6

Categorical Models of Personality Disorders

Igor Weinberg

Categorical models of personality disorders (PDs) have slowly fallen out of favor in the last decade. The burgeoning literature on the dimensional approaches to personality brought a new area and new enthusiasm in research on personality pathology, and in doing so it also raised questions about the viability of the categorical model. In fact, a survey of PD experts showed that 74 percent indicated that the categorical approach should be replaced, 87 percent stated that personality pathology is dimensional in nature, and 70 percent supported a mixed categorical–dimensional approach (Bernstein, Iscan, Maser, & Board of Directors of the Association for Research of Personality Disorders, 2007).

The ever increasing number of publications that focus on dimensional models of personality disorders could lead one to conclude that the categorical models are already a thing of the past. Or some might think that these models have only limited utility and should be reserved for the sidebars and historical sections of our professional textbooks. At least currently, categorical models remain the cornerstone of our current diagnostic approach, understanding, research, and treatment of personality pathology (and all psychiatric pathology for that matter) against which other – alternative – models are often held up for testing and validation. With this in mind, the present chapter provides a general overview of the categorical models of personality pathology focusing on historical developments in the conceptualizations underlying our understanding of PDs, changes in the categorical conceptualization of PDs in the DSM, examination of the strengths of the categorical model (examples of the disorder-specific constructs, documentation of the functional impairment, examination of the construct and predictive validity, as well as of clinical utility), as well as critical examination of the limitations of the model. The chapter concludes with general discussion of future directions.

History of the Concept

Categorical models of personality have a long and varied history. The first known descriptions of character or character type go back to ancient Greece. Hippocrates left us what seems to be the first account of four character types – the choleric, melancholic, sanguine, and phlegmatic. Those corresponded to excesses of four bodily humors: yellow bile, black bile, blood, and phlegm. Hippocrates postulated that imbalances of those four humors constitute the basis for any disorder, including personality types. These humors were bodily analogs of four basic components – earth, water, fire, and air. Those were proposed by the philosopher Empedocles as the basis for understanding the universe and natural phenomena. These four basic categories dominated early conceptualizations of personality, despite efforts to promote other approaches such as Aristotle’s suggestion that a person’s character could be judged based on their facial characteristics, leading to the term physiognomy.

Following the approach through the nineteenth century of viewing character, including normal character types, in terms of combinations of some basic elements including emotionality, activity, energy level, and energy direction, the twentieth century saw the emergence of temperament and the application of this concept to psychiatric patients (e.g., Hirt, 1902). Despite differences in the approaches, temperament-based descriptions followed the same notion of character being a combination of some basic elements. Attracting the interest of psychiatrists, character pathology began to be featured in clinical descriptions. This resulted in a number of now classical publications, such as Kraepelin (1919) and Schneider (1923) which describe different pathological character presentations and Kretchmer (1925) which related character to a person’s physical build.

Important advances in the twentieth-century conceptualization of personality can be attributed to Freud who set the stage for many early psychoanalysts to immerse themselves in study of various characters that were defined by stages of psychosexual development. These descriptions creatively used psychoanalytic thinking and clinical observations to account for presentations and symptoms of patients. The concept of character was first related to the characteristic stage of psychosexual development (Abraham, 1927; Ferenczi, 1938; Freud, 1908). Later, it was related to specific defense mechanisms (Rappaport, 1961; Reich, 1933), specific solutions to internal conflicts (Horney, 1937) – and with advent of object relation theory – specific ways the person experienced self and others (Fairbairn, 1952; Guntrip, 1968; Klein, 1948; Winnicott, 1965).

Later analytic authors incorporated developments in psychoanalysis and psychiatry, and came forth with further diagnostic classifications and subsequent theories and treatments for personality disorders (Gunderson, 1984; Kernberg, 1975; Kohut, 1971; Stone, 1980). Cognitive behavior theories expanded their interest to include personality pathology, resulting in competing and creative formulations (Beck, Freeman, Davis, & Associates, 2006; Young, Klosko, & Weishaar, 1993). Introduction of the biological model to psychiatry led to original formulations of personality in terms of specific biochemical mechanisms (Cloninger, Svrakic, & Przybeck, 1993, Siever & Davis, 1991), while theorists interested in expanding personality theory into understanding psychopathology proposed alternative, dimensional models (Costa & McCrae, 1985; Costa & Widiger, 1993). This rich history resulted in a voluminous literature on different personality types and different types of personality pathology.

It was in this intellectual climate of categorical classification that DSM was developed. Inheriting its conceptual basis from the categorical literature, earlier versions of DSM described disorders in terms of distinct categories, not dimensions. It was not until DSM-5 that the dimensional focus was added as an alternative approach.

Categorical Approach in DSM

In DSM, personality disorders are classified as enduring conditions. They were distinguished from episodic disorders (that ultimately found their place on Axis I) in terms of earlier age of onset, persistence, and resistance to treatment. DSM defined personality disorders in terms of polythetic sets of criteria. Each disorder has a cutoff score of number of criteria necessary to meet the classification for the disorder. Such a definition has a pragmatic role in identifying appropriate treatment, framing appropriate treatment foci and hierarchy of treatment targets, as well as facilitating communication.

DSM underwent a number of significant revisions, though preserving the essential spirit of categorical classification. DSM-I was published in 1952, specifying three categories of personality disorders: (i) personality pattern disturbance, which included inadequate, paranoid, cyclothymic, and schizoid personality disorders; (ii) personality trait disturbance, which included emotionally unstable, passive-aggressive dependent or aggressive types, and compulsive personality disorders; and (iii) sociopathic personality disturbance, which included antisocial dissocial types. These conditions were considered relatively stable and refractory to treatment.

Revision of the ICD system led to a second edition of DSM in 1968 that was less theory-based and more empirically grounded. DSM-II held many similarities with DSM-I in its presentation of the personality disorders, with notable revisions. While inadequate, paranoid, cyclothymic, and schizoid personalities, as well as antisocial sociopathic disturbance remained largely unchanged, emotionally unstable personality was reformulated as hysterical and passive-aggressive personality took the place of passive-aggressive personality, aggressive type. Additionally, compulsive personality was reformulated as obsessive-compulsive personality, and explosive and asthenic personality categories were added.

Further emphasis on reliability and the empirical basis of the diagnosis that evolved in 1970s contributed to the third edition of DSM, published in 1980. Of note, Axis II was created for personality disorders to designate persistent conditions with an early onset. Beginning with the perspective that personality disorders were psychosocial in nature, DSM-III heralded subdivision of the personality disorders into three clusters: (i) Cluster A (odd) including schizotypal (StPD), schizoid (SPD), and paranoid (PPD) personality disorders; (ii) Cluster B (dramatic) including histrionic (HPD), antisocial (ASPD), borderline (BPD), and narcissistic (NPD) personality disorders; and (iii) Cluster C (anxious) including compulsive, avoidant (AvPD), dependent (DPD), and passive-aggressive personality disorders. Within Cluster A, schizoid personality was divided into schizotypal, schizoid, and avoidant personality disorders, with the latter shifting to Cluster C. Within Cluster B, hysterical personality was reformulated as histrionic personality. Within Cluster C dependent personality was added to the list and asthenic personality was omitted. Cyclothymic personality and explosive personality were shifted to Axis I as cyclothymic disorder and intermittent explosive disorder, respectively.

Published in 1994, DSM-IV was based on an extensive set of literature reviews, data analyses, field trials, and feedback from the clinicians and researchers in the field. It preserved the general categorization of DSM-III, with an exception of the omission of the passive-aggressive personality disorder and renaming compulsive personality disorder as obsessive-compulsive personality disorder (OCPD). DSM-5 was published in 2013 after further extensive literature review, data analyses, field trials, and feedback from the clinicians and researchers in the field. DSM-5 also marked the introduction of the alternative trait and dimensional models that have the potential to ultimately supplement the categorical model. Initially, the DSM-5 working group on personality disorders intended to replace the categorical model with the dimensional one. Initially, the dimensional model was proposed by the working group as a conceptually attractive solution for the challenges the categorical model posed (see below). However, this proposal was criticized because it had not accumulated the same level of evidence as the categorical model. A complete transition to the dimensional model would have severed the ties with decades of research and clinical tradition without a comparable empirical basis. Consequently, the dimensional model was included as an alternative conceptualization of personality disorders that is worth further research, not as a primary diagnostic model (for a review of the controversy see Chapter 5). It is to the examination of the validity of the categorical model that we will now turn.

Disorder-Specific Concepts

One of the advantages of the categorical concept is the richness of descriptions and understanding of the individual personality disorders. Coming from a long tradition of conducting therapy with these patients, these descriptions offer empathic and experience-near accounts of the subjective experiences of the patients. Such descriptions not only enrich and broaden our understanding of the disorder, but also create language that helps empathically name experiences for the individual patient.

Stated in the simplest terms, categorical descriptions of the personality disorders create disorder-specific language to label various experiences. This approach also creates a separate semantic network that helps understand the disorder and the patient who suffers from it. In such a way these disorder-specific descriptions and concepts are the first step in bridging the gap between the nomothetic understanding of the disorder and the ideographic formulation of the individual patient. However, applying a more generic language borrowed from other domains, including personality research or other personality disorders runs the risk of misrepresenting (misunderstanding) these disorders. In other words, a more generic language runs the risk of obfuscating what is unique and specific about the individual personality disorders and forcing outside concepts on them and the individual patients. This point can be illustrated with a few examples of such concepts and how they apply to different disorders.

