Suicide is notoriously difficult to predict.
Warning signs may be subtle, fleeting or easy to mistake for ordinary changes in mood or behavior.
For Americans aged 10 to 34, suicide is the second-leading cause of death, and it’s the fourth-leading cause among individuals between the ages of 35 and 44. To intervene early and save lives before a crisis escalates, health experts are on the lookout for warning signs or indications that someone could be suicidal.
While typical suicide prediction models focus on the likelihood of a future event over the next several months, year or even decade, recent research from Harvard University suggests that a majority of attempts can be predicted within a much tighter timeframe.
Previous studies have shown that around 50% of people who died by suicide saw a clinician in their final month. However, much of this data came from self-reporting, with participants asked if they had a suicide attempt since their last survey.
“We need to understand what puts people at risk, and the ebb and flow of suicidal thoughts — how to better measure them when they occur, and how to better predict increases in risk,” Matthew Nock, the study’s lead author and a Harvard psychology professor, told The FAS Current.
Nock, who pioneered the technique of surveying people in real time, led a multiyear study of more than 600 high-risk adults and adolescents who presented to a hospital with either or both suicidal thoughts and behavior.
The findings, set to be published in the October issue of the Journal of Psychopathology and Clinical Science, predicted 75% of suicide attempts and 87% of suicide-related events — attempts as well as hospitalizations to prevent attempts — in the week before they occurred.
Participants included adults who received emergency room-based psychiatric treatment and teens ages 12 to 19 who had been treated at an inpatient clinic for suicidal thoughts or behavior.
After being released from treatment, participants received 20-item smartphone-based surveys, with an optional one sent six times a day for the first three months and once a day for the next three months. They answered on a sliding scale from zero to 10.
The questions asked about suicidal thoughts, including intent and resistance to the urge to attempt, as well as emotional states such as hopeless, isolated, angry, worried, energy and positivity.
These emotional responses played a significant role in helping the team better predict suicide risk, especially because suicidal ideation doesn’t always indicate a mental health disorder.
For every additional point a person indicated on the agitation scale, their odds of attempting suicide increased by 11%, making agitation a stronger indication of suicide risk than depression.
These findings reflected earlier research from Nock, which found that 90% of people who survived a suicide attempt reported feeling “an urge to alleviate psychological agitation and pain that turned out to be temporal.”
Additionally, Nock and his team used metadata to examine how participants engaged with the surveys, including how long they took to start and complete one.
A real-time alert system was also included to intervene if someone showed intent to attempt suicide. Nock said was the purpose of the study was “to build a system that’s scalable and reproducible — and accurate — so we can intervene before these events occur, to keep people safe.”
This study was just one part of a larger initiative from Nock’s lab to continue improving real-time monitoring for suicide risk, including the use of wearable sensors to track health metrics such as sleep, heart rate and voice signatures.
Additional interventions could include reminding someone of tools they learned in therapy or encouraging them to reach out to a loved one or their clinical team.
While Nock acknowledges the privacy precautions that will need to be taken, he’s enthusiastic about the prospect of bringing care directly to someone’s own environment.
“We have to be careful in how we do it, and we have to work hand in hand with clinicians and with the people with the illnesses and conditions we’re trying to treat, but there’s great possibility here,” he said.
