DSM-III Personality Research Articles

Longitudinal relationships between personality disorder dimensions and depression in a community sample

BackgroundLittle is known about the long-term relationships between specific personality disorder dimensions and onset of major depressive disorder (MDD) in the community. MethodsA stratified random sample of adult residents of eastern Baltimore was interviewed by psychiatrists in 1981 to assess DSM-III personality disorders. A total of 252 of these individuals were re-examined between 1994 and 1999 and were included in the present analysis. Relationships between scores on personality disorder scales and the onset of MDD in the 13–18 year follow-up period were evaluated using logistic regression models. ResultsForty-nine individuals had an episode of major depression during the follow-up period; 34 of them had their first episode during this period. Only histrionic personality disorder score was significantly related to the occurrence of MDD in the follow-up period. No significant associations were found between first-incidence MDD and any of the personality disorder scores. LimitationsThe study was conducted prior to DSM-5 criteria. ConclusionThe findings suggest that the risk of MDD in adults in the community increases with histrionic personality disorder score. Future prospective population-based studies with larger sample sizes are needed in order to verify the present finding.

An Investigation of Adherence to Diagnostic Criteria, Revisited: Clinical Diagnosis of the DSM-IV/DSM-5 Section II Personality Disorders.

In an initial investigation by Morey and Ochoa (1989), adherence to DSM-III personality disorder diagnostic criteria was examined as an agreement rate between clinician (global) diagnoses and diagnoses algorithmically generated from DSM-III criteria rules. Morey and Ochoa (1989) findings suggested significant clinician-criterion diagnostic incongruity, a result that cross-validated in a DSM-III-R replication performed by Blashfield and Herkov (1996). The current study examined such adherence, utilizing DSM-IV decision rules, in a national sample of 337 clinicians and their target patients. The results of the current study are largely consistent with the earlier findings, with clinician-criterion agreement rates comparable to those commonly reported for interdiagnostician reliability. Ramifications for the future of personality disorder diagnostic classification are discussed.

The stability of DSM personality disorders over twelve to eighteen years

Background Stability of personality disorders is assumed in most nomenclatures; however, the evidence for this is limited and inconsistent. The aim of this study is to investigate the stability of DSM-III personality disorders in a community sample of eastern Baltimore residents unselected for treatment. Methods Two hundred ninety four participants were examined on two occasions by psychiatrists using the same standardized examination twelve to eighteen years apart. All the DSM-III criteria for personality disorders were assessed. Item-response analysis was adapted into two approaches to assess the agreement between the personality measures on the two occasions. The first approach estimated stability in the underlying disorder, correcting for error in trait measurement, and the second approach estimated stability in the measured disorder, without correcting for item unreliability. Results Five of the ten personality disorders exhibited moderate stability in individuals: antisocial, avoidant, borderline, histrionic, and schizotypal. Associated estimated ICCs for stability of underlying disorder over time ranged between approximately 0.4 and 0.7–0.8. A sixth disorder, OCPD, exhibited appreciable stability with estimated ICC of approximately 0.2–0.3. Dependent, narcissistic, paranoid, and schizoid disorders were not demonstrably stable. Conclusions The findings suggest that six of the DSM personality disorder constructs themselves are stable, but that specific traits within the DSM categories are both of lesser importance than the constructs themselves and require additional specification.

Underlying dimensions of ICD-10 personality disorders: Risk factors, childhood antecedents, and adverse outcomes in adulthood

