Borderline pathology and the Personality Assessment Inventory (PAI): an evaluation of criterion and concurrent validity

In this study, we examined how patients diagnosed with borderline pathology (BP) would respond on the Personality Assessment Inventory (PAI; Morey, 1991) Borderline (BOR) scales in relation to patients without BP pathology. In addition, we examined whether the PAI BOR scales would be related to variables on the Social Cognition and Object Relations Scale (SCORS; Hilsenroth, Stein, & Pinsker, 2004; Westen, 1995) derived from early memory narratives. Results indicate that outpatients with a Diagnostic and Statistical Manual of Mental Disorders (4th ed. [DSM-IV]; American Psychiatric Association, 1994) diagnosis of BP scored significantly higher on the PAI BOR Total (BOR-Total) score, Identity Problems, and Self- Harm scales in comparison to a Non-BP clinical sample. The overall correct classification rate for the presence or absence of BP using the BOR Total scale (T >or= 70) was 73%. In addition, there were several significant relationships between dimensional PAI BOR scales and the presence versus absence of DSM-IV BP. Moreover, both the BOR-Total and Affect Instability scales were significantly related to the SCORS variable Complexity of Representations. We provide clinical examples to illustrate these research findings in an applied manner.     

Correspondence between self-report and interview-based assessments of antisocial personality disorder

Antisocial personality disorder (ASPD) is associated with suicide, violence, and risk-taking behavior and can slow response to first-line treatment for Axis I disorders. ASPD may be assessed infrequently because few efficient diagnostic tools are available. This study evaluated 2 promising self-report measures for assessing ASPD--the ASPD scale of the Personality Diagnostic Questionnaire-4 (PDQ-4; S. E. Hyler, 1994) and the Personality Assessment Inventory (PAI; L. Morey, 1991, 2007)--as well as the ASPD module of the Structured Clinical Interview for DSM-IV Axis II (SCID-II; M. B. First, R. L. Spitzer, M. Gibbon, J. B. W. Williams, & L. S. Benjamin, 1997). The measures were administered to 1,345 offenders in court-mandated residential substance abuse treatment programs and prisons. PDQ-4 and PAI scores related strongly to SCID-II symptom counts (rs = .67 and .51, respectively), indicating these measures convey useful clinical information about the severity of offenders' ASPD pathology. The dimensional association between the measures was relatively invariant across gender, race, and site, although differences in mean scores were observed. Levels of agreement of the SCID-II with the PDQ-4 (kappa = .31) and PAI (kappa = .32) in classifying participants as ASPD was limited. Alternative thresholds for both self-report measures were identified and cross-validated.     

Use of structured self-report assessment to diagnose borderline personality disorder during major depressive episodes

Diagnosis of borderline personality disorder (BPD) during episodes of major depression (MDE), although clinically important, is complicated in several respects when using self-report methods. Structured interview data were used to select a group of patients with comorbid BPD (n=21) from a sample of outpatients presenting with MDE. This group was compared with a group of MDE patients without BPD (n=24) and with a group of community controls (n=20) using self-report data from the Personality Assessment Inventory (PAI), the revised Personality Diagnostic Questionnaire (PDQ), and the Beck Depression Inventory (BDI). Analyses revealed that the BPD group obtained significantly higher scores on PAI and PDQ scales measuring features of BPD and on the PAI Negative Impression Management scale. The severity and type of MDE symptoms reported on the PAI and BDI did not differentiate the clinical groups. These data show that useful information for the diagnosis of BPD during depressive episodes can be gathered from self-report assessment instruments like the PAI.     

Childhood antisocial behaviors as predictors of psychotic symptoms and DSM-III-R borderline criteria among inpatients with borderline personality disorder.

Structured clinical interviews of 107 female inpatients diagnosed with borderline personality disorder (BPD) were used to determine whether antisocial personality disorder (APD) diagnostic criteria evident prior to age 15 could be used to predict current Axis I and Axis II psychopathology. Diagnostic information was gathered using the Structured Clinical Interview for DSM-III-R Personality Disorders (SCID-II) and the Structured Clinical Interview for DSM-III-R-Patient Version (SCID-P). Childhood APD criteria were subjected to principal-components analysis, and three factors--rule-breaking, assault, and sadism--emerged. The severity of the childhood APD criteria was related to psychotic symptoms, as well as to the unstable relationships and labile affect BPD criteria and the current overall severity of BPD criteria. Sadism predicted psychotic symptoms and BPD severity, while rule-breaking predicted unstable relationships and BPD severity. Childhood APD severity also had a larger effect on BPD severity than on psychotic symptoms. Possible explanations for these findings are explored and discussed.     

