Dr. Marshall is director of trauma studies and services at the New York State Psychiatric Institute and associate professor of clinical psychiatry at Columbia University College of Physicians and Surgeons, both in New York City.

Dr. Sinha is research fellow in the Department of Psychiatry at Columbia University College of Physicians and Surgeons.

Acknowledgments: This work was supported in part by National Institute of Mental Health Grant#MH01412. 



The history of the recognition of psychological impairment after severe trauma is, in many ways, a microcosm of the history of clinical psychiatry and psychology. Until recently, etiological assumptions about clinical manifestations have confounded the field. However, current research regarding the central nervous system’s plasticity in both the developing and adult human links severe psychological trauma with enduring alterations in brain function. These recent studies offer a rationale for administering medication to patients with posttraumatic stress disorder (PTSD). More importantly, adults diagnosed with PTSD can receive a number of different medications that have demonstrated efficacy. However, the real priority is to achieve full remission in patients rather than merely improvement in symptoms. Hence, further study of new compounds as well as longer and combined treatments will be necessary and will likely improve our management of patients with this often refractory disorder.



There is … no a priori reason why drugs should not be discovered which have a wholly beneficial effect. I have never myself observed any but good effects to flow from the drinking of tea, at any rate if it is China tea.

Bertrand Russell1

For several centuries, medical literature has documented the relationship between traumatic experiences and enduring, debilitating psychological suffering. However, only recently has this association been incorporated in the psychiatric literature (Diagnostic and Statistical Manual of Mental Disorders, Third Edition2 [DSM-III]). The history of this phenomenon is revealing in itself and pertinent to contemporary debate in the field.3 The structure of arguments proposing a connection between trauma and psychological stress seem consistent with the Hegelian model of the evolution of “thesis, antithesis, synthesis.” Initial theoretical explanations focused on the biological perspective in the 18th and 19th centuries. The early 20th century witnessed the psychoanalysis debate with psychological explanations brought to the forefront.

In the past 10 years, proof of both somatic and psychological approaches have emerged, along with evidence of biological vulnerabilities and a genetic predisposition to posttraumatic stress disorder (PTSD). Once all the data are synthesized, we suspect that a true biopsychosocial model of PTSD will emerge as the most accurate and clinically useful model.

Throughout this debate, propositions about the etiology of PTSD symptoms have often polarized between two models.4 A more biologically based model posits that vulnerability factors explain why some individuals develop PTSD while other persons, despite the same severity of trauma, spontaneously recover. A psychologically based model driven by stigmatization proposes that all individuals with PTSD possess a universal vulnerability to severe trauma and objects to the medicalization of posttraumatic reactions as “symptoms” rather than normal reactions to abnormal events.

In the 19th and early 20th centuries, a common belief was maintained that PTSD-like syndromes resulted from neurological damage (eg, minimal brain damage during an accident, damaged heart tissue, or concussion-like effects of explosive shells leading to “shell shock”).5 The alternative viewpoint—that clinical symptomatology might be better explained as a psychological consequence of trauma—came to the forefront during the psychoanalytic era, in which all mental illness was reconceptualized as having a psychological and developmentally rooted etiology. However, even this perspective presumed the existence of constitutional vulnerabilities. We now know that both biological alterations and vulnerability factors are associated with PTSD. For example, in a large study of United States Vietnam veterans, genetic models accounted for about one third of symptom liability.6

After World War II, psychoanalysis was the dominant paradigm. In 1947, after treating many war veterans, Kardiner and Spiegel7 noticed striking physiologic symptoms in patients who seemed to suffer from damage and/or dysregulation of the autonomic nervous system. They termed this condition “physioneurosis.”

During this era, physicians primarily treated psychiatric illness by analyzing neurotic conflicts that were rooted in the patients’ early development; physicians assumed that the patients’ symptoms represented regression under stress.

