Does Cognitive Therapy Enhance Exposure Therapy?

A neat study was published recently exploring this topic. In the OCD literature, research has typically compared exposure response prevention (ERP) treatment to cognitive therapy (CT). Both are shown to be effective, but the literature is larger for ERP interventions. However, what about combining them?

Here’s the reference:

Rector, N. A., Richter, M. A., Katz, D., & Leybman, M. (2018). Does the addition of cognitive therapy to exposure and response prevention for obsessive compulsive disorder enhance clinical efficacy? A randomized controlled trial in a community setting. British Journal of Clinical Psychology. doi: 10.1111/bjc.12188

ERP often involves “informal” cognitive work. For example, when I finish doing an exposure I will always ask “what did you learn from that?” and we will work to weave in what was helpful. However, the researchers considered more formal cognitive therapy to include structured interventions for negative appraisals of intrusions and obsessive beliefs. This usually involves things like thought records, learning about thinking traps, and working to restructure those thoughts.


I liked that this study was in a community sample from a large OCD clinic. People in the study had to have a YBOCS score, which assesses OCD severity, of more than 16. YBOCS stands for the Yale-Brown Obsessive Compulsive Inventory and is considered the gold-standard OCD assessment. Needing to have a score above 16 to enter the study just shows that the participants likely met OCD diagnostic criteria and were having a hard time dealing with it.

The authors also randomly assigned people into different groups, including an ERP group (n=62) and ERP+CT group (n=65).

Note: From the description of ERP they used a habituation model. That is, they exposed people to something they feared and waited for anxiety to come down. This is what most ERP therapists do, but some researchers are arguing that this model is out of date. Read more here.

Based on their description, the CT + ERP appeared to have people test out their cognitive appraisals very directly, gathering evidence, questioning beliefs, and then reframing beliefs including having a reframe about a person’s ability to cope in response to a trigger without the use of compulsions. Thus, they did some pure cognitive therapy, did an exposure around it, and then did some more cognitive therapy.

How Do We Know They Did A Good Job?

They recorded all the sessions and demonstrated good adherence to treatment intervention.

Post-treatment they also did “blind administration” of the YBOCS, which is awesome. This is good because it means the people doing the OCD assessment did not know what condition the participant was in, nor who their provider was.


The table shows the main results, with the score being OCD severity. Higher scores mean more impairment. As you can see, both interventions decreased symptoms.


Post Treatment




Exposure + Cognitive




They also did a series of analyses to show that the ERP + CT group improved significantly more than just ERP alone, both after treatment and 6 months post-treatment.

Of the ERP group, 45% achieved mild symptom status by the end of treatment. In the ERP+CT group, 65% achieved this status. This difference was not significant.

What to Make of This?

First, it seems like ERP+CT reduces OCD symptoms more than just ERP alone. However, there was no significant difference between the treatments in the number of people who achieved mild symptom status. Overall, it seems like there could be some promise in looking more into how CT and ERP can work together. Specifically, ERP could be boosted by formally examining beliefs about obsessions before, during, and after exposures. CT might be boosted through experiential learning that can only occur with ERP. They have mutual benefit!

What They Didn’t Discuss

These results are really interesting. It’s a cool study, worth reading and considering. It definitely needs replication at another site, especially since the level of gains shown were much larger than the typical literature. But here’s something to consider. Theoretically, it’s possible that doing ERP and CT could be contra-indicated. In fact, and I wish they had discussed this, it’s very possible that the ERP + CT group could have done worse than the ERP group alone! Here’s why:

If you’re doing CT then your anxiety during an exposure could be lower as you are working to create more helpful thoughts that reduce anxiety. But we don’t want anxiety to be lower during exposures as that limits our expectancy violation effect. This effect basically means we want people to be as shocked as possible by the outcome of the exposure. Trying to develop rational or more helpful thoughts before/during an exposure could therefore get in the way of this. Especially early on in treatment, thoughts generated through CT can actually become safety cues that interfere with response prevention. I’ve written about this in more detail here. I wish the authors had addressed the issue because it’s a theoretical one, and way more interesting to think about than the limitations they actually listed. Other than that, I enjoyed the read and will consider this in my practice moving forwards.


Leave a Reply