[Skip to Navigation]
July 10, 2019

Leveraging Digital Health and Machine Learning Toward Reducing Suicide—From Panacea to Practical Tool

Author Affiliations
  • 1Department of Psychiatry, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts
  • 2Department of Psychology, University of Houston, Houston, Texas
  • 3Web Editor, JAMA Psychiatry
JAMA Psychiatry. 2019;76(10):999-1000. doi:10.1001/jamapsychiatry.2019.1231

Because the rates of suicide attempts and deaths have recently increased to 50-year highs,1 new solutions are needed. The urgency to reverse this trend has brought attention to technology-based tools, such as text messaging, smartphone apps, smartphone sensors, electronic health records, and machine-learning algorithms, that can offer crucial data to improve the prognostication of suicide or immediate support for those at risk. This promise of real-time data from connected devices, large quantities of social-behavioral interactions from social media and internet, and longitudinal clinical trends from electronic health records, when paired with artificial intelligence to automatically identify risk, is often touted as a panacea. Yet, to date, this approach has found less clinical success than expected. The current, limited technological advances in suicide prevention do not reflect a failure of technology or big data but rather a need to realign research aims and clinical use with prevention research that addresses the upstream suicide risk that precedes suicide crisis. In a recent report,2 the National Action Alliance for Suicide Prevention outlined 3 gaps in health care that contribute to suicide death: failing to (1) proactively identify suicide risk, (2) act efficiently for safety, and (3) provide supportive contacts for people at risk of suicide. Focusing on these 3 specific gaps as examples by identifying risks (such as limited social connectedness, cognitive hopelessness, and poor problem solving) can have immediate effects. This Viewpoint aims to explore how technology can augment solutions to these challenges while simultaneously addressing current gaps in translational research and clinical care.

Add or change institution