Posttraumatic Stress Disorder: A History and a Critique

Nancy C. Andreasen, Annals of the New York Academy of Sciences
Posttraumatic Stress Disorder: A History and a Critique


Although posttraumatic stress disorder (PTSD) is sometimes considered to be a relatively new diagnostic concept, given that it was first described using that name in the third edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-III),1 the concept of the disorder is considerably older. In order to understand where we are now, it is helpful to consider where we have been in the past. The disorder that we now know as PTSD has had a long and interesting history.

Historical antecedents to diagnostic manuals

Early descriptive accounts of stress-related disorders are often linked to the history of warfare. Stephen Crane’s introspective accounts of a youth’s reaction to the stress of a battle during the Civil War provides an early example. The Youth (Henry Fleming), the main character in The Red Badge of Courage, describes a range of anxiety symptoms that he experiences during combat.2 The horrors of trench warfare during World War I, and their resultant psychological consequences, led to formulation of the concept of “shell shock,” initially thought to be a consequence of exposure to intense artillery. Subsequently clinicians realized that the symptoms were due to the stress of the combat experience.

Interest in shellshock waned as memories of World War I receded, but it was reawakened by the advent of World War II. As had happened previously, soldiers who were chronically exposed to combat experienced a syndrome characterized anxiety, intense autonomic arousal, reliving, and sensitivity to stimuli that are reminiscent of the original trauma.3 This syndrome was given a variety of different names: traumatic war neurosis, combat fatigue, battle stress, and gross stress reaction. When the war drew to its end, another type of stress was discovered: the experience of death camp survivors.4

In the pre-DSM era a literature also accumulated on psychiatric disorders that occurred as a consequence of exposure to noncombat injuries. Alexandra Adler wrote seminal papers on the psychological effects of stress in civilian settings, beginning with her work on the Cocoanut Grove fire, and described both the clinical picture and the epidemiology.5 She also compared the effects of stress reactions occurring as a consequence of head injuries with those that occurred because of psychological stress, thereby anticipating current discussions of the relationship between PTSD and traumatic brain injury (TBI) in the context of the conflicts in Iraq and Afghanistan.6,7 Nemiah8 wrote about the effects of industrial accidents, and Hamburg et al.9 wrote about the effects of burn injuries. The descriptions of the syndromes occurring as a consequence of these diverse stressors were surprisingly similar.

During this time conceptual frameworks for understanding the effects of stress as a predisposing factor for mental illness also developed and matured. Two main positions were articulated. The first position (the “biological school”), represented by thinkers such as Selye, emphasized the role of physical mechanisms.10 Selye coined the term “stress” and hypothesized that it was mediated by the hypothalamic–pituitary–adrenal (HPA) axis. He described the general adaptation syndrome as a healthy response to stress, and he considered the traumatic neuroses to be a consequence of chronic or severe stress. The second position (the “psychological school”) had its roots in the psychodynamic tradition.11 It emphasized the role of the unconscious, and of repressed memories and early childhood traumata. It led eventually to descriptions of mechanisms of defense and of their role in producing or preventing disease. These two conceptual frameworks set the stage for the history that was to follow.

Defining gross stress reaction

Because World War II brought together psychiatrists from all over the world and from all over the United States, it became clear that they could differ in training, conceptual framework, and in approaches to diagnosis and treatment. A consensus developed that some standardization was needed, and this challenge led to the creation of the first diagnostic manual, developed by the Veterans’ Administration. This provided an incentive to the American Psychiatric Association (APA) to develop its own manual: the first Diagnostic and Statistical Manual of the APA, or DSM-I, which appeared in 1952.12 This manual included a category called gross stress reaction. It was defined as a stress syndrome that is a response to an exceptional physical or mental stress, such as a natural catastrophe or battle; it occurs in people who are otherwise normal; and it must subside in days to weeks; if it persists, another diagnosis should be made.

The first revision of this manual, DSM-II, was published in 1968.13 Without any explanation, the diagnosis of gross stress reaction was omitted. The most plausible explanation for the omission is that the concept was closely linked to warfare and combat, and DSM-II was written in a peaceful era. Consequently, between 1968 and 1980 no official diagnosis for stress disorders was available.

This change in the official APA nosology and classification occurred just as the Vietnam War was beginning its escalation. The publication of DSM-II and the Tet Offensive occurred in the same year. The war refocused attention on postcombat stress disorders. As the war became more unpopular, returning veterans were often greeted with contempt. No accepted diagnosis was available for those who had psychiatric symptoms, and treatment facilities were minimal. Activists began to note the inequity created by sending men to war without recognizing the psychiatric consequences and the need to provide adequate treatment for them. When the DSMIII Task Force was assembled in the mid-1980s, the status of gross stress reaction was one of the issues that it faced. Should this diagnosis be included again? And if so, should the old DSM-I definition be modified in any way?

