Photo Credit: iStock.com/KanawatTH

Increased time spent at home predicted next-week suicidal thoughts or behaviors in high-risk teens, suggesting smartphone GPS data may aid suicide prevention.

More time spent at home, compared with an adolescent’s usual pattern, was associated with increased risk for suicidal thoughts and behaviors among high-risk youths over the next week, according to a study published in JAMA Network Open.

Researchers captured adolescents’ geolocation metrics through an app installed on their smartphones.

“Acquiring passive sensor data through personal smartphones holds promise to monitor risk and to deliver timely interventions, as ownership of these devices is nearly ubiquitous,” wrote corresponding author Randy P. Auerbach, PhD, and colleagues.

The case series study included 186 high-risk adolescents, aged 13 to 18 years, with a current affective or substance use disorder. Participants were recruited from the greater New York City and Pittsburgh communities, with recruitment oversampling for suicidal thoughts and behaviors.

The Effortless Assessment Research System app used the global positioning system (GPS) to continuously record geolocation metrics over the course of 6 months. In addition, the app assessed suicidal ideation and behaviors through a weekly user prompt. If participants reported frequent suicidal ideation, a suicide plan, or a suicide attempt, the clinical staff made contact to assess safety and bridge to emergency services when necessary.

Researchers then investigated data to see whether any geolocation features, such as entropy, homestay, or distance traveled, were associated with next-week suicide events or ideation in teens.

“Increased homestay on a given week, but not entropy or distance traveled, was associated with a near 2-fold greater risk in the occurrence of suicidal events in the subsequent week,” they reported. “Analyses also corroborated the association between homestay and suicidal events in the same week, suggesting some temporal specificity in that homestay 2 or more weeks prior was not associated with suicidal events.”

A leave-future-out validation approach, however, found only modest accuracy for the homestay model. The area under the receiver operating characteristic curve was 0.64.

“Although accuracy was modest, these findings offer a novel starting point for suicide prevention research,” researchers wrote, “particularly as smartphone sensor data may have the capacity to identify who is at risk while also providing an opportunity to deliver clinical tools when that risk is greatest.”