More Options

We Can’t Treat Soldiers’ PTSD without a Better Diagnosis


Peter Barglow

Volume 36.3, May/June 2012

Post-Traumatic Stress Disorder is a diagnosis fully accepted by the U.S. Veterans Administration, psychiatrists, and the American public. But PTSD does not meet the criteria for a real psychiatric-medical disease.

Since 1980 Post-Traumatic Stress Disorder (PTSD) has been a major mental illness category of the American Psychiatric Association’s (APA) Diagnostic and Statistical Manual of Mental Disorders (DSM). In this article I critically examine the use of this diagnosis to treat soldiers suffering from the aftermath of physical and emotional war trauma. The term PTSD, referring to a psychiatric disease or disorder, appeared in the New York Times 19,000 times in the ten years between August 14, 2000, and August 15, 2010, compared to only 450 times during the prior twenty years.

Today, 400,000 war veterans obtain financial assistance for this medical condition. Of two million Iraq and Afghan­istan veterans, 10 percent are estimated to have PTSD. It is the topic of hundreds of psychiatric and psychology articles per year and absorbs large annual national expenditures for treatment and re­search. Out of $3.8 billion awarded as a result of U.S. Congres­sional funding bill HR2638 to the U.S. Veterans Admini­stration (VA) in 2009 for mental illness, the single largest mental disease category funded was PTSD. Between 2004 and 2009, 20 percent of the estimated half a million Iraq-Afghanistan war veteran patients were treated for PTSD. The first year of this health care cost was 1.4 billion dollars (U.S. Congres­sional Budget Office 2012). How did it come about that the diagnosis of PTSD became so widely accepted by the Veterans Health System (VHA), the American Psychi­atric Association, and the American public?

This question first was raised in my mind when two decades ago I did a psychiatric evaluation of a fifty-year-old Vietnam veteran, a Purple Heart recipient with a voluminous official history of treatment for PTSD. A few weeks earlier he had threatened to kill both himself and his therapist in her office at a VA outpatient psychiatric clinic. He was disabled by use of a Taser gun by police, who had stormed the building.

They seized him while he was un­conscious, and transported him to a prison. The patient was an unemployed married Hispanic man, wounded during the war’s Tet offensive. His record indicated that he had suffered for many years from PTSD symptoms such as hyper-arousal, insomnia with nightmares about Viet Cong snipers, and paranoid fears; these increased during binges of alcohol use. I had anticipated a confrontation with a huge, menacing figure but found the man to be a mild-mannered little guy who resembled Woody Allen. He said that he had noted none of the above PTSD symptoms for many months, but he was deeply concerned about retaining maximum VA compensation for this diagnosis.

My surprise at the apparent cure of this former soldier’s mental disease prompted me to review several dozen medical-psychiatric records of Vietnam veterans diagnosed with PTSD by me or other VA and military psychiatrists. I also reinterviewed over a dozen patients. PTSD symptoms listed in DSM-IV are memory loss or distressing flashbacks referring to battle events, hyper-vigilance, poor sleep with recurrent nightmares, irritability, startles, and episodes of emotional numbness. (This last symptom appears to be the single most important one in verifying the PTSD diagnosis [Pietrzak 2009].) The current APA diagnosis re­quires appearance of characteristic symptoms after a latency period of time subsequent to a specific severe precipitating traumatic event—constituting “Criterion A,” discussed below.

Several patients also shared with me considerable discomfort with the label of PTSD, which to them signified an emasculating weakness or dishonesty rather than a genuine illness. A search for financial benefits did appear to be one important factor in shaping the narratives of both patients and clinicians. This suggested the advantages for veterans (“secondary gain”) of re­porting typical PTSD symptoms, but it also reflected compassion of VA evaluating staff toward patients who clearly had suffered severely during and after warfare. The record review showed considerable co-morbidity (when a disease category overlaps with one or more other major psychiatric diagnoses, such as Major Depressive Disorder or Acute Stress Disorder), and many of the patients used addicting drugs (alcohol, marijuana, pain killers, or amphetamine stimulants). Substance dependence was almost impossible to disentangle from PTSD symptoms.