Previous studies have identified specific underlying dimensions of personality disorders with remarkable consistency. This study aimed to replicate the factor structure of personality disorders using ICD-10 dimensional scores from a sample of 366 male offenders. Factor analysis yielded three factors. Factor 1 demonstrated dissocial, paranoid, histrionic, and impulsive personality traits. Factor 2 reflected borderline, anxious, and avoidant PD symptoms. Factor 3 showed high loadings on schizoid PD traits and an inverse relation to anankastic personality. The factor structure was identical to a recent study using ICD-10 personality disorder scores and showed strong similarities to previous studies using DSM-III personality disorder dimensions. Factor 1 was substantially correlated with the psychopathy total score. Elevated scores on this factor were associated with adverse childhood experiences, early behavioural problems (conduct/hyperkinetic disorder), co-morbid substance abuse in adulthood, an earlier start of criminal career, and a higher number of sentences/offences. No risk factors or childhood antecedents could be identified for Factor 2. However, this factor was distinctively characterised by co-morbid affective and neurotic mental disorders in adulthood. Factor 3 was negatively associated with measures of cognitive functioning. The findings suggest specific pathways, courses, and outcomes of the identified personality disorder dimensions.

Dimensions of personality disorders in offenders

Owing to criticisms of current concepts of personality disorders such as high comorbidity, criteria overlap and arbitrary thresholds of categorical diagnoses, a dimensional assessment is proposed that considers interrelations between different personality disorders. Results of previous factor analyses using dimensional personality disorder scores have indicated that one underlying dimension shows strong similarities to the concept of psychopathy and is similarly related to criminal recidivism. The authors examined the underlying dimensions of ICD-10 personality disorders, to analyse their association with criminal behaviour in general, and with specific criminal history variables. Study samples included 105 offenders and 80 non-criminal controls. Personality disorders were measured using a clinical structured interview (IPDE), measures of personality using self-report (NEO-FFI, IPC, HDHQ), and criminal history variables obtained from court records. Three underlying personality disorder factors could be identified, which showed identical structures in both the forensic and the non-forensic sample. Factor 1 comprised emotionally unstable, histrionic, paranoid and dissocial traits and showed strong similarity to the construct of psychopathy. Factor 2 was defined by anankastic personality disorder scores and an inverse relation to schizoid personality features. Factor 3 showed high negative loadings of anxious and dependent personality disorders. Self-report measures of personality and criminal history variables yielded different associations with the three PD dimensions. Offenders with high scores on factor 1 were highly aggressive, violent and impulsive. The findings generally replicated previous analyses using DSM-III personality disorder scores. Differences can be explained by the different constructs of personality disorders included in ICD-10. Although a diagnosis of psychopathy is not currently included in these diagnostic systems, the authors' findings indicate that a highly inter-related pattern of personality disorder scores constitutes psychopathic personality disorder and can be used to identify impulsive, hostile and violent offenders.

The longitudinal relationship between personality disorder dimensions and global functioning in a community-residing population

Little is known about the long-term outcome of personality disorder traits. The purpose of this study was to investigate, in a community-residing population, the longitudinal relationship between psychiatrist-assessed personality disorder scores and global functioning 13-18 years later. A stratified random sample of residents of east Baltimore were examined by psychiatrists in 1981 and asssessed for DSM-III personality disorders using a semi-structured instrument, the Standardized Psychiatric Examination. A total of 292 persons were re-examined by different psychiatrists during 1994-1999 using the Schedules for the Assessment of Neuropsychiatry (SCAN). After completion of the SCAN, the subjects' functional status was evaluated using the Global Assessment of Functioning (GAF). The relationships between personality dimensions and follow-up GAF scores were evaluated using linear regression models. All of the personality disorder scales measured in 1981 were inversely related to functioning 13-18 years later, with the exception of narcissistic and compulsive scales. After controlling for Axis I disorders diagnosed contemporaneously with GAF assessment, schizoid, antisocial, borderline, histrionic, and avoidant personality disorder scores significantly predicted GAF scores. Most dimensions of DSM-III personality disorder traits were significantly associated with global functioning after an interval of 15 years. However, only schizoid, antisocial, borderline, histrionic, and avoidant personality disorder traits had long-term effects on functioning when Axis I disorders at follow-up were controlled. This suggests that the functional effect of the other personality disorder traits may be mediated through their relationship with Axis I disorders. Future research is needed using more specific and sensitive outcome measures.

Cigarette smoking and psychiatric disorder in a community sample.