Pharmacotherapy for Physioneurosis

After World War II, experts advocated the use of sedatives in patients with chronic conditions to facilitate the psychotherapeutic process of dealing with memories of trauma. Only barbiturates were available then, and their disadvantages often outweighed their usefulness. Dr. Kardiner8 explained it thus:

Barbiturates have the advantage of being simultaneously an anesthetic and a device to render the subject accessible. It has however the disadvantage of blunting the sensorium, so that, although it is possible to dissolve a good many peripheral defenses, the chances of complete integration of the reconstructive efforts are diminished.

Although researchers recognized that biological factors played a role in physioneurosis, they did not believe that medications contributed toward recovery, which was presumably a psychological process.

Empirical Findings to Date

Pharmacotherapy for patients with PTSD has demonstrated direct usefulness over the past 10–15 years, despite the fact that the mechanism of action for most psychiatric medications remains unknown. Although medications are categorized by their pharmacodynamic properties (eg, selective serotonin reuptake inhibitor [SSRI]) and molecular structure (eg, heterocyclic antidepressants), none of these designations explain the mechanism of action. This is a very important research issue because premature closure regarding the relationship between the psychobiology of PTSD and the putative mechanisms of pharmacotherapy can be damaging to scientific progress in this area.

Early medication trials were conducted exclusively with war veterans who usually demonstrated modest results with the exception of a single trial of phenelzine.9 Tricyclic agents also seemed to have modest effectiveness in this population.9,10

Since the discovery of the SSRIs, the study of PTSD caused by combat or other types of trauma has shifted to progressively larger, more rigorous trials. Five large multicenter trials using the SSRIs sertraline, paroxetine, and fluoxetine have demonstrated efficacy compared with placebo in adults who had chronic PTSD due to a wide range of traumas.10-14 These findings replicated and extended the results of two previous single-site studies with fluoxetine.15,16

Negative trials with SSRIs have also been reported.17-19 Three of the four negative trials were conducted primarily using US war veterans as subjects. To date, the only multinational trial conducted in Africa and Europe showed efficacy for war veterans. This suggests that recruitment patterns in US studies probably differ from those abroad and may be biased toward enrolling treatment-refractory individuals who have failed several prior treatment attempts.14

Three earlier pioneering trials demonstrated superiority over placebo for the tricyclic antidepressants amitriptyline9 and imipramine,9 and the monoamine oxidase inhibitor phenelzine.9 Smaller early trials conducted with phenelzine20 and similar drugs (alprazolam21 and inositol22) produced negative results, although the studies were probably inadequately powered.

A large number of case reports have reported therapeutic efficacy for other drug categories including mood stabilizers such as valproate, topiramate, carbamazepine, gabapentin, and lithium, and α-agonists (eg, clonidine).23 Of course, double-blind, placebo-controlled trials must be conducted before a drug can be considered effective. The history of medicine is replete with humbling examples of the post hoc fallacy. The scientific literature is biased toward positive reports because negative trials are less likely to be submitted for publication.

Treatment-Response Patterns

Many clinically relevant observations have emerged from the new large-scale SSRI trials that answer longstanding questions, sometimes overturn previously held assumptions, and often point investigators toward important and new directions such as the following:

