Judgment research and the dimensional model of personality

Comments on the original article "Plate tectonics in the classification of personality disorder: Shifting to a dimensional model," by T. A. Widiger and T. J. Trull. The purpose of this comment is to address (a) whether psychologists know how personality traits are currently assessed by clinicians and (b) the reliability and validity of those evaluations. Although Widiger and Trull argued that the shift to a dimensional model will improve the diagnosis of personality disorders, we can also expect that it will lead to improvements in evaluating personality traits.     

The DSM-5 dimensional model of personality disorder: rationale and empirical support

Ideally, a dimensional model of normal and abnormal personality functioning within the forthcoming DSM-5 would represent a common ground among the existing alternative models and would be consistent with the substantial body of empirical research that has accumulated on such models. The DSM-5 Work Group had the opportunity to make an historic step toward the integration of the normal personality traits researched within psychology with the personality disorders of psychiatry. They chose instead to construct a cumbersome six-dimensional model of personality disorder that excludes normal personality traits, is inconsistent with the preponderance of the research, and is distinguished explicitly from the predominant dimensional model of general personality structure.     

Plate tectonics in the classification of personality disorder: shifting to a dimensional model

The diagnostic categories of the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders were developed in the spirit of a traditional medical model that considers mental disorders to be qualitatively distinct conditions (see, e.g., American Psychiatric Association, 2000). Work is now beginning on the fifth edition of this influential diagnostic manual. It is perhaps time to consider a fundamental shift in how psychopathology is conceptualized and diagnosed. More specifically, it may be time to consider a shift to a dimensional classification of personality disorder that would help address the failures of the existing diagnostic categories as well as contribute to an integration of the psychiatric diagnostic manual with psychology's research on general personality structure.     

An invariant dimensional liability model of gender differences in mental disorder prevalence: evidence from a national sample

Epidemiological studies of categorical mental disorders consistently report that gender differences exist in many disorder prevalence rates and that disorders are often comorbid. Can a dimensional multivariate liability model be developed to clarify how gender impacts diverse, comorbid mental disorders? We pursued this possibility in the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC; N = 43,093). Gender differences in prevalence were systematic such that women showed higher rates of mood and anxiety disorders, and men showed higher rates of antisocial personality and substance use disorders. We next investigated patterns of disorder comorbidity and found that a dimensional internalizing-externalizing liability model fit the data well, where internalizing is characterized by mood and anxiety disorders, and externalizing is characterized by antisocial personality and substance use disorders. This model was gender invariant, indicating that observed gender differences in prevalence rates originate from women and men's different average standings on latent internalizing and externalizing liability dimensions. As hypothesized, women showed a higher mean level of internalizing, while men showed a higher mean level of externalizing. We discuss implications of these findings for understanding gender differences in psychopathology and for classification and intervention.     

Alternative dimensional models of personality disorder: finding a common ground

The recognition of the many limitations of the categorical model of personality disorder classification has led to the development of quite a number of alternative proposals for a dimensional classification. The purpose of this article is to suggest that future research work toward the integration of these alternative proposals within a common hierarchical structure. An illustration of a potential integration is provided using the constructs assessed within existing dimensional models. Suggestions for future research that will help lead toward a common, integrative dimensional model of personality disorder are provided.     

A dimensional model of personality disorder: incorporating DSM Cluster A characteristics

The authors articulate an expanded dimensional model of personality pathology to better account for symptoms of DSM-defined Cluster A personality disorders. Two hundred forty participants (98 first-degree relatives of probands with schizophrenia or schizoaffective disorder, 92 community control participants, and 50 first-degree relatives of probands with bipolar disorder) completed a dimensional personality pathology questionnaire, a measure of schizotypal characteristics, and Chapman measures of psychosis proneness. Scales from all questionnaires were subjected to an exploratory factor analysis with varimax rotation. A 5-factor structure of personality pathology emerged from the analyses, with Peculiarity forming an additional factor to the common 4-factor structure of personality pathology (consisting of Introversion, Emotional Dysregulation, Antagonism, and Compulsivity). These results support a 5-factor dimensional model of personality pathology that better accounts for phenomena encompassed by the Cluster A personality disorders in DSM-IV-TR (4th ed., text revised; American Psychiatric Association, 2000). This study has implications for the consideration of a dimensional model of personality disorder in DSM-V by offering a more comprehensive structural model that builds on previous work in this area.     

