Digital phenotyping to improve prediction of suicidal urges in treatment: Study protocol

Lily A. Brown, Daniel J. Taylor, Craig Bryan, Joshua F. Wiley, Kristi Pruiksma, Lauren Khazem, Justin C. Baker, Johnnie Young, Kerrie O'Leary

Research output: Contribution to journalArticlepeer-review

6 Scopus citations


Digital phenotyping offers a powerful approach for forecasting risk for suicidal ideation or attempts over time. This method may serve as an especially useful strategy for understanding associations between sleep disorder symptoms and suicidal ideation and behavior. Sleep disorder symptoms predict suicidal ideation and behavior in cross-sectional research among active-duty military personnel, but few studies have examined longitudinal associations between sleep disorder symptoms and suicidal thoughts and behaviors in service members. In this study, we will use digital phenotyping to intensively assess Marines for 28 days through a combination of active and passive assessment strategies. Methods: Marines with suicidal ideation or a suicide attempt in the past month will be recruited from Camp Lejeune, NC and provided with a Fitbit device and receive ecological momentary assessments (EMA) of suicidal urges several times throughout the day. Using dynamic multilevel models, we will explore the impact of sleep disorder symptoms on next-day suicide urges, as well as mediators of these effects. Clinical implications: This study has the potential to inform optimal strategies to assess suicide risk in treatment. These findings will inform the development and implementation of real-time interventions to reduce risk for suicide among military personnel.

Original languageEnglish (US)
Article number101733
JournalAggression and Violent Behavior
StatePublished - Sep 1 2022


  • Digital phenotyping
  • Ecological momentary assessment
  • Insomnia
  • Military
  • Nightmares
  • Sleep
  • Suicide

ASJC Scopus subject areas

  • Pathology and Forensic Medicine
  • Clinical Psychology
  • Psychiatry and Mental health


Dive into the research topics of 'Digital phenotyping to improve prediction of suicidal urges in treatment: Study protocol'. Together they form a unique fingerprint.

Cite this