How New Technologies Are Changing the Way We Study and Treat Suicidal Behaviors

Available with English captions and subtitles in Spanish.

Matthew Nock, PhD, Harvard University, presents as part of the 2022 Suicide-Focused Assessment and Treatment: An Update for Professionals course.

Innovations in Technology

Suicide is a complex problem—one that humans have been trying to understand for thousands of years.

As the mortality rate for other health issues has dropped, we haven’t seen the same progress in the area of suicide. In this talk, Nock explains why this may be the case, and discusses new ways to determine suicide risk.

Watch now to learn more about:

  • Why suicide mortality rates have remained stagnant
  • How research may not adequately address suicide risk
  • How technology can lead to improvements in assessing and preventing suicide

Nock states that research on suicide has repeatedly looked at the same risk factors, such as demographics, life events, and internalizing and externalizing symptoms. Studies have been largely conducted with self-report surveys and interviews.

“If we’re using the same methods, we shouldn’t be surprised to be seeing the same results,” Nock says. “We need new technologies.”

Nock discusses the need for methods that better combine known risk factors for suicide. He points out that the strongest risk factors for suicide have about the same odds ratio. While there isn’t one factor that most strongly predicts suicide, 90% of research over the past 50 years has examined one risk factor at a time.

“This is a problem for our clinical predictions,” he states. “The human brain isn’t designed to assess dozens of risk factors, weigh them, combine those weights, and make a predictive probability of a suicide event. Yet this is what we’re asking clinicians to do in emergency, inpatient, and outpatient settings.”

This is where technology can come in, according to Nock. He describes a study in which researchers used machine learning applied to administrative medical data to create risk scores for patients in the year after psychiatric hospitalization.

View the Slides

He also describes how machine learning and patient self-reporting can be another way to improve prediction by combining sources of data.

Nock emphasizes the need to identify imminent risk factors. He points out that while most clinicians want to know which patients are currently at risk for suicide, most studies examine patient risk for suicide at least ten months into the future.

To address this gap, Nock and his colleagues have been conducting smartphone studies with participants who report suicidal thoughts daily for a month.

The researchers are exploring how technology can use machine learning to find people in distress, connect them to resources, or address reasons why an individual resists getting help.

The technology is fully automated, scalable, and led to a 23% increase in the use of crisis services.


You might also find this information useful:

About Matthew Nock

Dr. Matthew Nock is the Edgar Pierce Professor of Psychology at Harvard University and is considered one of the leading researchers on suicide. His research is aimed at advancing the understanding of why people behave in ways that are harmful to themselves, with an emphasis on suicide and other forms of self-harm.

Nock’s multidisciplinary research uses a range of methodological approaches to better understand how these behaviors develop, how to predict them, and how to prevent their occurrence.