Dependence

A large number of studies have focused on understanding the concept of dependence (Bornstein, 1993). Dependence appears in a number of personality disorders, though its meaning differs in each case. In patients with DPD, overreliance on others and the reported need for others’ support and assistance is related to low self-efficacy beliefs. Dependence for patients with BPD is related to fear of abandonment and intolerance of aloneness. Dependence, paradoxically, appears in NPD patients on context of idealization of significant others. In these patients dependence is connected to the need to rely on the idealized others, who are experienced as major sources of self-esteem. Therefore, in these patients, dependence is fueled by a specific way of regulating self-esteem.

These formulations demonstrate the discontinuity of the concept of dependence across various personality disorders, and also highlight the diagnostic and therapeutic significance of such discontinuity. From the diagnostic perspective, identifying these dynamics would prompt the clinician to make a careful assessment of the relevant disorder. From the treatment perspective, the therapist will intervene differently to address these patterns of dependence.

With a DPD patient, the therapist will frame therapy in the context of increasing self-reliance, problem-solving ability, and self-efficacy beliefs, while being vigilant regarding the risk of stepping into a more authoritarian role and solving the problems on the patient’s behalf. With a BPD patient, the therapist will need to help the patient identify patterns of avoidance of experiences of abandonment and help develop more efficient strategies to tolerate aloneness and connect with others. With NPD patients, the task of the therapist is helping the patient identify the role of reliance on others in regulating self-esteem and develop alternatives for self-esteem regulation.

Interpersonal Distancing

Distancing appears in a number of disorders, and, similar to the concept of dependence, has a different significance in different disorders. In AvPD, distancing stems from fear of judgment by others and it appears in the context of negative self-perception and a positive perception of others. In SPD, distancing appears in the context of disinterest in connection with others and lack of investment in real relationships with others. In NPD, distancing is usually connected with devaluation of others and the desire to preserve self-sufficiency; therefore – counter-dependence. In PPD, distancing is related to a suspicious stance about others. In PPD others are typically seen as negative and as having malicious intentions, while the self is seen in a more positive light.

Differences associated with dependence indicate differential treatment approaches to these patients and would suggest different prognoses for the symptoms of “distancing.” For instance, in AvPD, the therapist could intervene by encouraging exposure-style interventions and engaging the patient in confronting and testing negative expectations of others. With SPD, distancing will require a much longer treatment process that would involve helping the patient become aware of the need for others and create coping skills to make up for the difficulty connecting with others. In NPD patients the therapist will likely target distancing by identifying different ways in which the patient accomplishes that as well as exploring the function of it; encouraging the patient to develop curiosity about what he or she is avoiding is yet another possible intervention. In PPD patients, the therapist is likely to take a stance of promoting trust and understanding as preconditions for any therapeutic process and in this way will start helping with distancing. With this in mind, prognosis is likely to be better for AvPD patients, intermediate for NPD patients, and poor for SPD and PPD patients.

Complexities of Individual Disorders

Some concepts were specifically developed for individual personality disorders and are more likely to lose their specificity and explanatory power within other diagnostic approaches. For example, Ronningstam (2011) lists a long line of concepts central to the understanding of NPD that get lost in other models, but retain their meaning as long as NPD is recognized as a separate disorder. These concepts bridge the gap between the phenomenological, observable aspects of NPD and the subjective and functional aspect of it.

Ronningstam (2011, p. 253) proposes the following reformulation of the construct of narcissism: “enhanced or unrealistic, either overtly interpersonally or behaviorally expressed or internally hidden, sense of superiority and exaggeration of own achievement and capability vulnerability and vulnerable self-esteem with self-criticism and inferiority, and intense reactions to threat, criticism, or defeats; and self-preoccupation with self-enhancement and self-serving interpersonal behavior.” Such reformulation is clinically meaningful and humanizing in terms of understanding the individual patient with NPD. It also meaningfully helps facilitate the dialogue with the patient about the disorder and provides guidelines for intervention. Importantly, this formulation also explains the fluidity of the clinical presentation of some NPD patients who, at times, acquire a chameleon-like capacity to blend in and present without any symptoms, despite having the most severe forms of the disorder. While the trait approach might be able to capture some of the aspects of NPD, it misses important sectors of the pathology, including functional significance of behaviors, subjective experiences of the patients, or radically different aspects of the pathology that are simply not represented in the trait models for theoretical reasons.

On the other hand, description of NPD simply in terms of manipulativeness and callousness will misrepresent the NPD patient as too similar to patients with ASPD. Even though many NPD patients have comorbid ASPD (Gunderson & Ronningstam, 2001), their antisocial behaviors are typically guided by self-regulatory function, rather than calculated agenda to accomplish an incentivized goal. In this way application of global traits misrepresents what is specific about the disorder.

Schizoid personality disorder is yet another disorder that can be easily misrepresented if it is reduced to global traits. Many authors describe a seemingly contradictory conglomerate of traits that many SPD patients have that could be confusing at first glance, but perfectly understandable if some of them are understood as serving a self-protective function. For instance, Livesley and colleagues (Livesley, West, & Tanney, 1986) describe presence of active and passive social avoidance, the inner hypersensitivity and the callous persona as well as hypervigilant and absent-minded attitudes. Such schism in terms of various aspects of functioning and experiencing is self-protective and reflects not only facets of functioning, but also aspects of subjective experience.

The concept of the False Self (Winnicott, 1965) was described to characterize a particular aspect of functioning of people with SPD – an overly compliant persona that is detached from the core sense of self and that is protecting the person from the inner experiences of falling into pieces, being disconnected from one’s body and experience of disorientation (Winnicott, 1965). This is a complicated construct, but accurately describes subjective and very private experiences of patients with SPD. While many descriptions might capture the interpersonal distancing, it can be argued that only the categorical model has the conceptual power to uniquely associate these experiences of fragmentation and of the False Self to SPD.

Clinical descriptions of PPD present similar complexity. Akhtar (1992, p. 156) summarizes that this disorder has a “contradictory presence of (1) profound mistrust with naive gullibility, (2) arrogant demandingness with hidden inferiority, (3) emotional coldness with marked sensitivity, (4) superficial asexuality with vulnerability to erotomania, (5) moralistic stance with potential corrupt attitudes, (6) acutely vigilant attention with inability to use the whole picture.” Such complexity requires seeing the person beyond their individual behaviors and understanding the essence of the person’s functioning. This is a complicated matter because it requires the clinician to see the big picture and not necessarily get lost in individual features. Similarly, it requires understanding of the person, integrating and not simply compiling phenomenological features.

These examples emphasize the need for distinguishing the diagnostic aspects of the disorder from the aspect of understanding of the individual patient. Symptoms of the disorder represent nomothetic dimensions of the psychopathology and they were designed to identify significant personality disorders. Understanding of the individual patient, while relying on these signs and symptoms is different as it attempts to integrate all characteristics of the person to come up with a comprehensive and meaningful understanding of him or her. Diagnosing a specific personality disorder helps in the process of accomplishing such understanding. However, the diagnostic sets designed to diagnose each disorder do not aspire to represent an essential understanding of each individual patient. While diagnosis does not equal understanding of the patient with the disorder, it is an essential first step to such understanding. Meaningfully formulated understanding, communicated to the patient, paves the way toward alliance building. Finally, clinical concepts associated with each disorder create a conceptual map that guides the clinician through the process of developing understanding of the individual patient. This helps to bridge the gap between the nomothetic and the ideographic and translate generic characteristics of the disorder into personally meaningful concepts.

The difference between meaningful understanding and statistical prediction was emphasized by Holt (1970, p. 347) who warned that “the logic of statistical prediction does need require understanding of the behavior in question … The statistician’s interest ceases once he has found the most efficient and stable formula that combines scores for prediction. On the other hand understanding is concerned with questions regarding ‘how things work’.”

Taking these points together, the importance of seeing the person behind clinical descriptions is hard to overestimate. Such understanding helps with alliance building and treatment interventions, and it also contributes to clinical prediction (Holt, 1970). Statistical prediction is seeking individual variables that have the statistical power to predict an important outcome. Clinical prediction, on the other hand, relies on understanding how the person functions, experiences self, others and the world, or responds to significant stimuli such as stress or relationships to anticipate important outcomes. An example of this dichotomy relates to the prediction of suicide. Pokorny (1983) failed to predict suicide in a prospective study that relied on statistical prediction, using numerous predictors. On the other hand, Edwin Shneidman, a prominent suicide expert, blindly reviewed clinical material in the prospective study of gifted women conducted by Lewis Terman and successfully identified all completed suicides using clinical judgment – his understanding of women participants of the study (Orbach, personal communication, March 7, 2002).

Functional Impairment and the Categorical Model

Functional impairment is one of the cornerstones of psychopathology in DSM and it is one of the critical components of personality disorders. Functional impairment refers to difficulty performing tasks and roles in such areas as social or interpersonal, school or work, recreational or leisure, self-care, communication and mobility (Skodol, 2018).

The critical role of functional impairment as an important component of the diagnosis of personality disorders is highlighted by the study by Trull and colleagues (Trull, Jahng, Tomko, Wood, & Sher, 2010). Trull and his colleagues used data from the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) study (Grant, Moore, Shepard, & Kaplan, 2003) that assessed the prevalence of psychiatric disorders. The original NESARC study used only descriptive phenomenology – symptoms – to meet the criteria for personality disorders. Trull and his colleagues re-analyzed prevalence data from the NESARC and introduced a further requirement for personality disorder – functional impairment and/or distress. Table 6.1 shows prevalence of each personality disorder reported by the NESARC study and the prevalence reported by Trull and his colleagues (Trull et al., 2010).