We examined the lifetime prevalence of psychiatric disorders in cigarette smokers and nonsmokers in a non-patient sample. First-degree relatives of psychiatric patients (n = 697) and normal controls (n = 360) were interviewed with the Diagnostic Interview Schedule and the Structured Interview for DSM-III Personality Disorders. Using these interviews we diagnosed the major mental (Axis I) disorders and personality (Axis II) disorders. A cigarette smoker was defined as someone who smoked daily for a month or more at some time in their lives. We found that smokers more frequently had a lifetime history of major depression, alcohol and drug abuse/dependence, agoraphobia, unstable/acting out and anxious/fearful personality disorders. In a logistic regression analysis, the only significant variables independently associated with smoking status were the alcohol and drug use disorders. Age was an important modifying variable--the smoking-illness relationship was robust in the youngest age cohort and negligible in the oldest cohort. We conclude that cigarette smokers have increased rates of mood, anxiety, substance use, and personality disorders. However, after controlling for the comorbidity among the disorders only alcohol and drug abuse/dependence were independently associated with smoking. Young smokers had particularly high rates of substance use disorders. This age effect may reflect the impact of a quarter century of health education.

Assessing axis II disorders by informant interview.

Although much of personality disorder research depends on diagnostic data obtained directly from patients, this approach has rarely been compared to interviews with knowledgeable informants. The purpose of this study was to determine the diagnostic agreement between these two assessment methods, as well as their relative contribution to the formulation of consensus diagnoses. Sixty-two psychiatric patients were assessed directly with the Structured Interview for DSM-III Personality Disorders (SIDP), and were asked to nominate an informant--either a family member or friend--to provide information about the patient in an interview with the same instrument. Informant interviews were conducted blind to patient-based information whenever feasible, and diagnostic consensus was achieved by an independent review of all available data by a senior clinician. Diagnostic agreement between patient-based and informant-based personality disorder interview was poor, confirming the findings of two previous studies. Information obtained from patients tended to be given greater weight in formulating consensus diagnoses than information provided by informants. However, about one quarter of diagnostic disagreements were resolved in favor of informant-based information. In contrast to a previous study, the inclusion of informant information did not appear to reveal greater psychopathology in patients. We conclude that supplementing direct patient interview with data provided by a knowledgeable informant appears to enhance the resolution of some personality disorder diagnoses. The utility of informant interviews may depend on an analysis of the costs and benefits of this additional degree of descriptive refinement.

Pathological narcissism and narcissistic personality disorder in Axis I disorders.

This paper presents available information on the comorbidity of narcissistic personality disorder (NPD) and pathological narcissism with major mental illness. A review of empirical studies reporting on the prevalence of NPD in Axis I disorders, and of theoretical and clinical literature on narcissistic pathology in major mental illness, forms the basis for an analysis of this interface. The results show that prevalence rates of NPD in Axis I disorders rarely exceed those found in the general psychiatric or personality disorder populations (i.e., less than 22%). NPD was found at high rates in individuals with a substance use disorder (12-38%) or bipolar disorder (4-47%); it was present at very low rates or absent in persons with obsessive-compulsive disorder. Higher prevalence rates were reported in the studies that used the Millon Clinical Multiaxial Inventory I or II than in those that employed the Structured interview for DSM-III Personality Disorders or the Structured Interview for DSM-III-R Personality Disorders--Revised. There is no evidence implicating a significant relationship between NPD and any specific Axis I disorder. A comparison of theoretical and clinical studies with empirical ones reveals major differences in the views regarding the presence and significance of NPD in Axis I disorders. However, the results highlight trends of interacting comorbidity between NPD and substance use disorders, bipolar disorder, depression, and anorexia nervosa.

