Patterns of Change in User-Generated Social Media Behavior among Service Members Who Die by Suicide

Abstract

This is a secondary analysis of a retrospective study examining social media posts made by military Service members 1 year prior to their deaths. The sample consisted of 700 randomly selected military Service members who died by suicide and a demographically-matched control group of 700 military Service members who died of causes other than suicide. Social media data were previously coded by staff and students at the University of Utah's National Center for Veterans Studies (NCVS). The present study used data analytic methods consistent with dynamic systems modeling and were aimed at identifying suicide-specific change processes that could predict manner of death (suicide vs. control) and the imminence of suicide death (6 months prior to 1 day prior). Findings identified long-term, medium-term, and imminent "signatures" of emerging suicide death, suggesting there may be observable change processes in social media behavior that could signal eventual suicide.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
May 01, 2016
Accession Number
AD1036617

Entities

People

  • AnnaBelle O. Bryan
  • Craig J. Bryan
  • Jonathan E Butner
  • Sungchoon Sinclair

Tags

Communities of Interest

  • Biomedical

DTIC Thesaurus Topics

  • Brain Injuries
  • Data Analysis
  • Data Centers
  • Data Sets
  • Databases
  • Demographic Cohorts
  • Health Services
  • Media
  • Medical Personnel
  • Military Personnel
  • Psychology
  • Public Health
  • Security
  • Social Media
  • Social Networking Services
  • Social Networks
  • United States

Fields of Study

  • Psychology

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Psychological Intervention/Treatment for Stress, Anxiety, PTSD, and Related Emotional and Cognitive Health Symptoms.