Table 6.1Prevalence of personality disorders in NESARC: the role of functional impairment and distress

Personality disorder	NESARC results	Trull et al. (2010) results
Schizotypal PD	3.9%	0.6%
Schizoid PD	3.1%	0.6%
Paranoid PD	4.4%	1.9%
Histrionic PD	1.8%	0.3%
Antisocial PD	3.8%	3.8%
Borderline PD	5.9%	2.7%
Narcissistic PD	6.2%	1.0%
Obsessive-compulsive PD	7.9%	1.9%
Dependent PD	0.5%	0.3%
Avoidant PD	2.4%	1.2%

Studies uniformly document that PDs are associated with impairments in one or more areas of functioning. The Collaborative Longitudinal Personality Disorders Study (CLPS) reported that, compared to MDD, StPD and BPD were associated with a more than three-fold decreased chance of having only a high school education, and being less frequently employed; compared to MDD and OCPD, StPD and BPD were associated with lower vocational, social, and leisure functioning. Patients with AvPD fell between these two groups. After controlling for Axis I pathology and demographics, these results remained significant (Skodol et al., 2002).

The NESARC study reported that in Wave 1 PDs were associated with one of the following marital statuses: being divorced, separated, widowed, or never married. AvPD, DPD, SPD, PPD, and ASPD were associated with social impairment, even after controlling for age and Axis I pathology. Interestingly, there was no consistent relationship between OCPD and impairment. HPD was not associated with any impairment. In Wave 2, BPD correlated with unmarried status, lower income, and lower education level, after controlling for demographics and other psychiatric disorders. NPD was associated with not being married and with emotional role impairment; men but not women with NPD had global impairment. StPD was associated with unmarried status and global impairment (Grant et al., 2004, 2008; Penner-Goeke et al., 2015; Pulay et al., 2009; Stinson et al., 2008).

Results from the Children in the Community Study (CICS) indicated that all PDs were associated with social impairment and lower academic achievement after controlling for Axis I disorders. All PDs, and especially Cluster B PD, had poor quality of life at age 33 (Chen, Cohen, Crawford, Kasen, & Johnson, 2006; Crawford et al., 2008; Johnson et al., 2005).

The National Comorbidity Survey Replication reported that all PDs had impairments in the following domains: basic role functioning, instrumental role functioning, and social role. However, after controlling for Axis I disorders, most of these associations became not significant, except the association between Cluster B PD and social role functioning as well as between any PD and impairment in productive role functioning, social role functioning, and instrumental role functioning (Lenzenweger, Lane, Loranger, & Kessler, 2007).

The longitudinal course of impairment in personality disorders was examined in three studies. In the CLPS, functional impairment remained mostly stable, with the exception of improvement in social functioning. Even after remission of BPD symptoms (85 percent), only 20 percent displayed also functional remission; others continued to display significant functional impairment (Gunderson et al. 2011; Skodol et al., 2005). The McLean Study of Adult Development (MSAD) reported that among BPD patients, functioning improved so that the proportion of patients with good functioning increased from 26 percent at the baseline to 56 percent six years later. At ten-year follow-up, BPD patients had trouble gaining better functioning or regaining better functioning if they lost it at some point. Vocational impairment was the most persistent (Zanarini, Frankenburg, Hennen, Reich, & Silk, 2005; Zanarini, Frankenburg, Reich, & Fitzmaurice, 2010). In the CICS, PD persistence was associated with persistence of the functional impairment; even when the PD remitted, a mild functional impairment still persisted (Skodol, Johnson, Cohen, Sneed, & Crawford, 2007).

Taken together these studies validate the association of personality disorders and functional impairment – one of the central characteristics of the disorder. One possibility is that psychological symptoms of personality disorders, including problems with emotional regulation, relationship with others, impulsivity or cognition impair one’s ability to function. Another possibility is that functional impairment is yet another manifestation of personality disorders, not exclusively mediated by symptoms. Persistence of impairment despite improvement in symptoms supports the latter hypothesis. This suggests that treatments need to target functional impairment in personality disorders. This opens an avenue to the development of new treatments that help patients address various impairments in their functioning, in addition to their symptoms (for a similar suggestion, see Links, 1993).

Construct Validity of the Personality Disorders

The categorical model of personality disorders has the advantage of relying on the categorical conceptualization of other – syndromal – psychiatric disorders. Robins and Guze (1970) defined five requirements to establish construct validity of psychiatric disorders. Table 6.2 describes these requirements.

Table 6.2Requirements to validate psychiatric disorders

Requirement	Explanation
Clinical description	There is a set of clinical descriptors characteristic of the disorder
Laboratory studies	There are psychological, anatomical, chemical, biological, radiological etc. tests that are associated with the presence of the disorder
Delimitation from other disorders	There are criteria that differentiate the described disorder from other disorders
Follow-up studies	Follow-up studies demonstrate similarity of the longitudinal course of the disorder
Family studies	Family studies demonstrate familiality of the disorder – increased likelihood of “running in the family”

These research criteria have been the cornerstone of establishing the construct validity of any psychiatric disorders (Aboraya, France, Young, Curci, & Lepage, 2005). Thus, one of the strengths of the categorical model is the possibility of subjecting each personality disorder to such testing of construct validity using criteria spelled out by Robins and Guze (1970).

Tables 6.3–6.5 summarize empirical evidence relevant to construct validity of each personality disorder. Such PDs as schizotypal, antisocial, and borderline have the strongest level of evidence, while such PDs as schizoid, paranoid, histrionic, and dependent PD have the weakest level of evidence, partially because of the paucity of research, inadequate reliability of the diagnostic set, or use of non-clinical samples. In addition to improving methodology in these domains, future translational research needs to operationalize disorder-specific constructs and measure them in relevant disorders. Not only will such studies increase the concept validity of the disorders, but also, by testing experience-near constructs, they will inform clinical work.

Table 6.3General validity of Cluster A personality disorders

	Schizotypal PD	Schizoid PD	Paranoid PD
Clinical description	Excellent inter-rater and fair test/retest reliability of the diagnostic set (Zanarini et al., 2000)	Only fair reliability of the diagnostic set (Zanarini et al., 2000)	Good reliability of the diagnostic set (Zanarini et al., 2000)
Laboratory studies	Disturbances in: prepulse inhibition (Abel et al., 2004), eye tracking (Siever et al., 1989), P300 (Trestman et al., 1996), P50 suppression (Croft et al., 2001), catechol-o-methyltransferase gene (Minzenberg et al., 2006), proline dehydrogenase gene (Stefanis et al., 2007)	Not reported	Paranoid ideation was related to: limited hearing (Zimbardo et al., 1981), external attribution of failure (Bodner & Mikulincer, 1998)
Delimitation from other disorders	Clear differentiation from schizophrenia, schizoid or borderline PD (Chemerinski et al., 2013)	Clear differentiation from avoidant PD	Clear differentiation from other disorders
Follow-up studies	Slow improvement, though 17% develop schizophrenia (McGlashan et al., 2005; Fenton & McGlashan, 1989)	Stability of symptoms (Chanen et al., 2004)	Axis I comorbidity predicts worse outcome in PPD (Hong et al., 2005)
Family studies	Familiality was demonstrated (Siever et al., 1996)	Some (0.29) heritability (Torgersen et al., 2000)	Familiality was demonstrated (Triebwasser et al., 2013)
Conclusion	+++; StPD bears similarity to schizophrenia spectrum disorder – spectrum disorder	+	+

Note: 0: no evidence, +: modest evidence, ++: moderate evidence, +++: strong evidence

Table 6.4General validity of Cluster B personality disorders

	Antisocial PD	Narcissistic PD	Borderline PD	Histrionic PD
Clinical description	Excellent reliability of the diagnostic set (Zanarini et al., 2000)	Good relability of the diagnostic set (Zanarini et al., 2000)	Good to excellent reliability of the diagnostic set (Zanarini et al., 2000)	Fair to good reliability of the diagnostic set (Zanarini et al., 2000)
Laboratory studies	Specific alleles of 5-HTT and MAO genes (Ficks & Waldman, 2014), abnormalities in prefrontal; cortex (Yang & Raine, 2009), reduced low frequency fluctuations in EEG (Liu et al., 2014)	Emotional empathy deficits (Baskin-Sommers et al., 2014), gray matter abnormalities in the right hemisphere (Schulze et al., 2013)	Abnormalities in amygdala (Ma et al., 2016), prefrontal cortex (Ma et al., 2016), polymorphisms of genes related to HPA axis (Martín-Blanco et al., 2015), decreased level of oxytocin (Bertsch et al., 2013)	Not reported
Delimitation from other disorders	Clear differentiation from other disorders	Clear differentiation from other disorders	Clear differentiation from other disorders (Gunderson et al., 2017)	Clear differentiation from other disorders
Follow-up studies	Relatively stable, with a slow and modest improvement being possible (Black 2015)	Gradual and slow improvement (Plakun, 1990; Ronningstam et al., 1995)	Gradual improvement in symptoms, less in functioning (Gunderson et al., 2011)	Gradual and slow improvement (Nestadt et al., 2010)
Family studies	Some heritability (Ma et al., 2016)	High heritability (0.71; Torgersen et al., 2012)	High heritability (0.70; Torgersen et al., 2000)	Not reported
Conclusion	+++	++	+++	+

Note: 0: no evidence, +: modest evidence, ++: moderate evidence, +++: strong evidence

Table 6.5General validity of Cluster C personality disorders

	Dependent PD	Avoidant PD	Obsessive-compulsive PD
Clinical description	Fair to excellent reliability of the diagnostic set (Zanarini et al., 2000)	Good reliability of the diagnostic set (Zanarini et al., 2000)	Good reliability of the diagnostic set (Zanarini et al., 2000)
Laboratory studies	Correlates are anxiety, insecurity, fear of negative evaluations, loneliness (Bornstein et al., 2005)	Abnormalities in amygdala, prefrontal cortex and in connectivity between amygdala and insula (Koenigsberg et al., 2014)	Inconsistent findings (Diedrich & Voderholzer, 2015)
Delimitation from other disorders	Clear differentiation from other studies	Clear differentiation from other disorders	Clear differentiation from other disorders
Follow-up studies	No studies that documented outcome	Gradual improvement (Weinbrecht et al., 2013)	Gradual improvement (McGlashan et al., 2005)
Family studies	Some heritability (0.30; Torgersen et al., 2000)	Substantial heritability (0.64; Gjerde et al., 2012)	High heritability (0.78; Torgersen et al., 2000)
Conclusion	+	++	++

Note: 0: no evidence, +: modest evidence, ++: moderate evidence, +++: strong evidence

Predictive Utility of Categorical Models

Predictive validity of categorical models was compared to dimensional models. Overall, initial studies show that dimensional scores of personality outperform the categorical scores in predicting important clinical variables, including functional impairment. For example, incremental concurrent validity of the DSM-5 PD system was contrasted with the DSM-IV PD system in a recent cross-sectional study (Skodol, Bender, Gunderson, & Oldham, 2014). The predicted variables were psychosocial functioning, risk of self-harm, violence, criminal behaviors, optimal level of treatment intensity and prognosis. Compared to DSM-IV, DSM-5 showed stronger correlations with the predictors in 11 out of 12 comparisons. However, the level of personality functioning and total level of risk had a stronger association with DSM-IV, than DSM-5. The partial multiple correlations of controlling for the competitive version of DSM showed that DSM-5 was a stronger predictor of all assessed areas.

There also is evidence that points to predictive validity of categorical descriptions. Using data from the CLPS, Morey et al. compared predictive utility of the categorical model to that of the Five-Factor Model of Personality (FFM) and Schedule for Non-adaptive and Adaptive Personality (SNAP). These comparisons were tested for baseline functions and years 2 and 4 of the follow-up (Morey et al., 2007) and for years 6, 8, and 10 of the follow-up (Morey et al., 2012). Predictors were psychosocial functioning, work and social functioning, number of concurrent Axis I diagnosis, number of medications, and depression scores. First, the results consistently demonstrated that dimensional scores outperformed categorical diagnoses of the personality disorders in predicting important areas of outcome. Second, an interesting pattern emerged when categorical diagnoses were transformed into dimensional scores through criteria counting. Results showed that the dimensionalized scores of the categorical PD diagnoses outperformed FFM in terms of prediction of longer term outcome. These findings suggest that categorical diagnoses capture more enduring aspects of the disorder.

Taken together, these results suggest that both categorical and dimensional models have merit and predict different areas of functioning, including functional impairment. Along with many other similar reports, these findings paved the way for adding the dimensional approach to personality disorders. Overall, the hybrid model that incorporates both the categorical and dimensional approaches has the strongest predictive validity.

Clinical Utility of a Categorical Model of Personality Disorders

Clinical utility is a pragmatic construct pertaining to the efficiency of use of the clinical diagnosis for six main purposes: (i) communication with the patient, (ii) communication with other professionals, (iii) comprehensiveness, (iv) descriptiveness, (v) ease of use, and (vi) utility for treatment planning (First et al., 2004; Morey, Skodol, & Oldham, 2014).

Communication with the Patient

Communication of personality diagnosis, tested for BPD patients (Rubovzki, Gunderson, & Weinberg, 2006), coupled with psychoeducation for the disorder (Zanarini & Frankenburg, 2008) help with destigmatization, relief from shame, and provision of hope. Effective personality disorder descriptions empathically convey various aspects of the person’s functioning and explain his or her suffering and difficulties in meaningful ways. In doing so, they contribute to increased treatment alliance and treatment engagement.

Communication with Other Professionals

Effective communication with other professionals relies on simplicity, accuracy, and clarity of the communicated information. In other words, clinicians using categorical diagnoses uniformly understand what they denote and communicate. Use of categorical diagnoses allows quick and efficient documentation and communication of information related to typical clinical presentation and course as well as possible associated comorbidities and risks. For example, the diagnosis of borderline personality disorder captures patterns of emotional, behavioral, and interpersonal instability, possible risk of suicidal and other high risk behaviors, as well as the need to assess likely comorbid conditions – mood disorder, PTSD, and substance use disorders. Clinicians are likely to anticipate reactive and impulsive presentation and adjust their clinical style in service of more efficient clinical encounter. Similarly, the diagnosis of PPD will communicate to the clinician that the patient is likely to be suspicious, sensitive to injustice, and might hold grudges. Therefore, the clinician will be more likely to take a more patient and less confrontational approach.

Comprehensiveness

This aspect of utility pertains to the extent to which the disorder comprehensively describes all important personality problems of the individual. This includes domains of symptoms, but also possible areas of functional impairment – vocational, social, intimacy, leisure, self-care, etc.

Descriptiveness

Descriptiveness refers to the extent to which the disorder describes the individual’s global personality. That includes adaptive and maladaptive aspects, strengths and weaknesses, motivations and goals. Descriptiveness also captures features that represent core characteristics of the person that represent who he or she is.

Ease of Use

This has to do with ease of applying diagnostic concepts to the individual. In other words, it refers to the extent to which symptoms of the disorder can be easily deduced from the assessment. Difficulty of use, for example, would require a great deal of inference. Ease of use would rely on very little or no inference.

Treatment Planning Utility

A categorical approach to diagnosis also helps with treatment planning, especially if published and even evidence-based treatments are available for the disorder. Disorder-specific treatments, treatment recommendations, and treatment principles were published for some personality disorders. For example, the diagnosis of BPD is likely to lead to recommendations of evidence-based treatments such as DBT, MBT, GPM, or TFP (for a review, see Gunderson, Weinberg, & Choi-Kain, 2013). For ASPD clinicians are likely to incorporate contingency management as an evidence-based approach (Salekin, 2002). For such conditions as NPD, where no evidence-based treatments are available, the available literature on its treatment relies mostly on clinical expert opinion. Thus, for NPD, the writings of Kernberg (1975), Kohut (1971), Ronningstam (2013), or Young et al. (1993) can inform the treatment. Beck’s group published a CBT manual for personality disorder-specific treatment approaches (Beck et al., 2006), while Young developed disorder-specific treatments for BPD and NPD (Young et al., 1993). Principles for personality disorders treatments were published for Cluster A PDs (Williams, 2010), BPD (Gunderson et al., 2013), NPD (Ronningstam, 2013), HPD (Horowitz & Lerner, 2010), ASPD (Meloy & Yakeley, 2010), and Cluster C PDs (Stone, 2013; Svartberg & McCullough, 2010). In other words, the clinical utility of categorical diagnoses relies on many decades of clinical and empirical work on development, testing, validating, and dissemination of disorder-specific treatments. The unique value of categorical models stems from their ability to suggest disorder-specific treatments with demonstrated effectiveness.

The usefulness of the categorical approach was tested in a number of empirical studies with some mixed results. Morey et al. (2014) compared the clinical utility of categorical vs. alternative – dimensional trait – models of personality disorders in DSM-5 in a sample of 337 mental health clinicians. Clinicians evaluated the clinical utility of each model (categorical vs. dimensional) with respect to communication with patients and with other professionals, comprehensiveness, descriptiveness, ease of use, and utility for treatment planning. The categorical model was seen as more useful in terms of communication with other clinicians. In every other respect the alternative – dimensional – model was perceived as more useful. Crego and colleagues (Crego, Sleep, & Widiger 2016) assessed clinical utility of traits in capturing personality disorders. Traits were taken from the Five-Factor Model as well as each iteration of the trait set development for DSM-5 (37 item set and 25 item set). Participants were psychologists surveyed through section 42. Clinicians regarded the final, 25 version of traits assigned to the following disorders less acceptable than the 37 trait version assignment: AvPD, NPD, OCPD, and SPD. Clinicians thought that FFM assignments were more acceptable than the final 25-item version assignments to the following PDs: AvPD, NPD, OCPD, DPD, and HPD. The authors concluded that there are “potentially important limitations with respect to the dimensional trait descriptions of respective PDs.”

These studies reported perceived utility, rather than the utility of the actual clinical use of these different approaches and tests of the predictive utility of them. While predictive utility of the categorical diagnoses will be reviewed below, the testing of the relative utility of different approaches in actual clinical use – more valid than the test of the perceived utility – is waiting for further empirical investigations.

Limitations of the Categorical Model

Over time, the categorical model has accrued a number of criticisms (Morey, Benson, Busch, & Skodol, 2015; Widiger & Mullins-Sweatt, 2005). The four most commonly mentioned criticisms are (i) excessive diagnostic co-occurrence, (ii) phenomenological heterogeneity within the same diagnostic category, (iii) arbitrary diagnostic boundaries, and (iv) inadequate coverage.

Excessive Diagnostic Co-Occurrence

Comorbidity is a natural by-product of applying fixed diagnostic categories on the human complexity. In other words, the categorical diagnostic system anticipates comorbidity – co-occurrence of a number of disorders in the same individual. In fact, disorders are not distributed randomly as one disorder increases the likelihood of having yet another, and having two disorders increases the likelihood of having additional comorbid disorders, and so on (Kessler et al., 1994).

