Post-traumatic stress disorder (PTSD) is an increasingly recognized and potentially preventable condition. Certain factors, especially the severity of the trauma, perceived lack of social support and peri-traumatic dissociation have been associated with its development. In recent years, a more robust evidence base regarding the management of individuals involved in traumatic events has emerged. Immediately after a traumatic event, simple practical, pragmatic support provided in a sympathetic manner by non-mental health professionals seems most likely to help. For individuals who develop persisting PTSD, trauma-focused cognitive behavioural therapy (TFCBT) may be beneficial within a few months of the trauma. For those who develop chronic PTSD, TFCBT and eye movement desensitization and reprocessing are best supported by the current evidence. Some anti-depressants appear to have a modest beneficial effect and are recommended as a second-line treatment. The current evidence base has allowed the development of guidelines that now require implementation. This has major implications in terms of planning and developing services that allow appropriately qualified and trained individuals to be available to cater adequately for the needs of survivors of traumatic events.
Despite many accounts of apparent post-traumatic stress disorder (PTSD) over the last few centuries (e.g. [1,2]), it was only formally recognized as a psychiatric disorder in the third edition of the Diagnostic and Statistical Manual of Mental Disorders . The criteria for PTSD have been refined in subsequent editions and it was first included in the 10th edition of the International Classification of Diseases . Its characteristic features are displayed in Box 1. In order to satisfy the DSM IV criteria , an individual has to be exposed to a traumatic event that involves actual or threatened death or serious injury, or a threat to the physical integrity of self or others. It is also essential that the individual experience a response at the time that involves intense fear, helplessness or horror. The symptoms must have been present for at least 1 month (the 1 month does not apply in the ICD10 classification) and cause clinically significant distress or impairment in social, occupational or other important areas of functioning. Acute stress disorder (ASD) occurs within 1 month of a traumatic event and has similar symptom criteria to PTSD but with more emphasis on dissociation. Acute PTSD becomes chronic if it continues beyond 3 months. Symptoms usually begin shortly after the trauma but are said to have delayed onset if they commence at least 6 months later.
Re-experiencing Phenomena (at least one required)
Recurrent and intrusive distressing recollections
Recurrent distressing dreams
Acting or feeling as if events recurring
Intense psychological distress to cues
Physiological reactivity to cues
Avoidance and numbing (at least three required)
Avoidance of thoughts, feelings and conversations
Avoidance of reminders
Markedly diminished interest in significant activities
Detachment or estrangement feelings
Restricted range of affect
Sense of a foreshortened future
Increased arousal (at least two required)
Irritability or outbursts of anger
Exaggerated startle response
Prevalence and course
The United States National Co-Morbidity survey  found that of 5877, 15–54 year olds just >60% of males and just >50% of females have been exposed to traumatic events with lifetime prevalence of PTSD of just >10% for females and 5% for males. Over a third of individuals reported having PTSD 6 years after they developed it. There was a 50% chance of remission at 2 years. In the recently published replication of this work [7,8], the lifetime prevalence was 6.8% and the 12-month prevalence was 3.5%. PTSD is often co-morbid with other psychiatric disorders. Over 80% of participants with lifetime PTSD in the United States National Co-Morbidity survey  had at least one other disorder. The commonest co-morbid diagnoses are major depressive disorder, panic disorder, other anxiety disorder and substance abuse or dependence. In the National Vietnam Veterans readjustment survey , there was a 99% lifetime co-morbidity.
Immediate reactions following trauma
Exposure to a traumatic event can result in a wide range of reactions. The focus has often been on pathological reactions and the development of psychiatric disorders such as PTSD but more recently resilience has been increasingly recognized as a common response (e.g. ). The majority of distressing reactions settle over a matter of weeks or months with a minority developing into a diagnosable psychiatric disorder, such as PTSD but also others including depression, anxiety disorders and substance misuse. Higher impact trauma is more likely to precipitate a distressing response. Rothbaum and Foa  found that >90% of female rape victims satisfied the symptom criteria for PTSD within a week of the event and 40% at 6 months. Following the September 11th terrorist attacks in New York, Galea et al.  found that probable PTSD reduced from 7.5% after 1 month to 0.6% 6 months afterwards among over a 1000 residents of New York. The rates were highest among those who lived closest to the site of the attacks. This recovery trajectory is all important in considering how best to provide for individuals following traumatic events.
