Disorders associated with stress versus personality disorders

CPTSD has been included in ICD-11 as one of several diagnoses designated under the general category ‘disorders specifically associated with stress’.³ The CPTSD diagnosis requires ‘exposure to an event or series of events of an extremely threatening or horrific nature, most commonly prolonged or repetitive events from which escape is difficult or impossible. Such events include, but are not limited to, torture, concentration camps, slavery, genocide campaigns and other forms of organized violence, prolonged domestic violence, and repeated childhood sexual or physical abuse’. At a symptom level, CPTSD includes the core PTSD symptoms of (a) re-experiencing of the traumatic event in the present, (b) avoidance of traumatic reminders and (c) persistent perception of heightened current threat and the three symptom clusters of (a) pervasive problems in affect regulation, (b) negative self-concept and (c) relationship difficulties.

Borderline personality disorder has been somewhat reformulated in ICD-11 owing to the introduction of a fundamentally different approach to the classification of personality disorders.³ Instead of diagnosing them according to categorical types, ICD-11 now requires impairments of the self (e.g. identity, self-worth, accuracy of self-view, self-direction) and of interpersonal functioning as core features. A ‘borderline pattern’ specifier has been included, based on the nine DSM-5 diagnostic criteria for BPD, where the salient diagnostic features are instability in sense of self, relationships and affect and the marked presence of impulsivity (e.g. unsafe sex, excessive drinking, reckless driving, uncontrollable eating). These diagnostic features represent domains of problems and symptoms that overlap with some of those found in the identity and relational symptom clusters of CPTSD.

Even though CPTSD is a new diagnosis, it has been used extensively in research and clinical practice for several decades and the overlap between BPD and symptoms of CPTSD has been a subject of debate in recent years. There have been multiple attempts to determine whether CPTSD and BPD diagnoses differ substantially enough to warrant separate diagnostic classifications. Different solutions have been offered in the literature. Some have suggested that CPTSD and BPD are distinct disorders with similar pathways, including neurological^{Reference Amad, Radua, Vaiva, Williams and Fovet5} and anthropogenic.^{Reference Grant, Beck, Marques, Palyo and Clapp6} Others have suggested that CPTSD and BPD symptoms can only co-occur as one syndrome and the two conditions are not distinct.^{Reference Saraiya, Fitzpatrick, Zumberg-Smith, López-Castro, Back and Hien7} Finally, it has been proposed that CPTSD is the by-product of comorbid BPD and post-traumatic stress disorder (PTSD).^{Reference Kulkarni8} Nevertheless, as shown below, there is some emerging evidence suggesting that the two conditions can be distinguished.

The evidence base

To date there have been seven studies exploring the association between BPD and ICD-11 CPTSD using disorder-specific measures. These studies have been conducted in general population samples as well as in clinical samples of traumatised individuals and they include factor analysis,^{Reference Hyland, Karatzias, Shevlin and Cloitre9,Reference Powers, Petri, Sleep, Mekawi, Lathan and Shebuski10} latent class analysis^{Reference Cloitre, Garvert, Weiss, Carlson and Bryant11–Reference Jowett, Karatzias, Shevlin and Albert13} and network analysis^{Reference Knefel, Tran and Lueger-Schuster14,Reference Owczarek, Karatzias, McElroy, Hyland, Cloitre and Kratzer15} designs. All these studies concluded that there is a group of individuals who endorse criteria of both conditions, but CPTSD and BPD were generally found to be distinct disorders.

Differential diagnosis

If the two conditions are distinct, how can we accurately diagnose each of them in clinical practice? As indicated in Table 1, there are several differences in the diagnostic criteria that are clinically informative. Although exposure to traumatic life events can precipitate both conditions, a history of traumatic life events is not required to generate a diagnosis of BPD, whereas it is a prerequisite for consideration of a CPTSD diagnosis. Nevertheless, it is also important to highlight that a significant number of people with BPD report exposure to traumatic life events such as sexual abuse.^{Reference de Aquino Ferreira, Pereira, Benevides and Melo16} Diagnostic items related to affect dysregulation are often equally endorsed across the disorders and in network analyses appear to link CPTSD and BPD.^{Reference Owczarek, Karatzias, McElroy, Hyland, Cloitre and Kratzer15} However, BPD is associated with high rates of impulsive, suicidal and self-injurious behaviours, whereas in CPTSD these characteristics may be present but do not occur as frequently as other CPTSD symptoms nor as markedly or persistently as those seen in BPD.^{Reference Cloitre, Garvert, Weiss, Carlson and Bryant11} Indeed, addressing suicidal and self-injurious behaviours has been viewed as the defining concern and primary treatment target in BPD.^{Reference Linehan, Comtois, Murray, Brown, Gallop and Heard17–Reference DeCou, Comtois and Landes19} Our clinical observations with people with CPTSD suggest that difficulties in affect regulation are ego dystonic (i.e. incompatible with one's beliefs and personality), stressor specific and variable over time. In BPD, affect dysregulation and unstable mood seem to be ego syntonic (i.e. compatible with one's beliefs and personality) and persistent over time.^{Reference Biskin and Paris20} In BPD, self-concept difficulties reflect an unstable sense of self which includes changing goals and beliefs, whereas in CPTSD, individuals have a consistent and stable sense of self. Although it is frequently the case that individuals with either diagnosis will endorse feelings of low self-esteem, the additional endorsement of changing view of self would, in a differential diagnostic effort, rule out the diagnosis of CPTSD and support a BPD diagnosis. Relational difficulties in BPD are characterised by volatile patterns of interactions, whereas in CPTSD they consist of and are driven by difficulties in trusting others and relational avoidance and this is supported in factor analytic and other studies.^{Reference Hyland, Karatzias, Shevlin and Cloitre9,Reference Frost, Murphy, Hyland, Shevlin, Ben-Ezra and Hansen12}

Table 1 Comparison of diagnostic criteria of complex post-traumatic stress disorder (CPTSD) and borderline personality disorder (BPD)

Differential diagnosis may become challenging when assessing an individual with a history of trauma. In such cases, the differential diagnosis may be between CPTSD and BPD comorbid with PTSD (i.e. re-experiencing, avoidance and sense of threat). A diagnosis of BPD comorbid with PTSD versus a diagnosis of CPTSD is likely to be assigned if there is significant presence and continued risk of suicidal or self-injurious behaviours, unstable sense of self or instability in relationships. Below are two brief case vignettes that distinguish CPTSD from BPD with PTSD. It is also possible that a diagnosis of CPTSD can co-occur with BPD, as in the third vignette. Although instability in affect, self-identity and relationships is a salient feature of BPD and a helpful guide in differential diagnosis, it is possible for someone with CPTSD to have a diagnosis of BPD without these features. All three case vignettes were based on composite clinical material from general clinical experience that does not represent any particular clinical case.

Hierarchical diagnosis

An important consideration in diagnosis is to avoid over-pathologising the individual. For example, a symptom that is common to both disorders, such as emotional volatility, should be considered as part of each disorder when summing the totality of symptoms to determine whether the person meets criteria for a specific disorder. However, once a primary diagnosis has been made, the symptom should not be counted twice. The symptom should be counted once and designated to the diagnosis that has been identified as primary. This approach is a ‘hierarchical’ method to diagnosis, where a symptom assigned to the primary diagnosis is not repeated in other secondary diagnoses that are under consideration.

The clinical utility in carefully considering these two diagnoses is primarily as a means by which to guide treatment decisions and provide an intervention that optimises outcomes by addressing the most life-threatening or impairing features associated with each disorder. BPD is likely to be the more severe disorder with the greater impairment, owing to the presence of suicidality and self-injurious behaviours. It is possible but remains to be seen whether recovery from instability in affect, sense of self and relationships requires longer treatment than recovery in someone who has a stable but severely negative self-concept and severe relational avoidance. However, it is likely that the types of intervention needed will differ. There may be some intervention components in mental health programmes for each disorder that overlap, particularly addressing commonly endorsed symptoms (e.g. affect dysregulation).

Future research

We recommend that future research include surveying practitioners about what they find are the benefits and drawbacks of the current classification of CPTSD and BPD. In addition, the development of reliable and valid clinical interviews will further enable diagnostic accuracy of these two conditions. There are currently no validated clinician interview instruments for either ICD-11 CPTSD or personality disorders, but there is some emerging evidence for the usefulness of the International Trauma Interview (ITI) for the assessment of CPTSD (Test Version 3.2 was used, available from the authors, N.R., M.C., J.I.B. and C.R. Brewin, on request).^{Reference Gelezelyte, Roberts, Kvedaraite, Bisson, Brewin and Cloitre21}

Future research is also needed to assess the benefits of providing the same treatment programmes across the different diagnoses. Finally, there is a need to develop tailored treatments informed by the phenomenology and severity of these two conditions. A number of treatments with proven efficacy for PTSD, such as cognitive–behavioural therapy (CBT) and eye movement desensitisation and reprocessing (EMDR), might also be helpful for CPTSD.^{Reference Voorendonk, De Jongh, Rozendaal and Van Minnen22} It is also worth noting that dialectical behaviour therapy (DBT), a treatment that has been extensively used for people with BPD, has been modified and found effective for PTSD and comorbid BPD symptoms (DBT-PTSD^{Reference Bohus, Kleindienst, Hahn, Müller-Engelmann, Ludäscher and Steil23}), BPD with comorbid PTSD^{Reference Kleindienst, Steil, Priebe, Müller-Engelmann, Biermann and Fydrich24} and BPD alone.^{Reference Gillespie, Murphy and Joyce25} A trauma-informed modular approach has also been suggested as a helpful treatment model for CPTSD.^{Reference Karatzias and Cloitre26} The modular approach proposes that symptom clusters of CPTSD be targeted based on the individual's treatment goals and severity of their symptoms. Modular approaches such as Skills Training in Affective and Interpersonal Regulation (STAIR) have been found useful for those who have experienced childhood trauma and they can be a useful intervention for those with CPTSD.^{Reference Cloitre, Stovall-McClough, Nooner, Zorbas, Cherry and Jackson27} There is a need for more research on effective treatments for CPTSD. Perhaps more careful consideration should be given to those who report most or apparently contradictory symptoms of both conditions. Existing evidence suggesting co-occurrence of symptoms for some might reflect false positives as a result of using self-assessment measures. However, considering that traumatic stressors precipitate CPTSD and can be a risk factor for BPD, the two conditions can co-occur in some people and may represent a trauma continuum of psychopathology,^{Reference Hyland, Murphy, Shevlin, Bentall, Karatzias and Ho28} with BPD positioned at the more severe end of the spectrum. For those who present with both conditions, a trauma-informed approach may still be the best treatment option. There is an urgent need to explore the effectiveness of existing and new interventions for ICD-11 CPTSD, the new construct of personality disorder (including the new BPD identifier) and those who endorse symptoms of both conditions.