Chapter 20 Christopher J. Hopwood and Katherine M. Thomas Personality disorders (PDs) represent a major public health concern, as more than 1 in 10 adults in the community meet the diagnostic criteria for at least one PD (Torgersen, 2005; Torgersen, Kringlen, & Cramer, 2001) and PD diagnoses are associated with increased risk for hospitalization (Bender et al., 2001), criminal behavior (Johnson et al., 2000), suicidal behavior (Soloff, Lis, Kelly, & Cornelius, 1994), and dysfunction at work and in relationships (Grant et al., 2004; Skodol et al., 2002; Torgersen, 2005). Relatively few evidence-based treatments are available for PDs (Matusiewicz, Hopwood, Banducci, & Lejuez, 2010), which are notoriously difficult to treat and interfere with treatment of other kinds of disorders (Cyranowski et al., 2004; Feske et al., 2004; Reich, 2003). The wide-ranging clinical importance of normative personality traits thought to predispose personality and other forms of psychopathology is also well-documented (e.g., Lahey, 2009; Ozer & Benet-Martinez, 2006; Roberts, Kuncel, Shiner, Caspi, & Goldberg, 2007). For instance, Cuijpers et al. (2010) estimated that the direct and indirect medical costs for individuals in the top 5% of neuroticism scores are $12,362 per person per year, compared with $7,851 for individuals with mood disorders and $3,641 for the average person. Despite the clinical significance of personality and personality pathology, representing PDs in a manner that is scientifically valid and clinically useful continues to pose a major challenge. In the run-up to the publication of DSM-5, the view that the DSM-III and DSM-IV model of PDs that has guided recent research and practice is substantially flawed has been commonly advanced, and many alternatives have been suggested (Bornstein, 1997; Clark, 2007; Hopwood et al., 2011; Krueger, Eaton, Derringer, et al., 2011; Westen, Shedler, & Bradley, 2006; Widiger & Mullins-Sweat, 2009). Although it seemed that changes to the PDs in the DSM-5 would be dramatic (Bender, Morey, & Skodol, 2011; Krueger, Eaton, Clark, et al., 2011; Skodol et al., 2011), in the end the Board of Trustees of the American Psychiatric Association voted to retain the DSM-IV system in DSM-5. However, they also included an alternative formulation in DSM-5, Section III. The new framework has the potential to increase clinical utility and efficiency, more closely link the PDs with evidence-based models of personality and personality pathology, and ultimately lead to an increased focus on personality and PD among clinicians and scholars, with a corresponding improvement in practice and research. However, the proposed DSM-5 model is not without critics, who have expressed concerns both about its clinical utility (Bornstein, 2011b; Clarkin & Huprich, 2011; Shedler et al., 2010) and its evidentiary basis (Samuel, 2011; Widiger, 2011; Zimmerman, 2011). Addressing these concerns is the next major step in PD research, as this will facilitate the migration of an improved system, like the one in Section III, into the official diagnostic nomenclature. Nevertheless, given that DSM-5 PDs are essentially DSM-IV PDs, in this chapter we focus on the DSM-IV model. We begin this chapter by discussing several concepts relevant to the definition of personality pathology and PD. We next describe the DSM-IV PDs in a historical context, review research on the prevalence, etiology, and course of personality pathology, and describe several approaches to its assessment. We close with a review of the DSM-5 proposal, highlighting differences between the DSM-IV and DSM-5 approaches to PD diagnosis in the context of a clinical case. Personality is a broad concept with considerable room for theoretical variation in terms of emphasis on different, even nonoverlapping, components. One way to organize divergent theoretical perspectives is to differentiate those aspects of personality on which they focus. For instance, a protracted rivalry exists in academic personality psychology between those who focus on its more stable or dynamic aspects that extend to the problem of how to classify personality pathology (Wright, 2011). There are also vigorous debates between those who would prefer a more conservative approach to diagnosis involving the retention of categorical PD constructs (e.g., Bornstein, 2011a; Gunderson, 2010; Shedler et al., 2010) and those who would overhaul the model more dramatically by utilizing a completely dimensional approach (e.g., Widiger & Mullins-Sweat, 2009; Krueger et al., 2011). The DSM-III and DSM-IV circumvented conceptual debates somewhat by focusing on atheoretical descriptions of personality pathology. The DSM-IV defines PD as “an enduring pattern of inner experience and behavior that deviates markedly from the expectations of the individual’s culture, is pervasive and inflexible, has an onset in adolescence or early adulthood, is stable over time, and leads to distress or impairment” (APA, 2000, p. 685). The DSM-IV describes 10 instantiations of PD, organized into three clusters: Cluster A: schizotypal, schizoid, and paranoid; Cluster B: antisocial, borderline, histrionic, and narcissistic; and Cluster C: avoidant, dependent, and obsessive-compulsive. These specific PDs, their clustering, and their criteria were selected