Has DSM-5 saved PTSD from itself?

Summary

In 2007, Robert Spitzer considered validity challenges to the diagnosis of post-traumatic stress disorder (PTSD), a construct that originated when he was Chair of DSM-III. Spitzer suggested changes for DSM-5, then in its planning stages, for the purpose of ‘Saving PTSD from itself’. With years gone by, it can be asked if DSM-5 followed Spitzer's recommendations to advance our understanding of posttraumatic disorder.

Robert Spitzer (22 May 1932 – 25 December 2015) was one the most influential figures in 20th-century American psychiatry. In 1980, as Chair of the third edition of American psychiatry's Diagnostic and Statistical Manual of Mental Disorders, Spitzer played a seminal role in deliberations that led to the introduction of post-traumatic stress disorder (PTSD). ¹ Decades later, in 2007, Spitzer revisited that diagnosis and wrote with colleagues an article entitled, ‘Saving PTSD from itself in DSM-V’. ² In the article, Spitzer observed that no other diagnosis, with the exception of dissociative identity disorder, generated so much controversy as did PTSD (such as central assumptions, boundaries of the disorder, clinical utility). Responding to these controversies, Spitzer provided recommendations to improve the validity of diagnostic criteria. Following publication of DSM-5 in 2013, ³ it is now timely to consider how the diagnostic manual responded to Spitzer's proposals.

Controversies surrounding how PTSD is framed in DSM-5

From its inception, a major issue challenging PTSD was the construct's underlying assumption of a specific aetiology: namely, that a definable subset of stressors (criterion A) creates risk for a distinct clinical syndrome. This assumption has been critiqued for definitional ambiguities, the problem of ‘criterion creep,’ and research findings that question a distinct stressor–symptom linkage. ⁴ To address these issues, Spitzer recommended that the definition of traumatic stressors be tightened by dropping the term ‘confronted with’, and by specifying ‘experienced’ to mean ‘directly experienced’. DSM-5 adopted these recommendations, but included other changes that extended the range of potentially traumatic stressors: specifically, it became possible in DSM-5 to develop PTSD by viewing traumatic work-related content on television or other electronic media. ³

Spitzer also addressed the concern that several of PTSD's defining symptoms lacked specificity because they overlapped with other diagnoses (such as depression) and/or contained general descriptors of negative affect that might encompass normal responses. It is noteworthy that when Spitzer served as Chair for DSM-III, the PTSD syndrome was associated with 12 symptoms grouped into three clusters. With the introduction of DSM-IV (1994), three symptom clusters remained, but the number of associated symptoms grew to 17. DSM-5 reworded, but essentially retained, all but 1 of DSM-IV's 17 symptoms (sense of a foreshortened future); added 4 more; and grouped the resulting 20 criteria within four clusters. ³ This resulted in neither a tightening of PTSD's defining criteria, nor an improved factor structure. ⁵ Further, the listing of 20 symptoms grouped into four clusters created 636 120 possible presentations by which an individual could fulfil diagnostic requirements. ⁶ This situation is in sharp contrast to a more focused approach taken by the World Health Organization in its most recent edition of the International Classification of Diseases. In ICD-11, PTSD was operationalised with just six core symptoms, in line with Spitzer's recommendations, and a simple two-factor model was able to account for the latent structure of the proposed syndrome. ⁷ Still, the question of which symptom list provides the best means for ‘carving nature at its joints’ ⁸ remains undecided, with each system yielding different prevalence rates and case determinations. ⁹

In addition to addressing the validity of PTSD's defining criteria, Spitzer considered the validity of case presentations and observed that malingering was a long-standing concern in personal injury and disability determinations. ¹⁰ Spitzer noted that a guideline to rule out malingering had been provided in the differential diagnosis section of DSM-IV. To strengthen the import of this guideline, Spitzer recommended its inclusion among DSM-5's diagnostic criteria. Contrary to this recommendation, DSM-5 eliminated all mention of the need to rule out malingering. At the same time, DSM-5 introduced a new symptom, ‘Reckless or self-destructive behavior,’ thereby creating scenarios wherein malingering could be a concern (for example, criminal proceedings). ¹¹

Finally, and perhaps most foundational to how PTSD is conceptualised, DSM-5 moved the diagnosis from its listing among the anxiety disorders to a new header: ‘Trauma and Stressor Related Disorders’. Authorities have challenged the basis for this change and questioned how it contributes to a clinical understanding of post-traumatic disorder. ¹²

Conclusions

Inevitably, significant issues arise when changes are made to how PTSD is operationalised: new assessment instruments are required; research must determine how extant findings fit the redefined construct; and new underlying assumptions may falter. ^3,13 A recent study demonstrates the importance of these points. Using criteria from DSM-IV and DSM-5, the study found that overall rates of PTSD-caseness were equivalent across both editions, but 30% of individuals that met criteria in DSM-IV did not screen positive for DSM-5. Conversely, 27% of those meeting DSM-5 criteria did not screen positive for DSM-IV. ¹⁴

When Spitzer suggested that PTSD needed to be saved from itself, he was cautioning colleagues that American psychiatry was increasingly applying a singular disorder to account for human reactions to adversity, even in the face of challenges to that syndrome's validity. ¹⁵ Within the context of this expanding narrative, DSM-5 largely ignored Spitzer's criterial recommendations, introduced numerous changes and produced no meaningful improvement to issues of validity. Yet, the current state of affairs presents little challenge to the viability of PTSD. As observed by British historian, Ben Shephard: ‘If “trauma” could now be broken up into its constituent parts, it would return to its social contexts and be demedicalized … [but] it is now too late. Trauma has been vectored into the wider society by the law and the media’. ¹⁶ Perhaps, in this context, we should revisit Spitzer's goal of saving PTSD from itself, and ask how American psychiatry and the broader public can be saved from PTSD as currently framed in DSM-5.