Relationship between hypomania and personality disorders before and after successful treatment

To examine the effect of hypomanic states on maladaptive personality traits and personality disorders, the authors evaluated personality traits and disorders of patients during an episode of hypomania and after successful somatic treatment. The authors used the Structured Interview for DSM-III Personality Disorders to study 66 outpatients who had a lifetime diagnosis of bipolar disorder and who met the minimum Research Diagnostic Criteria for hypomania. All patients had a knowledgeable informant separately undergo the Structured Interview for DSM-III Personality Disorders during the patient's hypomanic state. Outpatients who successfully recovered from the hypomanic episode (N = 47) and their informants were read-ministered the interview 4-8 weeks after the initial assessment. During the hypomanic state, informants generally reported higher levels of maladaptive personality traits among patients than patients themselves. For the patients who recovered successfully from the hypomanic episode, a reduction in all maladaptive personality traits except schizoid and dependent traits was reported by both patients and their informants; however, the decrease reported by patients generally was much greater than that reported by informants. In addition, schizoid traits actually increased after successful treatment according to patient reports but were unchanged according to informant reports. Hypomania may be associated with an exacerbation of maladaptive personality traits, which may be attenuated after successful treatment. Even with the attainment of euthymic mood, however, about 50% of the cohort had at least one personality disorder, which suggests that a high degree of comorbidity may exist between bipolar disorders and maladaptive personality traits or personality disorders.

Longitudinal follow-up of amenorrhea in eating disorders.

The authors recruited 229 treatment-seeking anorexic and bulimic women for a prospective, longitudinal study. Telephone interviews were arranged every 3 months for at least 1 year for 225 patients. At intake, 132 subjects were menstruating, 34 subjects were taking oral contraceptives, 5 subjects had an organic cause for amenorrhea (e.g., hysterectomy), and 58 subjects were amenorrheic. Each patient met Diagnostic and Statistical Manual of Mental Disorders, third edition, revised (DSM-III-R) criteria for anorexia nervosa (AN, N = 41), bulimia nervosa (BN, N = 98), or AN/BN (N = 90). All subjects were interviewed with the Schedule for Affective Disorders and Schizophrenia-Lifetime Version, which was modified to include a section for DSM-III-R eating disorders, the Longitudinal Interval Follow-Up Evaluation, and the Structured Interview for DSM-III Personality Disorders. It was found that body weight was associated with menstrual status: those with amenorrhea had a mean percent ideal body weight (IBW, Metropolitan Life criteria) of 74 +/- 1% compared with 102 +/- 19% for menstruating patients (p < .01). Affective illness was more prevalent among patients with amenorrhea than among menstruating patients (75% vs. 56%, p < .05). Menses were regained within 1 year by 33% of amenorrheic patients. These patients gained an average of 7.3% of their IBW. Longer duration of eating disorder (p < .03) and the presence of an anxiety disorder (p < .05) were associated with persistent amenorrhea. Menses were lost within 1 year by 8% of menstruating patients. These patients lost an average of 5.0% of their IBW.(ABSTRACT TRUNCATED AT 250 WORDS)

Memories of Parental Rearing and Personality Pathology in Male Alcoholics

In this study, the relation between memories of parental rearing and personality pathology is investigated in 81 male inpatient alcohol addicts. Memories of parental rearing styles were measured with the EMBU, a questionnaire of 65 items about paternal and maternal rearing. Personality pathology was measured with the Structured Interview for DSM-III Personality Disorders (SIDP), on the basis of which dimensional scores were obtained. Results indicated a positive correlation between rejection and favoring by mother on the one hand and the total score for personality pathology as measured with the SIDP on the other hand. The significant correlations with maternal rearing styles were different for different clusters. Personality pathology in the eccentric cluster was associated with a mother who was remembered as emotionally cold and rejecting. In case of pathology in the dramatic cluster, an association was found with favoring and rejecting. In the anxious-avoidant cluster no significant correlation was found. Significant correlations between paternal rearing styles and personality pathology were only found in the dramatic cluster: a positive correlation with rejection and favoring.