as much based on clinical legacy in the medical and psychoanalytic perspectives on personality and the collective wisdom of a specific group of committee members with diverse theoretical perspectives, as they were for their empirical support or conceptual coherence (Widiger, 1993). While the atheoretical approach of the DSM-III and DSM-IV had many advantages, notably its apparently having affected a dramatic increase in PD research, it has become clear that important conceptual issues must be addressed more directly for PD classification to move forward. To clarify some of these conceptual issues, we begin by distinguishing four broad domains of personality that are relevant to debates on the nature of personality and related pathology: traits, dynamics, pathology, and disorders. These definitions provide a framework for discussing various perspectives on PD classification in the remainder of this chapter. Personality traits are enduring features of personality that are (a) cultural universal (McCrae & Terracciano, 2005); (b) heritable (Jang, Livesley, & Vernon, 1996); (c) linked to specific neurobiological structures (DeYoung, 2010) and pathways (Depue & Lenzenweger, 2005); (d) well-characterized in terms of content and course (Soldz & Valliant, 2002); (e) valid for predicting a host of important life outcomes (Roberts et al., 2007); and (f) amenable to reliable assessment, particularly via self-report questionnaires (Samuel & Widiger, 2006). The Five-Factor Model (FFM) currently represents the most viable model of normative personality traits, having the advantage of decades of empirical justification (Digman, 1990) and extensive theoretical articulation (Costa & McCrae, 2006). In the FFM, five normally distributed traits represent the broadest level of variation in personality: neuroticism, extraversion, openness to experience, agreeableness, and conscientiousness. These traits provide a context for all human behavior, and certain constellations of these traits make personality pathology in general and certain forms of PD more or less likely (Morey et al., 2002; Samuel & Widiger, 2008; Saulsman & Page, 2004). From the perspective of the DSM-IV, personality traits such as those of the FFM are enduring features with broad implications for behavior across many situations. These features become clinically relevant when they are “inflexible and maladaptive and cause significant functional impairment or subjective distress” (APA, 2000, p. 686). This distinction implies a discontinuity between traits and their maladaptive consequences under conditions of inflexibility and extremity. However, an explicit linkage between traits and associated dysfunction invites a dimensional view of personality pathology, in which differences between trait and PD are quantitative rather than qualitative. Indeed, evidence suggests that the cluster model of the DSM-IV is not valid (Lenzenweger & Clarkin, 2005) and that the disorders are overlapping blends of polythetic and potentially common traits (Widiger et al., 1991), suggesting scientific advantages to viewing personality pathology dimensionally. These potential advantages have led some theorists to suggest defining PDs as reflecting extreme or maladaptive variants of normative traits (Widiger, 1993). However, several research findings prescribe pause in equating personality pathology with extreme scores on normative traits (Hopwood, 2011). First, relations between normative traits and PDs are not unique or special; normative traits relate systematically to most forms of psychopathology (Cuijpers et al., 2010; Kotov, Gomez, Schmidt, & Watson, 2010), just as they relate to a wide variety of individual differences in human behavior (Ozer & Benet-Martinez, 2006). Second, there are potential structural differences between normal and pathological personality traits (Krueger, Eaton, Derringer, et al., 2011). Third, personality traits and disorders can be distinguished empirically in terms of stability and incremental validity (Hopwood & Zanarini, 2010; Morey et al., 2012; Morey & Hopwood, 2013; Morey & Zanarini, 2000). Stable trait concepts are also limited for conceptualizing dynamic and contingent aspects of social behavior and emotional experiences related to personality and related pathology, which we describe next. Mischel’s 1968 text Personality and Assessment initiated a major conflict in personality psychology between those who view personality as made up of stable traits and those who see personality primarily as a function of situational contingencies (see Journal of Research in Personality, 43[2]). Although it has been difficult and contentious, this debate has had several positive consequences. With regard to personality stability, trait psychologists were prompted to develop methods that could more convincingly show that traits can be reliably assessed, are stable, and are valid predictors of important behaviors. Equally relevant to contemporary models of PD, this debate led to the development of new models for understanding how traits and situations interact in if…then behavioral signatures (Mischel & Shoda, 1995). For instance, if a conscientious person is on the clock, then she will typically attend to her work. This logic extends to PD (Wright, 2011): If a person with borderline PD is