To ascertain true comorbidity, categorical diagnostic systems establish guidelines to rule out differential diagnoses. The process of ruling out the differential diagnoses reduces the number of overall diagnoses and, when conducted properly, establishes accurate diagnoses for the person.

While it is expected that a person might be diagnosed with more than one personality disorder, a number of research findings point out that the number of comorbid personality disorders seems high (Bornstein, 1998; Lilienfield, Waldman, & Israel, 1994; Oldham et al., 1992; Widiger & Sanderson, 1995). In other words, a sizable proportion of patients are being diagnosed with a large number of comorbid disorders. This could mean that the disorders themselves are not adequately capturing the complexity of the individual patients, thus raising a question of their construct validity.

This approach to the challenge of excessive diagnostic co-occurrence involves establishing hierarchical rules (Gunderson, 1992) that would guide the clinician to establish primacy of the personality disorders that are clinically most relevant. Such an approach is likely to make the actual diagnosis more clear and more useful for the purposes of understanding the patient and treatment planning. However, such an approach would not eliminate the actual number of the disorders. It will simply organize them in a more comprehensive picture.

From the standpoint of the categorical diagnostic picture, these findings can be interpreted as a sobering reminder that humans are complicated and so are their disorders. Our attempts to fit different individuals into pre-fixed categories will not account for their inherent complexity. Yet another aspect has to do with the purposes of diagnostic practice. Diagnoses cannot convey the full complexity of human nature and of the individual patient’s subjectivity and uniqueness. This would be the purpose of a detailed clinical report or formulation in which the person’s unique patterns can be described and spelled out. The purpose of the diagnostic process is to identify those patterns that are consistent, maladaptive, and causing substantial dysfunction and distress. Teasing apart which patterns are causing the dysfunction and distress is likely to take time and so the number of diagnosed disorders will likely diminish as the clinician gets to know the patient and follows him or her over a period of time.

Phenomenological Heterogeneity within the Same Diagnostic Category

The polythetic diagnostic approach taken by the categorical approach suggests that the patient can meet only a fraction of criteria out of the full list of criteria for the disorder. While clinically, such an approach allows for flexibility and accounts for variability of the clinical presentation, it also leads to yet another challenge – substantial heterogeneity in the possible clinical presentations. In fact, statistically speaking the multiplicity of clinical presentation for each personality disorder will be in the hundreds! Such an approach also allows two different patients to not share even one symptom. For instance, in order for a person to meet criteria for OCPD, he or she is required to meet four out of eight criteria, thus allowing for a possibility of two different patients meeting criteria for OCPD without sharing even a single common criterion. Similarly, it is theoretically possible for a patient to meet criteria for BPD without having the prototypical behaviors of affective instability, self-harm/suicidality, and unstable relationships.

One way to address the challenge of the heterogeneity of the clinical presentation is to identify necessary criteria for each disorder. For example, the Diagnostic Interview for Borderline Personality Disorder (Zanarini, Frankenburg, & Vujanovic, 2002) currently used for research purposes only requires the diagnostician-interview to ascertain presence of pathology in four sectors of BPD functioning: affective, behavioral, interpersonal, and cognitive. If a patient meets criteria for less than four sectors of pathology, the BPD diagnosis is not made. Such an approach increases internal consistency, but leaves a large number of patients that are diagnosed with BPD traits or PDNOS (not otherwise specified).

It is also the case that from the clinical perspective not all “types” of clinical presentation of a certain personality disorder that are possible theoretically, actually exist in clinical practice. Individual symptoms do not have an equal base rate and their probability of co-occurrence is contingent on each other. This fact reduces the number of these theoretically possible “types” that are plausibly in existence.

Finally, it is also important to make a distinction between signs and symptoms used to diagnose a condition and actual characteristics of individual patients. In medicine there are plenty of conditions that have a wide range of clinical presentations. However, the numerous presentations do not suggest differences in the actual disorder or differences in the intervention. Probably, the most well-known example is the clinical presentation of heart attacks. Symptoms of heart attacks can vary greatly from person to person and they differ significantly between men and women. Some individuals present with a prototypical picture of chest pain, profuse perspiration, and difficulty breathing, while others will complain only about shoulder or jaw pain. Still others – and many women belong to this category – present without significant symptoms. It is only men who have the “prototypical” clinical presentation, though the treatment for heart attacks will be the same irrespective of variations in clinical presentation. Many chronic conditions in medicine, such as diabetes, hypertension, fibromyalgia, or arthritis present differently from person to person. However, the clinical differences do not suggest a difference in the actual disorder or its treatment. With this in mind, can we hold personality disorder – yet another enduring condition – to the standard of phenomenological heterogeneity? In fact, diagnostic sets were designed to identify the disorder, not to convey everything we know about it. Thus heterogeneity is not posing a threat to conceptual integrity, but invites us to understand the disorder beyond its phenomenological presentation. In fact, in the example of the heart attack, the clinician will administer full treatment for the disorder, regardless of the variations in the clinical presentation and will be targeting the cause, not just the symptoms.

Arbitrary Diagnostic Boundaries

This challenge posed by a critique focused on arbitrary clinical boundaries has to do with two issues. First, the use of polythetic criteria requires a clinician to ascertain the presence of a certain number of criteria to establish a clinical diagnosis. However, the rationale for the cutoff or the number of required criteria, is not empirically established for most personality disorders, though it is suggested for practical use. The notable exceptions are guidelines for thresholds for BPD and StPD in DSM-III, which had an empirical rationale behind them (Spitzer, Endicott, & Gibbons, 1979). For BPD, a threshold of 5 out of 8 criteria was identified based on discrimination between clinically identified BPD patients (n = 234) from the non-BPD patients (n = 808). For StPD, a threshold of 4 out of 8 was identified based on discrimination between patients with the clinical diagnosis of “borderline schizophrenia” (n = 222) from controls (Spitzer et al., 1979). However, despite the revisions in the diagnostic criteria for the subsequent DSM edition, these thresholds remained the same.

The issue of the diagnostic threshold is further complicated by the taxonometric studies that showed that for all but StPD, PD symptoms have a continuous distribution (Edens, Marcus, & Morey, 2009; Edens, Marcus, & Ruiz, 2008; Everett & Linscott, 2015; Rothschild, Cleland, Haslam, & Zimmerman, 2003), thus defying the assumption of the discontinuity from the norm.

The lack of empirical basis for diagnostic thresholds is a common, yet at the same time, certainly an unfortunate occurrence in current diagnostic practice. Heterogeneity of the phenomenological presentation as well as debates as to what constitutes the disorder as opposed to the variability of personality traits complicate testing of this important question. Other features, related to core aspects of the disorder (such as false self or the concept of dependence on others because of the need for admiration) might be difficult to capture empirically, but are more likely to follow a more discontinuous distribution among disorders.

Morey and Skodol (2013) suggested a strategy well suited to diagnostic threshold for the dimensional approach that is similar to one used for the Diagnostic Interview for Borderline Personality Disorder (Zanarini et al., 2002). Accordingly, they identify in what areas the individual is impaired across identity, self-direction, empathy, and intimacy. They suggest requiring a certain number of areas of impairment for the person to meet criteria for a given disorder (e.g., the person must meet criteria for impairment in two out of four areas). They reported that such models have a high degree of sensitivity and specificity (Morey & Skodol, 2013), thus validating an approach to determining thresholds. This approach could be extended to the categorical model by requiring that the person meets criteria from a certain number of domains. For instance, BPD was divided into three sectors or factors – emotional dysregulation, interpersonal dysregulation, and behavioral dysregulation (Sanislow et al., 2002). Thus, it could be required that only those patients that have symptoms in all three sectors meet criteria for BPD. This would require factor division of all PDs. This approach has never been tested, but it has the potential to respond to the criticism of the diagnostic thresholds of PDs.

Inadequate Coverage

Inadequate coverage reflects the fact that there is a substantial number of individuals who meet general criteria for personality disorder, but not criteria for any one of the existing categories of individual personality disorders. In other words, currently described categorical personality diagnoses do not “explain” clinical presentations of a substantially large number of individuals.

Such clinical presentations are assigned the label of not otherwise specified (NOS), which, according to some researchers, is more prevalent than any other individual personality disorder and even any diagnostic category in DSM (Widiger & Mullins-Sweatt, 2005). For instance, a survey of practicing clinicians showed that 60.6 percent of the cases in their practices did not meet criteria for any specific personality disorder, despite meeting the general criteria for personality disorder (Westen & Arkowitz-Westen, 1998). Studies that diagnosed PD directly in the clinical samples reported a lower prevalence. For example, Wilberg and colleagues (Wilberg, Hummelen, Pedersen, & Karterud, 2008) reported that in their sample of patients with any personality disorders, 22 percent met criteria for PDNOS. This could suggest that the actual prevalence of the PDNOS is affected by the accuracy and the systematic nature of the diagnosis of personality disorders. This hypothesis has been confirmed by the finding that in structured interview studies PDNOS is the third most frequently used personality disorder diagnosis, whereas in non-structured interview studies, PDNOS is often the single most frequently used diagnosis (Verheul & Widiger, 2004). In their meta-analysis of 51 studies the absolute prevalence of PDNOS was between 8 and 13 percent (Verheul & Widiger, 2004), though when the absolute prevalence was corrected for the prevalence of all other personality disorders, these figures increased to 21–49 percent (Verheul & Widiger, 2004).

Possible solutions to the challenge of inadequate coverage depend on what we know about the category itself. Unfortunately, beyond reports of prevalence, little has been published regarding the actual composition of this category. While definition of the PDNOS category in terms of general PD criteria leads to higher prevalence, inclusions of the patients that are meeting subthreshold criteria for at least two PDs or having at least ten PD criteria result in lower prevalence rate and include a patient population with higher levels of distress and lower functioning (Wilberg et al., 2008).

When certain PDNOS patients meet subthreshold criteria for two or more PD, these patients can be better understood and treated in the context of existing knowledge of this combination of personality disorders. For patients that meet a wide variety of symptoms without meeting subthreshold criteria for any specific disorders a dimensional approach could be more appropriate (Widiger & Mullins-Sweatt, 2005). It has the power of describing the specific, unique, and atypical clinical presentation of the patients in this category.

Discussion

Categorical models of personality disorders have a long history. These models were influenced by more general philosophical trends of reducing personality to basic components. In this way they antedated the dimensional models of today. However, categorical models were fertile ground for deepening our understanding of individual patients and translating ideographic knowledge into generalized clinical categories. The personality disorders were defined to identify and treat these conditions because of their clinical significance in understanding the patients, explaining atypical clinical presentation and the course of other disorders as well as accounting for risks of suicide, violence, and functional impairment.

Validation of these disorders has made visible progress and supported validity of some of the disorders. Validation of others requires further empirical effort. Studies have validated the relationship between personality disorders and functional impairment, which seems to persist even when the symptoms of the personality disorders remit. The clinical utility of the categorical model has received mixed support, though it has a strong contribution to treatment selection, since most evidence-based treatments are validated in the context of the categorical model. An additional value of the categorical model is its focus on description of separate disorders with specific experiences and processes associated with each one of the disorders. This approach promotes understanding of the individual patient and the use of the initial diagnosis as a starting point to promote further collaboration and alliance building.

Recently, dimensional models are being increasingly suggested as alternatives and in some cases as full replacement for the categorical models. Research demonstrates the value of the dimensional models, although the best prediction is usually provided by a hybrid model that combines categorical and dimensional models. Two models seem to predict somewhat different aspects of the personality disorders and of the functioning. The dimensional model could supplement the categorical model, especially in atypical cases or to characterize PD NOS cases. Moving forward, more research is needed to support further adaptation of the categorical model and to better understand its role in the classification and treatment of personality disorders.

References

Abel, K., Jolley, S., Hemsley, D., & Geyer, M. (2004). The influence of schizotypy traits on prepulse inhibition in young healthy controls. Journal of Psychopharmacology, 18, 181–188.

Aboraya, A., France, C., Young, J., Curci, K., & Lepage, J. (2005). The validity of the psychiatric diagnosis revisited. Psychiatry, 2, 48–55.

Abraham, K. (1927). Selected Papers of Karl Abraham. London: Hogarth Press.

Akhtar, S. (1992). Broken Structures: Severe Personality Disorders and Their Treatment. New York: Aronson.

Baskin-Sommers, A., Kruzermark, E., & Ronningstam, E. F. (2014). Empathy in narcissistic personality disorder: From clinical and empirical perspectives. Personality Disorders, 5, 323–333.

Beck, A. T., Freeman, A., Davis, D. D., & Associates (2006). Cognitive Therapy of Personality Disorders (2nd ed.). New York: Guilford Press.

Bernstein, D. P., Iscan, C., Maser, J., & Board of Directors of the Association for Research of Personality Disorders, International Society for Study of Personality Disorders (2007). Opinions of personality disorder experts regarding the DSM-IV personality disorders classification system. Journal of Personality Disorders, 21, 536–551.

Bertsch, K., Schmidingerm, I., Neuman, I. D., & Herpertz, S. C. (2013). Reduced plasma cortisol levels in female patients with borderline personality disorder. Hormones and Behavior, 63, 424–429.

Black, D. W. (2015). The natural history of antisocial personality disorder. Canadian Journal of Psychiatry, 60, 309–314.

Bodner, E., & Mikulincer, M. (1998). Learned helplessness and occurrence of depressive-like and paranoid-like responses: The role of attentional focus. Journal of Personality and Social Psychology, 74, 1019–1023.

Bornstein, R. F. (1993). The Dependent Personality. New York: Guilford Press.

Bornstein, R. F. (1998). Reconceptualizing personality disorder diagnosis in the DSM-V: The discriminant validity challenge. Clinical Psychology, 5, 333–343.

Bornstein, R. F., Ng, H. M., Gallagher, H. R., Kloss, D. M., & Regier, N. G. (2005). Contrasting effects of self-schema priming on lexical decisions and interpersonal Stroop task performance: Evidence for a cognitive/interactionist model of interpersonal dependency. Journal of Personality, 73, 731–761.

Chanen, A. M., Jackson, H. J., McGorry, P. D., Allot, K. A., Clarkson, V., & Yuen, H. P. (2004). Two-year stability of personality disorders in older adolescent outpatients. Journal of Personality Disorders, 18, 526–541.

Chemerinski, E., Triebwasser, J., Roussos, P., & Siever, L. J. (2013). Schizotypal personality disorder. Journal of Personality Disorders, 27, 652–679.

Chen, H., Cohen, P., Crawford, T. N., Kasen, S., & Johnson, J. G. (2006). Relative impact of young adult personality disorders on subsequent quality of life: Findings of a community-based longitudinal study. Journal of Personality Disorders, 20, 510–523.

Cloninger, C. R., Svrakic, D. M., & Przybeck, T. R. (1993). A psychobiological model of temperament and character. Archives of General Psychiatry, 50, 975–990.

Costa, P. T., & McCrae, R. R. (1985). The NEO Personality Inventory Manual. Odessa, FL: Psychological Assessment Resources.

Costa, P. T., & Widiger, T. (Eds.) (1993). Personality Disorders and the Five-Factor Model of Personality. Washington, DC: American Psychological Association.

Crawford, T. N., Cohen, P., First, M. B., Skodol, A. E., Johnson, J. G., & Kasen, S. (2008). Comorbid Axis I and Axis II disorders in early adolescence: Outcomes 20 years later. Archives of General Psychiatry, 65, 641–648.

Crego, C., Sleep, C. E., & Widiger, T. A. (2016). Clinicians’ judgments of the clinical utility of personality disorder trait descriptions. Journal of Nervous and Mental Disease, 204, 49–56.

Croft, R., Lee, A., Bertolot, J., & Gruzelier, J. (2001). Association of P50 suppression and desensitization with perceptual and cognitive features of “unreality” in schizotypy. Biological Psychiatry, 50, 441–446.

Diedrich, A., & Voderholzer, U. (2015). Obsessive-compulsive personality disorder: A current review. Current Psychiatry Reports, 17, 1–10.

Edens, J. F., Marcus, D. K., & Morey, L. C. (2009). Paranoid personality has a dimensional latent structure: Taxometric analyses of community and clinical samples. Journal of Abnormal Psychology, 118, 545–553.

Edens, J. F., Marcus, D. K., & Ruiz, M. A. (2008). Taxometric analyses of borderline personality features in a large-scale male and female offender sample. Journal of Abnormal Psychology, 117, 705–711.

Everett, K. V., & Linscott, R. J. (2015). Dimensionality vs. taxonicity of schizotypy: Some new data and challenges ahead. Schizophrenia Bulletin, 41(Suppl. 2), S465–S474.

Fairbairn, W. R. D. (1952). Psychoanalytic Studies of Personality. London: Tavistock.

Fenton, W., & McGlashan, T. (1989). Risk of schizophrenia in character disorder patients. American Journal of Psychiatry, 146, 1280–1284.

Ferenczi, S. (1938). Thalassa: A Theory of Genitality. New York: Norton.

Ficks, C. A., & Waldman, I. D. (2014). Candidate genes for aggression and antisocial Peele, R. (2004). Clinical utility as a criterion for revisiting psychiatric diagnoses. American Journal of Psychiatry, 161, 946–954.

Freud, S. (1908). Character and anal erotism. In Collected Papers (Vol. 2). London: Hogarth Press.

Gjerde, L. C., Czajkowski, N., Røysamb, E., Ørstavik, R. E., Knudsen, G. P., Østby, K., … Reichborn-Kjennerud, T. (2012). The heritability of avoidant and dependent personality disorder assessed by personal interview and questionnaire. Acta Psychiatrica Scandinavica, 126, 448–457.

Grant, B. F., Chou, S. P., Goldstein, R. B., Huang, B., Stinson, F. S., Saha, T. D., … Ruan, W. J. (2008). Prevalence, correlates, disability, and comorbidity of DSM-IV borderline personality disorder: Results from the wave 2 National Epidemiologic Survey of Alcohol and Related Conditions. Journal of Clinical Psychiatry, 69, 533–545.

Grant, B. F., Hasin, D. S., Stinson, F. S., Dawson, D. A., Chou, S. P., Ruan, W. J., & Pickering, R. P. (2004). Prevalence, correlates, and disability of personality disorders in the United States: Results from the National Epidemiologic Survey of Alcohol and Related Disorders. Journal of Clinical Psychiatry, 65, 948–958.

Grant, B. F., Moore, T. C., Shepard, J., & Kaplan, K. (2003). Source and Accuracy Statement: Wave 1 National Epidemiological Survey on Alcohol and Related Conditions (NESARC). Bethesda, MD: National Institute on Alcohol Abuse and Alcoholism.

Gunderson, J. G. (1984). Borderline Personality Disorder. Washington, DC: American Psychiatric Press.

Gunderson, J. G. (1992). Diagnostic controversies. In A. Tasman & M. B. Riba (Eds.), American Psychiatric Press Review of Psychiatry (Vol. 11, pp. 9–24). Washington, DC: American Psychiatric Press.

