DSM-III Personality Disorders Research Articles

Differentiation between schizoid and avoidant personalities: an empirical test.

A number of authors have questioned the rationale for subdividing the DSM-II schizoid diagnosis into three separate personality disorders in DSM-III, the schizoid, avoidant, and schizotypal. The present study was designed to explore differences between psychiatric patients with schizoid and avoidant personalities as compared to psychiatric controls with no personality disorder. Differences were examined on demographic data, self-report measures, and clinical information. A Multivariate Analysis of Variance (MANOVA) revealed a significant overall effect for groups across MMPI subscales. However, subsequent univariate Analyses of Variance (ANOVA's) revealed that almost all differences were between the two personality disorder groups as compared to the psychiatric controls. Contrary to expectations, schizoid and avoidant personalities were found to display equivalent levels of anxiety, depression, and psychotic tendencies as compared to psychiatric control patients. No meaningful distinctions were found between the avoidant and the schizoid personalities. Results are discussed in terms of problems with the assessment methods and the diagnostic criteria.

Interpersonal Aspects of DSM-III Axis II: Theoretical Hypotheses and Empirical Findings

In recent years several authors have attempted to map correspondences between interpersonal models and psychiatric classifications. Morey was the first to try this empirically, using, for purposes of comparison, interpersonal (ICL) and DSM-III (MCMI) approaches to the classification of personality disorders. He showed that the two approaches did not converge as much as expected; in particular, it appeared that DSM-III personality disorders were not as differentiated with regard to affiliative needs as had been hypothesized. The present study set out to determine if Morey's findings could be replicated in a different population, using a more suitable assessment procedure for DSM-III personality disorders (SIDP). Our study did corroborate Morey s main findings, although we found that DSM-III personality disorders were even less differentiated with regard to affiliative needs than Morey's findings suggest. This study concludes that interpersonal and DSM-III approaches to personality taxonomy are complementar...

Residents’ Treatment of Clinic Patients Beyond Graduation

Twenty-one of 24 psychiatrists surveyed report that 51% (n=56) of their clinic patient panel entered their private care after they completed residency training. Patients with a DSM-III personality or depressive disorder were more likely than those with schizophrenia to continue in private treatment (p<.001). The practice of continued treatment of clinic patients beyond residency is reported to aid the professional development of the psychiatrist and the quality of patient care. Residency program directors may consider recommending this practice, if it is appropriate in their institution, to facilitate the psychiatrist's transition from training to career.

Update on instruments to measure DSM-III and DSM-III-R personality disorders.

In the last 2 years, researchers have revised personality instruments to encompass the DSM-III-R criteria for personality disorders. More recent, innovative instruments undergoing validity work include a self-report instrument that rederives DSM-III personality categories from dimensional measures, a self-report designed to be used in conjunction with a diagnostic interview, an interview designed to be given by trained lay people, and a family history measure of the DSM-III personality disorder clusters. The modifications of older instruments and those instruments more recently developed are described. Clearly instruments require additional validity testing and better test-retest reliability. Discrepancies between different measurement instruments also need to be addressed.

Impulsive aggression in personality disorder: Evidence for involvement of 5-HT-1 receptors

In order to explore relationships between 5-HT function and aggressive impulses in patients with personality disorder (PD), neuroendocrine (i.e. “net” preand postsynaptic: FEN and post-synaptic: m-CPP) indices of central 5-HT function were examined in patients with DSM-III PD. PRL responses to both FEN and m-CPP were inversely correlated with clinicianand self-rated measures of impulsive aggression in PD patients (e.g. Buss-Durkee “Assault + Irritability”: FEN: r = -0.77, N = 20, p = .0002; m-CPP: r = 55, N = 16, p < .05). Preliminary data examining the PRL response to the putative 5-HT-la receptor specific agent Buspirone revealed a similar finding (e.g. Buss-Durkee A+I: r = -0.69, N = 9, p < .05). These data suggest that central serotonergic function is reduced in PD patients with prominent histories of impulsive aggressive behavior and that a representation of the abnormality may be found at the level of hypothalamic post-synaptic 5-HT receptors similar to those of the 5-HT-1 subtype.

A comparison of clinical and self-report diagnoses of DSM-III personality disorders in 552 patients

This report examines the relationship between clinicians' diagnoses of personality disorder and self-report diagnoses of personality disorder obtained from the Personality Diagnostic Questionnaire (PDQ). The results from 552 patients showed general lack of agreement between clinical and self-report diagnoses of DSM-III personality diagnoses. The best agreement obtained was for Borderline Personality Disorder: κ = 0.46, and r = .51 for scaled ratings. Possible sources of disagreement including failings of the self-report questionnaire, difficulties in relying upon patients' self-reports, lack of reliability of clinical diagnoses of personality, and possible inherent lack of reliability of several of the DSM-III personality disorders are discussed.

