Algorithmic app could head off suicides of high-risk adolescents

Researchers have demonstrated the feasibility of AI-based smartphone interventions for young people who have attempted suicide and may try again.

While the subjects were simulated patients, the preclinical work also succeeded in showing that Bayesian network modeling is “an accurate and very efficient branch of AI in psychiatry and clinical psychology,” the authors state.

That’s a key finding in a specialty short on big datasets for algorithm training and validation, they suggest.

The work was conducted in France to accord with digital healthcare guidelines set by the American Psychiatric Association and is posted in the Journal of Medical Internet Research.

Olivier Bonnot, MD, PhD, of Nantes University Hospital Center and colleagues showed how a smartphone app that twice daily gauges a patient’s anxiety, mood and sleep patterns can appropriately urge action, offer advice or stand down so as not to needlessly increase anxiety.

The authors note that, within a year of a first suicide attempt, the subsequent-try rate among adolescents and young adults ranges from 15% to almost double that.  

“To date, no suicide prevention program is better than others, and all of them require the active participation of healthcare professionals,” Bonnot and co-authors write. “However, studies have shown that personalized brief-contact interventions reduce recurrence after a suicide attempt. A specific and personalized app could optimize these strategies.”

Acknowledging that their app needs to be both built out and clinically trialed, the team underscores the importance of showing Bayesian network modeling to be an efficient way to develop an algorithm for a digital assistant dedicated to suicidal crisis management.

They write:

We are convinced that using digital devices with efficient algorithms is crucial for medical treatment in terms of reliability and safety,” they write. “However, to date, very few devices meet accurate methodological requirements. Our work is a proof of concept that emphasizes the need for preclinical trials by algorithm development … [although] building the application and randomized controlled clinical trials are necessary to confirm our choices and the overall efficacy of our device.”

The study is available in full for free.

Dave Pearson

Dave P. has worked in journalism, marketing and public relations for more than 30 years, frequently concentrating on hospitals, healthcare technology and Catholic communications. He has also specialized in fundraising communications, ghostwriting for CEOs of local, national and global charities, nonprofits and foundations.

Trimed Popup
Trimed Popup