Gunderson, J. G., Herpertz, S., Skodol, A. E., Torgersen, S., & Zanarini, M. C. (2017). Borderline personality disorder. Nature Reviews Disease Primers, 4, 1–20.

Gunderson, J. G., & Ronningstam, E. F. (2001). Differentiating antisocial and narcissistic personality disorder. Journal of Personality Disorders, 15, 103–109.

Gunderson, J. G., Stout, R. L., McGlashan, T. H., Shea, M. T., Morey, L. C., Grilo, C. M., … Skodol, A. E. (2011). Ten-year course of borderline personality disorder: Psychopathology and function from the Collaborative Longitudinal Personality Disorders Study. Archives of General Psychiatry, 68, 827–837.

Gunderson, J. G., Weinberg, I., & Choi-Kain, L. (2013). Borderline personality disorder. In G. O. Gabbard (Ed.), Gabbard’s Treatments of Psychiatric Disorders (DSM-5 ed.). Washington, DC: American Psychiatric Publishing.

Guntrip, H. (1968). Schizoid Phenomena, Object Relations, and the Self. London: Karnac.

Hirt, E. (1902). Die temperamente. Leipzig: Barth.

Holt, R. R. (1970). Yet another look at clinical and statistical prediction: Or, is clinical psychology worthwhile? American Psychologist, 25, 337–349.

Hong, J. P., Samuels, J., Bienvenu, O. J., Hsu, F. C., Eaton, W. W., & Costa, P. T., Jr. (2005). The longitudinal relationship between personality disorder dimensions and global functioning in a community-residing population. Psychological Medicine, 35, 891–895.

Horney, K. (1937). The Neurotic Personality of Our Time. New York: Norton.

Horwitz, M. J. & Lerner, U. (2010). Treatment of histrionic personality disorder. In J. F. Clarkin, P. Fonagy, & G. Gabbard (Eds.), Psychodynamic Psychotherapy for Personality Disorders (pp. 289–310). Washington, DC: American Psychiatric Publishing.

Johnson, J. G., First, M. B., Cohen, P., Skodol, A. E., Kasen, S., & Brook, J. S. (2005). Adverse outcomes associated with personality disorders not otherwise specified in a community sample. American Journal of Psychiatry, 162, 1926–1932.

Kernberg, O. F. (1975). Borderline Conditions and Pathological Narcissism. New York: Jason Aronson.

Kessler, R. C., McGonagle, K. A., Zhao, S., Nelson, C. B., Hughes, M., Eshleman, S., … Kendler, K. S. (1994). Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States: Results from the National Comorbidity Survey. Archives of General Psychiatry, 51, 8–19.

Klein, M. H. (1948). Contributions to Psychoanalysis, 1921–1945. London: Hogarth Press.

Koenigsberg, H. W., Denny, B. T., Fan, J., Liu, X., Guerreri, S., Mayson, S. J., … Siever, L. J. (2014). The neural correlates of anomalous habituation to negative emotional pictures in borderline and avoidant personality disorder patients. American Journal of Psychiatry, 171, 82–90.

Kohut, H. (1971). The Analysis of the Self. New York: International Universities Press.

Kraepelin, E. (1919). Dementia Praecox and Paraphrenia. Edinburgh: Livingstone.

Kretchmer, E. (1925). Korperbau und character. Berlin: Springer Verlag.

Lenzenweger, M. F., Lane, M. C., Loranger, A. W., & Kessler, R. C. (2007). DSM-IV personality disorders in the National Comorbidity Survey Replication. British Journal of Psychiatry, 62, 533–564.

Lilienfield, S. O., Waldman, I. D., & Israel, A. C. (1994). A critical examination of the issue of the term “comorbidity” in psychopathology research. Clinical Psychology, 1, 71–83.

Links, P. S. (1993). Psychiatric rehabilitation model for borderline personality disorder. Canadian Journal of Psychiatry, 38(Suppl. 1), S35–S38.

Liu, H., Liao, J., Jiang, W., & Wang, W. (2014). Changes in low-frequency fluctuations in patients with antisocial personality disorder revealed by resting-state functional MRI. PLoS ONE, 9, e89790.

Livesley, W. J., West, M., & Tanney, A. (1986). Historical comment on DSM-III schizoid and avoidant personality disorders. American Journal of Psychiatry, 142, 1344–1346.

Ma, G., Fan, H., Shen, C., & Wang, W. (2016). Genetic and neuroimaging features of personality disorders: State of the art. Neuroscience Bulletin, 32, 286–306.

Martín-Blanco, A., Ferrer, M., Soler, J., Arranz, M. J., Vega, D., Calvo, N., … Pascual, J. C. (2015). The role of hypothalamus-pituitary-adrenal genes and childhood trauma in borderline personality disorder. European Archives of Psychiatry and Clinical Neuroscience, 266, 307–316.

McGlashan, T. H., Grilo, C. M., Sanislow, C. A., Ralevski, E., Morey, L. C., Gunderson, J. G., … Pagano, M. (2005). Two-year prevalence and stability of individual DSM-IV criteria for schizotypal, borderline, avoidant, and obsessive-compulsive personality disorders: Toward a hybrid model of Axis II disorders. American Journal of Psychiatry, 162, 883–889.

Meloy, J. R., & Yakeley, J. (2010). Psychodynamic treatment of antisocial personality disorder. In J. F. Clarkin, P. Fonagy, & G. Gabbard (Eds.), Psychodynamic Psychotherapy for Personality Disorders (pp. 311–336). Washington, DC: American Psychiatric Publishing.

Minzenberg, M., Xu, K., & Mitrolopou, V. (2006). Catechol-O-methyltransferase Val158Met genotype variation is associated with prefrontal-dependent task performance in schizotypal personality disorder patients and comparison groups. Psychiatric Genetics, 16, 117–124.

Morey, L. C., Benson, K. T., Busch, A. J., & Skodol, A. E. (2015). Personality disorders in DSM-5: Emerging research on the alternative model. Current Psychiatric Reports, 17, 1–9.

Morey, L. C., Hopwood, C. J., Gunderson, J. G., Shea M. T., Skodol, A. E., Grilo, C. M., … McGlashan, T. H. (2007). A comparison of personality disorder models. Psychological Medicine, 37, 983–994.

Morey, L. C., Hopwood, C. J., Markowitz, J. C., Gunderson, J. G., Grilo, C. M., McGlashan, T. H., … Skodol, A. E. (2012). Comparisons of alternative models of personality disorders, II: 6-, 8-, and 10-year follow-up. Psychological Medicine, 42, 1705–1713.

Morey, L. C., & Skodol, A. E. (2013). Convergence between DSM-IV-TR and DSM-V diagnostic models for personality disorders: Evaluation of strategies for establishing diagnostic thresholds. Journal of Psychiatric Practice, 19, 179–193.

Morey, L. C., Skodol, A. E., & Oldham, J. M. (2014). Clinician judgment of clinical utility: A comparison of DSM-IV-TR personality disorders and the alternative model for DSM-V personality disorders. Journal of Abnormal Psychology, 123, 398–405.

Nestadt, G., Samuels, J. F., Bienvenu, O. J., Reti I. M., Costa, P., Eaton, W. W., & Bandeen-Roche, K. (2010). The stability of the DSM personality disorders over ten to eighteen years. Journal of Psychiatric Research, 44, 1–7.

Oldham, J. M., Skodol, A. E., Kellman, H. D., Hyler, S. E., Rosnick, L., & Davies, M. (1992). Diagnosis of DSM-III-R personality disorders by two structured interviews: Patterns of comorbidity. American Journal of Psychiatry, 149, 213–220.

Penner-Goeke, K., Henrikson, C. A., Chateau, D., Latimer, E., Sareen, J., & Katz, L. Y. (2015). Reduction in quality of life associated with common mental disorders: Results from a national representative sample. Journal of Clinical Psychiatry, 76, 1506–1512.

Plakun, E. M. (1990). Narcissistic personality disorder: A validity study and comparison to borderline personality disorder. Psychiatric Clinics of North America, 12, 603–620.

Pokorny, A. D. (1983). Prediction of suicide in psychiatric patients: Report of a prospective study. Archives of General Psychiatry, 40, 249–257.

Pulay, A. J., Stinson, F. S., Dawson, D. A., Goldstein, R. B., Chou, S. P., Huang, B., … Grant, B. F. (2009). Prevalence, correlates, disability, and comorbidity of DSM-IV schizotypal personality disorder: Results from the wave 2 National Epidemiologic Survey of Alcohol and Related Conditions. Primary Care Companion to the Journal of Clinical Psychiatry, 11, 53–67.

Rappaport, D. (1961). The Collected Papers of David Rappaport. New York: Basic Books.

Reich, W. (1933). Character Analysis. New York: Farrar & Rinehart.

Robins, E., & Guze, S. B. (1970). Establishment of diagnostic validity in psychiatric illness: Its application to schizophrenia. American Journal of Psychiatry, 126, 983–987.

Ronningstam, E. (2011). Narcissistic personality disorder in DSM-V: In support of retaining a significant diagnosis. Journal of Personality Disorders, 25, 248–259.

Ronningstam, E. F. (2013). The update on narcissistic personality disorder. Current Opinion in Psychiatry, 26, 102–106.

Ronningstam, E. F., Gunderson, J. G., & Lyons, M. (1995). Changes in pathological narcissism. American Journal of Psychiatry, 152, 253–257.

Rothschild, L., Cleland, C., Haslam, N., & Zimmerman, M. (2003). A taxometric study of borderline personality disorder. Journal of Abnormal Psychology, 112, 657–666.