DSM-III Personality Disorder in Bariatric Clinic Patients

AbstractThirty-eight patients (30 women, 8 men) with morbid obesity who presented for vertical banded gastroplasty (VBG) at a bariatric clinic were each compared with two age- and sex-matched community controls for DSM-III personality disorders using the Personality Diagnostic Questionnaire (PDQ), a well-validated self-report instrument. Fifteen (39.5%) morbidly obese patients and 10 (13.2%) control subjects met criteria for at least one personality disorder (odds ratio = 5.8, p < 0.001). Morbidly obese patients were significantly more likely to display eccentric cluster and dramatic cluster personality disorders and traits, particularly borderline personality. The clinical implications of these findings are discussed.

The Reliability of Personality Disorder Diagnoses in a Nonpatient Sample

We examined the joint interview interrater reliability of the Structured Interview for DSM-III Personality Disorders (SIDP) in 104 relatives of normal controls and psychiatric patients. The absolute rate of agreement between the two raters for the presence of any personality disorder was 98.1% (? = .93, Yule's y = .94). The kappa and y values for all disorders diagnosed two or more times were above .65 and .78, respectively. We also examined the reliability of dimensional scores and found that the mean intraclass correlation coefficient was .91.

Prevalence of DSM-III personality disorders in the community

A community sample of adults with a standardized DSM-III, Axis II self-report instrument yielded an age adjusted community prevalence of 11.1% of DSM-III personality disorders. When those with personality disorders (PDs) (n = 26) were compared to those without personality traits (n = 167) the PD group had less education 14.9 (3.0) years vs 16.5 (3.3) years, p = 0.02) and a greater percentage with difficulty with alcohol (19% vs 0.6%, p = 0.0001). Of those married, more PDs reported marital difficulties (29% vs 3.5%, p = 0.002). There was a trend for the PD group to have longer unemployment (p = 0.07).

What's Wrong with DSM-III Personality Disorders?

DSM–III and DSM–III–R personality disorders represent a major advance on earlier classifications but are deficient in several major respects. A fundamental criticism is that the system for classifying mental state disorders has been applied uncritically to personality disorders with considerably less success. The gains made by the introduction of well–defined operational criteria to DSM–III mental state disorders are considerably less successful with personality disorders, and there is doubt whether they are apposite at all because of the pervasiveness and variability of behavior shown in personality disorders. Many of the criteria used for the diagnosis of DSM–III personality disorders are based, with the exception of antisocial personality disorder, on inadequate evidence. Too much reliance is placed on questions to the subject, which is suspect with personality disorder as it is largely the impact of the abnormal personality on other individuals that leads to the diagnosis being made. Subjective report...

A Multitrait-multimethod examination of Minnesota Multiphasic Personality Inventory (MMPI) and Millon Clinical Multiaxial Inventory (MCMI)

Recently, certain Minnesota Multiphasic Personality Inventory (MMPI) and Millon Clinical Multiaxial Inventory (MCMI) scales have seen increasing usage for the measurement of DSM-III personality disorders. The current study sought to identify the convergent and discriminant validity of these two sets of scales for this purpose. In general, the results indicated significant convergence across the two instruments. However, better convergent validity was found for scales representing those DSM-III disorders which are most consistent with the typology upon which the MCMI was based. In particular, convergent and discriminant validity results were poorest for Compulsive, Antisocial, and Passive-Aggressive personality scales.

A family history method for DSM-III anxiety and personality disorders

A family history method for DSM-III anxiety and personality disorders (FHPD) was validated by direct interview of 182 relatives. The categories of “any anxiety disorder,” “panic disorder,” and “any personality disorder” achieved sensitivities and specificities equivalent to currently accepted psychiatric family history methods. The family history results for the DSM-III dramatic and anxious personality disorder clusters showed good specificities, but only fair sensitivities. It is possible these cluster sensitivities could be improved in the future by using less stringent criteria. The FHPD is designed so that it can easily be integrated into the Family History-Research Diagnostic Criteria format and scoring.

Correlates of DSM-III personality disorder in panic disorder and agoraphobia

One hundred eighty-seven patients meeting DSM-III criteria for panic disorder (n = 26) or agoraphobia with panic (n = 161) were assessed with the Personality Diagnostic Questionnaire (PDQ), a self-rating scale designed to assess Axis II personality disorders and traits. Results replicated our earlier findings of a preponderance of dependent, avoidant, and histrionic features and the finding that patients exhibiting a greater number of personality traits were also significantly more symptomatic. Patients with the diagnosis of panic disorder did not differ on any personality disorder variables from patients with the diagnosis of agoraphobia with panic. Furthermore, none of the specific symptom dimensions, i.e., panic, anxiety, or agoraphobia, was selected as a unique predictor of any personality variables in the regression analyses. Rather, the most important correlates of personality disorder in these patients consisted of general factors such as dysphoric mood, social phobia, or interpersonal sensitivity, and Eysenck's neuroticism dimension. The results are discussed in light of recent findings suggesting a nonspecific link between panic disorder or agoraphobia and personality disorder.

