Keywords

1 Introduction

Classifying mental disorders, including traumatic stress disorders, in psychiatry has often been difficult because of the need to discriminate between normal and abnormal states. This situation is particularly difficult in the context of posttraumatic presentations because stress responses are common, and it raises issues of where a line should be drawn between normative reaction and disorder. This chapter reviews the current status of diagnostic systems for describing posttraumatic stress conditions. There has been much activity in this space in recent years because the major diagnostic systems have been undergoing significant reviews and modifications. In this context, this chapter reviews the current status of the major conditions, including posttraumatic stress disorder (PTSD), complex PTSD, acute stress disorder (ASD), acute stress reaction (ASR) and prolonged grief disorder.

2 History of DSM

The American Psychiatric Association’s diagnostic recognition of stress-related conditions can be traced back to the origins of the Diagnostic and Statistical Manual for Mental Disorders (DSM). In the initial iteration, DSM-I (American Psychiatric Association 1952) identified ‘gross stress reactions’, which was a loosely defined classification aimed to describe those affected by traumatic exposure. Arguably influenced by military conceptualizations that stress reactions were typically transient, this conceptualization was based on the premise that these reactions were temporary. In DSM-II (American Psychiatric Association 1968), this diagnosis was removed and replaced by ‘situational reaction’, which described reactions to the full range of severe and mild aversive experiences. The first significant recognition of posttraumatic stress reactions came in 1980 with the publication of DSM-III (American Psychiatric Association 1980). Strongly influenced by the need to formally recognize the mental health needs of Vietnam veterans, this diagnosis encompassed 17 symptoms that fell into three clusters: re-experiencing, avoidance and hyperarousal. This formulation remained for many years and set the framework by which PTSD has been understood since 1980. In DSM-IV (American Psychiatric Association 1994), it underwent minor revisions but essentially kept to the same formula as set by DSM-III. DSM-IV defined PTSD as having been exposed to or witnessing a severely threatening experience and responding with fear, horror or helplessness. This was the gatekeeper to the diagnosis because only if these experiences were present could one then consider the re-experiencing, avoidance and arousal symptoms.

3 History of ICD

The World Health Organization has traditionally recognized stress-related conditions in its International Classification of Diseases (ICD). ICD-8 described ‘transient situational disturbance’ that was a broad category that comprised adjustment problems, severe stress reactions and combat neurosis (World Health Organization 1965). In the next revision (ICD-9), acute stress reaction (ASR) and adjustment reaction (AR) were introduced, and another two were noted in ICD-10 (World Health Organization 1994): posttraumatic stress disorder (PTSD) and enduring personality change after catastrophic experiences (EPACE). The latter two disorders marked important changes from prior diagnoses, which had been conceptualized as transient reactions which normally subside after a period of time had elapsed since the trauma. It is worth noting that the ICD approach has often been influenced by military psychiatry and so an emphasis was placed on the temporary nature of stress reactions. It is also worth noting an important difference between ICD and DSM in terms of their missions. Whereas DSM is understandably focused on US health-care agendas, ICD is more globally focused and aims to address the mental health needs of people across the rest of the world. This focus has resulted in ICD diagnoses being more attuned to the needs of low-resource settings and those affected by conflict, disaster and war. Accordingly, an explicit goal of ICD has been to place the emphasis on practical applications, which includes having diagnoses that are (a) consistent with clinicians’ usual classifications, (b) simple diagnoses with minimal symptoms and (c) useful to allow distinctive decisions about treatment between conditions (Reed 2010).