Unlike my patient whom I de­scribed earlier, few veterans with PTSD im­proved very much during the many years that had elapsed since their initial clinical assessment. Often DSM-IV clinical criteria for PTSD had been carelessly applied to veterans whose post-war lives had been dominated by poverty, unemployment, homelessness, and family disruptions because of violence, drugs, or divorce.

A Brief History of PTSD

The DSM classification system was created in 1952. Its first two editions (I [1952] and II [1968]) were based upon Freud’s psychoanalytic formulations. The etiology of mental disorders was thought to originate in early-life traumatic experiences. However, a major change in thinking about the concept of a mental disorder and its etiology occurred in the 1970s, reflected in DSM-III (1980). There was a marked shift away from attention to early childhood histories. Valiant attempts were made to mimic mainstream medicine and surgery using their ancient etiological categories—trauma, cancer producing, infectious, toxic, degenerative, genetic, metabolic, and endocrine.

But psychiatric disorders proved difficult to classify with quantifiable chemical findings or specific identifying clinical signs. The revised DSM systems (in an effort to establish reliable guidelines for diagnosticians) still had to rely substantially upon self-reported descriptions of symptoms, not on measurable data. DSM I–III systems’ categorical decisions reflected literature reviews, some data analysis, periodic field trials, and the outcome of verbal debates between experts. That complex decision-making process used patients’ clinical information but had to rely upon fallible doctors’ judgments. A better approach to diagnosis creation was clearly needed, and so psychiatric research in the early 1990s was increasingly devoted to the human brain.

A 1889 line drawing of the Greek hero Herakles, afflicted with something akin to PTSD brought on by the violence of his twelve labors, by the artist August Baumeister. It originates in a Greek-Sicilian vase painting signed by the artist Asteas (350–320 BCE), depicting a theater performance in which the Herakles of the dramatist Euripides is about to immolate the first of his three children while his wife attempts to escape his psychotic wrath.

PTSD was first listed in the 1980 edition of the APA’s Diagnostic and Statistical Manual of Mental Disorders (DSM-III) and modified later in DSM-IV and DSM-IV-TR. Its ap­pearance in the official APA nomenclature followed years of intense lobbying effort by Vietnam veterans’ organizations, activist social workers, psychologists, and anti-war psychiatrists. Advo­cates for the PTSD diagnosis asserted that traumatic memories of war experiences were being revived in contemporary time, producing a new serious men­tal illness. Soldiers should be treated and compensated for a disorder attributable to events that took place many years earlier. This understanding necessitated a shift in attention away from the psychodynamics of individual veterans, and other risk variables, to a heavy emphasis upon a single major factor—the negative aftereffects of war trauma on later mental health.

The leaders of the American psychiatric profession who became midwives to the official birth of PTSD during the 1970s shared today’s almost universal belief that large-scale suffering of others matters universally and that it demands to be recognized and ameliorated. This moral value probably accounted for the diagnostic inclusion of Criterion A, which has generated heated debate since its original inclusion in the 1980 DSM-III. Criterion A for PTSD states that a patient diagnosed with PTSD was confronted with “events that involved actual or threatened death or serious injury. . .” and responded with “intense fear, helplessness, or horror.” Criterion A made PTSD the only DSM mental disorder that required a subjective appraisal of an external environmental stressor as part of its diagnosis. Retention of the trauma criterion has been supported by the observation that studies of symptoms unconnected to a specific precipitant have failed to identify any “characteristic set of symptoms” (North et al. 2009). Also this position is consistent with the conclusion that treatment concentrating on specific trauma memories and their meaning is more effective than nontrauma-focused therapy (Ehlers et al. 2010).

