PTSD NOT AN ANXIETY DISORDER? DSM COMMITTEE PROPOSAL TURNS BACK THE HANDS OF TIME

The proposed DSM-5 changes to posttraumatic stress disorder (PTSD), reviewed in detail in the last month's issue by Friedman et al.^[1] include the addition of symptoms, changes to the symptom clusters, and re-conceptualization of PTSD as a “trauma and stressor-related disorder” instead of an anxiety disorder. The rationale for this shift is unclear, under-developed, and unsupported. It is our strong opinion that, at this point, there is insufficient evidence for PTSD to be considered distinct from the anxiety disorders as outlined below.

FEAR IS A CRITICAL CONSTRUCT FOR THE DEVELOPMENT OF PTSD

Across the main theoretical models for understanding the development of PTSD is the centrality of classical conditioning of fear^[2–7] as a necessary, but not sufficient, mechanism for the development of PTSD. This emphasis is consistent with key models across the anxiety disorders^[8] and, in fact, underlies the DSM linkage of trauma exposure to subsequent PTSD symptoms. Empirical evidence also strongly supports this, with one of the best predictors of PTSD being if the person thought they would be killed or seriously injured.^[9–11] In many respects, PTSD is the quintessential anxiety disorder,^[12] where of all the anxiety disorders, PTSD consistently shows an empirical pattern (i.e., stronger, more consistent evidence than panic disorder, social anxiety disorder) of anxiety disorder-defining characteristics. Namely, self-reported symptoms of anxiety and fear, heightened anxiety and fear responding to threat/no threat signaling cues, elevated stress reactivity to threat-related stimuli, attentional biases to threat-relevant stimuli, and threat-based appraisals of ambiguous stimuli; and elevated amygdala responses to threat-relevant stimuli.^[12] Based on years of research, this is a compelling evidence base arguing that PTSD is an anxiety disorder.

TREATING TRAUMA-RELATED FEAR AND AVOIDANCE IS CENTRAL TO PTSD

In comparison to any other area in the PTSD field, more is understood neurobiologically about fear and extinction^[13–17] than any other, including the role of heightened fear acquisition, enhanced conditioned responding, and slower extinction of fear responses.^[17–20] In fact, the evidence Friedman et al.^[1] presents for PTSD being a fear circuitry disorder supports it remaining with the anxiety disorders, acknowledging that, “Anxiety disorders occur when fear conditioning persists and there is a failure of extinction learning” (p. 741). This empirical and theoretical work forms the foundation for treatment development in PTSD, applying conditioning and extinction principles to modern learning theory^[21–23] and even in understanding learned cognitive associations about conditioned stimuli and unconditioned stimuli expectancies (i.e., cognitive therapy).^[24] Indeed, the only treatment with a sufficient, well-developed empirical base for the treatment of PTSD is exposure therapy,^[25] which relies on these principles. Notably, these theoretical models and effective treatments are the same as those used in the treatment for other anxiety disorders (e.g., panic disorder, specific phobias, obsessive compulsive disorder). Quite simply, like others with anxiety disorders, patients with PTSD seek treatment to reduce their anxiety and avoidance.

Friedman et al.^[1] note, “... psychiatry has diverged from most other medical specialties’ emphasis on causation as a critical component of diagnosis, e.g., ‘myocardial infarction’ rather than ‘chest pain syndrome’...(p. 741)” This is certainly true, but other fields of medicine are better at personalizing treatment as well, and the etiology is critical to determine for proper treatment. In this cardiac example, cardiologists take into account genetic influence (i.e., family history), presentation, timing, etc. to prescribe a personalized intervention. Every person experiencing a myocardial infarction requires attention, although every person experiencing chest pain does not. Similarly, as the authors acknowledge (p. 745), “Most people exposed to a traumatic or nontraumatic stressor do not develop any mental disorder.” Therefore, all those experiencing a stressor would NOT require attention but those meeting the diagnostic criteria for PTSD would, supporting our current, symptom-based classification system. The shift toward an emphasis on the stressor for classification neglects what we as a field have learned over the years: The persistence of the reaction to the stressor is what defines PTSD, not the presence of or reaction to a stressor.