• Medication is effective for PTSD. Although this point may seem obvious to many readers, only a decade ago this controversial assertion was maintained by only a minority of experts in the field. The Food and Drug Administration (FDA) has approved sertraline and paroxetine for PTSD. The largest paroxetine trials to date have now demonstrated efficacy for all three symptom clusters analyzed.13,14 It was initially surprising that a medication could help to reverse entrenched behavioral avoidance patterns that had often been maintained for years.
• Medication is effective for PTSD arising from a wide range of trauma types. Early literature focused largely on the type of trauma, rather than the universal features that most trauma survivors shared. Thus, there was a railway spine syndrome, a rape trauma syndrome, battle fatigue, irritable heart syndrome, and so forth. A major contribution of the DSM was the focus on defining universal features of syndromes, such as longitudinal course.3 Thereafter, clinical trials demonstrated that drugs were effective for treating individuals with PTSD resulting from all types of traumas ranging from car accidents to domestic violence.
• SSRIs appear to be equally effective for men and women. A major controversy in the literature has been the question of whether there are gender differences in treatment response. Approximately twice as many women as men report a lifetime history of PTSD.24 However, until recently, the literature was confounded by the fact that most of the negative trials with SSRIs were conducted in US war veterans, who were almost entirely males. A recent trial13 with paroxetine demonstrated equal treatment efficacy in both genders, the majority of whom had experienced nonwar-related trauma.
For most individuals, medication alone will not lead to remission, at least in the short-term. On the whole, the mean reduction in symptoms in large SSRI trials is approximately 50% compared with a 30% reduction in symptoms observed in patients who received a placebo. Similar findings are seen in time-limited psychotherapy trials that use the same methodology for outcome analysis (ie, intent-to-treat analysis). This deserves emphasis because the psychological literature has traditionally published more complete analyses which tend to be more favorable than intent-to-treat analyses. These results suggest that additional treatment will be needed for most individuals. In double-blind, placebo-controlled trials, the response pattern suggests a flattening of the response curve at 6–8 weeks of treatment with adequate doses of an SSRI (in the higher range).

However, a recent report by Londborg and colleagues25 suggested that therapeutic benefits can continue for more than 12 weeks in at least a subgroup of individuals. Patients were offered the opportunity to receive open treatment with sertraline in a continuation phase (24 weeks) after completing a double-blind, placebo-controlled trial. The authors included in the analysis only those patients who had been randomized to receive the active medication. Of those patients who received sertraline during the trial and opted to receive open treatment (n=128), about half of those who had been rated as nonresponders (28/52) became responders during this phase over the next 24 weeks.

This response pattern stands in contrast to those observed in all acute controlled trials to date—a clear plateau in treatment response seen in patients who received medication alone after 6–8 weeks. It is possible that individuals who could not tolerate sertraline or had a minimal response did not enter the open trial, thereby altering the sample composition. Nevertheless, this study25 provides the first evidence that at least some patients can show ongoing improvement over 6 months, and that partial responders should probably continue medication beyond the usual 12 weeks if other treatments that might accelerate recovery are unavailable or are refused.
The clinical need for additional treatments exceeds our empirical knowledge. Case reports and open trials document the fact that many patients do not achieve full recovery after treatment with medications of established efficacy. This finding is apparent to most clinicians who manage and treat PTSD patients. Until further research is available, the recommended strategy is to target specific symptom clusters or problems that may be responsive to certain types of medications. For example, patients who have problems with affect dysregulation, irritability, and aggressive impulses and who have not responded to SSRIs might experience symptom reduction by taking a mood stabilizer. Patients with persistent reexperiencing in the form of nightmares and/or vivid imagery might respond to α-agonists. Patients with depressive symptoms that have only partially responded to the first antidepressant agent might improve with the addition of a second medication of complementary antidepressant action. In all clinical situations, the combination of medication with supportive psychotherapy has been recommended.26

Treatment Versus Referral

Most experts working with PTSD patients, including those involved in research in primary care settings, have stated that the primary care provider’s preferred role is identification, provision of education, and referral to a specialist.27

The most efficient and successful way to detect PTSD is to assess for lifetime traumatic experiences. Brief questionnaires such as the Life Events Questionnaire28 or Short PTSD Rating Interview (SPRINT)29 offer a natural lead into questions about the consequences of the traumatic experience. The physician’s assessment might include questions about reexperiencing (eg, “Was it on your mind a lot afterward?” or “Did you keep seeing it in your mind and dreaming about it?”), avoidance (eg, “Did you find yourself withdrawing from other people afterward?” or “Do you feel less interested in things you enjoyed before?”), and arousal (eg, “Were you very jumpy or jittery afterward?”, “Are you more cautious or fearful in general?”, “Are you having trouble sleeping?”, or “Do you startle easily?”). Relationship problems are also common (eg, “Did you find you were more irritable, having trouble getting along with others? If so, did this affect your sex life?”).

