Technologies for Assessing Behavioral and Cognitive Markers of Suicide Risk
Abstract
According to the Centers for Disease Control and Prevention, 39,518 suicides were reported in 2011, making suicide the 10th leading cause of death overall and the leading non-natural cause of death for Americans. Among military personnel, rates of suicide are even higher. One of the biggest challenges to improving the success of suicide prevention efforts in the military is the absence of reliable methods for predicting who will engage in suicidal behaviors and when these behaviors will occur. This limitation restricts our ability to identify at-risk military personnel and to make sure that they are getting the best available and most appropriate treatment for their psychological symptoms. Clinical interviews and surveys are the best available methods for determining which individuals are at increased risk for attempting suicide and for identifying when these individuals are at particularly heightened risk. However, these methods do not work well for measuring suicide risk in Soldiers. One limitation of clinical interviews conducted by medical personnel is that they require in-person interaction with a healthcare professional. The primary limitation of this method is that most Service members who die by suicide do not interact with a healthcare professional immediately prior to the event. Likewise, surveys about suicidal thoughts and feelings are not informative if Soldiers are unwilling to acknowledge or are unaware of their psychological distress. The lack of risk assessment methods that are not dependent on a Soldier s report of his or her psychological distress not only impacts military personnel but also their spouses who are at increased risk for a wide range of psychological and physical health symptoms. This project aims to address these limitations in existing methods for determining suicide risk by developing an automated, computer-driven suicide risk assessment tool that is not based on Soldiers reports of psychological distress. There are risk factors for suicide in the general population that can be assessed by measuring specific aspects for behavior. For example, one indicator of suicide risk is how much attention an individual pays to stimuli related to death. Another kind of behavioral risk factor for suicide can be observed when married Soldiers talk with their spouses. For example, high levels of self-denigration from the Solider and criticism from the spouse are associated with increased marital tension, psychological distress, and risk for suicide. Because married couples comprise a dyad, this effect is not unidirectional; the communication patterns of the Soldier directed towards his or her spouse can similarly contribute to marital tension, psychological distress, and risk for suicide. To date, however, there is limited information available regarding how transactional patterns within the context of military families contribute to and sustain mental health problems and suicide risk. The current project aims to fill this knowledge gap by evaluating communication patterns among military couples and developing an automated system for measuring these categories of behavioral indicators of suicide risk in Soldiers in the National Guard. More specifically, we propose to measure communication behaviors during conversations between Soldiers and their spouses using Behavioral Signal Processing (BSP). BSP uses cutting-edge computer algorithms to automatically measure human expression during recorded conversations in a manner consistent with human judgment. Additionally, we propose to measure death-related attention with computer tests that can be completed over the internet. Such a system has the potential to result in several important benefits for Soldiers, their spouses, and their healthcare teams. First, this system would enable measurement of suicide risk outside of healthcare settings. This aspect of the system could result in many more at-risk individuals being identified than i
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Apr 04, 2016
- Source ID
- W81XWH1510632
Entities
People
- Brian Baucom
Organizations
- United States Army
- University of Utah