Because I had studied stress disorders in burn patients,14-17 I was assigned the task of determining whether the diagnosis should be reintroduced and, if so, how it should be defined. The task required addressing three major issues.

The stressor

How severe should it be? Does the type of stressor matter? For example, if the stressor involves a cruelty inflicted by human beings, does that create more psychological stress than one occurring by accident? Does being in a death camp create more distress than experiencing a natural disaster such as a hurricane? Does the duration of the stressor matter? What is the relationship between the time of occurrence of the stressor and the onset of symptoms? Must the onset be immediate, or can it be delayed, as often occurs in postcombat situations? Should there be a specific diagnosis for each of the many different kinds of stressors? Should there be a specific diagnosis of “Post-Vietnam Syndrome,” as the pro-Vietnam veteran activists were advocating?

The stressed

DSM-I specified that gross stress reaction should only be diagnosed in individuals who were normal prior to experiencing the stress. If they had another psychiatric disorder, such as depression, the stress reaction would be treated as secondary to that and would not be given an independent diagnosis. Implicitly, this approach suggested that gross stress reaction was a diagnosis that should not confer any stigma; it implied that people who developed it were normal healthy individuals who had simply been temporarily overcome by a stress that was overwhelming. If this approach were to be adopted in DSM-III, it would ignore the accumulating evidence that protective and predisposing factors could play a role in response to stress. Might it not be the case that people with preexisting disorders, such as depression, are more vulnerable to stress and therefore more prone to develop an independent pathological response to it that should also be recognized and treated?

The symptoms

In the context of the biological versus psychodynamic interpretations of the stress response, how should the characteristic symptoms be described and defined? Should these emphasize the physical symptoms of the stress response, such as the intense autonomic arousal? Or should they emphasize the more psychological symptoms, such as dissociation, reliving, and psychic numbing? How should they be described in terms of time of onset? Should acute reactions be the only ones emphasized? Or should a delayed onset also be recognized? How long might the symptoms persist? DSM-I had specified that the disorder should subside in days to weeks, and that another diagnosis should be given if the symptoms did not remit. Should this also be the model for DSM-III?

DSM-III and the definition of PTSD

The decision to include a diagnosis that described a psychiatric syndrome occurring as a consequence of an exposure to a significant stressor was very easy to make. By the mid-1970s there was an extensive research literature describing both the epidemiology and symptomatology of syndromes occurring as a consequence of stress.3-9, 14-17 Although the absence of a diagnosis for stress disorders was a problem for government systems such as the Department of Veterans Affairs (VA), during the 1970s most of psychiatry in fact took little note of the DSM-II classification system. The absence of an “official” diagnosis for stress disorders did not preclude their study by researchers. Consequently, there was a great deal of evidence indicating that stress disorders were common, that they had characteristic symptoms, and that they were a final common pathway reached by experiencing a variety of different types of stressors: combat, death camps, industrial accidents, natural disasters, mass catastrophes, and violent acts against individuals. The issue was simply how to incorporate all this evidence into the definition and description of the disorder.

The stressor was defined relatively narrowly: as so severe that it would produce significant symptoms in almost anyone, as outside the range of normal human experience. It could be physical or psychological or both. In recognition that a stress syndrome is a final common pathway with many entry points reflecting the variety of stressors that can produce it, there was no specific “post-Vietnam syndrome.” Instead the new diagnosis was given the very general name of “posttraumatic stress disorder.” For the stressed, there was no requirement of preexisting normality; this decision was based on the recognition that individuals vary in vulnerability and resilience. The symptoms were divided into three general categories: reexperiencing (including dissociative-like states), numbing of responsiveness, and cognitive or autonomic symptoms. The onset could be either acute or delayed.

The reemergence of a diagnostic category for stress syndromes after 22 years of absence clearly filled a niche. The diagnosis soon became widely used clinically, and it also became the object of many research studies.18 The rapid and widespread acceptance also led to some unintended consequences. Despite the narrow definition of the stressor specified in the diagnostic criteria, the concept of was steadily broadened by clinicians (and also researchers) to include milder stressors that were not intended for inclusion (e.g., auto accidents, childhood abuse). The concept of dissociation was increasingly emphasized, which introduced a psychodynamic coloring that was not intended, given that DSM-III attempted to avoid alliances with any of the various competing models of disease mechanisms that were available. The temporal juxtaposition between the stressor and the symptoms was allowed to become longer and longer, so that a delayed onset became the norm. As a consequence, dissociative syndromes after “childhood abuse” (broadly defined and sometimes poorly documented) were reported with increasing frequency. The diagnosis, assumed to be relatively rare in peacetime, became much more common.