If the diagnosis of PTSD is made, it is crucial to mention that there are safe and effective treatments, in order to give the patient hope and motivation for seeking help. Referral to a specialist is preferable for patients who are willing. If patients are reluctant to meet with a mental health practitioner or if none are available, providing information about treatment options is still important because a subsequent partial response to medication may require revisiting the option of psychotherapy.

A number of typical questions can arise during medication treatment. Because trauma patients have a heightened sense of caution, they may have exaggerated fears about side effects of medications. Reviewing research findings with these patients may provide reassurance. Patients may misperceive the offer of medication treatment as an expression of unwillingness to talk about their traumatic experience.

On the other hand, many patients seek help based on the assumption that the medical model is most appropriate when they feel severely ill. In nonwestern countries, depression and anxiety primarily present with somatic symptoms. The fact that there is a biological model for PTSD can be experienced as legitimizing their suffering. Patients may experience the biological model as a positive connection to their physicians  and, hopefully, as a way of taking positive action to improve their lives. However, patients may sometimes rely heavily on the medication based on an idealized wish to erase the trauma from their memory. We believe it is important to present the medication as an adjunct to the patients’ efforts toward recovery.

If a first medication is only partially effective, the physician can again recommend the option of psychotherapy. Although guidelines have been recently published26 concerning the next steps in a treatment algorithm, a psychiatrist’s assistance is highly advisable at this point, if possible.


The history of treatment for chronic PTSD reflects the vicissitudes of our field. Although we still have no coherent and validated theory to explain PTSD symptoms, emerging ethological and clinical research strongly supports the power of pharmacotherapy to correct alterations in multiple systems that modulate affect and perception. The SSRIs offer the most optimal and validated empirical treatments to date. The FDA has approved both sertraline and paroxetine in the US.

Acute treatment trials of medication or psychotherapy demonstrate that a substantial proportion of patients remain symptomatic at the conclusion of the trial. Unfortunately, most individuals with trauma-related symptoms or problems never seek and/or receive effective treatment. Public health efforts and clinical trials are equally important in diminishing the suffering of trauma survivors.   PP