4 Classification in DSM-5

There were a number of core changes in DSM-5. One of the fundamental shifts was the location of trauma-related disorders in DSM. Traditionally, PTSD and ASD were classified with anxiety disorders because of the common phenomenology and presumed mechanisms. Leading up to DSM-5, there was considerable debate about creation of a fear circuitry section that would comprise PTSD, ASD, panic disorder, agoraphobia, social phobia and specific phobia (Andrews et al. 2009). This proposal rests on the notion that there is a common aetiology and neural circuitry underpinning these disorders. Building on fear conditioning models, it was proposed that these disorders commence when stimuli are paired with an inherently aversive event; subsequent exposure to the conditioned stimuli signals threat and results in anxiety (Milad et al. 2006). Although PTSD is the classic example of a disorder commencing after a conditioned aversive experience, there is also evidence that aversive experiences can precede onset of panic disorder (Faravelli 1985; Manfro et al. 1996) and social phobia (McCabe et al. 2003). In terms of neural circuitry, fear circuitry disorders tend to be characterized by excessive amygdala reactivity and, to a lesser extent, impaired regulation of that response by the medial prefrontal cortex (Rauch and Drevets 2009; Shin and Liberzon 2010), whereas different neural networks appear to be involved in non-fear circuitry anxiety disorders (Cannistraro et al. 2004; Rauch et al. 2007). This is supported by evidence that following trauma, fear circuitry disorders are characterized by elevated heart rate but non-fear circuitry disorders are not (Bryant et al. 2011a). Despite this overlap between PTSD and other fear circuitry disorders, other arguments were put forward to challenge the view that trauma-related disorders should be understood as anxiety disorders. First, the evidence that aversive experiences precipitate most fear circuitry disorders is mixed (Rapee et al. 1990, 2009). Second, many symptoms of PTSD can be found in other disorders; numbing, withdrawal and disinterest are common in depression (Blanchard and Penk 1998). Third, fear conditioning models cannot readily explain the guilt, anger and shame that often characterize PTSD, and so it is argued that this weakens the argument that PTSD is exclusively a fear circuitry disorder (Horowitz 2007). On this basis, the decision was made to include PTSD, ASD, adjustment disorder and dissociative disorders into a category of trauma and stressor-related disorders. The decision to not conceptualize PTSD as an anxiety disorder has been controversial, especially considering that the treatments for PTSD overlap very strongly with those for other fear circuitry disorders.

5 PTSD

5.1 DSM-5

A number of reasonably significant changes were introduced in the DSM-5 definition of PTSD (see Table 6.1). The major change to the entry point to the diagnosis was that the subjective aspect of the stressor (A2: ‘fear, horror or helplessness’) was removed. This had been initially introduced, in part, to ensure that minor reactions to events would not qualify for a PTSD diagnosis (Friedman et al. 2011). Studies indicate that this qualification to the stressor definition is poorly predictive of PTSD (Brewin et al. 2000) and that some people who would otherwise meet criteria for PTSD were excluded from the diagnosis (O’Donnell et al. 2010; Rizvi et al. 2008).

Table 6.1 Posttraumatic stress disorder definitions in DSM-5 and proposed for ICD-11
Full size table

DSM-5 has few changes to the re-experiencing cluster. In contrast, the avoidance conceptualization has been markedly altered. Whereas DSM-IV presumed PTSD comprised three factors, multiple factor analytic studies have indicated that the construct is better explained by four factors: re-experiencing, active avoidance, passive avoidance (including numbing) and arousal (Asmundson et al. 2000; King et al. 1998; Marshall 2004). Accordingly, DSM-5 now has a separate cluster that requires the person to satisfy at least one of two active avoidance symptoms (of either internal or external reminders). The major change has been the addition of a new cluster, termed negative alterations in cognitions and mood. This cluster recognizes that numbing is distinct from active avoidance, but also notes the importance of exaggerated negative appraisals about the trauma and the range of emotional responses that can be experienced in PTSD. This has led to the addition of new symptoms. On the basis that many people with PTSD blame themselves and feel guilty (Feiring and Cleland 2007), self-blame has been added to this new cluster. Given the abundant evidence that people with PTSD have negative evaluations about themselves and the world (e.g. ‘I am a bad person’) and that they will not enjoy positive future experiences (‘Nothing will ever work for me’) (Ehring et al. 2008), the DSM-IV symptom of foreshortened future has been replaced by a symptom that involves exaggerated negative appraisals about oneself and the world. Evidence that PTSD can also exist in association with diverse negative mood states, including anger, shame and guilt (Leskela et al. 2002; Orth and Wieland 2006), led to the inclusion of a symptom of pervasive negative mood states. The arousal cluster has remained largely the same in DSM-5 as it was in DSM-IV, with a few exceptions. Based on evidence that reckless or self-destructive behaviour has been observed in a range of PTSD populations (Fear et al. 2008), this has been added as an additional symptom to the arousal cluster. The only further modification to this cluster was altering irritable mood to aggressive behaviour because this is seen as more indicative of PTSD (Jakupcak et al. 2007).

What is the impact of the altered PTSD definition in DSM-5? One study of traumatic injury survivors found comparable rates of PTSD across both DSM-5 (6.7 %) and DSM-IV (5.9 %) definitions (O’Donnell et al. 2014). Further, this study found that comorbidity with depression was comparable across both DSM-5 and DSM-IV definitions (67 % vs. 69 %). One interesting outcome of the DSM-5 modifications is that it has greatly expanded the possible number of permutations by which PTSD can now be diagnosed; whereas in DSM-IV there were 79,794 possible combinations, the added cluster and the new symptoms in DSM-5 have resulted in 636,120 possible clinical presentations of PTSD (Galatzer-Levy and Bryant 2013). It is premature to cast judgement on how the DSM-5 definition is faring relative to the DSM-IV iteration of the condition because it will require multiple studies conducted in different settings to answer this question.

5.2 ICD-11

As noted above, ICD-11, which is expected to be published in 2017, proposes a considerably simpler definition than DSM-5 – and this is exemplified in the proposed definition of PTSD (see Table 6.1). It has been noted that PTSD was more readily diagnosed in ICD-10 than DSM-IV and that ICD-10 required an impairment requirement to raise the threshold for diagnostic criterion (Peters et al. 1999). ICD-11 is also introducing a formal stressor criterion to tighten the entry for the diagnosis (Maercker et al. 2013a). Arguably the biggest difference between DSM-5 and ICD-11 is the latter’s emphasis on re-experiencing symptoms. In an attempt to reduce comorbidity and focus PTSD on its core element (i.e. a memory-based disorder characterized by reliving of the traumatic experience), considerable weight was placed on the role of the distinctive types of memory for the trauma evident in PTSD (Maercker et al. 2013b). Specifically, whereas intrusive memories are evident across many disorders, the sense of reliving of a trauma is apparently distinctive to PTSD (Brewin et al. 2010; Bryant et al. 2011c). Accordingly, ICD-11 defines re-experiencing the traumatic event(s) in the present, reflected by either vivid intrusive memories, flashbacks or nightmares, accompanied by fear or horror; in this definition, flashbacks can range from transient experiences to a complete disconnection from one’s current state of awareness (Maercker et al. 2013b). ICD-11 also stresses avoidance of re-experiencing symptoms, which includes effortful avoiding of internal (e.g. thoughts, emotions) and external (e.g. situations) reminders. The third emphasis is an excessive sense of current threat, which can be reflected in hypervigilance or by exaggerated startle.

Overall, the ICD-11 definition is intended to simplify the diagnosis for clinicians and allow diagnosis to be made on the basis of satisfying two symptoms of each of the three central features of PTSD. This definition is clearly much simpler than the DSM-5 criteria and leads to much fewer potential permutations by which the diagnosis can be made. Some initial evidence has emerged about the relative performances of the DSM-5 and ICD-11 definitions of PTSD. In one study of 510 traumatically injured patients, PTSD current prevalence using DSM-5 criteria was markedly higher than the ICD-11 definition (6.7 % vs. 3.3 %), and ICD-11 tended to have lower comorbidity with depression (O’Donnell et al. 2014).

6 Acute Stress Disorder

DSM-5 and ICD-11 have two very different conceptualizations of acute stress responses, and they do not match onto each other. They are based on different premises, have very different timeframes and consequently are operationally defined in very distinct ways. In fact, ASD only exists in DSM and has never been a diagnosis in ICD, which instead has a construct termed acute stress reaction.

6.1 DSM-5

ASD was first introduced in DSM-IV for two stated reasons: (a) to describe severe acute stress reactions that predated the PTSD diagnosis (which can only be recognized 1 month after trauma exposure) and (b) as a means to identify people who are at high risk for developing subsequent PTSD (Spiegel et al. 1996). In DSM-IV, to meet criteria for ASD, one needed to experience a traumatic event and respond with fear, horror or helplessness (criterion A), and also dissociative (criterion B), re-experiencing (criterion C), avoidance (criterion D) and arousal (criterion E) symptom clusters. Whereas most clusters were similar to those in PTSD, although more loosely defined (Bryant and Harvey 1997), the exception was the dissociative cluster which required at least three of five possible symptoms (emotional numbing, derealization, depersonalization, reduced awareness of surroundings or dissociative amnesia). This emphasis resulted from arguments at the time that dissociative responses were central to posttraumatic response because they impeded emotional processing of the experience, and therefore were predictive of PTSD (Harvey and Bryant 2002).

In preparing the ASD diagnosis for DSM-5, a core question was: How well was ASD predicting PTSD? Longitudinal studies that indexed the relationship between ASD and later PTSD display a convergent pattern. Whereas the majority of individuals with a diagnosis of ASD do subsequently develop PTSD, most people who eventually experience PTSD do not initially display ASD (Bryant 2011). That is, although ASD is performing adequately in terms of most people who meet criteria are high risk for PTSD, it is performing poorly by not identifying most people who are high risk. For this reason, it was decided in DSM-5 that the ASD diagnosis should not be aiming to predict PTSD but rather simply describe severe stress reactions in the initial month (Bryant et al. 2011b). A driving reason for retaining the diagnosis was that a major utility of the ASD diagnosis is that within the US health-care system having a diagnosis can facilitate access to mental health services.

Recognizing that the requirement of dissociative symptoms was arguably too prescriptive in the DSM-IV definition and precluded many distressed people from being identified (Bryant et al. 2008; Dalgleish et al. 2008), the DSM-5 definition was modified such that to meet criteria one needs to satisfy at least 9 out of possible 14 symptoms without regard to any specific clusters (American Psychiatric Association 2013) (see Table 6.2). Although the diagnosis is structured in a way that does not require any specific symptoms or clusters, to meet criteria one nonetheless must display re-experiencing and/or avoidance symptoms. This retains the essential core of ASD as being comparable to PTSD. One study has reported that the DSM-5 (14 %) identifies more distressed people than the DSM-IV (8 %) definition (Bryant et al. in press). Interestingly, this study also reported that the DSM-5 definition also identified more participants who developed PTSD than DSM-IV criteria.

Table 6.2 DSM-5 criteria for acute stress disorder and proposed ICD-11 criteria for acute stress reaction
Full size table

6.2 ICD-11

Acute stress reactions (ASR) have always been conceptualized in ICD as transient responses that are not necessarily psychopathological (Table 6.2). It is a category that is meant to capture the initial distress that is commonly experienced after traumatic exposure, and it was expected that these reactions would subside within a week or soon after the threat has eased (Isserlin et al. 2008). In this way, ASR is qualitatively different from DSM-5’s ASD because it is neither a mental disorder in its own right nor a predictor of subsequent disorder. It is also worth noting that in ICD-11 there is no minimal time in which PTSD can be diagnosed, and so the issue of having a diagnostic ‘gap’ to describe posttraumatic stress responses (which existed in DSM prior to DSM-IV) does not apply to ICD.

In terms of its definition, ASR has never been limited to strict PTSD definitions because it is intended to encompass the broader array of reactions that can occur in the initial aftermath of trauma. Motivated by the need to be applicable to emergency workers, military personnel and disaster agencies who respond initially to trauma, especially large-scale events, the described symptoms are intentionally very broad and non-prescriptive. The symptoms may include shock, sense of confusion, sadness, anxiety, anger, despair, overactivity, stupor and social withdrawal. Underscoring the intent that ASR is not a mental disorder, it is coded as a ‘Z’ code, distinguishing it from mental disorders. ICD-11 proposes that if the symptoms of ASR persist beyond a week, one should consider a diagnosis of adjustment disorder or PTSD.

7 Complex PTSD

Perhaps the most difficult traumatic stress condition to categorize over the past 20 years has been the notion of complex PTSD. Dating back to the early 1990s, the notion of more complicated PTSD responses has been discussed at length, typically in the context of describing the more complex reactions suffered by survivors of prolonged, and often childhood, trauma. It was argued that those who had suffered sustained and severe trauma, such as childhood abuse, torture or domestic violence, can experience marked problems with their sense of identity and organization of emotions (Herman 1992). Termed disorders of extreme distress not otherwise specified (DESNOS), it was never well defined and accordingly not systematically studied.

In more recent years, the field has moved towards the construct of complex PTSD, which has enjoyed a tighter definition. This is a proposed condition that requires the PTSD symptoms noted above but also reflects the impact that trauma can have on systems of self-organization, specifically in affective, self-concept and relational domains. Unlike the PTSD symptoms in which reactions of fear or horror are tied to trauma-related stimuli, these three latter types of disturbances are pervasive and persistent and occur across various contexts and relationships regardless of proximity to traumatic reminders. Specifically, the construct has evolved to comprise three major sets of disturbances in addition to the core PTSD responses: affective regulation, self-construct and interpersonal. These have been identified both from studies of patients (Roth et al. 1997) and expert clinicians (Cloitre et al. 2011). Though not defined by exposure to prolonged trauma, this constellation of reactions is typically associated with very prolonged and severe traumatic experiences (van der Kolk et al. 2005).

7.1 DSM-5

The possibility of introducing complex PTSD in DSM-5 was debated; however, it was rejected. It was decided to not consider complex PTSD as a separate entity because in the DSM-IV field trials, only 8 % of those who displayed DESNOS did not also have PTSD; thus, it was suggested that it could only be considered as a subtype (Friedman et al. 2011). It was argued that it was premature to introduce this subtype because it had not been adequately defined, insufficient data existed to warrant its distinction from other disorders (including Borderline Personality Disorder), and there was no evidence that people with this presentation respond differentially to treatments that work effectively with PTSD (Resick et al. 2012). In contrast, DSM-5 did introduce a dissociative subtype of PTSD which was regarded as a viable alternative to complex PTSD. This subtype builds on evidence of two types of presentation of PTSD: one characterized by elevated arousal and one by blunting/dissociative responses. This division is largely based on some evidence that people who present with dissociative symptoms show less reactivity at both peripheral (Griffin et al. 1997) and neural (Felmingham et al. 2008; Lanius et al. 2012) levels relative to those with non-dissociative symptoms. Although other studies have reported that there is no difference in reactivity in dissociative and non-dissociative presentations of PTSD (Kaufman et al. 2002; Nixon et al. 2005), this subtype was nonetheless introduced into DSM-5 in recognition that it was a valid sub-entity.

7.2 ICD-11

It appears that a different approach is being taken in ICD-11. ICD has a different organizational structure than DSM, and so if accepted, complex PTSD may be a ‘sibling’ disorder to PTSD rather than a subtype. The proposal being put forward for ICD-11 is based on the core PTSD symptoms with the addition of affective, self and relational disturbance (see Table 6.3). Affective disturbances include emotional reactivity, extreme outbursts, self-destructive behaviour and potentially dissociative states. Disturbances in self may include the sense of worthlessness, or of being defeated or diminished. Difficulties in relations often involve deficits in maintaining a sense of intimacy with others, disinterest in social relations or oscillating between intimate relations and estrangement. Initial evidence supporting this proposal comes from a latent profile analysis that showed patients with affective, self and relational disturbances comprised a distinct class from PTSD patients who were low on these symptoms; further, the former class were more likely to have suffered chronic rather than discrete traumas (Cloitre et al. 2013). Further evidence for the complex PTSD construct has come from other studies that have found supporting confirmatory factor analyses of the proposed structure, and higher rates of the proposed symptoms in survivors of childhood abuse (Knefel and Lueger-Schuster 2013), as well as from other studies using latent class analysis indicating a distinct class of complex PTSD (Elkit et al. 2014). Whether complex PTSD is introduced into ICD-11 is yet to be determined, and if it is, what form it will ultimately take remains to be seen. At this stage, it is gathering more support than it received in DSM-5, arguably because the several years delay in production of DSM-5 and ICD-11 means the latter can benefit from targeted research that is developing a broader evidence base to influence the final decision.