Frequency of DSM-III personality disorders in patients with panic disorder: Comparison with psychiatric and normal control subjects

Three clinical populations—panic disorder ( n = 88), randomly selected outpatients ( n = 82), and normal control subjects ( n = 40)—were compared on three standardized DSM-III personality disorder instruments, the Structured Interview for DSM-III Personality Disorders (SIDP), the Millon Clinical Multiaxial Inventory (MCMI), and the Personality Diagnostic Questionnaire (PDQ). Significant differences were consistently found in presence of “any” personality disorder and DSM-III Cluster C (there were always more disorders in the outpatients). Logistic regression analysis revealed the important determinants predicting personality disorders, and therefore of differences between groups, were state depression, age, lifetime history of alcohol abuse, and presence of panic disorder.

The relationship between DSM-III symptom disorders (Axis I) and personality disorders (Axis II) in an outpatient population.

The relationship between symptom disorder and personality disorder according to DSM-III was studied in 289 consecutive outpatients. It was observed that personality disorders occurred frequently among the chronic affective and anxiety disorders. The "dramatic" personality disorders were observed especially frequently among patients with cyclothymic disorder, and the "eccentric" personality disorders among patients with a diagnosis of dysthymic disorder, social phobia and agoraphobia. Dramatic personality disorder was also common among patients with simple phobia. As expected, a close correspondence was observed between social phobia, agoraphobia and avoidant personality disorder, between substance use disorder and borderline personality disorder, and between obsessive-compulsive disorder and compulsive personality disorder. Even if a relationship was observed, it was not strong enough to warrant a combination of chronic symptom disorder diagnoses and personality disorder diagnoses.

DSM-III personality disorders and the outcome of treated panic disorder.

Fifty-two patients with panic disorder who had been receiving active benzodiazepine treatment for 8 weeks were assessed by using the outcome measures of spontaneous and situational panic attacks, scores on the Hamilton scales for anxiety and for depression, and scores on self-rated disability scales. Although spontaneous panic attacks were not affected by the presence of any personality disorder, the remaining outcome measures showed a strong and negative association with DSM-III antisocial, borderline, histrionic, and narcissistic personality disorders. There was also a mild negative association with avoidant personality disorder. A subgroup of patients with both major depression and panic disorder appeared more strongly affected.

Validity of MMPI Scales for DSM-III Personality Disorders: What Are They Measuring?

Morey, Waugh, and Blashfield (1985) derived 11 scales from the MMPI which purportedly measure personality disorders. We correlated these with the first 11 scales of the Millon Clinical Multiaxial Inventory (MCMI), which measure the same factors. Correlations between equivalent scales were low, and one case was negative. On six scales, the MMPI scores correlated higher with MCMI scales measuring different attributes than with MCMI scales measuring the same factor. We conclude that the MMPI scales may be measuring things other than what the scale names imply, and that they should not be used to assess DSM-III personality disorders.

Predicting alcoholism and personality disorder in a 33-year longitudinal study of children of alcoholics.

SummaryThe authors examine adolescent adjustments and the antecedents of DSM‐III alcoholism and DSM‐III personality disorder in a 33‐year longitudinal study of inner‐city, non‐delinquent sons of alcoholic men (COA's). In early adolescence, the 174 COA's (38% of the sample) had more alcoholic relatives (in addition to fathers), more non‐Mediterranean ethnicity, more environmental stresses, and poorer adjustments than their non‐COA peers. Poor adolescent adjustment among COA's was associated with having a poor relationship with one's mother. By midlife, 28% of the COA's had developed alcohol dependence, and 25% were diagnosed with at least one personality disorder. Personality disorder but not alcoholism had strong continuities with adolescent adjustment problems. Alcohol dependence was predicted by total number of alcoholic relatives, non‐Mediterranean ethnicity, low SES, and school behavior problems. Personality disorder was predicted by total environmental weaknesses, poor maternal relationships, low IQ, and feelings of inadequacy.

The DSM-III-R personality disorders: an overview.

The revision of DSM-III (DSM-III-R) includes substantial changes in the axis II personality disorders. The authors present and critically review these revisions and suggest directions for further research. The issues discussed include the multiaxial system, the use of a categorical rather than a dimensional format, the change from monothetic to polythetic criteria sets, the reduction in overlap among criteria sets, the decrease in the amount of unnecessary inferential clinical judgment required to make diagnoses, and the addition of two new personality disorders in an appendix. The criteria sets for many of the DSM-III personality disorders were also substantially revised in DSM-III-R. Changes in each of these are discussed in turn.

Personality disorders in DSM-III and DSM-III-R: convergence, coverage, and internal consistency.

Although many sections of DSM-III-R reflect few changes from DSM-III, the diagnostic criteria for the personality disorders involve more substantial revisions. This study examined the impact of DSM-III-R revisions in terms of convergence with DSM-III, the diagnostic coverage of patients with personality disorders, and the internal consistency of the personality disorder criteria sets. The results demonstrated a substantial divergence between DSM-III and DSM-III-R diagnoses, increased coverage accompanied by a large increase in overlap among personality disorders, and internal consistency estimates of DSM-III-R criterion sets comparable to those in DSM-III.