The relationship between borderline personality disorder and number of sexual partners

In this study, we examined the number of self-reported lifetime sexual partners between those with and without borderline personality disorder (BPD). Our hypothesis was that, due to the nature of the disorder, those with BPD would exhibit a greater number of lifetime sexual partners. Using a survey methodology in a consecutive sample of internal medicine outpatients (N = 354), we administered two self-report measures for BPD (the borderline personality scale of the Personality Diagnostic Questionnaire-4 and the Self-Harm Inventory), and queried participants, "How many sexual partners have you had in your lifetime?" Correlation coefficients and one-way ANOVAs were conducted to determine the relationship between BPD and the number of sexual partners. According to BPD status with the borderline personality scale of the Personality Diagnostic Questionnaire-4, we found that the mean number of sexual partners among BPD-positive participants was nearly double (p < .001) that of BPD-negative participants; according to BPD status with the Self-Harm Inventory, the mean number of sexual partners among BPD-positive participants was more than double (p < .001) that of BPD-negative participants. Compared to internal medicine outpatients without BPD, those with BPD exhibit a statistically significantly greater number of sexual partners.     

Validation of the Five-Factor Model Personality Disorder Prototype Counts with the Personality Assessment Inventory

Five-factor model (FFM) personality disorder (PD) prototype scores drawn from the revised NEO-Personality Inventory (NEO-PI-R) have demonstrated strong convergent validity with clinical measures of PD (Miller in Journal of Personality, 80, 1565–1591, 2012). However, an examination of the relations between the Personality Assessment Inventory (PAI; Morey 1991), which is widely used in personality assessment research, and the FFM PD prototype “count” method, has not been conducted. The present study examines the convergent validity of the FFM borderline, antisocial, and psychopathy prototype counts with the Borderline Features (BOR) and Antisocial Features (ANT) full scales and subscales from the PAI. Furthermore, the BOR and ANT scales are correlated with all 30 NEO-PI-3 facets to empirically evaluate the rationally-selected facet scales used in the borderline, antisocial, and psychopathy PD counts. Five hundred thirty-five undergraduate students completed both the NEO-PI-3 and the PAI. The PD counts demonstrated strong convergent and discriminant validity with the PAI clinical scales. Facet-level analyses generally supported the structure of the PD count formulas, and the exceptions align with previous evidence that the Competence, Dutifulness, and Self-Discipline facets (from the Conscientiousness domain of the FFM) associate strongly with the BOR scale.     

Borderline and avoidant personality disorders and the five-factor model of personality: a comparison between DSM-IV diagnoses and NEO-PI-R.

A self-report measure of the Five-Factor Model (FFM) of personality, NEO-PI-R, was administered to a sample of patients with borderline (BPD, N = 29) or avoidant PD (AVPD, N = 34), admitted to a day treatment program, to investigate the NEO-PI-R profiles of the disorders, and the ability of NEO-PI-R to discriminate between the two disorders. The diagnoses were assessed according to the LEAD standard. AVPD was associated with high levels of Neuroticism and Agreeableness, and low levels of Extraversion and Conscientiousness. BPD was associated with high levels of Neuroticism and low levels of Agreeableness, Extraversion, and Conscientiousness. Eighty-eight percent of the AVPD group had high scores on Neuroticism and low scores on Extraversion, whereas 65% of the BPD group were high on Neuroticism and low on Agreeableness. The Extraversion and Agreeableness scales of NEO-PI-R discriminated between patients with BPD and those with AVPD. Patients with BPD scored significantly higher on the Angry Hostility and Impulsiveness subscales of Neuroticism and significantly lower on three Extraversion subscales, three Agreeableness subscales, and one Conscientiousness subscale. At the DSM-IV criterion level, there were more significant relationships between the subscales of NEO-PI-R and the AVPD criteria than with the BPD criteria. The findings suggest that the FFM has good discriminating ability regarding BPD and AVPD. However, there may be a closer conceptual relationship between the FFM and AVPD than between the FFM and BPD.     

Mediators of the relationship between childhood sexual abuse and suicidal behavior in borderline personality disorder

A history of childhood sexual abuse (CSA) increases the risk of suicidal behavior and the lifetime number of suicide attempts in adults with BPD. Despite a strong association, specific variables mediating the relationship between CSA and adult suicidal behavior in BPD have not been identified. This study examined risk factors for suicidal behavior as potential mediators of the relationship between CSA and suicide attempts in 151 men and women with BPD. Diagnostic, clinical, and psychosocial risk factors, as well as CSA, and suicidal behaviors, were assessed by standardized interviews or self-rated measures. Psychotic and schizotypal symptoms, and poor social adjustment, were associated with both CSA and suicidal behavior, and partially mediated the relationship between CSA and attempter status. Psychotic and schizotypal symptoms in BPD define a vulnerability to cognitive and perceptual distortions under stress. They increase the likelihood of suicidal behavior in BPD, especially in the absence of mitigating social support. A theoretical model for suicidal behavior in BPD is presented which integrates direct, mediated, and moderated pathways between childhood sexual abuse and temperament, adolescent and adult risk factors, and suicidal behavior.     

Development and first results of the Borderline Personality Inventory: a self-report instrument for assessing borderline personality organization.

This article reports the development of a short 53-item true-false self-report instrument, the Borderline Personality Inventory (BPI). The BPI is based on Kernberg's (1984) concept of borderline personality organization. However, the diagnostic criteria are compatible with both the Diagnostic and Statistical Manual of Mental Disorders (4th ed.; American Psychiatric Association, 1994) and Gunderson's (Gunderson & Kolb, 1978) concept of borderline personality disorder (BPD). The BPI contains scales for assessing identity diffusion, primitive defense mechanisms, and reality testing. Another scale refers to the fear of closeness. The construction of the scales is based on the results of a factor analysis. As a result of several studies, a cutoff score is proposed to make the diagnosis of a BPD. Thus, the BPI combines dimensional and categorical models of BPD. The BPI was tested in several studies. According to the results, internal consistency and retest reliability are satisfactory (Cronbach's alpha = .68-.91, rtt = .73-.89). Results for sensitivity are .85 to .89, and results for specificity .78 to .89. The BPI identifies borderline patients in high agreement with Kernberg's criteria of borderline personality organization, Diagnostic and Statistical Manual of Mental Disorders (3rd ed., rev.; American Psychiatric Association, 1987) criteria of BPD and Gunderson's criteria of BPD. The BPI is recommended as a screening instrument for borderline personality organization and BPD and for dimensional research of borderline features in Axis I and Axis II disorders.     

Heterogeneity of borderline personality disorder: do the number of criteria met make a difference?

Many studies have compared the demographic and clinical characteristics of patients with and without borderline personality disorder (BPD), but there is limited knowledge on differences within the population of borderline patients. One potential index of heterogeneity is disorder severity. In the present report from the Rhode Island Methods to Improve Diagnostic Assessment and Services (MIDAS) project, we examined whether the severity of borderline personality disorder, as measured by the number of criteria present, is associated with co-morbidity of Axis I and Axis II diagnoses, as well as demographic factors and psychosocial functioning. Two thousand three hundred psychiatric outpatients were interviewed with the Structured Interview for DSM-IV Personality (SIDP-IV). Approximately ten percent (n = 237) of the patients were diagnosed with BPD, and they were divided into four groups based on the number of DSM-IV criteria met, 5 (n = 89), 6 (n = 70), 7 (n = 46), and 8 or 9 (n = 32). There were greater rates of drug use disorders and comorbid Axis II disorders, as well as a greater number of suicidal gestures, in patients meeting seven or more BPD criteria. There were no significant differences between the groups in the number and specific rates of other co-morbid Axis I disorders, other measures of psychosocial functioning, or demographic correlates. Sub-typing of borderline patients by the number of criteria met provides a limited explanation for heterogeneity within BPD patients.     

Borderline Personality Disorder and Posttraumatic Stress Disorder: Time for Integration?

The increasing prevalence of borderline personality disorder (BPD) and posttraumatic stress disorder (PTSD) diagnoses among women illustrates problems and limitations of the medical model system (Diagnostic and Statistical Manual of Mental Disorders, fourth edition; American Psychiatric Association, 1994). In particular, diagnoses of BPD continue the trend of overpathologizing women's issues and reinforcing a patriarchal system of diagnosis and treatment. A PTSD diagnosis, with similar criteria, is preferred for traumatized women because it portrays them in a more positive context. This article explores the overlapping relationship between BPD and PTSD and critiques how both are viewed within the mental health community. Previous research on BPD and PTSD is explored, as well as concerns and limitations regarding both diagnostic categories.     

Feigning combat-related posttraumatic stress disorder on the personality assessment inventory

This study examined whether individuals who were instructed on the Diagnostic and Statistical Manual of Mental Disorders (4th ed. [DSM-IV]; American Psychiatric Association, 1994) criteria for posttraumatic stress disorder (PTSD) could feign PTSD on the Personality Assessment Inventory (PAI; Morey, 1991). The study also investigated whether PAI indexes of symptom exaggeration, the Negative Impression Management (NIM) scale and the Malingering index, could identify individuals feigning PTSD. The diagnostic rule for PTSD (Morey, 1991, 1996) was applied to the profiles of a group of 23 veterans with combat-related PTSD and 23 male undergraduates instructed to malinger PTSD. Seventy percent of the student malingerers produced profiles that received diagnostic consideration for PTSD. The NIM cutting score (> or = 8) was highly effective in detecting simulation of PTSD but resulted in the misclassification of a large number of true PTSD cases. There were no significant differences in the overall efficiency of the test with various validity criteria. We discuss the implications of these findings for the use of the PAI in the diagnosis of combat-related PTSD.     

Assumptions in borderline personality disorder: specificity, stability and relationship with etiological factors.

The specificity and stability of a set of assumptions hypothesized to be characteristic of Borderline Personality Disorder (BPD) was investigated. BPD patients (n = 16) were compared to cluster-C personality disorder patients (n = 12) and to normal controls (n = 15). All subjects were female and diagnosed with SCID-I and -II. Subjects rated a short version of the Personality Disorder Beliefs Questionnaire (PDBQ), with six sets of 20 assumptions each, hypothesized to be characteristic of avoidant, dependent, obsessive-compulsive, paranoid, histrionic and borderline personality disorder. The BPD assumptions (Cronbach alpha = 0.95) proved to be the most specific to BPD patients. Subjects rated the shortened PDBQ again after viewing an emotional video fragment one week later. Despite increased negative emotions, the PDBQ ratings remained relatively stable. Confirming the cognitive hypothesis, regression analyses indicated that the BPD assumptions mediate the relationship between self-reported etiological factors from childhood (sexual abuse and emotional/physical abuse) and BPD pathology assessed with the SCID-II. It is suggested that a set of assumptions is characteristic of BPD, and is relatively stable despite the instability of the behaviour of people diagnosed as having BPD.     

Zanarini Rating Scale for Borderline Personality Disorder (ZAN-BPD): a continuous measure of DSM-IV borderline psychopathology

The purpose of this study was to assess the psychometric properties of the Zanarini Rating Scale for Borderline Personality Disorder (ZAN-BPD), the first clinician-administered scale for the assessment of change in DSM-IV borderline psychopathology. The questions for the measure were adapted from the BPD module of the Diagnostic Interview for DSM-IV Personality Disorders (DIPD-IV) to reflect a 1-week time frame and each of the nine criteria for BPD is rated on a five-point anchored rating scale of 0 to 4, yielding a total score of 0 to 36. Two diagnostic interviews that assess the presence of BPD were administered to 200 nonpsychotic patients: the BPD module of the DIPD-IV and the Revised Diagnostic Interview for Borderlines (DIB-R). The ZAN-BPD was also administered, blind to diagnostic information. In addition, each patient filled out a self-report measure of general psychopathology that is often used in borderline treatment studies, the Symptom Checklist 90 (SCL-90). The convergent validity of the ZAN-BPD and relevant scales of the SCL-90 and the DIB-R was assessed and found to be highly significant. The discriminant validity of the various scores of the ZAN-BPD was also found to be highly significant, easily discriminating the 139 patients who met the DSM-IV criteria for BPD from the 61 patients who did not. In addition, internal consistency of the ZAN-BPD was found to be high (Cronbach's alpha=0.85). The interrater reliability of the ZAN-BPD was assessed using 32 conjoint interviews, while same day test-retest reliability was assessed in a separate sample of 40 patients. All reliability raters were blind to all previously collected information concerning each subject. All intraclass correlations were in the good to excellent range. Finally, the sensitivity of the ZAN-BPD to change was assessed using a third sample of 41 patients who were reinterviewed by a blind rater 7 to 10 days after the ZAN-BPD was first administered. The SCL-90 was also readministered at this time. The correlations between difference scores of the ZAN-BPD and difference scores of the SCL-90 were found to be significant, indicating that the ZAN-BPD measures change in a clinically meaningful manner. Taken together, the results of this study suggest that the ZAN-BPD is a promising clinician-administered scale for the assessment of change in borderline psychopathology over time.     

Discriminant validity of the MMPI for borderline personality disorder

This study explores the discriminant validity of the MMPI in relationship to the DSM-III and the Diagnostic Interview for Borderlines (DIB) constructs of Borderline Personality Disorder (BPD). A two-way analysis of variance model assessed differences between inpatients (n = 42) and outpatients (n = 42) and between BPD and nonBPD patients. We compare the best discriminant model for the current samples with previous BPD discriminant functions. The present study assesses a wider range of psychopathology in which the MMPI appears to be relatively insensitive to the BPD construct while retaining the high specificity reported in previous studies. Implications for the use of the MMPI as a diagnostic instrument are discussed.     

Reliability and validity of three screening measures of borderline personality disorder in a nonclinical population

This study explored reliability and validity of three self-report screening measures of borderline personality disorder (BPD): the Personality Assessment Inventory-Borderline Features Scale (PAI-BOR; Morey, 1991), Personality Diagnostic Questionnaire-4-BPD scale (PDQ4-BPD; Hyler, 1994) and Mclean Screening Instrument for BPD (MSI-BPD; Zanarini et al., 2003). Participants (N = 523) were drawn predominantly from community and student populations. All three measures were internally consistent. Confirmatory Factor Analysis of the PAI-BOR did not support Jackson and Trull’s (2001) six-factor structure or Morey’s (1991) four-factor structure. The three measures converged highly, despite the broader theoretical basis of the PAI-BOR. Expected demographic differences on the measures supported construct validity, and concurrent validity in relation to theoretical-related criteria was also supported. Only the PAI-BOR and PDQ4-BPD showed incremental validity in predicting specific criteria, although, the amount of unique variance predicted was small.     

An application of item response theory to the DSM-III-R criteria for borderline personality disorder

This paper summarizes results from analyses of the DSM criteria for borderline personality disorder (BPD) using models from item response theory (IRT). The study sample consisted of 353 participants, the majority of whom were psychiatric patients. Confirmatory factor analysis showed that a one-factor model provided the best fit to the data. All the DSM BPD criteria had moderate or higher item discrimination parameters, indicating that all items contributed meaningful information in assessing BPD. Item information functions revealed that the BPD criteria as a whole were useful for capturing BPD traits in the moderately severe to severe range, but that they performed less well in the less severe range. The general conclusion is that the criteria do represent a coherent syndrome and that further research on the informational value of the individual criteria would be useful.     

Borderline personality and obsessive-compulsive symptoms

Diagnostic criteria for borderline personality disorder (BPD) are diverse, covering a broad range of symptoms. One criterion, self-mutilation, is a behavioral excess that may be considered a predictor of other psychopathological states. The present study sought to determine the extent to which two groups of BPD patients, those who mutilate and those who do not, differed on measures of general psychopathology, depression, anxiety, and obsessive-compulsive symptoms. Results indicate that the only source of significant variation was the level of obsessive-compulsive symptoms, with mutilators exhibiting greater symptomatology. The findings from this study support the idea that self-mutilation is a more severe form of psychopathology relative to the rest of the BPD population. These results are interpreted based on the affect regulation model of self-mutilation, and contrasted with contemporary models of impulse control in relation to obsessive-compulsive disorder.     

Construct Validity of the DSM–5 Section III Personality Trait Profile for Borderline Personality Disorder

This study evaluated the nomological network of the borderline personality disorder (BPD) trait profile in the Diagnostic and Statistical Manual of Mental Disorders (5th ed. [DSM–5]) