Smartphone language features may help identify adverse post-traumatic neuropsychiatric sequelae and their...

Area of Science:

• Psychiatry
• Computational Linguistics
• Digital Health

Background:

• Language use may reflect cognitive and emotional states post-trauma.
• Identifying early markers of posttraumatic symptoms is crucial for timely intervention.

Purpose of the Study:

• To determine if language features from smartphone use indicate concurrent and changing posttraumatic symptoms.
• To explore language as a potential digital biomarker for recovery after trauma.

Main Methods:

• Secondary analysis of the Advancing Understanding of RecOvery afteR traumA (AURA) study.
• Language features extracted from smartphone app data of 1744 trauma survivors.
• Bivariate linear mixed models used to identify language markers associated with symptom severity and change over six months.

Main Results:

• Fourteen language features correlated with posttraumatic symptom severity cross-sectionally.
• Five language features predicted symptom changes longitudinally.
• References to body/health predicted worsening symptoms; references to others/causation predicted improvement.

Conclusions:

• Smartphone-derived language features can serve as digital biomarkers for posttraumatic stress.
• These linguistic markers offer insights into patient functioning and recovery trajectories.
• Potential for clinical integration to identify at-risk individuals and monitor treatment response.