Personality disorders in offenders: categorical versus dimensional approaches.

The present study focused on the consequences of cut-off scores in personality disorder diagnoses for their association with criminal behavior. Using ICD-10 personality disorder criteria eliminating offence-related symptoms, we studied the distributions of categorically diagnosed personality disorders and of dimensional personality disorder scores in a group of offenders and a noncriminal control group. Whereas the dimensional scores of the offender group differed significantly from those of the control group for all personality disorders under study, the frequency of categorical diagnoses differed significantly for two personality for two personality disorders only. Moreover, prediction of group membership (offenders vs. nonoffenders) from personality disorder scores was substantially more precise than prediction from categorical diagnoses. It is concluded that a dimensional approach to personality disorder diagnosis is not only superior theoretically but also yields more precise information about the specific associations with criminal behavior.     

The relationship between five-factor personality measurements and ICD-10 personality disorder dimensions: results from a sample of 229 subjects

This article examines the relationship between the five-factor model (FFM) and dimensional ICD-10 personality disorders. In a follow-up study of a child and adolescent psychiatric cohort, former patients and controls were assessed with NEO-FFI and the IPDE interview (CD-10 personality disorder). Full data were available for 229 subjects (149 former patients, 80 controls). Multiple regression analysis showed that the five factors of the FFM as independent variables explained between 5% (schizoid personality disorder) and 32% (anxious personality disorder) of the variance of ICD-10 dimensional personality disorder scores. For the two types of emotionally unstable personality disorder dimension (impulsive and borderline), for anxious (avoidant) personality disorder dimension and for the total score of any personality disorder dimension, FFM explained between 17% and 32% of the variance with almost identical results for the former patient group and the control group. High neuroticism was a feature of paranoid, emotionally unstable, histrionic, anankastic, anxious (avoidant), and dependent personality disorder dimensions, whereas low agreeableness was found in dissocial, emotionally unstable and histrionic personality disorder dimensions. Low extraversion was found in schizoid, anxious (avoidant) and dependent personality disorder dimensions, whereas histrionic PD dimension correlated with high extraversion. We find that the FFM is valuable for the further understanding not only of DSM-IV but also of ICD-10 personality disorder dimensions. The differences between ICD-10 and DSM-IV in this respect seem to be small.     

Dimensional personality traits and the prediction of DSM-IV personality disorder symptom counts in a nonclinical sample

The third edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-III; APA, 1980) set forth a categorical system of personality psychopathology that is composed of discrete personality disorders (PDs), each with a distinct set of diagnostic criteria. Although this system is widely accepted and highly influential, alternative dimensional approaches to capturing personality psychopathology have been proposed. Three dimensional models of personality have garnered particular attention-the Five-Factor Model (FFM; Costa & McCrae, 1992), the Seven-Factor Psychobiological Model of Temperament and Character (Seven-Factor Model; Cloninger, Svrakic, & Przybeck, 1993); and the 18-factor model of personality pathology (18-factor model; Livesley, 1986). Although the personality traits from each of these models has been examined in relation to the ten personality disorders in the DSM-IV, no study has examined the comparative and incremental validity of these models in predicting PD symptoms for these ten disorders. Using self-report instruments that measure these models and the ten DSM-IV PDs, correlation and linear regression analyses indicate that traits from all three models had statistically significant associations with PD symptom counts. Hierarchical regressions revealed that the 18-factor model had incremental predictive validity over the FFM and Seven-Fac-tor Model in predicting symptom counts for all ten DSM-IV PDs. The FFM had incremental predictive validity over the Seven-Factor Model model for all ten disorders and the Seven-Factor was able to add incremental predictive validity over the 18-factor model for five of the ten PDs and for eight of the ten disorders relative to the FFM.     

Mapping the Personality Psychopathology Five Domains Onto DSM–IV Personality Disorders in Dutch Clinical and Forensic Samples: Implications for DSM–5

The Personality Psychopathology Five (PSY–5) model represents 5 broadband dimensional personality domains that align with the originally proposed DSM–5 personality trait system, which was eventually placed in Section III for further study. The main objective of this study was to examine the associations between the PSY–5 model and personality disorder criteria. More specifically, we aimed to determine if the PSY–5 domain scales converged with the alternative DSM–5 Section III model for personality disorders, with a particular emphasis on the personality trait profiles proposed for each of the specific personality disorder types. Two samples from The Netherlands consisting of clinical patients from a personality disorder treatment program (n = 190) and forensic psychiatric hospital (n = 162) were used. All patients had been administered the MMPI–2 (from which MMPI–2–RF PSY–5 scales were scored) and structured clinical interviews to assess personality disorder criteria. Results based on Poisson or negative binomial regression models showed statistically significant and meaningful associations for the hypothesized PSY–5 domains for each of the 6 personality disorders, with a few minor exceptions that are discussed in detail. Implications for these findings are also discussed.     

Important questions remain to be addressed before adopting a dimensional classification of mental disorders

Comments on the original article "Plate tectonics in the classification of personality disorder: Shifting to a dimensional model," by T. A. Widiger and T. J. Trull. Widiger and Trull raised important nosological issues that warrant serious consideration not only for the personality disorders but for all mental disorders as the Diagnostic and Statistical Manual of Mental Disorders (DSM) is revised during the next few years. As argued compellingly by these authors, dimensional classification may indeed offer substantial improvement over the present categorical system. Several questions remain to be addressed in evaluating whether this is the case and, if so, how dimensional classification can be implemented to best serve the disparate users of the DSM.     

Personality disorders in the 21st century

The diagnosis and classification of personality disorders have come a long way since the beginning of the 20th century. Perhaps by this time next century, the diagnostic manual of mental disorders will be using a dimensional model of classification that will provide more reliable and meaningful points of demarcation between normal and abnormal personality functioning. This article offers suggestions for the form, content, and placement of such a dimensional model. The particular model emphasized herein is the five-factor model of personality functioning, but the optimal diagnostic system will probably involve an integration of alternative dimensional models.     

Personality assessment in DSM-5: empirical support for rating severity, style, and traits

Despite a general consensus that dimensional models are superior to the categorical representations of personality disorders in DSM-IV, proposals for how to depict personality pathology dimensions vary substantially. One important question involves how to separate clinical severity from the style of expression through which personality pathology manifests. This study empirically distinguished stylistic elements of personality pathology symptoms from the overall severity of personality disorder in a large, longitudinally assessed clinical sample (N = 605). Data suggest that generalized severity is the most important single predictor of current and prospective dysfunction, but that stylistic elements also indicate specific areas of difficulty. Normative personality traits tend to relate to the general propensity for personality pathology, but not stylistic elements of personality disorders. Overall, findings support a three-stage diagnostic strategy involving a global rating of personality disorder severity, ratings of parsimonious and discriminant valid stylistic elements of personality disorder, and ratings of normative personality traits.     

Behavioral and molecular genetic contributions to a dimensional classification of personality disorder

This article examines the possible contribution of behavioral and molecular genetic research to the development of a dimensional classification of personality disorder. It is argued that the results of molecular studies are too preliminary to have immediate nosological significance. However, behavioral genetic methods could play a useful role in constructing a classification that reflects the genetic architecture of personality disorder. It is also argued that the best approach to constructing a valid classification would be to integrate behavioral genetic methods with the construct validation framework used in test construction. An integrative approach is proposed that seeks to combine constructs from alternative dimensional models. It is suggested that strong evidence of a four-dimensional structure to personality disorder provides a way to organize a preliminary model. An initial set of primary traits to define these secondary domains would then be compiled from existing models and refined using a combination of traditional psychometric analyses and behavioral genetic methods. It is concluded that an etiologically based classification is feasible for the DSM-V.     