Predicting the development of PTSD
It is difficult to predict exactly who will go on to develop PTSD after a traumatic event. Two large systematic reviews [13,14] have found relatively weak but positive associations of PTSD with the factors shown in Box 2. Those most associated with PTSD were perceived lack of social support and peri-traumatic dissociation, although even these had an effect size of <0.5. Other possible predictors such as increased heart rate after trauma have been shown to be associated with the development of PTSD but are not very discriminating (e.g. ). The possibility of detecting individuals who will go on to develop PTSD has led to attempts to predictively screen shortly after a traumatic event. Several screening instruments for chronic PTSD have been developed (see  for review). The 10-item Trauma Screening Questionnaire (TSQ)  is one of the best validated. Walters et al.  considered the TSQ as a predictive screening instrument with victims of violent crime 1–3 weeks after the assault. Very high rates of sensitivity (0.85) and specificity (0.89) were found but a much lower positive predictive value (0.48) meaning that although it detected the vast majority of PTSD sufferers at 1 month, 50% of those who screened positive did not develop PTSD. It remains to be seen whether a predictive screening instrument can be developed that will be practical and acceptable in the future.
Pre-traumatic risk factors
Previous psychiatric disorder
Gender (female greater than male)
Personality (external locus of control greater than internal locus of control)
Lower socioeconomic status
Lack of education
Race (minority status)
Family psychiatric history
Peri-traumatic risk factors
Perceived life threat
Post-traumatic risk factors
Perceived lack of social support
Subsequent life stress
Prevention of PTSD
Several systematic reviews of early interventions following traumatic events have been published in the last few years (e.g. [19,20]). Two main approaches have been considered—interventions for everyone involved and interventions targeted at individuals who remain symptomatic after a certain amount of time.
Interventions for everyone
Single session psychological interventions
Twelve randomized controlled trials of single session psychological interventions have been published, most commonly variants of critical incident stress debriefing . Meta-analysis provides no evidence of a positive overall effect. Some studies have raised the possibility that single session individual debriefing may cause harm to some individuals [22,23]. The only study of group debriefing  gave neutral results. A recently published dismantling study showed that individuals who received an ‘emotional debriefing’ fared worse than those who received a debriefing that did not contain the emotional components but focused on education, particularly in individuals who were more markedly hyper aroused . A trend for individuals who were more traumatized to do worse with debriefing has been found in other studies and two of the debriefing studies that produced positive outcomes excluded more traumatized individuals [26,27].
Early pharmacological interventions
There have been three randomized controlled trials of early pharmacological interventions following traumatic events. Schelling et al.  found that early administration of intravenous hydrocortisone in 20 septic shock victims on an intensive care unit in Switzerland resulted in lower rates of PTSD at 31-month follow-up compared to placebo. These results are particularly interesting given the suggestion that low cortisol levels may mediate the development of PTSD. Propranolol  and temazepam  did not reduce PTSD development.
Given the disappointing results of one-off interventions, several researchers have advocated using more complex interventions for individuals who develop more symptoms. The most researched interventions have been trauma-focused cognitive behavioural therapy (TFCBT) occurring >4–12 sessions containing components such as education, relaxation, imaginal exposure, in vivo exposure and cognitive restructuring. Overall, the studies have produced positive results for sufferers of ASD [31,32], acute PTSD  and symptoms of PTSD .
Clinical implications for prevention
At present there is no convincing evidence for any intervention for everyone involved in a traumatic event. There is emerging evidence for TFCBT provided 1–3 months following the trauma to individuals who are symptomatic. This evidence resulted in the UK's National Institute for Health and Clinical Excellence's guidelines (NICE)  recommending that immediate practical, social and emotional support are offered to individuals following traumatic events but that individuals are not debriefed. The guidelines state that acute phase symptomatic pharmacological management could be considered using hypnotics or anti-depressants, for example for marked insomnia. They also recommend that TFCBT be offered to individuals within 1 month if they are suffering from severe symptoms of PTSD or within 3 months if they are suffering from acute PTSD, and that the sessions should normally be over 8–12 sessions, with some sessions of 90 min duration.