Personality Disorder in Later Life: A Community Study

This exploratory study compares the prevalence of personality disorders and traits in people over and under 55 years of age. The comorbidity between personality and other psychiatric disorders is also examined. Psychiatrists examined 810 subjects in a two-stage community survey. The semi-structured Standardized Psychiatric Examination was used to diagnose all DSM-III personality disorders and other psychiatric disorders. The older subjects were significantly less likely than the younger subjects to have any personality disorder (6.6% v. 10.5%; relative odds = 0.42, 95% confidence interval = 0.25-0.70, P < 0.001). Antisocial and histrionic personality disorders were much less prevalent in the older than younger subjects (P < 0.05). The older subjects also had significantly fewer maladaptive personality traits (chi 2 = 88.9, d.f. = 3, P < 0.001). The patterns of comorbidity between personality disorders and other psychiatric disorders were different in the two age groups. It is important to evaluate personality in patients of all ages. While some older patients no longer meet criteria for personality disorder, maladaptive traits may become evident during times of stress.

Personality disorder traits in generalized anxiety and panic disorder patients.

Eight-four panic disorder (PD) and 29 generalized anxiety disorder (GAD) patients were compared with respect to abnormal personality traits assessed by a structured interview (Structured Interview for DSM-III Personality [SIDP]) and a self-report inventory (Personality Diagnostic Questionnaire [PDQ]). An earlier study using many of the same patients by Noyes et al. found PD patients to have more extensive axis I psychopathology than GAD patients. However, in this study it was the GAD patients who appeared to have greater axis II pathology. In particular, when using a subset of patients who had been matched for age and gender, the GAD patients reported more antisocial traits. This finding is particularly interesting, since the matched samples consisted primarily of women in their forties and fifties.

DSM-III personality disorders in the community.

The aims of this study were to estimate the prevalence and investigate the comorbidity and potential consequences of DSM-III personality disorders in the community. A total of 810 adults were examined in the second stage of the Eastern Baltimore Mental Health Survey in 1981, part of the National Institute of Mental Health Epidemiologic Catchment Area program. The subjects were directly examined by psychiatrists using a semi-structured method that allowed diagnosis of all DSM-III personality disorders as well as other DSM-III psychiatric disorders. The prevalence of personality disorders in these adults was 5.9% (9.3% when provisional cases were included). Men had higher rates than women, and subjects who were separated or divorced had the highest rates. There was little comorbidity among specific personality disorders. Subjects with personality disorders were significantly more likely to have a history of sexual dysfunctions, alcohol use disorders, and drug use disorders as well as suicidal thoughts and attempts. In addition, they reported significantly more life events in the past year. Among subjects with any axis I disorder, those with personality disorders were judged by the psychiatrists to be more in need of treatment; however, only 21% were receiving treatment. Personality disorders are relatively common in the community. They are associated with axis I disorders and life events. Only one-fifth of the individuals who qualify for diagnoses of personality disorders in the community are receiving treatment.

Perceptions of Nonshared Childhood Environment in Bulimia Nervosa

The aim of this research was to identify perceptions of nonshared childhood environmental factors associated with bulimia nervosa and the comorbid personality traits associated with this disorder. Twenty-nine individuals with bulimia nervosa and 27 normal controls compared themselves to a sibling on a measure specifically designed to assess perceptions of nonshared environmental factors in parental, peer, and sibling relationships. Also, bulimic individuals' primary clinicians completed a series of ratings for each of the DSM-III personality disorders. Bulimic individuals were more likely than controls to rate their fathers as showing less affection and more control toward them than toward their sibling. Among bulimic individuals, borderline personality disorder ratings were associated with the perception that both mother and father showed less affection toward the bulimic individual than toward her sibling. Avoidant and depressive personality ratings were associated with perceptions of unique experiences with peers. These data provide preliminary evidence suggesting that the paternal relationship may be a source of nonshared environmental experience associated with bulimia nervosa. Borderline personality disorder among bulimics appears related to more pervasive within-family experiences across both the maternal and paternal relationship, while depressive and avoidant personality seems related to unique peer experiences.

