1 - (SYM 82) Predictors of Dropout in Cognitive Processing Therapy for PTSD: An Examination of In-session Treatment Processes

Background: Dropout rates in trauma-focused treatments for adults with posttraumatic stress disorder (PTSD) are high. Most research to date has focused on fixed and relatively immutable characteristics such as demographics and pretreatment symptomology, but findings have been inconsistent. Process research can reveal client factors that unfold during treatment that might be related to dropout. We used an observational coding system to code client processes in audio-recorded early sessions of cognitive processing therapy (CPT).

Methods: Data are from a randomized controlled noninferiority trial of CPT and written exposure therapy (WET) in which CPT showed significantly higher dropout rates than WET (39.7% vs. 6.4%). Participants were 53 treatment-seeking adults with a primary diagnosis of PTSD, who completed at least one session of CPT. Sixteen (28.3%) of this subsample were classified as dropouts (completing < 10 sessions) and 37 as completers. The CHANGE coding system (Hayes et al., 2007) was used to code audio-recorded sessions for maladaptive trauma-related responses (overgeneralized beliefs, avoidance, and ruminative processing), affective engagement (negative emotions and physiological distress responses), and adaptive processing (cognitive emotional processing). Each CHANGE coding variable is coded on a four-point scale from 0 (absent) to 3 (high).

Results: Averaging across the first five available session recordings, binary logistic regressions showed that more physiological distress (OR = .18, 95% CI [.04-.81] p=.026) and more cognitive emotional processing (OR = .123, 95% CI [.018-.857] p=.034) expressed in the sessions predicted lower dropout, whereas more avoidance was associated with higher dropout (OR = 4.19, 95% CI [1.07-.16.39] p=.039). Negative emotion, ruminative processing, and overgeneralization did not significantly predict dropout.

Conclusion: Expression of more physiological distress predicted lower dropout rate, which might suggest more engagement with the traumatic memory. More adaptive processing also predicted lower risk of dropout, perhaps because the client was able to approach, rather than avoid, the traumatic memory and make meaning of the experiences. This sense of progress might have increased engagement. In contrast, early avoidance increased the risk of dropout. Together, these findings highlight potential early indicators of treatment engagement that could be targeted to reduce the risk of dropout and perhaps facilitate therapeutic change.