Chart: PTSD benefits for vets vary

But the criterion does have many problems. One statistical piece of evidence against inclusion of documentation of a quantitative trigger to facilitate making a diagnosis is that most soldiers do not develop an anxiety disorder or any major psychiatric disorder even when exposed to the most horrific trauma. The widespread application of Criterion A ignores vast individual variations in patients’ resilience and capacity to adapt. Its reliance upon subjective reports rather than objective eyewitness evidence further weakens its scientific status. For the preceding reasons, Criterion A does not appear in the PTSD diagnosis category of the International Classification of Mental Diseases (ICD-10).

The importance of this diagnostic criterion in determining the size of disability benefits for veterans with PTSD has been diminished by new VA standards issued by the Obama admini­stration in 2010. The VA policy now states that VA psychiatrists need not require proof of the quantitative impact of a traumatic precipitant. This VA policy change was inspired by a deep concern for the suffering of victims and largely ignored the APA’s PTSD Criterion A, which may be deleted from DSM-V’s definition.

Does PTSD Meet the Criteria for a Valid Psychiatric Diagnosis?

The diagnosis of PTSD has always had many critics, ranging from McHugh and Treisman (2007), who boldly consider PTSD to be a “faddish postulate” that “creates a medical condition out of normal distress,” to the meticulous scholars Rosen and Lilienfeld (2007), who concluded that the disorder’s “core assumptions and hypothesized mechanisms lack compelling or consistent empirical support.” Robins and Guze (1970) proposed five research areas in which a psychiatric diagnosis might be validated: (1) clinical description including precipitants and diagnostic stability over time; (2) biological, hormonal, and radiological quantitative evidence; (3) distinct boundaries be­tween the disorder’s characteristics and other psychiatric conditions; (4) family or genetic statistical connections be­tween patients in the diagnostic category; (5) treatment relevance and success related to precise diagnosis. Schizophrenia, major depression, and alcohol dependence are examples of mental disorders that have achieved considerable legitimacy through this process, but PTSD as a diagnosis for war veterans has not yet attained comparable validity.

The above five research domains constitute fertile ground for further re-examining PTSD’s diagnostic weaknesses and generating potential remedies:

Clinical Description and Precipitants. The VA patient I described earlier is an example of diagnostic instability. The weakness of Criterion A suggests a problematic relation between PTSD and specific precipitants. These days it is obvious that patients suffering from the emotional aftermath of warfare have not demonstrated consistent symptom patterns. Perhaps, then, we can better comprehend the emotional toll of recent American wars by applying the sociological concept that throughout history powerful professional and political communities have constructed truth, established definitions, and generated rules for the interpretation of trauma’s impact. This viewpoint provides insight into the history of emotional war trauma and explains its massively varying conceptualizations and manifestations. Like “Historical Critical Psychopathology” (Bald­win et al. 2004), it emphasizes“his­­torically situated and contingent aspects of mental disorders.”

Chart: Iraq and Afghan war veterans

Using such a framework of understanding, the earliest PTSD portrayal may be the fourth-century character Herakles, created by the dramatist Euripides. Driven insane by a Greek goddess, he suffered a transitory murderous frenzy precipitated by the violence of his twelve labors. A messenger inquires of him, “Has the blood of the men you recently killed driven you out of your wits?” Over two millennia later, emotional victims of American Civil War trauma were said to be afflicted with “Soldier’s Heart” (1864–1868), and those suffering from “Railway Spine” in America (circa 1886) showed psychological and physical characteristics quite different from those used currently to identify PTSD. Such patients in the latter half of the nineteenth century experienced bodily shaking and tremors of arms and legs, stuttering, and limping, but they did not report anger, numbing sensations, or flashback symp­toms as contemporary PTSD victims often do.

Civil War victims’ clinical presentations resembled those reported during the Russo-Japanese War (1904–1905), when emotional casualties began to be treated with considerable respect. Many Russian psychiatrists argued that afflicted soldiers had a “real illness.” English and German “shell shock” victims and Russian “contusion” casualties during World War I (1914–1918) also demonstrated a quite different symptom constellation from that characteristic of PTSD in twenty-first-century America. European patients manifested multiple sensory-motor signs such as deafness, muteness, and blindness. Such striking historical dissimilarities in the psychological clinical phenomena of post-war emotional syndromes suggest they resemble cultural constructions more than disease categories. Cultural factors may also influence treatment outcome; PTSD victims in Kenya found help in their religious community, in contrast to Okla­homa City bombing survivors who received benefit from medical treatments (North 2009).

