Researchers at the University of Virginia hope to use text messages to help clinicians detect an increased risk of suicide attempts in real-time.
With software that gauges a person’s mood according to the frequency of positive or negative words sent in a text 鈥� like happy, joyful, hate or mad 鈥� lead author Jeff Glenn and others aim to use digital data to move suicide prevention beyond relying on patients to self-report suicidal thoughts that can sometimes be fleeting or concealed.
鈥淲hen the clinician is doing a risk assessment, we鈥檙e only getting a really narrow snapshot in time during that face-to-face encounter,鈥� Glenn said Monday. 鈥淲hat we tried to do is design a study to learn if we could see signs of increased risk through text messaging, which is something that a lot of people do every day.”
Glenn and his team collected nearly 200,000 messages from 33 individuals who had attempted suicide in the past, comparing messages sent in the weeks leading up to an attempt to lower-risk periods to determine if there was a discernible change in their tone.
鈥淲e tried to reconstruct their timeline,” Glenn said. 鈥淲hat we鈥檙e trying to do is identify unique signs within that period of time before suicide attempts, which would give us a sense that they鈥檙e close to a crisis or a crisis might be coming up.鈥�
The research seemed to support Glenn’s hypothesis: Negative emotions like anger were expressed more often than positive ones in the days leading up to a suicide attempt.
Though he said more research needs to be done before an implementation in the real world, Glenn is optimistic that these findings could form the foundation for a web or phone-based early warning system.
With technology becoming more integrated in every facet of society, his research aligns with the health industry’s shift toward meeting people where they are, leveraging mountains of available data to help clinicians diagnose and monitor outside an office.
鈥淭he big take-away would be that text messaging is one of many potential ways to gain insight into one鈥檚 suicide risk level,鈥� Glenn said. 鈥淎ll this data is accumulating normally as we鈥檙e messaging, which provides unique opportunities to better understand how behaviors are occurring in the real world.鈥�
The U.Va. team’s full findings will be published in the journal .