Deconstructing dependency in a five-factor world: a meta-analytic review

Meta-analysis of studies assessing the relation between interpersonal dependency test scores and five-factor model (FFM) domain scores revealed that dependency scores are positively correlated with FFM Neuroticism and Agreeableness scores and negatively correlated with FFM Extraversion, Openness, and Conscientiousness scores. The magnitudes of these correlations were all in the small-to-moderate range, and comparable score intercorrelations were obtained when participants' dependency levels were assessed by means of a trait dependency questionnaire, dependent personality disorder questionnaire, or dependent personality disorder interview. These findings have implications for researchers' efforts to deconstruct dependency into its basic trait elements and for the dimensional approaches to personality disorders being considered for future versions of the Diagnostic and Statistical Manual of Mental Disorders.     

A framework for integrating dimensional and categorical classifications of personality disorder

Although empirical evidence strongly supports a dimensional representation of personality disorder, there is strong resistance to dimensional classification due in part to concerns about clinical utility. Acceptance of an evidence-based dimensional classification would be facilitated by information on how such a system would map onto existing diagnoses. With this objective in mind, an integrated framework is proposed that combines categorical and dimensional diagnoses. A two-component classification is adopted that distinguishes between the diagnosis of general personality disorder and the assessment of individual differences in the form the disorder takes. Then, the DSM definition of personality disorders is extended by defining individual disorders as categories of trait dimensions. This makes it possible to develop an integrated classification organized around a set of empirically derived primary traits. Assessments of these traits may then be combined to generate categorical and dimensional diagnoses. It is argued that this approach would introduce an etiological perspective into the classification of personality disorder and improve categorical classification by providing an explicit definition of each diagnosis. The clinical utility of incorporating a dimensional classification is discussed in terms of convenience and acceptability, value in predicting outcomes and treatment planning, and usefulness in organizing and selecting interventions.     

A study of temperament and personality in anorexia and bulimia nervosa

Although temperament and personality traits could influence the development and course of eating disorders, only a few studies examined the similarities and differences in personality between anorexia and bulimia nervosa. We compared 72 patients with DSM-IV eating disorders and 30 healthy controls. Dimensions of personality and personality disorders were evaluated with the Eysenck's EPQ, Cloninger's TCI, and the SCID-II questionnaires. The rates of impulsivity and clinical features were evaluated using specific rating scales. A comorbid personality disorder was found in 61.8% of patients with eating disorder. Avoidant personality disorder appeared was relatively common in anorexia nervosa restricting type; borderline personality disorder was most frequent in bulimia nervosa and the binge eating-purging type of anorexia nervosa. From a dimensional perspective, anorexic patients presented high scores in the dimension of persistence. Higher harm avoidance and impulsivity was found in bulimic patients. The overall eating disorders group presented high scores in neuroticism and low scores in self-directedness. Eating disorder patients have heterogeneous features of temperament and personality traits. Cluster C personality disorders seem more common in anorexia nervosa restricting type and impulsive personality features are associated with bulimic symptoms. Impulsivity seems to be a key aspect of temperament of bulimic patients, whereas anorexic symptoms are linked to persistent temperament traits.     

Personality disorders in a Dutch forensic psychiatric sample: convergence of interview and self-report measures

Convergence of PDQ-R- and SIDP-R-derived personality disorder diagnoses was studied in a sample of 85 forensic psychiatric patients. For categorical diagnoses, the mean kappa was .34, but on a dimensional level convergence was somewhat higher. Paranoid, antisocial and borderline personality disorders had prevalence rates around 40%; the other personality disorders occurred with much lower frequency. The PDQ-R yielded more diagnoses, except for antisocial, histrionic, narcissistic, and sadistic personality disorder. Because the latter disorders are among the most prevalent in forensic settings, and because they have important risk and treatment implications, the PDQ-R is not suitable as a screening device in forensic populations. Semistructured interviews that make use of collateral information are recommended for diagnosing personality disorders in forensic subjects.