI tool helped to identify signs of distress in conversations from healthcare workers (HCWs) with their therapists during the pandemic, according to a study from NYU Grossman School of Medicine, EurekAlert! October 24 reports. The findings were published in Journal of Medical Internet Research AI. 

Researchers analyzed treatment transcripts from more than 800 physicians, nurses, and emergency medical staff, as well as transcripts from 820 people receiving psychotherapy during the first US wave of COVID-19 who did not work in healthcare. They used digital session transcripts to identify common phrases used by patients, then associated these with mental illness using natural language processing (NLP) in which a computer algorithm pinpoints keywords. 

Findings include:

• HCWs who spoke to their therapists about working in a hospital unit, lack of sleep, or mood issues were more likely to be diagnosed with anxiety and depression compared with HCWs who did not discuss these topics
• these same risks were not seen in workers from other fields who discussed the pandemic or their jobs.

The study reinforced that HCWs working on the hospital floor during the height of the pandemic faced extraordinary challenges that put them at risk for serious mental health concerns.