Harvard study claims real-time surveys can foresee most suicide attempts a week ahead

Researchers at Harvard University monitored over 600 individuals, aged 12 to adults, who had been treated for suicidal thoughts or behaviors in emergency rooms or inpatient settings. After discharge, participants answered a 20-item smartphone survey up to six times daily for three months, then once daily for the next three months, rating feelings such as hopelessness, isolation, anger, and agitation on a 0-10 scale. The analysis showed that a rise in agitation scores increased the odds of an attempt by 11% per point, allowing the team to predict 75% of suicide attempts and 87% of related events a week before they happened.

The researchers also examined response timing and built an alert system to notify clinicians when intent rose. Matthew Nock emphasized the need for privacy safeguards while highlighting the potential to deliver preventive tools, like therapy reminders, directly to users' devices. The findings will appear in the October issue of the Journal of Psychopathology and Clinical Science and are part of a broader effort to incorporate wearables and other real-time data into suicide-prevention strategies.