Management of chronic PTSD
Most chronic PTSD sufferers presenting for treatment will receive medication . Many will also receive some form of psychological treatment, although there are often long waits for such treatment.
Evidence for psychological treatment
Various psychological treatments have been tested in randomized controlled trials. The NICE guidelines  separated the treatments into individual TFCBT, group CBT, stress management, eye movement desensitization and reprocessing (EMDR) and other therapies which included supportive therapy, non-directive counselling, psychodynamic therapies and hypnotherapy. TFCBT and EMDR fared better than the other therapies. Fewer individuals were included in the EMDR studies but the studies that directly compared the two found no evidence that one of these trauma-focused therapies was better than the other. Stress management fared slightly worse than the trauma-focused therapies with the other therapies and group cognitive behavioural therapy faring worse still. There was a trend for all therapies to perform better than a waiting list control but only TFCBT and EMDR reached a pre-determined threshold set by the NICE guideline development group as being clinically significant.
Evidence for pharmacological treatment
Various drug treatments have been tested in randomized controlled trials with more chronic PTSD sufferers entered into randomized controlled drug trials than psychological ones. Paroxetine, mirtazapine, amitryptyline and phenelzine all fared statistically significantly better than placebo, although the last three only included small sample sizes. No drug reached a pre-determined threshold set by the NICE guideline development group as being clinically significant. The other drugs were not statistically significantly better than placebo (sertraline, fluoxetine, imipramine, venlafaxine and olanzapine). However, there was some evidence that olanzapine, if added to an anti-depressant, was better than adding a placebo to augment treatment in chronic PTSD sufferers who had not fully responded to anti-depressant medication alone.
Clinical implications for the management of chronic PTSD
The NICE guidelines recommended that all chronic PTSD sufferers should be offered a course of TFCBT or EMDR, normally on an individual outpatient basis regardless of time since trauma. Again, they recommended 8–12 sessions with some at 90 min if the trauma is considered during the session. The guidelines also acknowledged that the number of sessions may need to be increased, particularly following multiple traumatic events and if there was co-morbidity or traumatic bereavement. If individuals did not improve or showed only little improvement, they were advised to consider an alternative trauma-focused treatment or augmentation with pharmacological treatment. Drug treatments were not recommended as routine first-line treatment. A limited role for paroxetine and mirtazapine was suggested if prescribed by non-specialists, with amytriptyline and phenelzine being prescribed by mental health specialists. Other issues that may precipitate prescription of medication include patient choice and serious ongoing threat. In reality, medication is often prescribed as a result of lack of immediate availability of psychological treatment but caution is required, not least because of the well-documented issues with paroxetine in recent years.
It is encouraging that there now exist effective trauma-focused psychological treatments for chronic PTSD and promising trauma-focused psychological treatments for acute PTSD. Unfortunately, no existing treatment is ideal and there is clearly a great need to develop more effective and more tolerable treatments for PTSD. The results of early interventions directed at everybody have been very disappointing and should encourage anybody involved in providing a response following a traumatic event to exercise extreme caution before providing a formal intervention. Simple practical, pragmatic support provided in a sympathetic manner by non-mental health professionals seems most likely to be the best first-line response but needs better evaluation. Lessons provided by the existing research should be heeded and inform our approach in the future. There is emerging evidence that targeted trauma-focused interventions are effective for individuals with PTSD symptoms within a few months of the trauma. This approach should be refined, as should the detection of symptomatic individuals. Indeed, as hoped with the trauma risk management model , the optimal way of detecting and treating most people may be to educate those who are most likely to be in contact with them about the recognition of problematic responses such as friends, families, work colleagues, managers, general practitioners and occupational health practitioners.
Conflicts of interest