When DSM-III-R appeared just 7 years later, in 1987, many of these unintended modifications were reified in new diagnostic criteria.19 It was now 42 years after the conclusion of World War II and 24 years after the end of the Vietnam War, both of which had shaped the conceptualization of PTSD in DSM-III. Clinicians were more interested in the problems of here-and-now, and so the diagnosis of PTSD was steadily changing in that direction. DSM-III-R broadened the definition of the stressor; it was no longer defined as so severe that it would produce symptoms in almost anyone. It emphasized the psychological nature of the stressor and minimized physical components. It expanded the range of symptoms to include a stronger emphasis on dissociation, and it eliminated the acute form of the disorder. These revisions raised several concerns about the degree and rapidity of the change in conceptualization. The plight of Nazi death camp victims and the combat stress of Normandy or Iwo Jima had been the prototype for the DSM-III definition of PTSD. Had the diagnosis become too broad? Had the diagnosis become too psychodynamic at the expense of its biological underpinnings? Does using the same diagnosis for death camp survivors and victims of auto accidents trivialize the diagnosis?

The process of change continued when DSM-IV was completed in 1994.20 It too was written in a time of relative peace, and this was reflected in the modifications made to the definition of PTSD. The definition of the stressor was further modified, but still open to a broad interpretation; for example, it was expanded so that the stress was no longer limited to one experienced by the patient himself; it could be “a threat to the physical integrity of self or others” (p. 427).20 Acute stress disorder was added, but with an emphasis on dissociative symptoms. Consequently, these changes raised concerns similar to those created by DSM-III-R.

PTSD at present

The occurrence of 9/11 and other acts of international terrorism have changed the context for conceptualizing stressors. The United States is now at war again in both Iraq and Afghanistan, and consequently combat-induced PTSD is now very much on the public and psychiatric radar screen. The biological aspects of the disorder have also reemerged in importance. For example, the relationship between TBI and PTSD requires exploration.7 Furthermore, advances in neuroscience have facilitated the identification of stress circuitry in the brain through neuroimaging and animal studies.21 The rising rates of PTSD in military personnel have set off alarm bells, and they have also caused some to question the validity of the diagnosis. Others point to valid reasons for the rise, such as prolonged deployments, the frustrations inherent in counterterrorist warfare, and the recruitment of reservists and National Guard members who did not expect to engage in combat when they joined. The burden on VA Hospitals has become so heavy that an Institute of Medicine study was requested and completed. This study, in three volumes, supports the validity of the diagnosis and the increased need for services.22-24

At present the existence of a valid syndrome occurring as a consequence of severe stress can not be questioned. The diagnosis of PTSD fills an important niche in psychiatric nosology. But there are also still many ambiguities that must be resolved and gray areas that must be clarified. These include the interplay between physical and psychological components, the perils of over diagnosing versus under diagnosing, and the complex interaction between the severity and duration of the stressor and the ego strength and coping mechanisms of the individual who is stressed.