Rubovzki, G., Gunderson, J. G., & Weinberg, I. (2006). Patients’ acceptance and emotional reactions to disclosure of borderline personality disorder diagnosis. Paper presented at the APA Conference, Toronto, Canada.

Salekin, R. T. (2002). Psychopathy and therapeutic pessimism: Clinical lore or clinical reality? Clinical Psychology Review, 22, 79–112.

Sanislow, C. A., Grilo, C. M., Morey, L. C., Bender, D. S., Skodol, A. E., Gunderson, J. G., … McGlashan, T. H. (2002). Confirmatory factor analysis of DSM-IV criteria for borderline personality disorder: Findings from the collaborative longitudinal personality disorders study. American Journal of Psychiatry, 159, 284–290.

Schneider, K. (1923). Psychopathic Personalities (9th ed.). London: Cassell.

Schulze, L., Dziobek, I., Vater, A., Heekeren, H. R., Bajbouj, M., Renneberg, B., … Roepke, S. (2013). Grey matter abnormalities in patients with narcissistic personality disorder. Journal of Psychiatric Research, 47, 1363–1369.

Siever, L., Amin, F., Coccaro, E., Trestman, R. Silverman, J., & Hovath, T. (1996). Schizotypal personality disorder. In T. Widiger, A. Frances, & H. Pincus (Eds.). DSM-IV Sourcebook (vol. 2, pp. 685–701). Washington, DC: American Psychiatric Association.

Siever, L. J., Coursey, R. D., Alterman, I. S., Zahn, T., Brody, L., Bernad, P., … Murphy, D. L. (1989). Clinical, psychophysiological, and neurological characteristics of volunteers with impaired smooth pursuit eye movements. Biological Psychiatry, 26, 35–51.

Siever, L. J., & Davis, K. L. (1991). A psychobiological perspective on the personality disorders. American Journal of Psychiatry, 148, 1647–1658.,

Skodol, A. E. (2018). Impact of personality pathology on psychosocial functioning. Current Opinion in Psychology, 21, 33–38.

Skodol, A. E., Bender, D. S., Gunderson, J. G., & Oldham, J. M. (2014). Personality disorders. In R. E. Hales, S. C. Yudofsky, & L. Roberts (Eds.), American Psychiatric Publishing Textbook of Psychiatry (6th ed., pp. 851–894). Washington, DC: American Psychiatric Publishing.

Skodol, A. E., Gunderson, J. G., McGlashan, T. H., Dyck, I. R., Stout, R. L., Bender, D. S., … Oldham, J. M. (2002). Functional impairment in patients with schizotypal, borderline, avoidant, or obsessive-compulsive personality disorder. American Journal of Psychiatry, 159, 276–283.

Skodol, A. E., Johnson, J. G., Cohen, P., Sneed, J. R., & Crawford, T. N. (2007). Personality disorder and impaired functioning from adolescence to adulthood. British Journal of Psychiatry, 190, 415–420.

Skodol, A. E., Oldham, J. E., Bender, D. S, Dyck, I. R., Stout, R. L., Morey, L. C., … Gunderson, J. G. (2005). Dimensional representations of DSM-IV personality disorders: Relationship to functional impairment. American Journal of Psychiatry, 162, 1919–1925.

Spitzer, R. L., Endicott, J., & Gibbons, M. (1979). Crossing the border into the borderline personality and borderline schizophrenia: The development of criteria. Archives of General Psychiatry, 36, 17–24.

Stefanis, N. C., Trikalinos, T. A., Avramopoulos, D., Smyrnis, N., Evdokimidis, I., Ntzani, E. E., … Stefanis, C. N. (2007). Impact of schizophrenia candidate genes on schizotypy and cognitive endophenotypes at the population level. Biological Psychiatry, 62, 784–792.

Stinson, F. S., Dawson, D. A., Goldstein, R. B., Chou, S. P., Huang, B., Smith, S. M., … Grant, B. F. (2008). Prevalence, correlates, disability, and comorbidity of DSM-IV narcissistic personality disorder: Results from the Wave 2 National Epidemiologic Survey on Alcohol and Related Conditions. Journal of Clinical Psychiatry, 69, 1033–1045.

Stone, M. (1980). The Borderline Syndromes. New York: McGraw-Hill.

Stone, M. H. (2013). Paranoid, schizotypal, and schizoid personality disorders. In G. O. Gabbard (Ed.), Gabbard’s Treatments of Psychiatric Disorders (DSM-5 ed.). Washington, DC: American Psychiatric Publishing.

Svartberg, M., & McCullough, L. (2010). Cluster C personality disorders: Prevalence, phenomenology, treatment effects, and principles of treatment. In J. F. Clarkin, P. Fonagy, & G. Gabbard (Eds.), Psychodynamic Psychotherapy for Personality Disorders (pp. 337–368). Washington, DC: American Psychiatric Publishing.

Torgersen, S., Lygren, S., Oien, P. A., Skre, I., Onstad, S., Edvardsen, J., … Kringlen, E. (2000). A twin study of personality disorders. Comprehensive Psychiatry, 41, 416–425.

Torgersen, S., Myers, J., Reichborn-Kjennerud, T., Roysamb, E., Kubarych, T. S., & Kendler, K. S. (2012). The heritability of Cluster B personality disorder assessed both by personal interview and questionnaire. Journal of Personality Disorders, 26, 848–866.

Trestman, R. L., Horvath, T., Kalus, O., Peterson, A. E., Coccaro, E., Mitropoulou, V., … Siever, L. J. (1996). Event-related potentials in schizotypal personality disorder. Journal of Neuropsychiatry and Clinical Neuroscience, 8, 33–40.

Triebwasser, J., Chemerinski, E., Roussos, P., & Siever, L.J. (2013). Paranoid personality disorder. Journal of Personality Disorders, 27, 795–805.

Trull, T. J., Jahng, S., Tomko, R. L., Wood, P. K. & Sher, K. J. (2010). Revised NESARC personality disorder diagnoses: Gender, prevalence, and comorbidity with substance use disorders. Journal of Personality Disorders, 24, 412–426.

Verheul, R., & Widiger, T. A. (2004). A meta-analysis of the prevalence and usage of the personality disorder not otherwise specified (PDNOS) diagnosis. Journal of Personality Disorders, 18, 309–319.

Weinbrecht, A., Schulze, L., Boetther, J., & Renneberg, B. (2013). Avoidant personality disorder. Current Psychiatric Reports, 18, 1–8.

Westen, D., & Arkowitz-Westen, L. (1998). Limitations of Axis II in diagnosing personality pathology in clinical practice. American Journal of Psychiatry, 155, 1767–1771.

Widiger, T. A., & Mullins-Sweatt, S. N. (2005). Categorical and dimensional models of personality disorders. In J. M. Oldham, A. E. Skodol, & D. S. Bender (Eds.), The American Psychiatric Publishing Textbook of Personality Disorders (pp. 35–53) Washington, DC: American Psychiatric Publishing.

Widiger, T. A., & Sanderson, C. J. (1995). Towards a dimensional model of personality disorders in DSM-IV and DSM-V. In W. J. Livesley (Ed.), The DSM-IV Personality Disorders (pp. 380–394). New York: Guilford Press.

Wilberg, T., Hummelen, B., Pedersen, G., & Karterud, S. (2008). A study of patients with personality disorder not otherwise specified. Comprehensive Psychiatry, 49, 460–468.

Williams, P. (2010). Psychotherapeutic treatment of Cluster A personality disorders. In J. F. Clarkin, P. Fonagy, & G. Gabbard (Eds.), Psychodynamic Psychotherapy for Personality Disorders (pp. 165–186). Washington, DC: American Psychiatric Publishing.

Winnicott, D. W. (1965). The Maturational Process and the Facilitating Environment. New York: International Universities Press.

Yang, Y., & Raine, A. (2009). Prefrontal structural and functional brain imaging findings in antisocial, violent, and psychopathic individuals: A meta-analysis. Psychiatric Research, 17, 81–88.

Young, J. E., Klosko, J. S., & Weishaar, M. E. (1993). Schema Therapy: A Practitioner’s Guide. New York: Guilford Press.

Zanarini, M. C., & Frankenburg, F. R. (2008). A preliminary, randomized trial of psychoeducation for women with borderline personality disorder. Journal of Personality Disorders, 22, 284–290.

Zanarini, M. C., Frankenburg, F. R., Hennen, J. J., Reich, D. B., & Silk, K. R. (2005). Psychosocial functioning of borderline patients and Axis II comparison subjects followed prospectively for six years. Journal of Personality Disorders, 19, 19–29.

Zanarini, M. C., Frankenburg, F. R., Reich, D. B., & Fitzmaurice, G. (2010). The 10-year course of psychosocial functioning among patients with borderline personality disorder and Axis II comparison subjects. Acta Psychiatrica Scandinavica, 122, 103–109.

Zanarini, M. C., Frankenburg, F. R., & Vujanovic, A. A. (2002). Inter-rater and test-retest reliability of the diagnostic interview for borderlines. Journal of Personality Disorders, 16, 270–276.

Zanarini, M. C., Skodol, A. E., Bender, D. S., Dolan, R., Sanislow, C. A., Schaefer, E., … Gunderson, J. G. (2000). The Collaborative Longitudinal Personality Disorders Study: Reliability of Axis I and II diagnoses. Journal of Personality Disorders, 14, 291–299.

Zimbardo, P. G., Andersen, S. M., & Kabat, L. G. (1981). Induced hearing deficit generates experimental paranoia. Science, 212, 1529–1531.