When Is One Psychotherapy Better Than Another?

One behavior of therapists that can address the science-practice gap.

Posted Sep 02, 2019

Wikimedia Commons
Source: Wikimedia Commons

A “science-practice gap” is a situation in which something that has been proven by research fails to be accepted. Going back hundreds of years, scientists demonstrated repeatedly that the earth revolved around the sun, but many failed to believe that; instead, they kept believing that the sun revolved around the earth. In psychotherapy, researchers have demonstrated the effectiveness of psychotherapy protocols with randomized clinical trials—yet private practitioners often fail to adopt these evidence-based practices (EBP).

Psychologist Jennifer Doran and colleagues from Yale University published a study recently to try to crack the stubborn mystery of why EBPs so often fail to be adopted in the real world (Doran et al., 2019). Doran and her colleagues conducted focus group interviews with eight clinicians working in the Veterans Administration system. Implementation of EBPs for PTSD is mandatory in the VA system, so these clinicians were all experienced with at least one EBP protocol (either cognitive processing therapy or prolonged exposure). 

They were asked 13 open-ended questions. Seven core domains were identified from the focus group responses:

  1. EBP Strengths
  2. EBP Weaknesses
  3. Challenges Specific to the Veteran Population
  4. Perceived EBP Effectiveness
  5. Active Ingredients for Treating PTSD
  6. Treatment Structure and Process
  7. Suggested Changes/Improvements to EBPs

To me, the most striking aspect of the findings was the problem of flexibility. Lack of flexibility was a cross-cutting concern that was mentioned in six of the seven domains.

In Doran’s study, the therapists complained that lack of flexibility was a problem in the EBP protocols. They perceived that the protocols were overly-structured and did not provide them with guidance on how to work with other issues that arose. They felt that protocols fit some cases well but not others. The protocols were not comprehensive enough to address all of the life issues or comorbid disorders of their patients. The therapists felt that they were not allowed to individualize the protocols. In other words, the therapists seemed to believe that they were not allowed to be flexible.

Overall, I think this article provides an important contribution to the field, as the issue of flexibility has not been written about much in a systematic fashion. However, I felt it was inaccurate that flexibility was perceived as a major problem of the protocols. I have used, developed, and tested EBPs for over 20 years. I already knew that flexibility was a key issue. I do not view flexibility as a problem of protocols. I view flexibility as a problem of therapists.

What is Meant by Flexibility?

The examples that Doran provided in the paper of issues that required flexibility during implementation of EBPs were all issues that were trauma-related—i.e., multiple traumas, residual trauma-related symptoms that did not improve, resistance to cooperate, trauma-related comorbid disorders, and needing more time to process trauma narratives. 

Based on my clinical experience and research, I would add many other types of non-trauma issues. In fact, the vast majority of time, flexibility is needed for issues that have nothing to do with trauma. An example of mild flexibility would be a young adult who needed to talk about a difficult experience during the previous week of dealing with her in-laws; she just needed 15 minutes to tell the story and process her guilty feelings. Examples of medium flexibility include repetitive interpersonal dramas that trigger catastrophizing thinking; reflections on childhood insults that are intrusive in adult lives; eating disorder issues that were initially kept secret but gradually become revealed; dealing with a spouse’s anger; discovery of mild autistic traits after the first several sessions; dealing with comorbid gambling issues; processing family dynamics left over from childhood abuse—and so on, for hundreds of other issues. There are very few issues that would make us completely abandon a protocol once we started because the issues that contraindicate using a protocol would have been detected before starting.

The therapists in Doran’s study felt stymied by complex trauma and comorbid disorders. I do not view complex trauma as a real thing. All traumas are complex. If therapists do not understand that car accidents and natural disasters are complicated traumas, I suspect they have never treated them. If therapists believe that comorbidity requires too much flexibility, I would suggest they are easily overwhelmed; research has shown that 80 to 90 percent of all individuals with PTSD have comorbid disorders, and nearly every patient requires flexibility to handle their comorbid disorders.

In summary, I disagree with the therapists in Doran’s study who think lack of flexibility is a problem of EBP protocols. I think the problem is the lack of flexibility of therapists. I think flexibility is required for EVERY case. The problem of flexibility is not a problem of EBT protocols; it is a problem of skill, which is required to know how to weave flexibility into an EBT protocol. 

Being flexible does not diminish the power of the EBPs. In fact, the authors cited a study which showed that flexible administration by therapists, such as adding sessions to an EBP protocol, had no negative impact on treatment outcome (Galovski et al., 2012).

In Doran’s paper, one therapist was quoted as saying, “At the end of the day, you really just have to be a good clinician. And that has nothing to do with the protocols or manuals.” I couldn’t have said it better. 


Doran JM, O’Shea M, & Harpaz-Rotem I (2019). In Their Own Words: Clinician Experiences and Challenges in Administering Evidence-Based Treatments for PTSD in the Veterans Health Administration.  Psychiatric Quarterly 90:11–27. https://doi.org/10.1007/s11126-018-9604-5

Galovski TE, Blain LM, Mott JM, Elwood L, Houle T (2012). Manualized therapy for PTSD: flexing the structure of cognitive processing therapy. Psychology 80(6):968–81.  https://doi.org/10.1037/a0030600.