Brain Structure and Neurophysiologic Studies. Since DSM-IV appeared in 1994, there has been a massive increase in U.S. research efforts to demonstrate that an organic central nervous system disturbance causes PTSD. This campaign was named “embodiment” by skeptics who deplored the use of inappropriate comparison groups and the contaminating role of “cultural ex­pectancy” in the studies mobilized to create a more precise and useful PTSD diagnostic category (Baldwin et al. 2004). The search for organic brain changes as the source of PTSD symptoms was fueled by the belief that war’s emotional stress could affect brain physiology and chemistry negatively, producing permanent and characteristic post-trauma symptom patterns. The first major effort to argue this position originated during the 1905 Russo-Japanese war through studies of soldiers who suffered from “contusion,” a puzzling nervous breakdown during or after combat. The psychology versus brain change etiological debate became even more active during World War I (1914–1918) in Europe with research about “shell shock.”

Since the Korean War (1950–1953), multiple research efforts in English-speaking countries have sought distinct patterns of psycho-physiological arousal in PTSD. But measurements of neurotransmitter levels across most populations with the disease showed marked differences of quantity, including a complete failure to show heightened re­sponses in a quarter of cases. Acute stress has been shown to activate the central nervous system to release catechola­mines, norepinephrine and epinephrine, into the bloodstream. But patients with acute PTSD symptoms have normal plasma concentrations of these substances. Only those with chronic PTSD have increased cerebrospinal fluid norepinephrine levels that correlate with the severity of PTSD symptoms. But all situations that produce stress for human primates—not just emotional war trauma—are associated with catecholamine release.

Malfunction of the hypothalamus-pituitary-adrenal hormonal axis has been postulated in those PTSD victims who have lowered blood cortisol levels and who demonstrate enhanced cortisol suppression with dexamethasone and exaggerated cortisol secretion during stressful exposure. But findings have been inconsistent or subtle, and many hormonal studies lack hormone level data pre-exposure to stress. Recent prospective longitudinal studies have shown that low cortisol levels in PTSD patients may actually have preceded, not followed, a traumatic event. Finally, cortisol levels vary dramatically secondary to numerous biological, temporal, and psychological variables, making them an unreliable biological marker.

Another major effort to find a neurobiological correlate to PTSD has focused upon brain morphology studied through brain scanning. Extrapo­lating from animal stress models, it was suggested that elevated corticosteroids (adrenocorticotropic hormone, or cortisol) in patients with PTSD could produce toxic damage to the hippocampus. The function of memory is thought to reside there, and the structure’s compromise might be related to flashbacks and nightmare symptoms. But there is only weak evidence for this link in humans. Also, most human studies have failed to control for other psychiatric disorders associated with hippocampal atrophy, including alcoholism. Smaller hippocampal volume is a relatively nonspecific finding that also has been reported to be associated with depression and even borderline personality disorder. Statistical associations between emotional war trauma, PTSD, and size of brain structures remain highly speculative and unproven.

PTSD Boundaries with Other Psychiatric Disorders. The diagnosis of PTSD today is made by the use of standardized tests, such as the self-report PTSD Symptoms Checklist, the Clini­cian Administered PTSD Scale, and a Clinical Interview for DSM Disorders. But still, PTSD is poorly separated from other major psychological disorders, and its diagnostic stability is low, as illustrated by my patient vignette in the introduction of this article. Using current diagnostic criteria, PTSD has consistently been found to overlap other major psychiatric disorders. Kessler et al. (1996) found that 88 percent of a large sample of U.S. patients with PTSD met DSM-III or IV clinical-case data criteria for at least one other major mental disorder, most often chemical dependence. PTSD does have convergent validity with major depressive disorder (95 percent lifetime and 50 percent co-occurring) in combat veterans. Further research on the centrality of “numbing” may help establish PTSD as an independent diagnostic entity (Pietrzak 2009).

The medical condition traumatic brain injury (TBI) is also difficult to distinguish from PTSD. TBI is a brain concussion caused by a blow to the head that changes a soldier’s consciousness, resulting in amnesia and neurological abnormalities. TBI can result from damage sustained from bullets, bombs, falls, or vehicle accidents. TBIs may be conceptually related to the above historical concepts of English “shell shock” and Russian “contusion” traumata.

Genetic Studies Related to Emotional Consequences of Battle Trauma. Such research remains suggestive rather than convincing. Molecular genetics has yet to identify specific genes that confer a vulnerability to PTSD. Although genetic research does support a biological component of PTSD, it has failed to distinguish it from other related (co-morbid) psychiatric diagnoses. One study showed that shared genetics accounted for 30 percent of the variance in PTSD symptoms in Vietnam War veteran twins, even after taking into account different levels of combat exposure. Stein et al. (2002) proved in 1999 that genetics contributes to both the tendency for a subject to be exposed to traumas involving assault and a vulnerability to develop PTSD after exposure. Such results have led to the theory that a combination of genes and environment are necessary to develop PTSD symptoms, a proposal that contradicts the assumption that trauma is the core etiological agent. The preceding observations are consistent with the idea that PTSD may be distinguishable from other common disorders.

The PTSD Diagnosis Used to Guide Treat­ment of the Illness. Experts today do not yet know if the prolonged emotional suffering following physical or emotional war trauma can be ameliorated, let alone cured. The Institute of Medicine carefully evaluated fifty-three drug studies and thirty-seven psychotherapy studies (National Acad­emy of Sciences 2008). Their team of academic specialists examined the evidence (or lack of evidence) for the success of anticonvulsants, benzodiazepines, MAOIs, SSRIs and other antidepressants, and other remedies such as Eye Movement Desensitization and Reprocessing (EM­DR). Their summary concluded that there was insufficient scientific evidence to determine that any treatment (except possibly Prolonged Exposure Therapy) had beneficial long-term effects for PTSD. Their report noted that drug manufacturers had funded many studies, possibly biasing their outcomes, and that too many research patients had dropped out of experiments to allow robust conclusions about efficacy. Little is known about what kind of patients benefit from which treatments.

On April 26, 2010, the New York Times published an article that de­scribed the plight of veterans who were physical and emotional casualties, slated for discharge, temporarily hospitalized at Fort Carson’s Warrior Transi­tion Battalion unit. Inmates and their close family members depicted absent or poor psychiatric treatment, overmedication, bureaucratic delays, and prescription of drugs that fostered ad­diction to heroin. Poly-pharmacy treatment for PTSD (combining anti-anxiety, antidepressant, antipsychotic, and anti-insomnia agents) recently has been associated with accidental death in U.S. veterans with this condition (New York Times, February 14, 2011). Such reports are painful reminders that neither the U.S. military establishment nor the Veterans Administration has learned quite enough to treat victims of emotional war trauma effectively.

“Prolonged Exposure Therapy,” originated and proven efficacious by Edna Foa, remains the most evidence-supported successful intervention. It helps traumatized soldiers to approach trauma-related thoughts, feelings, and situations previously avoided because they cause distress. The technique uses education, breathing exercises, safe re-exposure to painful avoided or un­avoidable war experiences, and “talking through” a patient’s individual history of trauma. A couple of my own PTSD patients who tried this method felt that it reawakened and reactivated traumatic memories that appeared to have been forgotten, making their emotional distress even worse. This possible disadvantage for a portion of the war-traumatized population may be comparable to the reaction that some alcoholic addicts have experienced during Alco­holics Anonymous meetings. The ex­plicit sharing with others of vivid ex­periences associated with the ravages of the disease stimulates them to drink right after a seemingly successful meeting. Also evidence for effectiveness of exposure therapy is not as strong for veterans as it is for civilians.