exposed to rejection, then he will tend to react in a self-damaging manner. Notably, this formulation, while inconsistent with a simple trait-behavior formulation of personality (Clarkin & Huprich, 2011), is consonant with several previous models, including Lewin’s (1936) classic equation that personality is a function of the person and the environment, or object-relations models that assert that behavior is influenced by the elicitation of self-other dyad units by the parameters of current, actual social situations (Kernberg & Caligor, 2005). As such, the concept of the behavioral signature both reemphasizes and builds upon a long tradition in personality and clinical psychology that emphasizes the moderation of trait-relevant behavior by situational contingencies. Understanding dynamic elements of personality may be particularly important given recent research suggesting that PD symptoms vary in their stabilities (McGlashan et al., 2005). However, research aimed at conceptualizing the behavioral signatures associated with PDs is just beginning. Most of the work in this area has involved borderline PD, likely because intraindividual variability in emotion and behavior is thematic of the disorder (Schmideberg, 1959). Research using ecological momentary assessment—in which assessments occur several times per day over several days—has generally showed increased variability in mood, interpersonal behavior, and self-esteem among borderline individuals (Sadikaj, Russell, Moskowitz, & Paris, 2010; Trull et al., 2008; Zeigler-Hill & Abraham, 2006). Ongoing and future work exploring the intraindividual dynamics associated with particular kinds of personality problems will be important for assimilating dynamic behavioral signatures into conceptualizations of personality pathology and PD. Several PD theorists have distinguished defining features of personality pathology from the stylistic manifestations of personality disorders (Bornstein, 2011a; Kernberg, 1984; Livesley, 1998; Pincus, 2005; Pincus & Hopwood, 2012). From this perspective, personality pathology indicates whether a person has a clinically significant PD diagnosis as well as the overall level or severity of personality-related dysfunction, whereas PDs (described later) reflect symptom constellations that vary across individuals, independent of the severity of their overall personality pathology. For example, the DSM-IV distinguishes the defining characteristics of PD in general from symptom criteria for 10 specific PD types. However, in the DSM-IV model, personality pathology is not quantified, and the PD symptoms conflate aspects of pathological severity and its stylistic manifestations. This conflation likely contributes to unnecessarily high comorbidity among the PDs (Parker et al., 1998). Research supports the distinction between personality pathology and stylistic aspects of PDs. Parker et al. (2004) derived two higher-order factors from an assessment of the basic elements of personality pathology, which they labeled cooperativeness (ability to love) and coping (ability to work). These factors correlated nonspecifically with PDs and differentiated clinical and nonclinical samples. Hopwood et al. (2011) factor-analyzed PD symptoms after variance in each symptom associated with a general pathology factor was removed. The severity composite explained most of the variance in functional outcomes, but the five stylistic dimensions, which were labeled peculiarity, deliberateness, instability, withdrawal, and fearfulness, incremented this composite for predicting several specific outcomes. Importantly, these stylistic dimensions were completely independent of the overall level of personality pathology severity and were mostly independent of normative traits. Morey et al. (2011) assessed personality pathology with items from questionnaires designed to assess global personality dysfunction. By refining these item sets using a host of psychometric procedures, they showed, in two large and diverse samples, that greater severity was associated with greater likelihood of PD diagnosis and higher rates of comorbidity. Delineating stylistic aspects of PD would represent a significant challenge even if global personality pathology were effectively separated from PD features. The DSM-IV proposes 10 PDs, but given the conflation of severity and style in the DSM-IV (Parker et al., 1998), it is possible that fewer than 10 stylistic dimensions would be sufficient for depicting the stylistic variability in PD expression. The Hopwood et al. (2011) study discussed previously identified five reliable PD dimensions, although that study was constrained by the DSM-IV content that was factor-analyzed. To the degree that the DSM-IV symptom criteria are not comprehensive (e.g., they do not include symptoms from appendicized diagnoses and may not fully capture the content of some disorders [e.g., Pincus, Wright, Hopwood, & Krueger, 2011]), it is possible that other important dimensions exist. Thus, an important question for ongoing research is: How many PD dimensions are there? We will discuss this issue in more detail, but for now we suggest that decisions about what PDs to keep or remove and whether PDs should be conceptualized as polythetic syndromes or trait constellations should be based on quantitative evidence. Given that evidence is currently insufficient for such decisions, contemporary decisions about how to slice up the stylistic variance in PD will be necessarily questionable and temporary. For this reason, it is wise that the DSM-5 is being conceived of as a living document subject to ongoing empirical and conceptual refinement (Krueger, Eaton, Derringer, et al., 2011). From our perspective, the process for making decisions about which PD constructs are sufficiently valid for routine clinical consideration should be regarded as a psychometric matter (Loevinger, 1957). The PDs listed in the DSM-IV and its appendix offer a reasonable starting point as a list of hypothetical constructs in the PD domain (MacCorquodale & Meehl, 1948). The first step in establishing their construct validity would involve describing their theoretical contents thoroughly. Next, these contents would be measured using multiple assessment methods in diverse samples. The constructs would then be refined based on the psychometric considerations, including (a) replicability of factor structure; (b) discriminant validity relative to one another, personality pathology, and normative traits; (c) freedom from bias; (d) reliability; and (e) criterion validity. Clinically efficient methods for their assessment would then be developed and field tested, permitting subjugation to further refinement through psychometric procedures. Ideally, the resulting constructs would provide a means for developing a coherent theory of PD style that could facilitate clinical formulations, future research, and testable inferences about dynamic processes. Several investigators have undertaken projects along these lines, leading to the development of Clark’s (1993) Schedule for Nonadaptive and Adaptive Personality, Livesley and Jackson’s (2006) Dimensional Assessment of Personality Problems, and the DSM-5 trait proposal (Krueger, Eaton, Clark, et al., 2011). This line of work provides an important foundation for the process of depicting PD and delineates areas needing further study. For example, because all of these research programs have relied nearly exclusively on self-report questionnaires, which may be limited in some respects for assessing PD (Huprich & Bornstein, 2007), future investigations should employ multiple assessment methods. Second, variance in the structure of each of these models needs to be resolved to build consensual models of PD. Third, each of these models has been guided by an underlying trait perspective on PD classification, and the integration of these approaches with other theoretical approaches, including those that emphasize dynamic or potentially discontinuous aspects of PD or that separate personality pathology from PD, remains unclear. Separating severity and style, as is done in DSM-5 Section III, has the potential to improve diagnostic efficiency and predictive validity. This two-part model of personality pathology and PD is analogous to common conceptions of intelligence involving a general component (i.e., g or IQ) and specific components (e.g., verbal versus nonverbal abilities). Clinical diagnosis is determined by the standing on the general component; just as mental retardation is defined by a particularly low IQ score, the diagnosis of personality pathology could be defined by a particularly low score on a measure of general personality functioning. More specific predictions about impairment can be made when severity and stylistic elements are distinguished. For example, clinicians would predict that any individual with intellectual disabilities would do poorly in schoolwork relative to most other students, but they would further predict that individuals with personal strengths in verbal versus nonverbal abilities would perform relatively better in reading than mathematics classes. Analogously, severity of personality pathology may permit predictions about the overall level of treatment needed (e.g., inpatient versus outpatient), whereas PD style permits predictions about how pathology will manifest (e.g., as impulsive social behavior or social withdrawal) and what treatment techniques might be most effective (e.g., group versus individual therapy, pharmacotherapy). One general issue that requires reconciliation in distinguishing personality pathology severity from style involves their differential association with some PD constructs. Specifically, the terms borderline and narcissism are treated as distinct disorders in the DSM-IV, but they have historically been employed as a general term for personality pathology in several major theories (e.g., Kernberg, 1984; Kohut, 1971). Indeed, empirical models of personality organization appear to relate, conceptually and empirically, to these two PDs (Morey, 2005; Morey et al., 2011). So are narcissistic and borderline PDs discrete, stylistic elements of PD, or are they proxies for personality pathology? These are the kinds of theoretical questions that require resolution through empirical procedures if the field is to make progress toward a more scientifically valid and clinically useful model of personality pathology and disorder. We will return to contemporary issues in PD classification at the end of the chapter. First, we review the historical context of current operationalizations and empirical evidence relating to the prevalence, etiology, and course of PDs as defined by the DSM-IV. Clinicians have been interested in pathological manifestations of personality for as long as they have been deriving psychopathology taxonomies. Among the first material approaches on record occurred in the fourth century B.C., when Hippocrates translated the philosophies of ancient Mesopotamia (Sudhoff, 1926) into a taxonomy consisting of four temperaments that he believed corresponded to imbalance in bodily humors: choleric (irritable), melancholic (sad), sanguine (optimistic), and phlegmatic (apathetic). It is notable how similar these temperaments are to contemporary models of human personality (i.e., irritable ∼ disagreeable, sad ∼ neurotic, sanguine ∼ extraverted, and phlegmatic ∼ [un]conscientious). Hippocrates developed a taxonomy for psychiatric conditions based on these temperamental factors and other conditions, which included six classes of disease: phrenitis, mania, melancholia, epilepsy, hysteria, and Scythian disease (Menninger, 1963). Clinical theorists such as Galen added complexity to early Greek models throughout the Middle Ages and Renaissance, but the quasi-medical approach to classification and basic categories remained fairly similar and continued to be influenced somewhat by supernatural assumptions. In the 17th century, scientific approaches began to supplant concepts that were rooted in clinical descriptions colored by metaphysical theories. The enhanced focus on falsifiable methods from the 17th century onward paved the way for contemporary models in descriptive psychiatry. Emil Kraepelin, who produced nine volumes of clinical psychiatry textbooks from 1883 to 1927 that represented a standard text on psychiatric classification during his lifetime and for many years to follow, is widely regarded as the pioneer of this movement. The aspect of his approach that set him apart from previous theorists was his focus on the course of disorders, in addition to their signs and symptoms. The concept of course is particularly important for PDs, which have been distinguished from other disorders based on the presumption that they are relatively more enduring. Given his focus on course and the historical importance of personality pathology, it is not surprising that many of the concepts in Kraepelin’s textbook are easily identified in the DSM PDs. This link is also due to Kraepelin’s influence on early 20th-century efforts to categorize mental disorders in the United States (Menninger, 1963). In the middle of that century, Kraepelinian concepts were blended with psychoanalytic ideas by major figures such as Adolf Meyer and William Alanson White, who contributed significantly to the conceptual models of psychopathology underlying the first DSM.1 PDs have appeared in every edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM; American Psychiatric Association [APA], 1952, 1968, 1980, 1987, 1994, 2000). In the DSM-I, they were characterized by developmental defects or pathological trends in personality structure, with minimal subjective anxiety and little or no sense of distress. In most instances, the disorder is manifested by a lifelong pattern of action or behavior, rather than by mental or emotional symptoms (APA, 1952, p. 34). These three pillars of PD definition have persisted in subsequent editions: PDs are thought to be developmental, stable, and ego-syntonic. The DSM-I employed a narrative rating system for diagnosing PDs, meaning that clinicians were expected to determine a diagnosis based on the perceived match between a patient’s behavior and a description of pathological prototypes in the manual. PDs in DSM-I were separated into distinct groups. Personality pattern disturbances were regarded as “deep-seated,” “with little room for regression.” These included the inadequate, schizoid, cyclothymic, and paranoid types. Personality trait disturbances referred to conditions brought about by stress, which were thought to indicate latent weaknesses in the underlying personality structure. These included emotional instability, passive aggression, compulsivity, and an “other”category. Sociopathic personality disturbances were similar to personality trait disturbances, but their manifestation was thought to be driven primarily by a mismatch between an individual’s behavior and cultural norms. Such disturbances included antisocial and dyssocial reactions, sexual deviance, and substance abuse. Although the second edition of the DSM (APA, 1968) brought with it some theoretical and classificatory changes to the PDs, it remained very similar in underlying approach and in content to the first edition. The primary change was the removal of the three diagnostic subcategories in favor of the more straightforward depiction of 10 distinct types, largely carried over from DSM-I. Specifically, dependent and aggressive subtypes of passive aggressive personality were collapsed, compulsive personality was reconceptualized and renamed obsessive-compulsive personality, dyssocial personality was renamed explosive personality, and asthenic personality, conceptualized as involving dependency and compromised character strength, was added. Finally, although PDs continued to be differentiated from other disorders in terms of their supposed ego-syntonicity, the assertion that patients with PDs routinely did not experience distress as a result of their personality pathology was tempered (Oldham, 2005). In order to improve diagnostic reliability, DSM-III categories were rated based on atheoretical, behavioral symptom criteria rather than prototype descriptions that were rooted in the formulations of specific theories and required relatively more clinical inference. The DSM-III also introduced the multiaxial system, in which a distinction was made between Axis I conditions (thought to be acute, ego-dystonic, and relatively amenable to treatment) and Axis II conditions (thought to be enduring, ego-syntonic, and relatively resistant to treatment), with PDs belonging to Axis II. Four DSM-II PDs were either eliminated (inadequate and asthenic) or moved to Axis I (cyclothymic and explosive). Schizoid PD was separated into schizoid (defined by interpersonal aloofness), schizotypy (defined by odd behavior), and avoidant (defined by fear of interpersonal criticism/embarrassment) PDs. Two new PD diagnoses, borderline and narcissistic, were introduced in DSM-III. Finally, the PDs were grouped into three clusters based on their degree of shared phenomenology: Cluster A consisted of schizoid, schizotypal, and paranoid; Cluster B of borderline, histrionic, narcissistic, and antisocial; and Cluster C of obsessive-compulsive, avoidant, dependent, and passive-aggressive. The DSM-IV largely retained this system, with three major exceptions: (1) the antisocial criteria were simplified somewhat; (2) a paranoid/dissociation criterion was added to borderline; and (3) passive-aggressive PD was appendicized. The rationale for removing passive-aggressive PD was that it referred to a narrow behavioral tendency, rather than a broad personality syndrome (Millon & Radovanov, 1995), although this view has been challenged on rational (Wetzler & Morey, 1999) and empirical (Hopwood et al., 2009) grounds. Finally, personality pathology was formally operationalized, separately from the criteria of each specific PD. It was defined as (a) an enduring pattern of inner experience and behavior that deviates markedly from the expectations of the individual’s culture and is manifested in at least two of the following areas: cognition, affectivity, interpersonal functioning, or impulsive control; (b) pervasive across a broad range of personal and social situations; (c) leading to clinically significant impairment in social, occupational, or other important areas of functioning; (d) stable with an onset that can be traced back at least to adolescence or early adulthood; (e) not better accounted for by another mental disorder; and (f) not due to the direct physiological effects of a substance or medical condition. Criteria sets were polythetic for all PDs, whereas in the DSM-III criteria for some PDs had been more typological and impressionistic due to limited theoretical and empirical understanding. For example, a patient would have to meet all three of the DSM-III-R dependent PD criteria for the diagnosis, including passively allowing others to assume responsibility, subordinating needs to others, and lacking self-confidence. In the DSM-IV, a patient would need to meet five of eight symptoms, including difficulties initiating projects on one’s own and urgently seeking new relationships for support when old relationships end. While this change enhanced the reliability of the PDs (Pfohl, Coryell, Zimmerman, & Stangl, 1986), the increased use of polythetic criteria may have also worsened the problems of construct heterogeneity and diagnostic overlap (Gunderson, 2010). Having described the clinical importance of personality pathology, defined the central concepts of personality and related pathology, and reviewed the history of PD taxonomy, we will now focus on clinical manifestations of the PDs listed in Section III of DSM-5 (APA, 2013). Cluster A consists of paranoid, schizoid, and schizotypal PDs, which are thought to share odd and eccentric features. These disorders are associated with psychotic disorders phenomenologically and etiologically (Maier, Lichtermann, Minges, & Heun, 1994) but are distinguished by their lack of persistent psychotic symptoms (i.e., hallucinations and delusions). Paranoid PD is defined by a pervasive pattern of distrust and beliefs that others’ motives are malevolent. Symptoms involve suspiciousness and consequent social dysfunction, loose and hypervigilant thinking, and resentment. Schizoid PD is characterized by a pervasive pattern of social detachment and restricted emotional expression. Symptoms include disinterest in relationships and preference for solitude, limited pleasure in sex or other activities commonly regarded as pleasurable, and emotional flatness. The defining feature of schizotypal PD is a pervasive pattern of interpersonal deficits, cognitive or perceptual distortions, and eccentric behavior. It is diagnosed by symptoms related to loose or eccentric perceptions and cognitions, flat affect, mistrustfulness, and profound social dysfunction. Antisocial, borderline, histrionic, and narcissistic PDs constitute Cluster B, which is regarded as the “dramatic, erratic, and emotional” group (APA, 2000). Individuals with these disorders tend to experience emotional dysregulation and behave impulsively. Antisocial PD is marked by a pervasive pattern of disregard for the rights and wishes of others (APA, 2000). The DSM-IV requires evidence of childhood conduct disorder for a diagnosis of antisocial PD, and additionally includes symptoms of socially nonnormative behavior, dishonesty, impulsivity, aggression, lack of empathy, and irresponsibility. Borderline PD is characterized by “stable instability” (Schmideberg, 1959) in emotions, interpersonal behavior, and identity. Emotional dysregulation, including anger and emptiness, is thought to be triggered by concerns about abandonment, which is followed by maladaptive coping, including impulsive and suicidal behavior (Zanarini & Frankenburg, 2007). Histrionic PD is characterized by excessive emotionality and attempts to obtain attention from others. This desire to be the center of attention often comes at the cost of deep and meaningful interpersonal relationships, as histrionic individuals tend to have relatively superficial interpersonal interactions and shallow emotions. The core of narcissistic PD involves grandiose thoughts and behaviors, a need for excessive admiration from others, and a lack of empathy. It is commonly believed that arrogant and haughty behavior is undergirded by feelings of vulnerability and inadequacy. Cluster C includes avoidant, dependent, and obsessive-compulsive PDs, which are grouped together based on their common thread of anxiety and fearfulness. Avoidant PD is characterized by social inhibition rooted in feelings of inadequacy and fears of negative evaluations from others (APA, 2000). Symptoms include avoidance of social and occupational opportunities, fears of shame and ridicule, and negative self-concept. Dependent PD is defined by an excessive need to be cared for by others that leads to submissive, clingy behavior. Symptoms include difficulties making autonomous decisions or expressing disagreement with others, nonassertiveness, preoccupation with abandonment, and maladaptive or self-defeating efforts to seek and maintain relationships. Obsessive-compulsive PD is defined by a preoccupation with order, perfection, and control in which flexibility, efficiency, and even task completion are often sacrificed. Symptoms include preoccupation with rules and order, perfectionism, workaholism, interpersonal inflexibility, frugality, and stubbornness. One of the principal advantages of the DSM-III and DSM-IV model of PDs that is being carried over to DSM-5, Section II has been its contribution to the increase in PD research. However, to the extent that a majority of this research has been based on a flawed conception of PDs, its utility is constrained by the validity of the model. Epidemiology instantiates this paradox. Although the development of reliable PD criteria has made it more possible to evaluate the prevalence of PDs, calculating the prevalence of PDs assumes that they are categorical taxa, even though the weight of evidence suggests they are not (Trull & Durrett, 2005), and diagnostic cutoffs in the DSM-IV are thus essentially arbitrary (Cooper, Balsis, & Zimmerman, 2010). As such, it is not clear what to make of PD prevalence rates. With this caveat in mind, we review the results of several epidemiological studies on PD based on DSM diagnostic cutoffs. Overall prevalence rates estimate that more than 10% of individuals suffer from a PD during their lifetime (Grant et al., 2004; Lenzenweger, Loranger, Korfine, & Neff, 1997; Samuels et al., 2002; Torgersen, 2005; Torgersen et al., 2001). Although rates for individual PDs are more variable across studies, research suggests that most PDs have prevalence rates between 0.5% and 5%, with paranoid, avoidant, and obsessive-compulsive PDs being relatively common and dependent and narcissistic PDs being relatively uncommon. PD prevalence is considerably higher in psychiatric settings: Research indicates that nearly half of clinical outpatients and more than half of clinical inpatients meet the diagnostic criteria for a PD (Molinari, Ames, & Essa, 1994; Zimmerman, Rothschild, & Chelminski, 2005), making PDs among the most commonly encountered disorders in psychiatric settings. The two most commonly occurring PDs among psychiatric patients are borderline, 10% to 20% (APA, 2000; Zimmerman et al., 2005), and dependent PD. Rates of dependent PD are particularly high among inpatients, 15% to 25% (Jackson et al., 1991; Oldham, 2005), relative to outpatients, 0% to 7% (Mezzich, Fabrega, & Coffman, 1987; Poldrugo & Forti, 1988; Zimmerman et al., 2005). In contrast, somewhat lower occurrence rates have been observed for paranoid (2% to 4%; Zimmerman et al., 2005), schizoid, and schizotypal (1% to 2%; Stuart et al., 1998; Zimmerman et al., 2005) patients in clinical settings. These relatively low rates may relate to the impact of the paranoia and social avoidance that characterizes these disorders on treatment seeking. Indeed, prevalence estimates for schizoid PD among a homeless population are as high as 14% (Rouff, 2000). Likewise, antisocial PD is seen in 1% to 4% of individuals in a clinical population (Zimmerman et al., 2005), but estimates are considerably higher in prison and substance abuse populations, indicating that individuals with this type of pathology are unlikely to initiate