  1. American PsychiatricAssociationCommittee onNomenclature and Statistics. 1980. Diagnostic and StatisticalManual of Mental Disorders, 3rd ed. American Psychiatric Association. Washington, DC.
  2. Crane, S. 1977. The red badge of courage. In The Portable Stephen Crane. Penguin Books. New York.
  3. Grinker, R.R. & J.P. Spiegel. 1944. Brief psychotherapy in war neuroses. Psychsom. Med. 6: 123–131.
  4. Kral, V.A. 1951. Psychiatric observations under severe chronic stress. Am. J. Psychiatry 108: 185–192.
  5. Adler, A. 1943. Neuropsychiatric complications in victims of Boston’s Cocoanut Grove disaster. JAMA 123: 1098– 1101.
  6. Adler, A. 1945. Two different types of post-traumatic neuroses. Am. J. Psychiatry 102: 237–240.
  7. Bhattacharjee, Y. 2008. Shell shock revisited: solving the puzzle of blast trauma. Science 319: 406–408.
  8. Nemiah, J.C. 1963. Psychological complications in industrial injuries. Arch. Environ. Health 7: 481–487.
  9. Hamburg, D.A., C.P. Artz, E. Reiss, et al. 1953. Clinical importance of emotional problems in the care of patients with burns.N. Engl. J. Med. 248: 355–359.
  10. Selye, H. 1956. Stress and psychiatry. Am. J. Psychiatry 113: 423–427.
  11. Fenichel,O. 1996. The PsychoanalyticTheory ofNeurosis, 2nd ed. Routledge. London.
  12. American PsychiatricAssociationCommittee onNomenclature and Statistics. 1952. Diagnostic and StatisticalManual of Mental Disorders. American Psychiatric Association. Washington, DC.
  13. American PsychiatricAssociationCommittee onNomenclature and Statistics. 1968. Diagnostic and StatisticalManual of Mental Disorders, 2nd ed. American Psychiatric Association. Washington, DC.
  14. Andreasen, N.J.C., R.Noyes, C.E.Hartford, et al. 1972.Management of emotional reactions in seriously burned adults. N. Engl. J. Med. 286: 65–69.
  15. Andreasen, N.J.C. 1974. Neuropsychiatric complications in burn patients. Int. J. Psychiatry Med. 5: 161–171.
  16. Andreasen, N.J.C., R. Noyes & C.E. Hartford. 1972. Factors influencing adjustment of burn patients during hospitalization. Psychosom. Med. 34: 517–525.
  17. Andreasen, N.J.C., A.S. Norris & C.E. Hartford. 1971. Incidence of long-term psychiatric complications in severely burned adults. Ann. Surg. 174: 785–793.
  18. Andreasen, N.C. 1985. Posttraumatic stress disorder. In Comprehensive Textbook of Psychiatry, 3rd ed. H.I. Kaplan & B.J. Sadock, Eds.: 918–924.Williams andWilkins.New York.
  19. American PsychiatricAssociationCommittee onNomenclature and Statistics. 1987. Diagnostic and Statistical Manual of Mental Disorders, 3rd ed. revised. American Psychiatric Association.Washington, DC.
  20. American PsychiatricAssociationCommittee onNomenclature and Statistics. 1994. Diagnostic and StatisticalManual of Mental Disorders, 4rd ed. American Psychiatric Association. Washington, DC.
  21. Etkin, A & T.D. Wager. 2007. Functional neuroimaging of anxiety: a meta-analysis of emotional processing in PTSD, social anxiety disorder, and specific phobia.Am. J. Psychiatry 164: 1476–1488.
  22. Institute of Medicine. 2006. Posttraumatic stress disorder: diagnosis and assessment. Washington, DC.
  23. Institute of Medicine. 2007. Treatment of PTSD: an assessment of the evidence. Washington, DC.
  24. Institute of Medicine. 2007. PTSD compensation and military service. Washington, DC.
Posted on BrainLine January 12, 2011

From the Annals of the New York Academy of Sciences, October 2010 issue. Used with permission.


Bravo! As I develop a Series of posts for PTSD Awareness Month, this is one of the most unusual and full-featured articles I've run across in my web-research.  I'm not sure how I'll reference it or where I'll link it in my articles to come, but I will most certainly find a place or two for it.

I am one who wants to see, at the very least, the addition of C-PTSD [Complex] to whatever version of the DSM comes next [DSM-5-R *pretty please*???].  I can only answer YES to "Does using the same diagnosis for death camp survivors and victims of auto accidents trivialize the diagnosis?" 

More to the point, it offers precious little help in developing treatment plans, and is most likely to lead to challenges as to the legitimacy of the disorder itself (as it adds the appearance of legitimacy to misguided theories that long-term sufferers are "malingerers," somehow scamming the disabilities system).

The more I investigate PTSD, the more I am convinced that we need to clearly differentiate PTSD as a result of time-delimited episodes from the reaction to traumas that are repeated and ongoing. Treatment protocols and time-to-recovery frames will most certainly be quite different, and the non-expert "doctor down the street" that the majority of sufferers will see need clearer, more specific diagnostic guidelines (for some help with differentials).

I also feel the urge to remind ALL treatment professionals and pundits that any focus on "the here and now" must *include* the Viet Nam war.  Those combat veterans have received the least respect of any soldiers from any war, and are suffering -- NOW.  We must all stop throwing them under the bus - inadvertently or otherwise!  Their lives still matter.

As science marches on and we discover more about the overlap and connection between genetics, neurological and psychological disorders, surely any field that is not willing or able to set aside 'either-or' turf wars will eventually become obsolete. 

Perhaps the APA might ask itself whether that time has already come for any DSM attempting to be all things to all disorders. Trauma is not the only diagnostic arena where differentiation was sacrificed to simplicity.  
(Madelyn Griffith-Haynie - ADDandSoMuchMore dot com)
- ADD Coach Training Field founder; ADD Coaching co-founder -
"It takes a village to educate a world!"

The journal was released in 1968, the peace came before 1968 not after

I sometimes fear that the DSM will be the most important book ever written. Society is headed for a mass breakdown of social, psychological and physiological proportions. The stigma of mental illness has no place in our world anymore. Between PTSD and Depression alone, I believe we are reaching "epidemic" proportions. It's time to recognize the need for greater research, safer and more functional pharmacology and global acceptance of all mental illnesses. Existing with mental illness is not a choice and people should not be shunned, ridiculed or condemned because they suffer from it.

1968-75 was not a peaceful era. Vietnam War?

Add new comment

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.