National Suicide Prevention Month: Using AI to enhance precision psychiatry - World Medical Innovation Forum
WMIF Main Site

National Suicide Prevention Month: Using AI to enhance precision psychiatry

September is recognized as National Suicide Prevention Month, a time dedicated to raising awareness about the prevalence of suicide and its devastating impact on individuals and communities.

Jordan Smoller, MD, ScD, Associate Chief for Research and Director, Center for Precision Psychiatry, Department of Psychiatry, MGH; Tepper Family MGH Research Scholar; Professor of Psychiatry, HMS’s First Look presentation, “Using big data and AI to advance precision psychiatry and suicide prevention”, is dedicated to unraveling the factors, both genetic and environmental, that contribute to psychiatric disorders. The goal, “…is to transform how we manage suicidality which unlike many areas of medicine has seen far too little progress for far too long and we believe that we can bend the curve on this tragic outcome for patients and families and ultimately we hope save lives,” said Smoller during his First Look presentation from the World Medical Innovation Forum this past June.

A significant aspect of this research involves utilizing the power of artificial intelligence (AI) and leveraging real-world health data to enhance the prediction of risks and the selection of treatments for psychiatric illnesses and suicide.

In the United States, an alarming 1.7 million people attempt suicide each year, making it the second leading cause of death among young individuals. Shockingly, suicide rates have surged by more than 30% in the past two decades. Most individuals who attempt or die by suicide had recently interacted with a healthcare provider, offering a crucial opportunity for risk assessment and intervention in healthcare settings. However, studies conducted by Smoller’s group and others have revealed that clinicians struggle to predict suicide-related behaviors more effectively than random chance.

To address this pressing issue, the research team turned to electronic health record (EHR) data and applied AI algorithms to identify individuals at high risk of suicide attempts and deaths. Their groundbreaking work, based on longitudinal data from 1.7 million patients, led to the development of an algorithm that successfully identified 45% of suicide attempts and deaths, with an impressive 90% specificity, on average two to three years in advance. These results were subsequently validated in five different health systems, encompassing a total of 3.7 million patients.

Looking ahead, Smoller’s team plans to refine their tool further through clinician input, develop a clinical decision support application for EHR integration, and ultimately scale up its implementation to address the critical need for improved suicide prevention.

Watch more from Jordan Smoller, MD, ScD’s First Look presentation.

This article is intended to provide information and resources related to suicide prevention and mental health. If you or someone you know is in crisis or experiencing thoughts of suicide, please seek immediate help. Contact a mental health professional, a crisis hotline, or go to the nearest emergency room. In the United States, you can call the National Suicide Prevention Lifeline at 1-800-273-8255 for confidential support.

the Future of Medicine and AI

May 11–13, 2020 | Boston, MA