psychological treatment. Somewhat low rates have also been observed for narcissistic (2%; Torgersen et al., 2001) and obsessive-compulsive (3% to 9%; Zimmerman et al., 2005) PDs in clinical settings, perhaps owing to the limited functional impact of these PDs relative to others. Rates of histrionic and avoidant PD in clinical patients have been estimated at 10% to 15% (APA, 2000; Zimmerman et al., 2005). Despite extensive theorizing, empirical evidence regarding the etiology of PDs is quite limited (Paris, 2011; Skodol et al., 2011). Factors contributing to this gap between theory and evidence include a history of underfunding for PD research relative to research on other psychiatric conditions and various conceptual problems discussed throughout this chapter. Nevertheless, we briefly review etiological contributions related to genes, neurobiology, learning, and cognition presently. While the broad heritability of personality traits is well-established (McGue, Bacon, & Lykken, 1993; Plomin, DeFries, Craig, & McGuggin, 2003), the heritability of personality pathology and disorders remains more ambiguous (Lenzenweger & Clarkin, 2005). A twin study by Torgersen and colleagues (2000) indicated that the overall 58% of the variance in PDs was due to genes, with specific heritability estimates for each PD as follows: paranoid (0.30), schizoid (0.31), schizotypal (0.62), borderline (0.69), histrionic (0.67), narcissistic (0.77), avoidant (0.31), dependent (0.55), and obsessive-compulsive (0.78). Notably, shared family environment influences were also particularly important in predicting borderline PD. These results led the authors to conclude that PDs may have even stronger genetic influences than most other disorders, similar to broad dimensions of normative personality. Rates of antisocial PD were too low to be included in the Torgersen et al. (2000) study; however, work by other researchers has broadly evidenced that the heritability of aggression is between 44% and 72% (Siever, 2008). A meta-analysis suggests the importance of both genetic and environmental influences on antisocial behavior in men and women, though this influence was measured on antisocial behaviors as opposed to antisocial PD (Rhee & Waldman, 2002). Evidence also suggests higher rates of schizotypal (Kendler et al., 2006) and borderline (Links, Steiner, & Huxley, 1988) PDs among family members of individuals with those disorders, and increased rates of Cluster C PDs among individuals who have relatives with Axis I anxiety disorders (Reich, 1991). Evidence also suggests interactive effects between genes and the environment, such as the finding that associations between a polymorphism on the MAOA gene is associated with antisocial traits only in individuals who have been exposed to trauma (Caspi et al., 2002). Evidence suggesting an interaction between genes and the early attachment environment may also be crucial in subsequent development of PDs (Siever & Weinstein, 2009). The interplay of genes and environment in the genesis of psychopathology is quite complicated (e.g., Burt, 2009), and much more research is needed with respect to the etiology of PDs. Beyond a basic decomposition of the etiological components of personality pathology and specific PDs, research should begin focusing more on molecular models and the interplay between behavior genetic and environmental risk factors. As with etiology and despite a recent rise in interest in understanding neurological risk factors for the development of PDs (for a review, see Siever & Weinstein, 2009), potential neurobiological endophenotypes for PDs are largely unknown (Paris, 2011; Skodol et al., 2011). It is widely presumed that endophenotypes are more likely to reflect neurobiological dimensions that underlie personality pathology rather than point to pathways to specific PDs. Potential dimensions include those related to cognitive dysregulation, impulsivity, and emotional dysregulation (Depue & Lenzenweger, 2005; Siever & Weinstein, 2009). We will briefly review these three classes of potential endophenotypes and their implications for PDs. Cluster A PDs, and particularly schizotypal, share features of cognitive dysregulation with schizophrenia, including distorted perception and disrupted attention. Cognitive dysregulation is thought to relate to reduced dopamine reactivity in the frontal cortex (Abi-Dargham et al., 1998; Seiver & Weinstein, 2009), as well as structural anomalies found in psychotic disorders such as increased ventricular volume (Hazlett et al., 2008). The influence of common endophenotypes represents a promising explanation for descriptive and phenomenological similarities between Cluster A PDs and psychotic disorders (Depue & Lenzenweger, 2005). Impulsivity
Personality Disorders
Defining Personality: Traits, Dynamics, Pathology, and Disorders
Personality Traits
Personality Dynamics
Severity of Personality Pathology
Personality Disorder Style
The Separation of Personality Pathology and Disorder
Diagnostic Considerations: A Historical Backdrop to the DSM-IV
DSM-I and DSM-II
DSM-III and DSM-IV
Clinical Picture: The DSM-5 Section II Personality Disorders
Research on the DSM-5 Section II Personality Disorders
Epidemiology
Etiology
Genetics
Neurobiology
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
