Diagnosing personality disorders in psychiatric inpatients

Personality disorder, symptoms and dexamethasone suppression in depression

The presence of Personality Disorder (PD) in depressed patients admitted to the acute wards of a psychiatric hospital was assessed using Tyrer's Personality Assessment Schedule, and its relationship to depressive symptoms, improvement after 2 weeks, and the dexamethasone suppression test was examined. 75.4% of the sample met Tyrer's criteria for a PD diagnosis. The initial symptoms of patients with personality disorder did not differ from those without PD. However patients with PD improved more slowly, and had significantly higher symptom scores after 2 weeks. Patients with PD also showed previous history of more depressive episodes, a younger first age of onset, and more frequent and severe previous suicide attempts, in addition to history of unstable marriage. No difference in dexamethasone suppression was found. These findings suggest that presence of personality disorder is related primarily to course of illness and life history, rather than symptom pattern.

An investigation into the internal structure of DSM-III antisocial personality disorder

The present study examined the DSM-III antisocial personality disorder by examining endorsements of each of the DSM-III criteria to obtain various indices, including interrater reliability, sensitivity, specificity, positive and negative predictive power, alpha levels, and item (criterion)-total correlations. 112 psychiatric inpatients were rated on the Structured Interview for DSM-III Personality. 11 patients were accorded a diagnosis of antisocial personality disorder, 65 had other forms of personality disorders, and 36 received no personality disorder diagnosis. The antisocial criteria successfully discriminated patients with antisocial personality disorder from those without the diagnosis. Strong interrater reliability and reasonable alpha levels were achieved. Although some criteria were successful in identifying patients with and without a diagnosis of antisocial personality disorder, the criteria were generally better at predicting the absence of antisocial personality disorder.

Morbid risks for major disorders and frequencies of personality disorders among spouses of psychiatric inpatients and controls

Three hundred fifty-three psychiatric inpatients and their 192 living spouses and 98 control subjects and their 54 living spouses were examined and interviewed for affective, schizoaffective, schizophrenic (Research Diagnostic Criteria [RDC]), and personality disorders (DSM-III-R) using the Lifetime Version of the Schedule for Affective Disorders and Schizophrenia (SADS-L) and the Structured Clinical Interview for DSM-III-Personality Disorders (SCID). The morbid risks of spouses for unipolar depression were between .15 and .25, and those for other major disorders were below .03. The morbid risks of spouses of bipolar patients for unipolar depression exceeded those of other spouses by 50% without reaching statistical significance. Personality disorders were found in 44.6% of patients, in 8.4% of patients' spouses, and in 9.8% of healthy controls. There was only one couple in which the husband and wife had each had a major disorder before marriage. Only four husband-wife pairs suffered the same personality disorder. Spouses of patients do not have significantly more psychiatric disorders than healthy controls; therefore, assortative mating can only be of minor relevance in family studies.

MCMI-II personality disorders in recent-onset bipolar disorders

This study investigated the personality disorders of 21 recent-onset Bipolar Disorder patients using the revised Million Clinical Multiaxial Inventory (MCMI-II; Millon, 1987). Personality disorder assessments, conducted after patients' clinical symptoms had settled, indicated that 17 patients received at least one MCMI-II personality disorder diagnosis with a trend toward multiple diagnoses. Narcissistic, Antisocial, and Histrionic personality disorders were diagnosed most frequently and were the scales most elevated. Schizoid and Compulsive personality disorders were the scales least elevated. Diagnostic concordance between the MCMI-II and the Structured Interview for DSM-III Personality (SIDP; Pfohl, Stangl, & Zimmerman, 1983) was poor; the MCMI-II made more multiple diagnoses. Implications of the discrepancies between these instruments and suggestions for future research are discussed.

The structure of personality disorders in DSM-III

In DSM-III it is proposed that there are 11 personality disorders (PDs) that fall into 3 superordinate clusters: cluster A, labelled the odd or eccentric; cluster B, labelled the emotional, dramatic or erratic; and cluster C, described as the anxious or fearful. To check this proposal, data were obtained from a sample of 112 state hospital psychiatric inpatients via the Structured Interview for DSM-III Personality (SIDP). Various statistical techniques were applied to the data, ranging from the least constrained method of multidimensional scaling to the most constrained and statistically rigorous approach of confirmatory factor analysis. In the latter approach, the models suggested by the various statistical approaches were contrasted directly. A 3-cluster solution was accepted as the best representation of the data, although PD membership of the 3 clusters varied in some ways from those suggested by DSM-III. An exploratory correlation analysis showed that many of the items were more strongly correlated with PDs other than the ones specified by DSM-III, but generally items clustered within the 3 superordinate clusters. This issue deserves further investigation.