PTSD and the DSM-5: A conversation with Dr. Matt Friedman

More than thirty-five years after the 1980 recognition of PTSD in the Diagnostic and Statistical Manual of Mental Disorders (DSM), the data are unequivocal: Today there can be no doubt about the validity of PTSD as a diagnostic entity. Yet, the disorder remains steeped in controversy, and more recently, there have been growing complaints that PTSD is being overemphasized in our society and over diagnosed in our hospitals.

PTSD can be found in its current avatar in the 2013 version of the DSM, called the DSM-5. Dr. Friedman is founder and former Executive Director of the National Center for Posttraumatic Stress Disorder and Professor of Psychiatry and of Pharmacology and Toxicology at the Geisel School of Medicine at Dartmouth. He was a member of the American Psychiatric Association’s (APA) DSM-5 Anxiety Disorders Work Group and chair of the Trauma and Dissociative Disorders Sub-Work Group.

Recently, I asked him about his work for the DSM-5 sub-work group.


Dr. Jain: The first question that I had was about the DSM-5 work group. The reason I ask about this is because I think, for a lay audience or scientists outside of our field, they may not be familiar with how much work went into the DSM-5. I was hoping you could speak a little bit about what that entailed, the scope of that mission, and the timeline of that whole effort.


Dr. Friedman: Basically, the DSM-5 had a work group for each diagnostic cluster. In the DSM-4, PTSD was an anxiety disorder, and the anxiety disorder group is the largest group of diagnoses. There were initially about a dozen of us who were asked to work on the anxiety disorder diagnoses. These were not just Americans, we had Europeans and people from Asia and other continents working with us.

When we looked at all these different anxiety disorder diagnoses, everything from panic disorder to obsessive compulsive disorder, and from tick disorders to PTSD, we felt that there was such a broad swath of diagnoses, and frankly, many of these diagnoses did not seem to belong together. So we divided ourselves into three work groups. There was the anxiety disorder work group, which looked at “classic” anxiety disorders (i.e., panic disorders, social anxiety, generalized anxiety disorder, agoraphobia, etc). Then there was a second group that looked at obsessive compulsive spectrum disorders. Finally, the third group, that I was asked to chair, was looking at what we came to call trauma and stress related disorders, and we were also asked to look at dissociative disorders.

We basically addressed the following: PTSD, acute stress disorder, reactive attachment disorder, and disinhibited social engagement disorder. We also looked at dissociative disorders, particularly de-realization and depersonalization disorder, dissociative amnesia, fugue states, and dissociative identity disorder.


Dr. Jain: Can you give me a sense of the timeline? When did you start work on this?


Dr. Friedman: It was a 5-year process. We began in 2008, and we knew that the DSM-5 had to be published in time for the May 2013 American Psychiatric Association meeting.


Dr. Jain: What was the process? Were you combing through all of the literature and picking out the higher quality studies? What were the sources for the work group?


Dr. Friedman: It was very conservative. It other words, there had to be very compelling research evidence to make any changes to the DSM-4. In terms of the process itself, for each diagnosis position papers were written and published in appropriate scientific journals. The position paper was a very thorough and careful literature review. Furthermore, we tried to make the process as open and transparent as possible, so on two separate occasions, the working criteria for each diagnosis of the different DSM-5 work groups were published on the American Psychiatric Association website. The entire clinical scientific community was invited to comment to raise objections, and these were taken very seriously. There were also field trials that were conducted. In our group, the field trials were really more about clinical utility. How easy was it for practitioners to use these criteria? What was the inter-rater reliability between one clinician and another who was looking at the same patient? It turned out PTSD emerged as having one of the highest inter-rater reliabilities. Also, our work group decided to do some research on its own.


Dr, Jain: Can you summarize the most important changes that eventually ended up being made for PTSD? I know your work group was looking at many disorders, but please comment specifically about PTSD.


Dr. Friedman: I think that the most important change was moving PTSD out of the anxiety disorders cluster. Our review of the data indicated that all posttraumatic clinically significant psychiatric symptoms don’t fit neatly into a DSM-III/DSM-IV fear-related anxiety disorder paradigm. A second important change was opening up the A criterion so that it was not a narrow fear-based anxiety disorder. The way we did that was by eliminating the A2 criterion, which required the following response after exposure to a traumatic event: fear, horror, or helplessness. By getting rid of this A2 criterion, we really opened things up to other acute posttraumatic emotional reactions. Also, in DSM-IV, you could have PTSD without reporting a single avoidance symptom. Now you have to have at least one avoidance symptom in order to have PTSD. We also stipulated that if PTSD was secondary to witnessing the sudden death of a loved one, that death had to have been under accidental or violent circumstances.


Dr. Jain: For me as a clinician there is this issue of partial PTSD that comes up all the time. We are seeing patients who do not necessarily meet the criteria in the DSM-5, but there is something going on, and they have considerable dysfunction. How do you think clinicians like me should approach those cases?


Dr. Friedman: I agree with you completely. The DSM-5 work group agrees with you completely. We very much wanted to have a sub threshold PTSD diagnosis. When we went into the literature, there are about 60 papers, including some that I have done with Paula Schnurr, looking at partial PTSD. The problem for DSM-5 is that there has never been a standard case definition of partial PTSD that has been used widely. Different investigators had different definitions of sub threshold PTSD.

As I told you earlier, given the very strong empirical rational that was required for any addition to be made, the APA concluded there was not sufficient evidence of a specific sub threshold PTSD diagnosis that had been validated in the literature. I do hope that we can soon develop a widely accepted case definition for subthreshold PTSD so that the needed research can be conducted to guide the next DSM consideration of subthreshold PTSD.

I agree with you that we see these patients who have been exposed to major traumatic events. They have many PTSD symptoms, it is clearly a trauma-related problem, but they do not exceed the diagnostic threshold for PTSD. What are we supposed to call them? In DSM-5, the diagnosis “other specified trauma or stress related disorder” would be the proper diagnosis. So there is a place for such cases in the DSM-5. Once we get a good case definition and we have the data, I think that there will be an actual sub threshold diagnosis in DSM.


Dr. Jain: One thing that I have noticed is there is a fundamental assumption about the DSM-5 which gets forgotten: it is a manual meant to be used by people who have specific clinical training. What scares me is when I see people finding bits of the DSM-5 and diagnosing themselves (or others) when they don’t have such training. Can you comment?


Dr. Friedman: You are correct. The DSM-5 is an ongoing manual under the aegis of the American Psychiatric Association. Diagnosis should only be done by people that are qualified to make diagnoses. That is why people go to medical school, whether they are making a psychiatric diagnosis or a surgical diagnosis or a pediatric diagnosis. The diagnostic criteria are to guide qualified people to be able to make diagnostic distinctions. The whole reason to do that is because diagnosis A may be better served by treatment A, and diagnosis B may actually have better outcomes if you use treatment B.

These distinctions are extremely important, and you need a professional in order to make these. On the other hand, I think it is important and useful for lay people, policy makers, journalists, and family members to have some idea about specific diagnoses so that if they, or a loved one, is exhibiting a specific set of symptoms, they can go to a qualified professional to see whether or not they are correct.


Dr. Jain: There are growing complains that PTSD is being overemphasized in our society and maybe over diagnosed in our hospitals. I think people who are not appropriately qualified might be making the diagnosis. There is a huge amount of variability in the mental health training of clinicians throughout any given hospital system. A classic example that I run into all the time is when someone has assumed that just because somebody served in a warzone, they must have PTSD. We know that that is an erroneous assumption. Actually, the opposite is true. The majority of people who served in a warzone will not develop PTSD. I am curious. Do you feel it is overemphasized and over diagnosed, and if so, what do you think the reasons might be?


Clipboard.svgDr. Friedman: There are different contexts in which the PTSD term is used. There is the way that you and I have been discussing it, in terms of the very strict DSM-5 diagnostic criteria. There is also the public health context. For example, following a terrorist attack, we know that the risk for developing PTSD or other problems is heightened, so it makes sense to be looking for PTSD. But we need to look with very accurate instruments and with well-trained individuals making that diagnosis. We have screening measures, like the Primary Care PTSD Screen (PC-PTSD), but that is where the diagnostic evaluation should begin, not where it should be concluded.

So, if a person comes in and says, “I am having terrible nightmares about the murder that I witnessed yesterday.” Well, that is normal during the immediate aftermath of a traumatic event. Most people are going to have PTSD symptoms after experiencing such an event, which will usually dissipate.  In that case the clinician can be reassuring but should also caution the patient that if such symptoms persist, further evaluation is necessary.

I think one of the major roles that we clinicians play is we are teachers. When someone comes into our office, we need to make that into a teachable moment. It does not bother me if someone comes in and he or she is a thoughtful person who looked at the Internet, maybe read your blog, and said, “I think I have PTSD.” That is okay with me. It is my job to make a careful evaluation and either validate or refute that assumption. It is my responsibility to tell them why they do or don’t have PTSD, and then make some recommendation for what they should do.

I think there is a rush to judgment sometimes. It is certainly reasonable for a clinician who has seen a traumatized patient, to ask the question, is this PTSD? But then they either have to follow through with a careful diagnostic evaluation or refer the patient to a mental health professional who will do the diagnostic assessment.

We developed the Clinician-Administered PTSD Scale (CAPS), which is the gold standard for making a diagnosis. The CAPS is only to be administered by a specifically trained clinician. It is not something that you can hire a college graduate to do. The important thing is the CAPS has behavioral anchors. If a patient said, “I am not sleeping well,” you want to say, “How many hours of sleep are you getting? What wakes you up? When are you sleeping? When are you not sleeping?” Only a trained clinician is going to have the awareness and the training and the skill to really use these diagnostic criteria in the way they are meant to be used.


Dr. Jain: You brought up that point about the positive PTSD screens. Sometimes, I think there is a misperception that because it is a positive screen, it automatically means someone has PTSD. Another issue is, as you know, we have a shortage of mental health professionals across the country. Sometimes, I think PTSD gets over diagnosed because there are not enough mental health professionals to see patients when they screen positive.


Dr. Friedman: I want to say something about screens. What many people do not understand is that screens are designed to be biased toward false positives. What you want to do with a screen is you want to include all the people who might have the diagnosis, whether it is high blood pressure, cervical cancer, PTSD etc. The expectation with the screen is that many of the people who screened positive are not going to have the diagnosis. You do not want to leave out anybody who might.


Dr. Jain: Like you said, interpreting that screen—that is the piece that I feel sometimes contributes to this overemphasis of the diagnosis.


Dr. Friedman: It’s understanding the utility and the limitations of the screen or of a self-report instrument. We’ve got some really good assessment instruments in mental health. We have the PCL and the PHQ-9. Again, these are all self-report. They do not have behavioral anchors like the CAPS (which is a structured interview administered by a trained clinician). For example, if a patient checks a four and states, “I am having terrible sleep,” the clinician has to use the behavioral anchors to drill down and clarify exactly what does that individual mean. One person’s four might be someone else’s two.


Dr. Jain: Any final thoughts?


Dr. Friedman: We do not have a stake in preserving the DSM-5 diagnosis in perpetuity. If there is better data that comes along, that is great. If the DSM-5 has raised questions that have motivated people to generate that data, that is great too. That is what science and clinical science is all about. I think that is an important contextual issue that people often miss.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s