Suicide is a leading cause of death in the United States and is the second leading cause of death in the U.S. military. Previous research suggests that data obtained from social media networks may provide important clues for identifying at-risk individuals. To test this possibility, the social media profiles from 315 military personnel who died by suicide (n = 157) or other causes (n = 158) were coded for the presence of stressful life situations (i.e., triggers), somatic complaints or health issues (i.e., physical), maladaptive or avoidant coping strategies (i.e., behaviors), negative mood states (i.e., emotion), and/or negative cognitive appraisals (cognition). Content codes were subsequently analyzed using multilevel models from a dynamical systems perspective to identify temporal change processes characteristic of suicide death. Results identified temporal sequences unique to suicide, notably social media posts about triggers followed by more posts about cognitions, posts about cognitions followed by more posts about triggers, and posts about behaviors followed by fewer posts about cognitions. Results suggest that certain sequences in social media content may predict cause of death and provide an estimate of when a social media user is likely to die by suicide.

The public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing the burden, to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number.

The public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing the burden, to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS.1. The PERSEREC Crime Self-Reporting Study covers criminal record checks conducted in CY00 on 14,470 subjects of DoD security clearance investigations, including uniformed military, civilian, and contractor personnel. Information found in these record checks was compared with subjects' responses to questions pertaining to criminal charges, arrests, and convictions on the Standard Form 86: Questionnaire for National Security Positions. Differences in rates of self-reporting were evaluated based on military status (military vs. nonmilitary), gender, age of subject at time of offense, age of subject at time of application, time elapsed since offense, type of investigation, type of offense to be reported (felonies and nonfelony alcohol, drug, firearms or explosives, miscellaneous other), level of investigation, and whether subjects were undergoing initial investigation or periodic reinvestigation. Variables that were most strongly associated with self-reporting of offense information were type of offense to be reported, military status, level of investigation, age at time of self-report, and recency of offenses. Overall, subjects reported alcohol-related offenses at higher rates than other types of offenses. The rate of selfreporting was lower among subjects undergoing NACLC investigations than those designated as NAC or SSBI, though the higher rate of reporting among NAC subjects was an artifact of the investigative process subjects were undergoing. Military subjects were less likely than nonmilitary subjects to self-disclose reportable offenses. Females tended to be less likely to report than males. Offenses that occurred between 3 months and a year after the date of self-report were among the least likely to be reported in NACLC and SSBI investigations, but were not significantly different in NAC investigations, suggesting this finding may be due to delays in investigations rather than subjects' reluctance to report relatively recent offenses. For offenses where subjects were asked "have you ever" been arrested, charged, or convicted, offenses that were more than 10 years old were reported at lower rates than more...

No abstract

The public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing the burden, to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number.