A recent issue of the Journal of Anxiety Disorders (Vol. 21, 2007) focused on the troubling possibility that the PTSD (posttraumatic stress disorder) construct is not nearly as valid as has been assumed. The articles are both fascinating and provocative and are well worth reading.
The journal kicks off with a bombshell of a study by Bodkin and his colleagues at McLean Hospital. The researchers enrolled 103 subjects who had originally been recruited for clinical trials of antidepressants. As part of the original study protocol, all patients were administered the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID). The SCID has a section on PTSD, which instructs raters to first ask patients if they have ever experienced a traumatic event (“criterion A”). If the patient responds positively, you ask about the remaining criteria (B-F) in turn, but if there has not been a trauma, you are to skip the questions and code the patient as not having the diagnosis.
In this fiendishly clever study, however, raters asked about criteria B-F even if there was no traumatic event. This is not as easy as it sounds; for example, how does one ask about flashbacks or nightmares in the absence of a traumatic event? To get around this, researchers asked subjects to think about something they had been worrying about, and then referred to this worry when they asked the questions. For example: “Have you had any nightmares about the possibility that you might have to declare bankruptcy,” and so on. What were the results? Of the 103 subjects, 54 had experienced a traumatic event, and of these, 42 (78%) of them also met symptomatic criteria for PTSD. Thirty six patients had never experienced trauma, and when these non-traumatized patients were interviewed, fully 28 (also 78%) met all the remaining criteria (B through F) for PTSD (Bodkin JA et al., J Anx Dis 2007;21:176-182).
The implication is that PTSD is not necessarily a “post-traumatic” disorder, but rather a non-specific cluster of symptoms that often occur with or without trauma. In the authors’ words: “It would follow, therefore, that in patients manifesting the symptom cluster of PTSD, it may be hazardous to assume that these symptoms were caused by trauma, even if an unequivocal traumatic event occurred.”
But if PTSD is so nonspecific, why did the diagnosis become so popular? In another article in the same issue, McHugh and Treisman of Johns Hopkins trace the history and genesis of the PTSD concept (J Anx Dis 2007;21:211-222). Surprisingly, the first formal definition of PTSD did not occur until 1980, with the publication of DSM-3. However, army doctors had known for decades that traumatic events often led to a syndrome that included emotional numbness, anxiety, flashbacks, and nightmares. World War I military doctors treated “shell shock” with brief removal from the combat zone and with psychotherapy emphasizing that the reactions were normal responses to combat and would soon dissipate.
The authors trace the ascendancy of “PTSD” to the Vietnam war. In their opinion, the diagnosis of PTSD in Vietnam veterans served several purposes at once. Veterans found a PTSD diagnosis less stigmatizing than alternative diagnoses, such as alcoholism and personality disorders. As an institution, the Veteran’s Administration found the diagnosis useful in expanding its own bureaucracy and ensuring continued funding for specialized PTSD treatment units.
After Vietnam, the ranks of PTSD patients swelled, and in 1983, Congress, alarmed at how much government money was being spent on PTSD treatment, commissioned a special study of its prevalence. The results, released in 1988, showed that almost a million of the 3.14 million men who served in Vietnam had suffered PTSD at some point. This number astounded many, particularly since only 20% of veterans had been assigned to combat units.
Apparently, some type of amplification of symptoms was going on, to put it charitably. This issue was taken up in another article in this issue, bluntly entitled “Pseudo-PTSD,” by Rosen and Taylor (J Anx Dis 2007;21:201-210). They note that PTSD is particularly vulnerable to malingering, both because the diagnosis is so often used in litigation, and because the symptom checklist is easily memorized and easily feigned. Clinicians are often fooled: in one study, even when clinicians were made aware of the possibility of malingering, many were unable to tell “real” from “fake” PTSD. Nobody knows how common malingered PTSD is, but the authors cite the forensic expert Philip Resnick’s estimate that it is as high as 50% of all PTSD presentations.
So where does this leave us? Are we to abandon the PTSD diagnosis altogether? Certainly not – we all have patients who are clearly suffering some type of posttraumatic syndrome, and for these patients the category is needed, both for diagnosis and for directing treatment decisions. Furthermore, patients who have suffered trauma often find the PTSD criteria reassuring, since they help to normalize their reactions.
The final article offers some suggestions. Written by the folks who “invented” the original DSM-3 PTSD diagnosis (including Robert Spitzer and Michael First), it suggests some significant tightening up of the diagnosis for the upcoming DSM-V (Spitzer RL et al., J Anx Dis 2007; 21:233-241). The definition of trauma would require “directly experiencing” rather than simply “experiencing” an event; some of the more nonspecific symptom criteria would be eliminated (including insomnia, irritability, poor concentration, and diminished interest); and the definitions of some core symptoms like flashbacks and avoidance would be changed to sound more extreme, in order to prevent people with milder symptoms from being included in the diagnosis.
In addition, a new “V” code would be introduced, entitled “Acute Stress Reaction,” for patients who have suffered some sort of trauma, but whose symptoms are not at the severity required for PTSD.
None of these changes are certain, however. The DSM-V is just entering the workgroup phase and the new PTSD criteria won’t be published until 2011. And some would argue that eliminating the “non-specific” criteria would actually detract from the validity of the diagnosis, since it is clear that so many patients suffer from them.
The bottom line is that in some people, trauma is associated with severe symptoms. The most useful aspect of these controversial articles is that they encourage us to be more thorough and cautious as we probe for PTSD symptoms. And no one would argue with that.
TCPR: The PTSD diagnosis: Valuable, but may need some tinkering