1. Russell B. The Scientific Outlook. New York, NY: W.W. Norton; 1931:184.
2.    Diagnostic and Statistical Manual of Mental Disorders. 3rd ed. Washington, DC:?American Psychiatric Association; 1980.
3.    Marshall RD, Klein DF. Diagnostic classification of anxiety disorders: historical context and implications for neurobiology. In: Charney DS, Nestler EJ, Bunney BS, eds. Neurobiology of Mental Illness. Oxford, England: Oxford University Press; 1999:437-450.
4.    Yehuda R, McFarlane AL. Conflict between current knowledge about posttraumatic stress disorder and its original conceptual basis. Am J Psychiatry. 1999;152:1705-1713.
5.    Kinzie JD, Goetz RR. A century of controversy surrounding posttraumatic stress-spectrum syndromes: the impact on DSM-III and DSM-IV. J Traumatic Stress. 1996;9:159-179.
6.    True WR, Rice J, Eisen SA, et al. A twin study of genetic and environmental contributions to liability for posttraumatic stress symptoms. Arch Gen Psychiatry. 1993;50:257-264.
7.    Kardiner A, Spiegel H. War Stress and Neurotic Illness. 2nd ed. New York, NY: Paul B. Hoeber, Inc; 1947.
8. Kardiner A, Spiegel H. War Stress and Neurotic Illness. 2nd ed. New York, NY: Paul B. Hoeber, Inc; 1947:364.
9.    Kosten TR, Frank JB, Dan E, McDougle CJ, Giller EL. Pharmacotherapy for posttraumatic stress disorder using phenelzine or imipramine. J Nerv Ment Dis. 1991;179:366-370.
10.    Davidson J, Kudler H, Smith R, et al. Treatment of posttraumatic stress disorder with amitriptyline and placebo. Arch Gen Psychiatry. 1990;47:259-266.
11.    Brady K, Pearlstein T, Asnis GM, et al. Efficacy and safety of sertraline treatment of posttraumatic stress disorder: a randomized controlled trial. JAMA. 2000;283:1837-1844.
12.    Davidson JRT, Rothbaum BO, van der Kolk B, Sikes CR, Farfel GM. Multi-center, double-blind comparison of sertraline and placebo in the treatment of posttraumatic stress disorder. Arch Gen Psychiatry. 2001;58:485-492.
13.    Marshall RD, Beebe KL, Oldham M, Zaninelli R. Efficacy and safety of paroxetine treatment of chronic PTSD: a fixed-dosage, multi-center, placebo-controlled study. Am J Psychiatry. 2001;158:1982-1988.
14.    Martenyi F, Brown E, Zhang H, Prakash A, Koke S. Fluoxetine vs placebo in posttraumatic stress disorder. Poster presented at: Annual Meeting of the American Psychiatric Association; May 2001; New Orleans, La.
15.    Connor KM, Sutherland SM, Tupler LA, Malik ML, Davidson JR. Fluoxetine in post-traumatic stress disorder. Randomised, double-blind study. Brit J Psychiatry. 1999;175:17-22.
16.    van der Kolk BA, Dreyfuss D, Michaels M, et al. Fluoxetine in posttraumatic stress disorder. J Clin Psychiatry. 1994;55:517-522.
17.    Nagy LM, Southwick SM, Charney DS. Placebo-controlled trial of fluoxetine in PTSD. Poster presented at: International Society for Traumatic Stress Studies 12th Annual Meeting; November 11, 1996; San Francisco, Calif.
18.    Hertzberg MA, Feldman ME, Beckham JC, Kudler HS, Davidson JR. Lack of efficacy for fluoxetine in PTSD: a placebo controlled trial in combat veterans. Ann Clin Psychiatry. 2000;12:101-105.
19.    Physicians’ Desk Reference. 55th ed. Montvale, NJ: Medical Economics Company, Inc; 2001.
20.    Shestatzky M, Greenberg D, Lerer B. A controlled trial of phenelzine in posttraumatic stress disorder. Psychiatric Res. 1988;24:149-155.
21.    Braun P, Greenberg D, Dasberg H, Lerer B. Core symptoms of posttraumatic stress disorder unimproved by alprazolam treatment. J Clin Psychiatry. 1990;51:236-238.
22.    Kaplan Z, Amir M, Swartz M, Levine J. Inositol treatment of post-traumatic stress disorder. Anxiety. 1996;2:51-52.
23.    Yehuda R, Marshall RM, Wong CM, Penkower A. Psychopharmacological treatment of PTSD. In: Nathan PE, Gorman JM, eds. A Guide to Treatments that Work. 2nd ed. Oxford, England: Oxford University Press. In press.
24.    Kessler RC, Sonnega A, Bromet E, Hughes M, Nelson CB. Posttraumatic stress disorder in the National Comorbidity Survey. Arch Gen Psychiatry. 1995;52:1048-1060.
25.    Londborg PD, Hegel MT, Goldstein S, et al. Setraline treatment of posttraumatic stress disorder: results of 24 weeks of open-label continuation treatment. J Clin Psychiatry. 2001;62:325-331.
26.    Foa EB, Keane TM, Friedman MJ. Effective Treatment for PTSD: Practice Guidelines From the International Society for Traumatic Stress Studies. Gilford Press: New York, NY; 2000.
27.    Doyle B, Foa EB, Keane TM, Marshall RD. Posttraumatic stress disorder: changing the perspective of primarys care. Med Crossfire. 2001; 33:33-43.
28.    Johnson JH, McCutcheon S. Assessing life stress in older children and adolescents: preliminary findings with the Life Events Checklist. In: Sarason IG, Spielberger CD, eds. Stress and Anxiety. Washington, DC: Hemisphere; 1980:111-125.
29.    Connor KM, Davidson JR. SPRINT: a brief global assessment of post-traumatic stress disorder. Int Clin Psychopharmacol. 2001;5:279-284.