01 - 12 Personality disorders

12 Personality disorders

517 Personality disorders The concept of personality disorder  518 ‘Normal’ personality  520 Classification of personality disorder  522 Psychopathy and ‘severe’ personality disorder  524 Aetiology of personality disorder  526 Epidemiology of personality disorder  528 Relationship between personality disorder and other mental disorders  530 Assessment of personality disorder  532 Management of personality disorder 1: general aspects  534 Management of personality disorder 2: social and pharmacological  536 Management of personality disorder 3: psychotherapy  538 Outcome of personality disorder  540 Chapter 12

518 Chapter 12  Personality disorders The concept of personality disorder Essence Personality describes the innate and enduring characteristics of an individual which shape their attitudes, thoughts, and behaviours in response to situ­ ations. We all recognize, among people we know well, some who manifest certain characteristics more than others: shyness, confidence, anger, gen­ erosity, tendency to display emotions, sensitivity, and being pernickety, to name but a few. When these enduring characteristics of an individual are such as to cause distress or difficulties for themselves or in their relation­ ships with others, then they can be said to be suffering from personality disorder (PD). PD is separate from mental illness, although the two interact. Definition The following definition is based on ICD-​10 and DSM-​5 (both are very similar). PD are enduring (starting in childhood or adolescence and con­ tinuing into adulthood), persistent, and pervasive disorders of inner experi­ ence and behaviour that cause distress or significant impairment in social functioning. PD manifests as problems in cognition (ways of perceiving and thinking about self and others), affect (range, intensity, and appropriate­ ness of emotional response), and behaviour (interpersonal functioning, oc­ cupational and social functioning, and impulse control). To diagnose PD, the manifest abnormalities should not be due to other conditions (such as psychosis, affective disorder, substance misuse, or organic disorder) and should be out of keeping with social and cultural norms. Development of the concept The development of clinical concepts of conditions which would today be recognized as PD started in the early nineteenth century, at a time when the main two groups of mental conditions acknowledged by psychiatrists were insanity and idiocy. It became clear that there were individuals who were neither insane (i.e. suffering from delusions or hallucinations) nor clearly idiots, imbeciles, or morons (to use the then contemporary terminology for ID), but who nevertheless had abnormalities in their behaviour. In 1801, Pinel described non-​psychotic patients with disturbed behaviour and thinking as ‘manie sans délire’, while the term ‘moral insanity’ was intro­ duced by Prichard in 1835. ‘Moral’ then meant ‘psychological’ (rather than the modern meaning concerning ethics), and among the patients described were people who had affective disorders, as well as people who were personality-​disordered. Koch in 1873 described ‘psychopathic inferiority’, making the socially maladaptive nature of the disorder the key to diagnosis. Kraeplin is reported as finding ‘the classification of PD defeating’. Nonetheless, he attempted to find a place for the description of its sub­ types within his evolving classification system. In 1921, he postulated that PDs, as they were then described, were biologically related to the major psychotic and affective illnesses. In 1927, Schneider introduced a classification system which can be seen as a forerunner of the current categorical approaches in DSM-​5 and ICD-​10. He did not use a spectrum concept but saw PD as representing

The concept of personality disorder a pronounced and maladaptive variation of normal personality traits and used social deviance as a diagnostic marker for his ten subtypes. The individual PD subtypes in use today derive from a number of dif­ ferent academic and theoretical backgrounds:  antisocial (dissocial) PD from child psychiatric follow-​up studies; borderline, histrionic, and narcis­ sistic PDs from dynamic theory and psychotherapeutic practice; schizoid and anankastic PDs from European phenomenology; and avoidant PD from academic psychology. Notably absent from the list of academic sources is the psychological study of normal personality, which has developed a trait model of normal personality along a varying number of axes (E Is per­ sonality stable?, p. 521) Despite major moves to significantly revise DSM-​5 to reflect this trait approach, the changes did not make the final version but are included in Section III ‘for further study’. ICD-​11 proposes using a primary dimension of severity (mild, moderate, or severe) and five trait do­ mains: negative affectivity (the tendency to manifest distressing emotions), dissociality (the tendency to disregard social conventions and the rights of others), disinhibition (the tendency to act impulsively), anankastia (the ten­ dency to control one’s own and others’ behaviour), and detachment (the tendency to maintain emotional and interpersonal distance). In this chapter, we hold to PD subtypes—​for the time being. Controversy A frequently repeated criticism of the present clinical concept has been the problem of tautology, i.e. the same features displayed by a patient, which suggest a diagnosis of PD, are then ‘explained’ by the presence of that diag­ nosis. For example, a patient may, among other features, display ‘an incap­ acity to experience guilt’ and ‘a low threshold for discharge of frustration, including violence’. This may lead to an ICD-​10 diagnosis of dissocial PD. It is then illogical to use that same diagnosis to ‘explain’ a subsequent episode of violence without remorse in that individual. Some psychiatrists believe that psychiatry has no role in the treatment of people with PDs. They argue that: personality is, by definition, unchange­ able; there is no evidence that psychiatry helps individuals with PD; these people are disruptive and impinge negatively on the treatment of other patients; these people are not ill and are responsible for their behaviour; and psychiatry is being asked to deal with something that is essentially a social problem. On the other hand, there are those who believe that people with PD clearly fall within the remit of psychiatry, arguing that: people with PD suffer from symptoms related to their disorder; they have high rates of suicide, other forms of premature death, and other mental illnesses; there are treat­ ment approaches which are effective; their opponents are rejecting patients because they dislike them; and the problem is not that these people cannot be helped, but that traditional psychiatric services do not provide the type of approach and services that are necessary.

520 Chapter 12  Personality disorders ‘Normal’ personality Psychologists have sought to conceptualize and describe the variations in normal personality. There are two main approaches: nomothetic and ideo­ graphic. In general, these approaches have developed separately from con­ cepts of abnormal personality and PD. Nomothetic approaches Personality seen in terms of attributes shared by individuals. Two subdi­ visions:  type (or categorical) approaches (discrete categories of person­ ality); and trait (or dimensional) approaches (a limited number of qualities, or traits, account for personality variation). Type approaches dominate the description and classification of PD, but trait approaches are pre-​eminent in modern personality psychology. Type approaches These describe individual personality by similarity to a variable number of predefined archetypes. These may attempt to include all aspects of personality and behaviour—​the ‘broad’ models—​or they may describe one aspect of personality—​the ‘narrow’ models. An example of the former is the humoral model of Hippocrates which described four fundamental personality types (choleric, sanguine, melancholic, and phleg­ matic); an example of the latter is type A vs type B model which describes groups of behaviours exhibited by people at higher and lower risk of car­ diac disease. Trait approaches These view a variable number of traits as continuous scales, along which each person will have a particular position; the posi­ tions on all the traits represent a number of dimensions which describe personality. Examples include:  Eysenck’s three-​factor theory (neuroticism, extraversion, psychoticism); Costa and McCrae’s five-​factor model (neuroti­ cism, extraversion, openness, agreeableness, conscientiousness); Cloninger’s seven-​factor model (novelty-​seeking, harm avoidance, reward dependence, persistence, self-​directedness, cooperativeness, self-​transcendence; ori­ ginally only first three factors); and Cattell’s 16-​factor theory. A consensus has emerged from personality questionnaire research and from lexical ap­ proaches that there are five fundamental traits (the ‘big five’) similar to those of Costa and McCrae. The heritability of personality traits in twin and adoptive studies has been found to be moderately large (about 30%). Ideographic approaches Unlike nomothetic approaches, these emphasize individuality and seek to understand an individual’s personality by understanding that individual and their development, rather than by reference to common factors. Examples are psychoanalytic, humanistic, and cognitive–​behavioural approaches. The first two have little scientific validity, and the last has compromised with trait theorists.

‘Normal’ personality Is personality stable? Are there traits which are persistent and predict a person’s behaviour over time in a number of situations? Situationists have argued that the situation was a stronger determinant of behaviour than personality traits. However, more recent research has demonstrated the long-​term stability of a number of personality traits, and, perhaps unsurprisingly, most now agree that both the situation and personality traits are important in determining behaviour.

522 Chapter 12  Personality disorders Classification of personality disorder It is largely accepted that normal personality is best described and classified in terms of dimensions or traits. Although this also applies to PD, our cur­ rent psychiatric classifications are categorical. The various categories of PD described in ICD-​10 and DSM-​5 have a number of origins: psychodynamic theory, apparent similarities between certain PDs and certain mental illnesses, and descriptions of stereotypical personality types. The various categories used come together in a piecemeal and arbitrary fashion and do not represent any systematic understanding or study of PD. The categorical classification of PD is psychiatric classification at its worst. There are a number of important points to bear in mind when using standard categorical approaches in the diagnosis of PDs: • Due to their heterogenous origins, there is overlap between the criteria for some categories. • It is more common for individuals to meet the criteria for >1 category of PD than to meet only the criteria for a single category. • When making a diagnosis, one should use all the categories for which a person meets the criteria. • If a person meets the criteria for >1 category, then they do not suffer from >1 actual disorder. A person has a personality, and this may or may not be disordered. If it is disordered, it may have various features which are rarely described adequately by a particular category. • Clinically, it is more important to understand and describe the specific features of a person’s personality than it is to assign them to a particular category. • The diagnosis of PD is a particular area where one may believe, wrongly, that one has a better understanding of a person by assigning them to a specific category (an example of ‘tautology’).1 ICD-​10 and DSM-​5 The PD categories in ICD-​10 and DSM-​5 are set out in Table 12.1. The two schemes are similar, but there are categories that appear in one but not the other, and for some categories, different terms are used. Each category has a list of features, a number of which should be present for the person to be diagnosed as manifesting that particular aspect of PD. DSM-​5 has lost the multi-​axial approach of DSM III (and other subsequent editions), and now PD is not diagnosed separately from other mental illnesses or reasons for consulting a psychiatrist (E The ICD-​10 multi-​axial system, p. 1118). 1  Tautology (the restatement of the same information using different words) is a particular danger in psychiatry generally, and the diagnosis of PD in particular. For example, saying that someone has ‘bor­ derline’ traits gives a gloss of understanding to the simple fact that a person repeatedly self-​harms, without actually communicating any new information (except perhaps the ‘therapeutic despair’ of the psychiatrist!).

Classification of personality disorder Table 12.1  ICD-​10 and DSM-​5 classifications of personality disorder ICD-​10 DSM-​5* Description Paranoid Paranoid Sensitive, suspicious, preoccupied with conspiratorial explanations, self-​referential, distrust of others Schizoid Schizoid Emotionally cold, detachment, lack of interest in others, excessive introspection, and fantasy (Schizotypal disorder classified with schizophrenia and related disorders) Schizotypal Interpersonal discomfort with peculiar ideas, perceptions, appearance, and behaviour Dissocial Antisocial Callous lack of concern for others, irresponsibility, irritability, aggression, inability to maintain enduring relationships, disregard and violation of others’ rights, evidence of childhood conduct disorder Emotionally unstable—​impulsive type –​ Inability to control anger or plan with unpredictable affect and behaviour Emotionally unstable—​borderline type Borderline Unclear identity, intense and unstable relationships, unpredictable affect, threats or acts of self-​harm, impulsivity Histrionic Histrionic Self-​dramatization, shallow affect, egocentricity, craving attention and excitement, manipulative behaviour –​ Narcissistic Grandiosity, lack of empathy, need for admiration Anxious (avoidant) Avoidant Tension, self-​consciousness, fear of negative evaluation by others, timid, insecure Anankastic Obsessive–​ compulsive Doubt, indecisiveness, caution, pedantry, rigidity, perfectionism, preoccupation with orderliness and control Dependent Dependent Clinging, submissive, excess need for care, feels helpless when not in relationship

• DSM-​5 uses three broader clusters to organize the categories of PD: cluster A (odd/​ eccentric)–​–​paranoid, schizoid, schizotypal; cluster B (emotional/​dramatic)—​antisocial, histrionic, narcissistic, borderline; and cluster C (fearful/​anxious)—​avoidant, dependent, obsessive–​compulsive. Although this may seem sensible, there is no particular validity to this clustering.

524 Chapter 12  Personality disorders Psychopathy and ‘severe’ personality disorder Psychopathy The terms ’psychopathy’, ‘psychopathic PD’, ‘psychopathic disorder’, and ‘psychopath’ have dominated much of the PD literature until relatively re­ cently. In England and Wales, the 2007 revision to the 1983 MHA has re­ moved ‘psychopathic disorder’ as a subcategory of mental disorder and included it within a single definition of mental disorder. Other jurisdictions have no category or legal diagnosis of psychopathy in their mental health legislation. The term ‘psychopathy’ should probably now be reserved for individuals meeting criteria as defined by the gold-​standard instrument for psychopathy assessment—​the Psychopathy Checklist–​Revised (PCL-​R) (see Table 12.2). Epidemiological studies report that psychopathy occurs in about 0.6% of the general population and in 7.7% of ♂ prisoners in the UK. It may occur in childhood and remain relatively stable throughout ado­ lescence and into adulthood.2 Psychopathy Checklist–​Revised In The Mask of Sanity (1941),3 Cleckley described various features of psychopathy referring to cold, callous, self-​centred, predatory, and para­ sitic individuals. This concept has led to the development of the PCL-​R,4 which measures the extent to which a person manifests the features of this prototypical psychopath. The items of the PCL-​R are listed in Table 12.2. Psychopathy, as defined by the PCL-​R, is strongly correlated with a risk of future violence. It defines a narrower group of individuals than antisocial or dissocial PD, and individuals scoring highly commonly fulfil the criteria for antisocial, narcissistic, histrionic, paranoid, and perhaps borderline cat­ egories in DSM-​5. Severe personality disorder The term ‘severe personality disorder’5 is often used but has no clear meaning or definition. The severity of PD has been defined in various ways: • In terms of severe impact on social functioning. • By using the PCL-​R cut-​off and being synonymous with psychopathy. • By defining severity as the presence of features fulfilling the criteria for multiple categories of DSM-​5 or ICD-​10 PDs (sometimes this is further defined by stating that the categories should be from at least two DSM-​5 clusters, and perhaps that one must be from cluster B). 2  Sarkar S, Clark BS, Deeley Q (2011) Differences between psychopathy and other personality dis­ orders: evidence from neuroimaging. Adv Psychiatr Treat 17:191–​200. 3  Cleckley H (1941) The Mask of Sanity. London: Henry Klimpton. 4  Hare RD (2003) Manual For The Revised Psychopathy Checklist, 2nd edn. Toronto: Multi-​Health Systems (first edition published in 1991). 5  Tyrer P (2004) Getting to grips with severe personality disorder. Crim Behav Ment Hlth 14:1–​4.

Psychopathy and ‘severe’ personality disorder None of these approaches is entirely satisfactory, and each defines dif­ ferent, but overlapping, groups of individuals. ICD-​11 severity specifiers may prove to be useful in this respect (E ICD-​11 proposals vs. DSM-​5, p. 1121). Moral responsibility? The exempting view that psychopaths lack the ability to function as moral agents is more often found in philosophical arguments than in court.6 Most clinicians are more comfortable with the mitigating view, which concedes that any impairment in moral understanding in psychopathy is insufficient to be completely exempting of the consequences of their (criminal) behaviour. Table 12.2  Notes on the PCL-​R Factor 1 Factor 2 Interpersonal •  Glibness—​superficial charm •  Grandiose sense of self-​worth •  Pathological lying •  Conning—​manipulative Lifestyle •  Need for stimulation •  Parasitic lifestyle •  Lack of realistic, long-​term goals •  Impulsivity •  Irresponsibility Affective •  Lack of remorse or guilt •  Shallow affect •  Callous—​lack of empathy •  Failure to accept responsibility Antisocial •  Poor behavioural control •  Early behavioural problems •  Juvenile delinquency •  Revocation of conditional release •  Criminal versatility Additional items: •  Promiscuous sexual behaviour •  Many short-​term marital relationships The 20 items of the PCL-​R fall broadly into two dimensions. Factor 1 items are mostly emotional or interpersonal traits, while Factor 2 items cover the behavioural manifestations of psychopathy. Characteristics from both factors are required for psychopathy to be diagnosed. Each item is rated 0 (absent), 1 (some evidence, but not enough to be clearly present), or 2 (definitely present). Each item has detailed descriptions in the coding manual. The total score (out of 40) gives an indication of the extent to which a person is psychopathic and may be converted into a percentile using reference tables for different populations. In the USA, a score of 30 or above is used as cut-​off to diagnose psychopathy; in the UK, a score of 25 is generally used as the cut-​off score. 6  Ramplin S, Ayob G (2017) Moral responsibility in psychopathy: a clinicophilosophical case discus­ sion. BJPsych Advances 23:187–​95.

526 Chapter 12  Personality disorders Aetiology of personality disorder While there is no single, convincing theory explaining the genesis of PD, the following observations are suggestive of possible contributing factors. Genetic Evidence of heritability of ‘normal’ personality traits; some evidence of her­ itability of cluster B PDs; familial relationship between schizotypal PD and schizophrenia, between paranoid PD and delusional disorder, and between borderline PD and affective disorder. There is no good evidence for a rela­ tionship between the XYY genotype and psychopathy. Neurophysiology ‘Immature’ EEG (posterior temporal slow waves) in psychopathy; functional imaging abnormalities in psychopathy (e.g. d activity in the amygdala during affective processing tasks); low 5-​HT levels in impulsive, violent individuals; autonomic abnormalities in psychopathy (slowed galvanic skin response). Childhood development Difficult infant temperament may proceed to conduct disorder in childhood and PD; ADHD may be a risk factor for later antisocial PD; insecure attach­ ment may predict later PD (particularly disorganized attachment); harsh and inconsistent parenting and family pathology are related to conduct disorder and may therefore be related to later antisocial PD; severe trauma in child­ hood (such as sexual abuse) may be a risk factor for borderline PD and other cluster B disorders. Psychodynamic theories Freudian explanations of arrested development at oral, anal, and genital stages, leading to dependent, obsessional, and histrionic personalities; ‘bor­ derline personality organization’ described by Kernberg (diffuse, unfiltered reaction to experience prevents individuals from putting adversity into per­ spective, leading to repeated crises); narcissistic and borderline personal­ ities seen as displaying primitive defence mechanisms such as splitting and projective identification; some see antisocial personalities as lacking aspects of superego, but a more sophisticated explanation is in terms of a reac­ tion to an overly harsh superego (representing internalization of parental abuse). Cognitive–​behavioural theories There are maladaptive schemata (stable cognitive, affective, and behav­ ioural structures representing specific rules that affect information pro­ cessing). These schemata represent core beliefs which are derived from an interaction between childhood experience and pre-​programmed patterns of behaviour and environmental responses. Schemata are unconditional, compared with those found in affective disorders (e.g. ‘I am unlovable’, ra­ ther than ‘If someone important criticizes me, then I am unlovable’) and are formed early, often pre-​verbally.

Aetiology of personality disorder Theories synthesizing cognitive–​behavioural and psychodynamic aspects The following are two quite similar models that underlie relatively recently introduced therapies for borderline PD. Cognitive–​analytical model (E Cognitive analytic therapy, p.  918) Borderline patients experience a range of partially dissociated ‘self-​states’, which arise initially as a response to unmanageable external threats and are maintained by repeated threats or internal cues (such as memories). Abusive experiences in childhood lead to internalization of the harsh par­ ental object, leading to intrapsychic conflict which is repressed or produces symptomatic behaviours. Deficits in self-​reflection, poor emotional vocabu­ lary, and narrow focus of attention lead to incoherent sense of self and others. Dialectical behavioural model (E Dialectical behaviour therapy, p. 916) Innate temperamental vulnerability interacts with certain dysfunctional (‘invalidating’) environments, leading to problems with emotional regula­ tion. Abnormal behaviours which are manifested represent products of this emotional dysregulation or attempts to regulate intense emotional states by maladaptive problem-​solving.

528 Chapter 12  Personality disorders Epidemiology of personality disorder Measurement of the prevalence of PD of any type and of specific cat­ egories of PD in any population has a number of problems; in earlier studies, PD and other mental disorders were mutually exclusive, not allowing for the recording of comorbidity; studies differ in the method used to make a diagnosis (interviews/​case notes/​informants; clinical diagnosis vs research instruments; emphasis on current presentation or on life history); and in some studies, subjects were only allowed to be­ long to one category of PD.7 Findings regarding PD of any type will be considered separately from find­ ings related to specific PD categories (see Table 12.3). Personality disorder of any type • Community: a weighted prevalence for a diagnosis of any PD was found to be 4.4% in a general population study of British households. Comorbidity within PD was also found to be common—​patients with PD are likely to meet the criteria for >1 subtype of PD.8 It is more prevalent in younger adults and generally more prevalent in ♂. • Primary care: prevalence of PD is around 10–​12%, consisting mainly of patients presenting with depressive and somatizing symptoms. • Psychiatric patients: 33% in general psychiatric outpatients. The prevalence of PD rises to roughly 40% in eating disorder services, and to 60% in substance misuse services.9 • Other populations: 65% of ♂ and 42% of ♀ prisoners have a PD, predominantly antisocial.10 7  Casey P (2000) The epidemiology of personality disorder. In:  Tyrer, P (ed). Personality Disorders: Diagnosis, Management and Course, pp. 71–​9. Oxford: Butterworth Heinemann. 8  Coid J, Yang M, Tyrer P, Roberts A, Ullrich S (2006) Prevalence and correlates of personality disorder in Great Britain. Br J Psychiatry 188:423–​31. 9  Adshead G, Sarkar J (2012) The nature of personality disorder. Adv Psychiatr Treat 18:162–​72. 10  Fazel S, Danesh J (2002) Serious mental disorder in 23000 prisoners: a systematic review of 62 surveys. Lancet 359:545–​50.

Epidemiology of personality disorder Table 12.3  Specific categories of personality disorder DSM Prevalence (%) Paranoid 0.5–​3 Schizoid 0.5–​7 Schizotypal 0.5–​5 Antisocial 2–​3.5 Borderline 1.5–​2 Histrionic 2–​3 Narcissistic 0.5–​1 Avoidant 0.5–​1 Dependent 0.5–​5 Obsessive–​compulsive 1–​2 The prevalence rates of the categories of PD (most studies have used DSM categories, so these are used here) in the general population are approximately as shown in the table.

530 Chapter 12  Personality disorders Relationship between personality disorder and other mental disorders The current state of classification and understanding of the aetiology and pathogenesis of mental disorders is such that most psychiatric diagnoses are based on descriptive criteria. It is common to find that an individual meets the criteria for one or more mental disorders, as well as a PD. At one ex­ treme, these may be a manifestation of the same underlying condition; at the other, they may represent completely separate aetiopathogenic entities. The relationship between PD and other mental disorders may be: • Mutually exclusive PD cannot be diagnosed in an individual with another mental disorder. The personality pathology displayed is a manifestation of the other mental disorder, and giving a separate personality diagnosis has no purpose. This approach is not favoured by current classification systems, even where the two appear to be manifestations of the same condition. • Coincidental In an individual, PD and another disorder may come together by chance. However, epidemiologically, there is support for an association between PD and other mental disorders. • Associative Both in individual cases and epidemiologically, there are a number of reasons why the coexistence of PD and other mental disorders may be more than just coincidental: •​ Sharing common aetiology (but separate disorder). •​ Prodromal (part of the development of another mental disorder). •​ Part of a spectrum (a ‘partial’ manifestation of a mental disorder). •​ Vulnerability (a separate disorder, manifestations of which make an individual more likely to suffer from another mental disorder). Problems in assessing personality in patients with other mental disorders A number of problems may arise in the diagnosis of PD in people who ap­ pear to have other specific mental disorders: • Underlying PD may be missed, as assessment may focus on the current mental state disorder. • PD may be misdiagnosed as another mental disorder, and vice versa. • In an individual with PD, another specific mental disorder may be missed or misconstrued as being part of the PD. In such cases, it is important to remember that other comorbid mental dis­ orders are common in people with PDs, and any change in the presentation of a patient with PD may be due to this. Equally, it is important to base the assessment of personality on information (preferably from a number of sources) on the premorbid functioning of an individual, rather than on their current functioning or just their own account of their previous functioning (their memory or interpretation of which may be coloured by their current mental state).

Relationship between personality disorder and other mental disorders RELATIONSHIP BETWEEN PD AND OTHER MENTAL DISORDERS Comorbidity between personality disorder and other specific mental disorders Strong associations • Cluster B PDs and psychotic, affective, and anxiety disorders. • Cluster C PDs and affective and anxiety disorders. • Avoidant PD and social phobia (possibly because they both describe a group of people with the same condition). • Substance misuse and cluster B PDs. • Eating disorders and cluster B and C PDs (particularly bulimia nervosa and cluster B). • Neurotic disorders and cluster C PDs (it has been suggested that these individuals have a ‘general neurotic syndrome’). • Somatoform disorders and cluster B and C PDs. • Habit and impulse disorders and cluster B PDs (unsurprisingly). • PTSD and borderline PD (this is not borderline PD redefined as chronic PTSD, but it is probably due to the i rate of life events and vulnerability of such individuals).11,12 Moderate associations • Schizotypal PD and schizophrenia (also a weaker association between schizophrenia and antisocial PD). • Depression and cluster B and C PDs. • Delusional disorder and paranoid PD. Impact of personality disorders on manifestation, treatment, and outcome of other mental disorders Although the concept of ‘comorbid PD’ may seem spurious from an aetiopathological perspective, its presence has an impact on the presenta­ tion, treatment, and outcome of other mental disorders, and it is therefore useful to recognize such comorbidity from a clinical perspective. • Presentation Another mental disorder’s presentation may be distorted, exaggerated, or masked by the presence of an underlying PD. • Treatment and outcome The presence of comorbid PD will usually make treatment more difficult and worsens the outcome of other mental disorders. This may be due to problems in the following areas: help-​seeking behaviours, compliance with treatment, coping styles, risk-​taking, lifestyle, social support networks, therapeutic alliance, and alcohol and substance misuse. Some contend that it is the presence of this comorbidity that makes it more likely for a person to fail to respond to standard primary care treatment ap­ proaches, therefore necessitating referral to psychiatric services. 11  Tyrer P (2000) Comorbidity of personality disorder and mental state disorders. In: Tyrer P (ed). Personality Disorders: Diagnosis, Management and Course, pp. 71–​9. Oxford: Butterworth Heinemann. 12  Coid J, Yang M, Tyrer P, Roberts A, Ullrich S (2006) Prevalence and correlates of personality disorder in Great Britain. Br J Psychiatry 188:423–​31.

532 Chapter 12  Personality disorders Assessment of personality disorder Potential pitfalls • Relying on diagnoses made by others (psychiatrists are notoriously poor at diagnosing PD).13,14 • Failing to recognize comorbidity. • Misdiagnosing PD as a mental illness, and vice versa. • Inadequate information. • Negative countertransference (basing the diagnosis on a negative reaction to a patient, rather than on an objective assessment; transference and countertransference may be a part of this, but negative feelings towards an individual should not be the primary basis for a diagnosis of PD). • Applying ICD-​10 or DSM-​5 categories without a broader assessment of personality. Diagnosing personality disorder • History-​taking A good psychiatric history should be obtained and include how long the problem has been present, variations in the difficulties, and any previous treatment and its efficacy, if applicable. It is also very useful to obtain education, employment, and relationship histories, to gain further understanding of interpersonal difficulties, as well as details of previous or current mental health problems and substance misuse. • Presentation It is often helpful to carry out the assessment over several interviews. This will allow the assessor to be more confident that the patient’s presentation reflects personality traits, rather than their mental state during the interview. A person’s presentation can vary significantly, depending on their current mental state or the presence of symptoms of mental illness. However, it is important to note that this fluctuation in presentation may also be a characteristic of PD, e.g. affective lability in borderline PD. • Clinical interview During a clinical interview, the patient’s interaction with the interviewer can be observed. The content of the response, emotional expression, and non-​verbal communication can be observed and reflected upon by the interviewer. The patient’s response to the interviewer (transference) and the feelings evoked in the interviewer (countertransference) also provide clues of the patient’s interpersonal functioning and difficulties. • Other sources of information Patients often have difficulty recognizing which aspects of themselves are the most problematic; sometimes friends or family are better able to identify these issues. This can be quite useful, in addition to information from the clinical interview and structured assessment. 13  Gunn J (2000) Personality disorder:  a clinical suggestion. In:  Tyrer P (ed). Personality Disorders: Diagnosis, Management and Cause, pp. 44–​50. Oxford: Butterworth Heinemann. 14  Banerjee PJM, Gibbon S, Huband N (2009) Assessment of personality disorder. Adv Psychiatr Treat 15:389–​97.

Assessment of personality disorder Assessment instruments There is currently no accepted gold standard measure of the assessment of personality, which makes it difficult to assess the validity of any instru­ ments. However, structured clinical interviews are generally regarded as more robust and detailed than self-​reported questionnaires which tend to over-​report symptoms. Structured categorical (diagnostic) assessments • Observer-​rated structured interviews International Personality Disorder Examination (IPDE), Diagnostic Interview for DSM-​IV Personality Disorders (DIPD-​IV), Structured Interview for DSM-​5 Personality Disorders (SCID-​5-​PD), Structured Clinical Interview for DSM-​IV Axis I Disorders, Personality Disorder Interview-​IV. • Self-​rated questionnaires Personality Diagnostic Questionnaire, Structured interview—​other sources, Standardized Assessment of Personality, Personality Assessment Schedule. Structured dimensional assessments • Observer-​rated structured interview Schedule for Normal and Abnormal Personality. • Self-​rated questionnaires Personality Assessment Inventory, Minnesota Multiphasic Personality Inventory-​2, Millon Clinical Multi-​axial Inventory-​ III, Eysenck Inventory Questionnaire, NEO Five-​Factor inventory-​3. Unstructured assessments • Interview-​based Clinical interview, psychodynamic formulation. • Other Rorschach test, Thematic Apperception Test. Additional assessment Comorbidity The presence of comorbidity should be explored, as patients with one diagnosed PD will often have additional PD(s) and psychiatric problems. Comorbidities can be identified during history-​taking and using assessment instruments. Severity The concept of the severity of PD is perhaps more relevant in specialized PD services and in forensic psychiatry. There is no standard way of recording this. From literature, people with a greater number of PD diag­ noses tend to be regarded as having more severe PD. Also individuals with PDs in >1 cluster are generally considered to have more severe PD. It is also useful to consider the degree of distress experienced by the individual, as well as the interference with functioning—​occupational, family and relation­ ships, offending/​violence, etc. Treatability Making an assessment whether an individual would benefit from a particular treatment is worthwhile, especially since many patients with PD disengage from services. Treatability with CBTs depends on the level of the individual’s intellectual ability, which, in turn, is affected by their current mental state, education, and cultural background.

534 Chapter 12  Personality disorders Management of personality disorder 1: general aspects It is generally felt that PD is resistant to specific psychiatric treatment. However, there is no good evidence to either refute or support this state­ ment. Patients often present at a time of crisis and/​or when they develop a comorbid axis I disorder. Although some may wish to, psychiatrists cannot avoid having to manage patients with PD. Principles of successful management plans A successful management plan in PD is tailored to the individual’s needs and explicitly states jointly agreed and realistic goals.15 The approach to these patients should be consistent and agreed across the services having con­ tact with the patient. Plans should take a long-​term view, recognizing that change, if it comes, will only be observable over a long period. Possible management goals Potential management goals include: psychological and practical support; monitoring and supervision; intervening in crises; increasing motivation and compliance; increasing understanding of difficulties; building a therapeutic relationship; limiting harm; reducing distress; treating comorbid axis I dis­ orders; treating specific areas (e.g. anger, self-​harm, social skills); and giving practical support (e.g. housing, finance, childcare). Managing comorbid mental disorders It is important to recognize and treat comorbidity in patients with PD. Standard treatment approaches should be used, taking into account aspects of the patient’s personality (e.g. impulsivity and an anti-​authoritarian atti­ tude may lead to non-​compliance with medication). Understanding and managing the relationship between the patient and staff16 Rejection for treatment of patients with PD (even when they present with mental illness) is often due to the intense negative feelings these patients may engender and the disruptive and uneasy relationships they form with those who try to help them. Just as they do in many of their interpersonal relationships, patients with PDs display disordered attachment in their re­ lationships with staff (whether with individuals or with a service). When dealing with such patients, this needs to be recognized, acknowledged, and managed. An acceptance of, and tolerance for, these difficulties need to be combined with continuing commitment to the patient. However, pa­ tients, staff, and other agencies need to realize there are no instant solu­ tions and that psychiatric services cannot take responsibility for all adverse behaviours. 15  Davison SE (2002) Principles of managing patients with personality disorder. Adv Psychiatr Treat 8:1–​9. 16  Adshead G (1998) Psychiatric staff as attachment figures. Understanding management problems in psychiatric services in the light of attachment theory. Br J Psychiatry 172:64–​9.

Management of personality disorder 1: general aspects Maintaining boundaries It is important for staff to maintain boundaried relationships with the pa­ tients, as this provides the context for recovery for them. Staff can be supported in achieving this through supervision, including group reflective practice and peer supervision.17 Admission to hospital Patients with PD benefit little from prolonged admissions to conventional psychiatric units. Admission to such units may be necessary when there is a specific crisis (usually in the short term) or when the patient presents with another specific mental disorder. Longer-​term admission for the treat­ ment of PD could be undertaken in a therapeutic community. Involuntary long-​term hospitalization of patients with PD primarily to prevent harm to others where there is little prospect of clinical benefit to the patient is ethically dubious. Managing crises Individuals with PD often present in crisis. This may follow life events or relationship problems, or occur in the context of the development of comorbid mental illness. In some cases, the crisis may follow what appears to the outside observer to be a relatively minor or non-​existent stressor. Where patients repeatedly present in crisis, it can be helpful for the various professionals involved to plan what the response should be in such situ­ ations. A consistent response is important, but there should be sufficient flexibility to deal with changes in circumstances. For example, where a pa­ tient repeatedly presents with self-​harm, it may be appropriate for out­ patient treatment to continue, following any necessary medical treatment; however, if this patient presents threatening suicide following the death of a partner, then it may be appropriate to arrange admission to hospital. Other approaches to individuals presenting with threats of self-​harm or of violence and to manipulative patients are covered in E The manipulative patient 1, p. 1056. 17  Moore E (2012) Personality disorder: its impact on staff and the role of supervision. Adv Psychiatr Treat 18:44–​55.

536 Chapter 12  Personality disorders Management of personality disorder 2: social and pharmacological Therapeutic communities A therapeutic community18 is a consciously designed social environment and programme within a residential or day unit, in which the social and group process is harnessed with therapeutic intent. It is an intense form of psychosocial treatment in which every aspect of the environment is part of the treatment setting, in which interpersonal behaviour can be challenged and modified. The main principles are democratization, permissiveness, communalism, and reality confrontation. There are various interactions between patients and staff both individually and in groups, particularly in daily community groups, which contribute towards achieving these prin­ ciples. There is some evidence that such treatment is effective with some patients with PDs. Medication The main indication for medication in patients with PD is the development of comorbid mental illness.19 There is no good evidence that medication has any effect on PD itself. The positive findings from studies have been short term, and probably due to the effects of medication on comorbid disorders, rather than on the PD itself. Bearing this in mind, the following have been suggested: • Antipsychotics may be of some benefit in cluster B, particularly borderline PD; however, the strength of evidence is low, as it is based mostly on single small studies. Aripiprazole has been demonstrated to have beneficial effects in treating impulsivity in those with borderline PD. Both aripiprazole and olanzapine have shown some benefit in treating patients with cognitive or perceptual symptoms, including suspiciousness and depersonalization. Aripiprazole, olanzapine, and haloperidol may also be useful for managing affect dysregulation.20 • Antidepressants may be of benefit in impulsive, depressed, or self-​ harming patients (particularly borderline) and in cluster C (particularly avoidant and obsessive–​compulsive) disorders. • Mood stabilizers, such as valproate (semisodium), lamotrigine, and topiramate, have demonstrated some benefit in patients with affect dysregulation.21 18  Pearce S, Scott L, Attwood G, et al. (2017) Democratic therapeutic community treatment for personality disorder: randomised controlled trial. Br J Psychiatry 210:149–​56. 19  Tyrer P (2000) Drug treatment of personality disorder. In:  Tyrer P (ed). Personality Disorders: Diagnosis, Management and Cause, pp. 126–​32. Oxford: Butterworth Heinemann. 20  Lieb K, Völlm B, Rücker G, Timmers A, Stoffers JM (2010) Pharmacotherapy for borderline per­ sonality disorder: Cochrane systematic review of randomized trials. Br J Psychiatry 196:4–​12. 21  Kerr IB, Bennett D, Mirapeix C (2012) Cognitive analytic therapy for borderline person­ ality disorder. In:  Sarkar J, Adshead G (eds). Clinical Topics in Personality Disorder, pp.  286–​306. London: RCPsych Publications.

537 MANAGEMENT OF PD 2: SOCIAL AND PHARMACOLOGICAL NICE guidelines on the treatment of antisocial/​borderline personality dis­ orders advise that medication should not be used in an attempt to treat borderline or antisocial personality disorders.22,23 Should medication be considered, it would be wise to use conservatively, as there is evidence that in specialist services for people with PD, clinicians are more likely to be involved in helping people to stop, rather than start psychotropic medi­ cation, due to polypharmacy, poor adherence to medication, and the risk of self-​poisoning.24 22  National Institute for Health and Care Excellence. Antisocial personality disorder: prevention and management. Clinical guideline [CG77]. 2009 (updated 2013). M https://​www.nice.org.uk/​guid­ ance/​cg77 [accessed 8 July 2018]. 23  National Institute for Health and Care Excellence. Borderline personality disorder: recognition and management. Clinical guideline [CG78]. 2009. M https://​www.nice.org.uk/​guidance/​cg78 [ac­ cessed 8 July 2018]. 24  Crawford MJ, Rutter D, Price K, et al. (2007) Learning the lessons: a multi-​method evaluation of dedicated community-​based services for people with personality disorder. London:  National Coordinating Centre for the Service Delivery and Organisation (NCCSDO) Research Programme.

538 Chapter 12  Personality disorders Management of personality disorder 3: psychotherapy Dialectical behavioural therapy (E Dialectical behavioural therapy, p. 916.)25 Dialectical behavioural therapy (DBT) was designed for women in the community who self-​harm. It is a structured and long-​term intervention (1–​ 2yrs or more) with a cognitive–​behavioural approach intended to address the difficulties of borderline PD. The therapy is a combination of individual and group sessions: • Individual therapy focuses initially on reducing behaviour, as well as ‘therapy-​interfering behaviours’. Acceptance strategies, through ‘validation’, are used to help patients understand and accept themselves. Problem-​solving strategies are used to effect change. • Group work aims to increase adaptive behavioural skills, including interpersonal effectiveness, emotion regulation, distress tolerance, and core mindfulness. Individuals are also instructed to telephone their therapists for skills coaching if they have urges to hurt themselves outside scheduled time. This serves to help keep the patient safe and to strengthen their skills by talking through the problem and exploring alternatives to self-​harm or suicidal be­ haviours. Results for studies have shown benefit of DBT in treating people with borderline PD. Cognitive analytic therapy (E Cognitive analytic therapy, p. 918.)26 May be appropriate for some patients with borderline PD. Aims to iden­ tify different ‘self-​states’ and associated ‘reciprocal role procedures’ (pat­ terns of relationships learnt in early childhood). Patients are helped to observe and change thinking and behaviour related to these self-​states. Countertransference helps provide useful information about ‘reciprocal role relationships’, either through identification with the patient or reacting to their projections. The aim is for patients to be able to recognize their various ‘self-​states’ and to be aware of them without dissociating. Psychodynamic therapy (E Psychodynamic psychotherapy, p. 902.)21,26 The psychodynamic and transference-​focused approach is relevant in the treatment of people with borderline and narcissistic PDs. This kind of therapeutic work can help to minimize the externalization of ‘unbearable self-​states’, i.e. the patient will manage their own internalized and distressing self-​perceptions by generating those same feelings in others. Early develop­ mental experiences will also be explored to link to presenting problems. 25  Evershed S (2011) Treatment of personality disorder: skills-​based therapies. Adv Psychiatr Treat 17:206–​13. 26  Bateman A, Tyrer P (2012) Treating personality disorder: methods and outcomes. In: Sarkar J, Adshead G (eds). Clinical Topics in Personality Disorder, pp. 213–​33. London: RCPsych Publications.

Management of personality disorder 3: psychotherapy Mentalization-​based therapy A form of psychodynamic psychotherapy specifically designed and manualized for individuals with borderline PD. The therapy seeks to address disorganized attachment and the individual’s failure to develop mentalizing capacities as a result of early attachment experiences. During times of stress, these ‘non-​mentalizing’ states may then appear—​‘psychic equiva­ lence’ (‘I think, therefore it is’), ‘pretend mode’ (where the individual is dis­ sociated from real thoughts and emotions), and ‘teleological thinking’ (the experience is only valid to the individual if there is tangible evidence of it). Mentalization-​based therapy has been shown in studies to be effective in the management of borderline PD, one in the context of a partial hospitaliza­ tion programme and the other in an outpatient setting. Cognitive behavioural therapy (E Cognitive behavioural therapy 1, p. 910.)27 Cognitive techniques used emphasize changing core beliefs about the self and the work. Three key ways are used to confront core schema once they are accessed: • ‘Schema restructuring’ enables the individual to change a maladaptive schema to an adaptive one. • ‘Schema modification’ aims to modify dysfunctional schemas in order to reduce their impact and their effect on patients’ responses. • ‘Schema reinterpretation’ seeks to make minor changes to existing schema, so patients reinterpret them and manage dysfunctionality better. Behavioural techniques are employed to cause a reduction in self-​harm and other maladaptive behaviours and also to help the individual develop better ways of coping with difficulties. 27  Kerry B, Gordon N (2012) Insight-​oriented therapies for personality disorder. In:  Sarkar J, Adshead G (eds). Clinical Topics in Personality Disorder, pp. 247–​60. London: RCPsych Publications.

540 Chapter 12  Personality disorders Outcome of personality disorder Morbidity and mortality High rates of accidents, suicide, and violent death, particularly where cluster B features are prominent. As mentioned already, there are high rates of other mental disorders. Outcome of other disorders in patients with personality disorder The outcome of mental illness and physical illness is worse in patients with PDs.28 Persistence of personality disorder Some contend that PD is, by definition, lifelong and therefore has a poor prognosis, but the evidence for this is far from conclusive. PDs are best conceptualized as long-​term and chronic disorders, manifesting with varying degrees of severity over time. Some may present with a relapsing and re­ mitting course, depending on environmental factors and comorbidity.29 Comparison between different age groups PD is less prevalent in older adults than younger adults, particularly for cluster B disorders. In terms of ‘normal’ personality, compared with young adults, the elderly are more likely to be cautious and rigid, and less likely to be impulsive and aggressive. However, cross­sectional studies looking at different age groups at one point in time tell us little about the development of personality in individuals over time. Follow-​up of individuals over time Antisocial/​dissocial Children presenting to child services with antisocial be­ haviour are 5–​7 times as likely to develop antisocial PD as those presenting with other problems. May show some improvement in antisocial behaviour by fifth decade. However, may just change with time from ‘overt’ criminal behaviour to more ‘covert’ antisocial behaviour such as domestic violence and child abuse. There is contradictory evidence as to whether ‘burnout’ or ‘maturation’ in later life really does occur. Borderline A third to a half of patients fulfilling the criteria for borderline PD do not have PD at all when followed up after 10–​20yrs. About a third continue to have borderline PD, and others have other predominating PDs. Poor prognostic indicators are severe, repeated self-​harm and a ‘comorbid’ antisocial personality; a good prognostic indicator may be an initial presen­ tation with a comorbid affective disorder. Schizotypal Generally have a poorer prognosis than borderline patients. About 50% may develop schizophrenia. Obsessional May worsen with age. More likely to develop depression than OCD. Clusters There is some evidence that cluster A traits worsen with age, cluster B traits improve, and cluster C traits remain unchanged. 28  Tyrer P, Seivewright H (2000) Outcome of personality disorder. In:  Tyrer P (ed). Personality Disorders: Diagnosis, Management and Cause, pp. 105–​25. Oxford: Butterworth Heinemann. 29  Adshead G, Sarkar J (2012) The nature of personality disorder. Adv Psychiatr Treat 18:162–​72.