Recent empirical research studies indicate that cognitive processing therapy (CPT) is also often effective in treating the emotional aftermath of warfare. The method tries to modify perceptions of a specific war trauma and its reproduction.


Currently there is little benefit in directing treatment interventions toward patients diagnosed with PTSD, an amorphous disease category with indistinct conceptual boundaries and without a firm biological foundation. Money might be more prudently spent on im­mediate post-trauma intervention ad­dressing individual patients’ symptoms. Treatment of concurrent depression, addiction, and other anxiety conditions may be far more valuable than targeting chronic symptoms and decades-later psychological aftermaths. All ap­proaches must ameliorate veterans’ social conditions—poverty, homelessness, and marital and familial friction. Acute post-traumatic symptoms must not be transformed into a chronic compensated disability, set in stone by its designation as an official major DSM disease syndrome.

The use of the PTSD diagnosis may contribute to treatment failures because it fabricates a spurious invalid category of illness, rather than seeing a unique sufferer. A humane society must compensate and reward all military victims with generosity, but strict application of Criterion A of the PTSD diagnosis does not accomplish this purpose. PTSD appears to be more of a social construction than a medical brain disease, and as of this date can best be considered “as encompassing a broad range of possible reactions to adverse events” (Rosen and Lilienfeld 2007, 858). Finally, the stigma associated with this diagnosis of a mental illness may keep some veterans from seeking care. The shortcomings of the current PTSD diagnosis jeopardize the treatment of the terrible aftermath of war’s emotional trauma.


I thank Dave Wilson, MD, (an internal medicine resident at Stanford Medical School) for assistance with review of the biological substratum of PTSD.


Baldwin, S.A., D.C. Williams, A.C. Houts. 2004. The creation, expansion, and embodiment of posttraumatic stress disorder. The Scientific Review of Mental Health Practice 3(1): 1–39.

Ehlers, A., D.M. Clark, M. Creamer, et al. 2010. Do all psychological treatments really work the same in posttraumatic stress disorder? Clinical Psychology Review 30: 269–76.

Kessler, R.C., C.B. Nelson, K.A. McGonagle. 1996. The epidemiology of co-occuring ad­dictive and mental disorders. American Journal of Orthopsychiatry 66(1): 17–31.

McHugh, P.R., and G. Treisman. 2007. PTSD, a problematic diagnostic category. Journal of Anxiety Disorders 21(2): 211–22.

National Academy of Sciences, Institute of Medicine. 2008. Treatment of Posttraumatic Stress Disorder. Published by Institute of Medicine, National Academic Press.

North, C.S., A.M. Suris, M. Davis, et al. 2009. Toward validation of the diagnosis of posttraumatic stress disorder. American Journal of Psychiatry 166(1): 34–40.

Pietrzak, R.R. 2009. The importance of four-factor emotional numbing and dysphoria models in PTSD. American Journal of Psychiatry 166(1): 40–41.

Robins, E., and S.B. Guze. 1970. Establishment of diagnostic validity in psychiatric illness. American Journal of Psychiatry 126: 983–87.

Rosen, G.M., and S.O. Lilienfeld. 2007. Post­traumatic stress disorder: An empirical evaluation of core assumptions. Clinical Psychology Review 28: 837–68.

Stein, M.B., J.L. Jang, and S. Taylor. 2002. Genetic and environmental influences on trauma exposure and posttraumatic stress disorder symptoms: A twin study. American Journal of Psychiatry 159: 1675–81.

U.S. Congressional Budget Office. 2012. The Veterans Health Administration’s treatment of PTSD and traumatic injury among recent combat veterans (CBO study, February 2012).

Peter Barglow

Peter Barglow, MD, is a psychiatrist who has taught at Northwestern, University of Chicago, and University of California at Davis medical schools. Currently he is a Clinical Professor of Medicine and Psychiatry at UC Davis. He treated active duty military personnel while in the U.S. Navy and veterans through the VA Northern California Health Care System. Email: