LEARNING OBJECTIVES

After reading this article you will be able to:

• • understand the relationship between CPTSD and precipitating events

• • distinguish CPTSD from PTSD and borderline personality disorder

• • recognise current issues and practices in psychological treatment.

In July 2018 the World Health Organization (WHO) formally issued ICD-11, the latest revision of the International Classification of Diseases, for consultation with member states (World Health Organization 2018). It contains major changes to the diagnosis of post-traumatic stress disorder (PTSD), which is now replaced by two diagnoses, PTSD and complex PTSD (CPTSD) (Maercker Reference Maercker, Brewin and Bryant2013). Consistent with the ICD-11 principle that diagnoses should be simple and have the maximum clinical utility, requirements for PTSD include evidence of the re-experiencing of traumatic events in the present, deliberate avoidance, a current sense of threat and functional impairment. CPTSD requires the same but, in addition, evidence of disturbances in self-organisation (DSO), consisting of affect dysregulation, negative self-concept and disturbances in relationships. A more detailed rationale for the divergence from PTSD as defined in ICD-10, DSM-IV and DSM-5 has been presented elsewhere (Brewin Reference Brewin2013, Reference Brewin, Cloitre and Hyland2017).

The term ‘complex PTSD’ was first used to describe a syndrome experienced by survivors of repeated, prolonged trauma and involving alterations in affect regulation, consciousness, self-perception and relationships to the perpetrator and to others (Herman Reference Herman1992). Other similar diagnoses have previously been put forward. One is the ICD-10 diagnosis ‘enduring personality change after catastrophic experience’ (EPCACE), which described the disturbances in self-organisation that can sometimes result from multiple, chronic or repeated traumas from which escape is difficult or impossible (e.g. childhood abuse, domestic violence, torture, war, imprisonment) (World Health Organization 1992). Another is ‘disorders of extreme stress not otherwise specified’ (DESNOS), which was included in the Appendix to DSM-IV (American Psychiatric Association 1994). The DESNOS diagnosis has been operationalised using 48 possible symptoms, organised into 6 scales and 27 subscales (Pelcovitz Reference Pelcovitz, van der Kolk and Roth1997). A comparable diagnosis for children is developmental trauma disorder (DTD) (Ford Reference Ford, Spinazzola and van der Kolk2018).

A number of practical difficulties have been identified with these earlier formulations (Resick Reference Resick, Bovin and Calloway2012). These include the large number of candidate symptoms and the substantial symptom overlap with other diagnoses, such as DSM-IV PTSD, major depression and borderline personality disorder (American Psychiatric Association 1994). Evidence has also been lacking that these formulations are uniquely linked to chronic or repeated trauma. There has been uncertainty over whether these formulations represent a complex and severe form of PTSD or a syndrome distinct from (although often comorbid with) PTSD (Ford Reference Ford1999). Relatedly, it has been pointed out that complex PTSD might simply represent one end of a spectrum of severity in post-traumatic reactions, rather than being a qualitatively separate disorder.

ICD-11's CPTSD (Box 1) is related to Herman's (Reference Herman1992) concept of complex PTSD, ICD-10's EPCACE and DSM-IV-TR's DESNOS, and shares with them all an emphasis on enduring changes in self-organisation and the expectation that these changes typically result from exposure to sustained or multiple traumas from which escape is difficult or impossible. In contrast to EPCACE, CPTSD does not describe these symptoms as personality changes and, in contrast to DESNOS, the number of symptoms is relatively small. CPTSD also differs from all these previous formulations in three crucial ways:

• • it requires the three symptom clusters of re-experiencing, avoidance and sense of threat that characterise PTSD

• • it is based on the symptom profile, not on the type of trauma exposure, and

• • functional impairment is explicitly identified as a requirement for the disorder.

This greater definitional clarity resolves a number of the problems that have hindered ‘complex PTSD’ from being recognised as a diagnosis in its own right. First, ICD-11 clearly positions CPTSD as a separate diagnosis from PTSD (a person can be diagnosed with PTSD or CPTSD but not both). At the same time, CPTSD must include the same evidence for re-experiencing in the present, avoidance and sense of threat that is part of PTSD. Thus, it shares with PTSD an explicit focus on specific, identifiable traumatic events that are prominent in consciousness, rather than being a non-specific response to extreme trauma. Finally, chronic or repeated trauma is a risk factor, not a requirement, for CPTSD. There is an acknowledgment that there is no necessary connection with severe trauma, meaning that it can be diagnosed after a single traumatic event (although this will be less likely).

Epidemiology

One of the first studies using the new ICD-11 diagnosis in a nationally representative sample of trauma-exposed adults was conducted in Israel, finding a 1-month prevalence of 9% for ICD-11 PTSD and 2.6% for CPTSD. Women reported a higher rate of PTSD than men but did not differ in the CPTSD rate (Ben-Ezra Reference Ben-Ezra, Karatzias and Hyland2018). A nationwide sample in Germany, in contrast, found a 1-month prevalence of 1.5% for PTSD and 0.5% for CPTSD, with no significant gender differences (Maercker Reference Maercker, Hecker and Augsburger2018). A nationally representative survey in the USA has reported total rates for both diagnoses in between these studies, with PTSD at 3.4% and CPTSD at 3.5% (Cloitre Reference Cloitre, Hyland and Bisson2019); women reported higher rates of both PTSD and CPTSD.

The requirements for PTSD in ICD-10 were less specific than in ICD-11 and did not include evidence of functional impairment. In DSM-5 the PTSD diagnosis is broader, in that it includes many of the symptoms that belong to both ICD-11 PTSD and CPTSD. A substantial number of investigators have now compared rates within the same community or treatment-seeking samples of ICD-10 PTSD, ICD-11 PTSD/CPTSD combined, and PTSD diagnosed using DSM-IV or DSM-5. A review of these studies found that in adult samples the combined ICD-11 PTSD/CPTSD rate is reliably lower than both the ICD-10 and DSM-IV/DSM-5 rates, consistent with the wish to define the disorder more narrowly (Brewin Reference Brewin, Cloitre and Hyland2017).

At present there is only very preliminary evidence to support the existence of CPTSD in children and adolescents, according to a position paper published by the International Society for Traumatic Stress Studies (ISTSS Guidelines Committee 2019). In community samples, however, the few studies available do not suggest any difference in prevalence rates among children and young people between ICD-11 PTSD/CPTSD combined and DSM-IV/DSM-5 PTSD (Brewin Reference Brewin, Cloitre and Hyland2017).

Differential diagnosis

A number of studies have conducted latent profile analysis or latent class analysis to test the assumption that there are different clinical groups corresponding to PTSD and CPTSD. With an occasional exception, these have consistently found one group of patients who report re-experiencing in the present, avoidance and sense of threat, and another group who report elevated levels of these same symptoms but in addition report problems in affect regulation, social relationships and a disturbed sense of self (Brewin Reference Brewin, Cloitre and Hyland2017). Similar findings have been reported for children and adolescents. Factor analytic studies have also consistently found evidence for six clusters of symptoms, with three related to a PTSD higher-order factor and three to a DSO higher-order factor in the expected way (Brewin Reference Brewin, Cloitre and Hyland2017). The cross-cultural validity of the proposals have been tested in a number of countries, including Austria, Denmark, Germany, Israel and Lithuania, and the distinction also appears to be applicable to samples of refugees (Vallières Reference Vallières, Ceannt and Daccache2018; Frost Reference Frost, Hyland and McCarthy2019).

Consistent with the idea that chronic or multiple trauma is a risk factor for CPTSD, studies have shown that childhood physical or sexual abuse, particularly within the family, is more strongly related to CPTSD than PTSD (Cloitre Reference Cloitre, Hyland and Bisson2019). CPTSD is also associated with higher levels of psychiatric burden than PTSD, including greater depression and dissociation (Hyland Reference Hyland, Shevlin and Fyvie2018; Cloitre Reference Cloitre, Hyland and Bisson2019).

Questions have been raised about the potential overlap between CPTSD and other disorders for which prolonged or repeated trauma is thought to be a risk factor, such as borderline personality disorder (BPD). Research is in its early stages but two studies employing latent class analysis on samples reporting a history of childhood abuse (Cloitre Reference Cloitre, Garvert and Weiss2014; Frost Reference Frost, Hyland and Shevlin2018) and one study that used network analysis on an institutionally abused sample (Knefel Reference Knefel, Tran and Lueger-Schuster2016) have suggested that the two disorders can be meaningfully distinguished. The first point to note is that trauma exposure and PTSD symptoms are required for a CPTSD diagnosis but not a BPD diagnosis. Further, the symptoms that are more characteristic of BPD than of CPTSD are being frantic about being abandoned, having an unstable sense of self, having unstable relationships, impulsiveness, and self-harm and suicidal behaviour. The symptoms that are more characteristic of CPTSD than of BPD are an extremely negative sense of self and avoidance of relationships with no significant shifts in identity.

Some individuals with a history of prolonged or repeated trauma may present with comorbid BPD and CPTSD. In such people the utility of the CPTSD diagnosis is primarily to identify active trauma symptoms that are affecting mental state and behaviour, whereas the utility of the BPD diagnosis is to identify that safety considerations are prominent and are likely to become more so if trauma symptoms are confronted directly without proper preparation.

Assessment of CPTSD

ICD-11 PTSD and CPTSD are more tightly defined diagnoses than DSM-5 PTSD, requiring a smaller number of more specific symptoms to be diagnosed. For example, DSM-5 includes five re-experiencing symptoms involving any kind of intrusive memory, as well as any kind of emotional or physiological reaction on encountering reminders of the event (American Psychiatric Association 2013). Because several of these symptoms, including intrusive memories, are found in many other psychiatric disorders, ICD-11 requires that there should be an element of re-experiencing in the present. This involves one of two symptoms: either a nightmare that recapitulates some aspect of the event (but does not have to be an exact replay) or a daytime flashback in which the event is vividly replayed. DSM-5 and ICD-11 now both define flashbacks as existing on a continuum: at one end is total absorption in the traumatic memory, with a complete loss of awareness of the current environment, and at the other is a vivid intrusive memory of the traumatic event in which the person does not lose contact with their current surroundings but has a sense, however fleeting, that the event is happening again in the here and now. This requirement is important to differentiate PTSD from other conditions – such as major depression – in which people have intrusive memories of distressing events but experience them as belonging to the past. If the person has no conscious memory of the event (perhaps because of a head injury or intoxication) ICD-11 allows this criterion to be met by an emotional response to reminders of it.

ICD-11 also requires evidence (again, at least one symptom) of both avoidance and a sense of threat. Although the deliberate avoidance symptoms are the same as in DSM-5, the hyperarousal symptom cluster is replaced in ICD-11 by the more specific construct of a continuing sense of threat despite the event being in the past. This can be manifested either by hypervigilance or an exaggerated startle reaction. The requirement for functional impairment in some important aspect of the person's life also differentiates PTSD in ICD-11 from the equivalent diagnosis in ICD-10.

For CPTSD to be diagnosed the person must also demonstrate pervasive, long-standing disturbances in self-organisation (DSO). DSO consist of three components, and again there must be evidence, in the form of at least one symptom, that all three are simultaneously present. Affective dysregulation may take the form of hyperactivation, the tendency to experience intense emotions that cannot readily be moderated, or of hypoactivation, in which there is an absence of normal feeling states, or of both. Negative self-concept refers to feelings of worthlessness or being a failure, while the ‘disturbances in relationships’ component focuses on detachment and withdrawal from others. These symptoms too must be accompanied by evidence of impairment in important life roles.

Different aspects of CPTSD may be more salient than others, depending on the type of stressor. For example, uncontrollable anger tends to have relatively low endorsement as part of the affect dysregulation cluster among adults with childhood sexual and/or physical abuse (Cloitre Reference Cloitre, Garvert and Weiss2014), but much higher endorsement among those who were exposed to armed conflict and abducted into child soldiering (Murphy Reference Murphy, Elklit and Dokkedahl2016). ICD-11 further recognises that many other symptoms (Box 2) commonly accompany a PTSD presentation, even though they do not discriminate it from other disorders.

Although there are a few instruments that assess different forms of complex PTSD (Pelcovitz Reference Pelcovitz, van der Kolk and Roth1997; Litvin Reference Litvin, Kaminski and Riggs2017), at present only one instrument is available that specifically assesses ICD-11 CPTSD, the International Trauma Questionnaire (Cloitre Reference Cloitre, Shevlin and Brewin2018). The questionnaire is in use in at least 29 countries on six continents. A companion structured interview measure is under development.

Treatment of complex PTSD

An important debate has been taking place over whether complex PTSD requires a different kind of psychological treatment to PTSD. The International Society for Traumatic Stress Studies produced a position paper (ISTSS Guidelines Committee 2019) recommending three stages or phases of treatment, each with a distinct function. Phase 1 focuses on ensuring the individual's safety, reducing symptoms, and increasing important emotional, social and psychological competencies. Phase 2 focuses on processing the unresolved aspects of the individual's memories of traumatic experiences so that these are integrated into an adaptive representation of self, relationships and the world, using standard or slightly adapted methods taken from conventional trauma-focused cognitive–behavioural therapy. Phase 3 involves consolidation of treatment gains to facilitate the transition from treatment into a greater engagement with the outside world.

Currently validated treatments include Skills Training for Affective and Interpersonal Regulation combined with Modified Prolonged Exposure (STAIR/MPE) (Cloitre Reference Cloitre, Stovall-McClough and Nooner2010), which includes a first phase emphasising the acquisition of affective and interpersonal regulation skills followed by a modified version of prolonged exposure to address the traumatic memories. Another alternative that could be considered for Phase 1 is compassion-focused therapy (Karatzias Reference Karatzias, Hyland and Bradley2019a). This addresses the frequent tendency for those with CPTSD to feel intense shame and to be highly self-blaming and self-denigratory, reactions that are risk factors for PTSD and would be expected to make exposure to the traumatic memories too painful to tolerate. Treatment of children and adolescents with a phase-based approach has been found to achieve similar gains in those diagnosed with CPTSD as in those with PTSD, although those with CPTSD started and finished therapy with higher symptom levels (Sachser Reference Sachser, Keller and Goldbeck2017).

However, the need for a phased approach to treatment has been challenged by other experts (de Jongh Reference de Jongh, Resick and Zoellner2016). They pointed to the lack of direct evidence for the superiority of a phase-based approach, as well numerous indications that immediate trauma-focused treatment (i.e. bypassing phase 1) could be effective for many patients with histories of multiple traumatisation, including childhood abuse. For example, an intensive form of trauma-focused treatment that involved twelve 90-minute sessions of prolonged exposure over 4 days, followed by four weekly 90-minute booster sessions, achieved large treatment effects that persisted over 6 months (Hendriks Reference Hendriks, de Kleine and Broekman2018). A recent meta-analysis (Karatzias Reference Karatzias, Murphy and Cloitre2019b) has confirmed that standard treatments for PTSD do reduce CPTSD symptoms of negative self-concept and disturbances in relationships, although little evidence is available for affective dysregulation. The analysis also found that treatment gains were reduced when trauma exposure dated from childhood.

This debate largely preceded the current conceptualisation of CPTSD in ICD-11, which has refocused attention on the presenting symptoms rather than the nature of the trauma. Rigorous comparisons of alternative treatment approaches are lacking but it is unlikely that diagnostic concerns alone will prove to be decisive. Theoretically, the factors that are thought to undermine direct work with trauma memories generally involve cognitive and emotional reactions that prevent the person from holding the most traumatic material in consciousness and keeping a degree of detachment and reflection as they do so (Brewin Reference Brewin, Gregory and Lipton2010a). For example, loss of trust very frequently accompanies CPTSD (Ebert Reference Ebert and Dyck2004) and may impede the formation of a therapeutic relationship strong enough to allow the patient to share critical experiences or even revisit them privately. This is to be expected, given that trauma survivors not infrequently have the experience of being disbelieved or denigrated, or are betrayed by individuals or organisations who have a duty of care towards them. Some other factors that affect CPTSD treatment are discussed further below.

What is important is that, for some patients, a stabilisation phase prior to trauma-focused treatment that directly addresses their traumatic memories will strengthen the therapeutic relationship and prevent drop-out. Consistent with this, the recently updated NICE guideline on PTSD (National Institute for Health and Care Excellence 2018) notes that trauma-focused cognitive–behavioural interventions should normally be provided over 8–12 sessions but may need to be extended for those with more complex presentations. Specific recommendations for such presentations are listed in Box 3. NICE does not give any indication of how much additional time might be needed, but therapeutic experience indicates that, although 20–30 sessions will be sufficient for many, 1–2 years of weekly therapy may be needed for the more complex cases.

Factors affecting the treatment of CPTSD

There are numerous clinical challenges that are regularly encountered in the management of CPTSD and that interfere with psychological treatment even in patients who are well-motivated and engaged. Of these, chronic dissociation and/or voice-hearing are among the most common.