THERE IS A LACK OF EVIDENCE FOR A STRESSOR META-CONSTRUCT SEPARATE FROM THE ANXIETY DISORDERS

Although factor analyses of symptom presentation are not sufficient for understanding underlying constructs, not a single of the major DSM factor analytic studies^[26–30] show PTSD loading as its own distinct meta-construct, although none have explicitly tested this. That is, there is not an evidence base for a distinct construct. PTSD clearly shares symptom features with the other anxiety disorders (e.g., specific phobia, panic, social anxiety, generalized anxiety, obsessive compulsive disorder) and has a high comorbidity with them.^[31] And, just like the rest of the anxiety disorders,^[31,32] it has a high comorbidity with depression.^[33] These overlapping characteristics clearly suggests shared core features^[34] and are much more consistent with symptom-based factorial models arguing for a higher-order internalizing factor^[35] than a new classification of trauma and stressor-related disorders. Without compelling empirical evidence, the presence of ancillary symptoms, termed a “wider range of emotions than fear-based anxiety” by Friedman et al.^[1] (p. 742), commonly seen across the anxiety and depressive disorders (and not just PTSD) such as numbing (i.e., lack of reactivity to positive stimuli), alienation, detachment, guilt, anger, and shame should not be used as support for the creation of a new classification of disorders. Kihlstrom^[36] likened the reliance of symptom presentation solely for classification to “... rearranging deckchairs on the Titanic.”

THIS SHIFT IGNORES CUMULATIVE EVIDENCE AND MOVES THE FIELD BACKWARD

As recently pointed out,^[37] the current proposed configuration yields 2,800 minimal and 5,800,410 possible combinations in which an individual could qualify for the PTSD diagnosis as currently proposed. Notably, in the DSM-IV, PTSD was criticized as being too heterogeneous,^[38,39] with a possible 79,794 combinations. This is in stark contrast with a major depression disorder in the DSM-IV, which has 126 minimal and 256 possible combinations. This astronomical degree of heterogeneity has the potential for even more broadening on the construct and moving the field backward rather than forward (e.g., when the disorder was classified according to the trauma type, such as shell shock, rape trauma syndrome, etc.), rather than focusing on key underlying commonalities. Along these lines, the broader shift toward creating a new overarching category “trauma and stressor-related disorders” has the potential for obscuring the strong translational and neurobiological research that under-girds this diagnosis and inadvertently suggesting to clinicians and researchers alike that these literatures on fear and anxiety are not critical in understanding PTSD. This would be a profound mistake for the field.

CONCLUSION

Changes to the posttraumatic stress disorder (PTSD) diagnosis have significant public health implications. Friedman et al.^[1] seem to carefully consider “‘goodness of fit’: whether PTSD is best classified as an anxiety disorder, a stress-related fear-circuitry, an internalizing disorder, or whether it should be classified elsewhere.” (p. 739). The strongest, most compelling evidence they present supports remaining an anxiety disorder, but the DSM-5 committee proposes to re-categorize PTSD as a “trauma and stressor-related disorder” instead of an anxiety disorder. The rationale for this shift is underdeveloped and negates the critical role of fear and anxiety in PTSD. Four arguments for retaining PTSD as an anxiety disorder were outlined with theoretical and empirical support: fear is a central construct for the development of PTSD; trauma-related fear and avoidance are critical in the treatment of PTSD; evidence supports its classification as an anxiety disorder; and this shift moves the field away from its well-developed knowledge base. In summary, the DSM-5 was to be a conservative revision,^[40] based on research evidence and, when possible, maintaining continuity with the previous edition. We strongly oppose shifting PTSD into the new category “trauma and stressor-related disorders” with what we feel is insufficient new evidence to warrant a major re-classification. Based on the overwhelming theoretical and empirical grounds, PTSD ought to remain classified as an anxiety disorder.