Familial Correlates of Reduced Central Serotonergic System Function in Patients With Personality Disorders

To test the hypothesis that evidence of reduced central serotonergic (5-HT) system function in probands with personality disorders is associated with an elevated morbid risk of psychopathological conditions putatively associated with 5-HT dysfunction in first-degree relatives of these probands. Data were collected during a study of the 5-HT correlates of behavior in male patients with DSM-III personality disorders conducted at a Veterans Affairs medical center. Probands in this study were selected from those patients who had undergone both a fenfluramine hydrochloride challenge and a family history assessment. Axis II diagnosis were made according to DSM-III criteria after a structured interview of the proband, using the Structured Interview for Diagnosing Personality Disorders, given by two raters and a similar interview with a knowledgeable informant by another rater. Reduced prolactin responses to the 5-HT releasing/uptake inhibiting agent fenfluramine was associated with an elevated morbid risk of impulsive personality disorder traits in the first-degree relatives of patients with a primary DSM-III diagnosis of a personality disorder. Quantitative scores on assessments of impulsive aggression in the probands were not correlated with an increased morbid risk for impulsive personality disorder traits. A trend in the same direction was noted for affective personality disorder traits and alcoholism. These results suggest that a central 5-HT system abnormality in probands is associated with an increased risk of impulsive aggression in their first-degree relatives, and that assessment of central 5-HT system function in probands may be a more sensitive parameter for identification of this familial trait than the presence of impulsive aggressive behaviors in the proband.

Assessment of DSM-III personality structure in a general-population survey

The object of this study is to assess the internal validity of DSM-III personality constructs and to explore whether the constituent elements are better explained by an alternate internally coherent classification. A two-stage stratified random sample of subjects identified at the Baltimore site of the Epidemiologic Catchment Area (ECA) program were examined by psychiatrists for DSM-III personality attributes using a semistructured instrument. Dichotomous factor analysis was used in the confirmatory mode to test whether a single factor explained each of the 11 DSM-III personality disorders. This approach rejected a single explanatory factor for all but compulsive personality disorder. Exploratory factor analysis showed that these DSM-III personality features are parsimoniously described by a five-factor model. These factors are warmth, animation, timidity, trust, and scrupulousness.

Personality disorder in obsessive-compulsive volunteers, well comparison subjects, and their first-degree relatives

The authors sought to determine the frequency of DSM-III personality disorders in subjects with obsessive-compulsive disorder, well comparison subjects, and their first-degree relatives. Thirty-two subjects with obsessive-compulsive disorder and 33 age-(within 5 years) and gender-matched well comparison subjects were assessed with the Structured Interview for DSM-III Personality Disorder and several self-report instruments. Their first-degree relatives were assessed in a blind manner with the same instruments. Subjects with obsessive-compulsive disorder were more likely than well comparison subjects to have a personality disorder, but compulsive personality was not the most frequent. No significant differences were found in the prevalence of personality disorders among first-degree relatives, including compulsive personality, or in obsessional, hysterical, or oral character traits. Personality disorders are highly prevalent among patients with obsessive-compulsive disorder, but the data do not support a relationship between obsessive-compulsive disorder and compulsive personality.

DSM-III personality disorders in generalized anxiety, panic/agoraphobia, and obsessive-compulsive disorders

In an earlier report, we stated that personality profiles of patients with panic disorder/agoraphobia (n = 187) and obsessive-compulsive disorder ([ODC] n = 51) were similar, albeit more pronounced in OCD, suggesting that the link between panic disorder/agoraphobia and DSM-III personality disorders (PDs) or traits may be nonspecific. The present report extends the comparative study of DSM-III PDs/traits, as assessed by the Personality Diagnostic Questionnaire (PDQ), by adding a third diagnostic group of 39 patients with generalized anxiety disorder (GAD). The personality assessment of panic disorder/agoraphobia and GAD patients yielded virtually identical results on the PDQ and Eysenck Personality Inventory (EPI). Because GAD lacks the prominent panic, phobic, and obsessivecompulsive symptoms of other anxiety disorders, the present findings provide strong support for a nonspecific link between panic disorder/agoraphobia and DSM-III PDs/traits and for the presence of common personality characteristics in anxiety disorders.