Harvard Study Finds Smartphone Check-Ins Can Predict Suicide Risk Days in Advance
A new study by Harvard researchers indicates that frequent smartphone check-ins can accurately predict suicide risk days before an attempt. The technique, developed by Harvard psychology professor Matthew Nock, identified 75% of suicide attempts and 87% of suicide-related events (like hospitalizations) in the week prior to their occurrence. Researchers followed 619 adults and adolescents seeking care for suicidal thoughts or behavior at two Boston-area hospitals. Participants received 20-question surveys on their smartphones six times a day for three months, rating their feelings of hopelessness, agitation, suicidal urges, and ability to resist those urges. This data, combined with information on survey completion patterns, was fed into machine-learning models. The study found that agitation, a strong feeling of irritability and discomfort, was a particularly strong predictor of near-term suicide attempt risk, more so than depression alone. For adults, every one-point increase in agitation on a 0-to-10 sca...