Table 6.3 Proposed ICD-11 criteria for complex PTSD
Full size table

8 Prolonged Grief Disorder

8.1 DSM-5

One of the vehemently debated diagnoses in DSM-5 was the issue of introducing a diagnosis describing complicated grief reactions. DSM has traditionally not recognized grief as a mental disorder because it is concerned about pathologizing a normal response to bereavement. Much work has focused on the condition over the past decade, resulting in a much greater body of evidence than previously existed. Although most normative grief reactions subside after the initial period of mourning, this condition refers to the proportion of grief reactions that persist. There are mixed opinions about the optimal term for the condition. Whereas some prefer the term ‘complicated grief’ to reflect the fact that the symptoms are qualitatively different from normal grief reactions (Shear et al. 2011), others support the term ‘prolonged grief’ in recognition of the condition being a persistence of the same symptoms observed in the acute bereavement phase (Prigerson et al. 2009). Although there is disagreement about the finer details, the generally accepted definition involves intense yearning or emotional pain that persists beyond 6 months after the death and potentially having difficulty accepting the death, anger over the loss, a diminished sense of one’s identity, feeling that life is empty and problems in engaging in new relationships or activities (Bryant 2012). Studies estimate that 10–15 % of bereaved people may suffer this condition, depending on the nature of the death and the relationship to the person (Shear et al. 2011). It was finally decided to not introduce the diagnosis on the basis that insufficient evidence exists to warrant its introduction as a separate diagnosis, instead relegating it to the Appendix as an area for future study.

8.2 ICD-11

Again, in contrast to DSM-5 a different approach appears to be taken in ICD-11. Termed prolonged grief disorder, it has been proposed that a new diagnosis be introduced to recognize the disabling nature of severe and persistent grief reactions (see Table 6.4). This disorder would be defined as severe and persisting yearning for the deceased or a persistent preoccupation with the deceased; this reaction may be compounded by difficulty accepting the death, feelings of loss of a part of oneself, anger about the loss, guilt or blame regarding the death or difficulty in engaging with new social or other activities. The diagnosis can only be made if the symptoms are impairing functioning and persist for over 6 months since the death (Maercker et al. 2013b). The evidence put forward to support this initiative includes multiple studies attesting to persistent yearning being central to the condition in adults (Simon et al. 2011) and children (Melhem et al. 2011). The disorder is distinct from anxiety and depression (Boelen and van den Bout 2005; Golden and Dalgleish 2010), and it contributes to a range of psychological, behavioural, medical and functional problems (Boelen and Prigerson 2007; Bonanno et al. 2007; Simon et al. 2007). From an ICD-11 perspective, it is important that these patterns have been observed across western and non-western cultures (Fujisawa et al. 2010; Morina et al. 2010). Further support for the ICD-11 diagnosis comes from evidence that targeted treatments for prolonged grief are effective relative to those that have shown efficacy for depression (Shear et al. 2005).

Table 6.4 Proposed ICD-11 criteria for prolonged grief disorder
Full size table

Once again, it remains to be seen whether ICD-11 introduces prolonged grief disorder as a new diagnosis. The decision by DSM-5 to not formally recognize this condition as a disorder raises the possibility that many bereaved patients who suffer ongoing and disabling distress may be misdiagnosed with depression or anxiety, and not directed to the optimal intervention. Of course, the ICD-11 does bring a degree of risk of overdiagnosis of grief responses because some grief responses may take longer than 6 months to resolve (Wakefield 2013). ICD-11 will emphasize that the diagnosis can only be made when the response is beyond what is culturally normative because it is sensitive to the problems of inappropriately prescribing a diagnosis across cultural contexts.

9 Summary

Posttraumatic psychiatric diagnoses have evolved significantly since they were introduced decades ago. Interesting distinctions are developing between the two major systems currently available. Whereas DSM is moving towards more complex and broader conceptualizations of PTSD, ICD in contrast is shifting towards a simpler and more focused definition. Beyond PTSD, ICD appears to be adopting a more lateral approach by considering complex PTSD and prolonged grief disorder. This is understandable as the different diagnostic systems have distinct agendas that they are addressing, guided by the respective needs of the American Psychiatric Association and the World Health Organization. With different nomenclatures operating across the world, there is the possibility of confusion and discrepancies in how traumatic stress is operationalized across the world. Time will tell how these respective systems will serve the field, facilitate identification of conditions and lead affected people to appropriate treatments.