In-session predictors of response to cognitive processing therapy for PTSD
This week we look at a study that examined in-session client and therapist behaviours and their influence on the outcome of cognitive processing therapy (CPT) for PTSD. Alpert and colleagues (2023) rated video sessions of 70 adults who began CPT of interpersonal violence-related PTSD. Client variables coded included: cognitive flexibility, the ability to change perspective based on new information; clients’ expression of fear, sadness and anger; and avoidance of engagement with therapist (e.g., providing minimal responses). Therapist behaviours coded included: use of Socratic dialogue, encouragement to express affect, and empathy expression.
The study employed quite a severe sample by research standards. Only 67% completed treatment (at least 12 CPT sessions). Only 49% were White, only 19% were in a relationship, only 23% had university education and 69% reported annual income < $20,000, although 86% were female, a good prognostic indicator. Statistical models appropriately adjusted for age, education, gender, and baseline PTSD symptoms.
Client factors generally have a stronger influence on therapy outcome than therapist factors. In this study, a difference between clients of 1 standard deviation units in expressed fear was associated with 1/3 standard deviation increase in PTSD symptoms posttreatment. A similar magnitude of association was evident for client avoidance of engagement with therapist. The role of client cognitive flexibility is a little unclear. The authors used two models, a “conservative” approach in which the dependent variable was a posttreatment PTSD score and an arguably more sensitive approach in which the dependent variable was the last available PTSD measure (the timing of which would vary by individual depending on when they dropped out). Cognitive flexibility did not predict posttreatment PTSD severity in the conservative analysis but did so in the sensitive analysis, where again, 1 standard deviation increase in cognitive flexibility between participants was associated with approximately 1/3 deviation lower PTSD symptom totals. Client-expressed sadness and anger were not associated with posttreatment PTSD levels.
Therapist in-session behaviours were not related to PTSD outcomes but were related to retention in treatment. Therapist behaviours were rated on 7-point scales from 1=poor to 7=excellent. For every 1-point increase in quality of therapist’s use of Socratic dialogue, clients were 6.75 times more likely to complete treatment. Conversely, for every 1-point in quality of therapist’s encouragement of the client’s natural affect, there was an 89% decrease in the likelihood of the client completing therapy. Therapist empathy made no difference to retention, although this may be because there was less variation between therapists, with all therapists showing high levels of empathy. Client behaviours were not associated with treatment completion although the trends for anger and sadness suggest that with a larger sample, these may have been significant predictors.
These findings highlight the dilemma for therapists in trying to facilitate positive outcomes for clients in therapies which involve clients confronting avoided aversive experiences and questioning beliefs forged in adversity and which at some level, are intended to protect themselves. Both client disengagement and excessive fear elevation across sessions are associated with worse outcomes. Even when observers rated that therapists were doing a better quality (NB: not just quantity) job of facilitating clients experiencing their natural emotions, this was associated with a much higher risk of dropout. The authors explained this pattern by emphasizing the CPT distinction between genuine emotions and “manufactured” emotions, the latter being the result of dysfunctional thinking. They argue that therapists (and so, presumably observers) may have difficulty discerning between genuine and manufactured emotions and may inadvertently encourage clients to continue to express the latter, thereby prolonging their psychopathology. The pattern of results in this study strikes me as being consistent with the idea that therapists can’t make the client do anything (of long-term value); therapists can create (or ruin) conditions that allow clients to see the value of therapeutic actions and gain confidence from practicing them. I, and no doubt many other well-meaning therapists, have tried to exhort, persuade, cajole, warn, or bribe their clients into exposure-like activities, driven by the belief that this is being a “good therapist” and fear of “not helping”. However, I don’t recall any time that a client’s emotional activation that occurred without their consent and willingness was therapeutic, even if apparently technically consistent with treatment guidelines. There are no reliable shortcuts - that I know of - to help fast-track change in an unwilling client. Conversely, Socratic questioning is a reliable way to help those who are slow to agree with the rationale for exposure-like activities or examination of dysfunction-prolonging beliefs (i.e., disengaged), to “come around” to see their value. Practicing in an efficient, directive way, will help those who come to therapy willing and ready, but is likely to push people with any substantial degree of ambivalence out early, without helping them. Good Socratic technique, like a good defensive technique in cricket (it is Ashes season after all), ensures the client stays “in” to help them face their Mark Wood, I mean, trauma, another day.
Clinician implications
Unless clients present as extremely motivated and appear to understand and agree with the rationale for exposure-based activities or evaluating their beliefs, resist the urge to simply provide didactic psychoeducation and instructions.
If clients present with any degree of reluctance, apprehension, or disagreement with the need for exposure or challenging their beliefs, invest time in the process of guided discovery. Help clients come to their own realisation that the costs of their avoidance and dysfunctional beliefs outweigh their benefits, and that the benefits of overcoming avoidance and building new, healthier beliefs outweigh the pain.
Set the bar low for detecting emotional activation. Appreciate that clients who have devoted years to inhibiting emotional expression might be quite good at hiding how distressed they are from you. If you’re unsure, ask them. If the client is not dissociating, detaching, or distracting themselves and is reporting feeling congruent emotions of a greater intensity than they were before beginning the task, you can trust that they are emotionally activated enough to benefit.
Try to see yourself more like a personal trainer than a surgeon. The client gets stronger by lifting the weights themselves. You can’t do that for them, although you may be able to support them by taking some of the weight if they discover they’ve picked up a barbell they’re not ready for. You do this by prompting them to use their arousal reduction strategies when necessary to avoid them being flooded, by taking breaks from immersive exercises so the client doesn’t feel out of control or violated, by helping them remember the rationale and the cost-benefit ratio of doing the task versus avoiding it, and listening to them when re-grading tasks to find tasks that are challenging but not overwhelming.
For the original article, go to: https://onlinelibrary.wiley.com/doi/full/10.1002/jts.22924
REFERENCES
Alpert, E., Carpenter, J. K., Smith, B. N., Woolley, M. G., Raterman, C., Farmer, C. C., Kehle-Forbes, S. M., & Galovski, T. E. (2023). Leveraging observational data to identify in-session patient and therapist predictors of cognitive processing therapy response and completion. Journal of Traumatic Stress, 36, 397– 408.