Linguistic predictors of response to internet-based written CBT for PTSD symptoms of 9/11 world trade center recovery workers and survivors.

Trauma-focused cognitive-behavioral therapies (CBT) are the mainstay of posttraumatic stress disorder (PTSD) treatment, but treatment response is heterogeneous. We used automated linguistic analysis to identify markers of PTSD symptom improvement in response to integrative testimonial therapy (ITT), an internet-based, therapist-assisted written CBT for PTSD, for 9/11 World Trade Center rescue and recovery workers and survivors with PTSD symptoms. Therapy comprised 11 written narratives, separated into biographical reconstruction, exposure, and cognitive reappraisal narrative modules. Narrative data of 35 ITT completers in a larger clinical trial were analyzed using the Linguistic Inquiry and Word Count (LIWC) software and the Valence Aware Dictionary and sEntiment Reasoner (VADER) toolkit. Separate regression analyses were conducted to predict treatment response, assessed with the PTSD Checklist for DSM-5 (PCL-5) at baseline and post-treatment. Four LIWC word categories collectively explained 59 % of the variance in PCL-5 score improvement: higher use of death, causation, and motion words in biographical narratives, and lower use of social behavior words in exposure narratives, were associated with greater improvement. Higher VADER negative sentiment score in biographical narratives was also associated with greater improvement. Findings provide insight into potential factors driving therapeutic response to written CBT for PTSD and may help advance personalized treatment for individuals with this disorder.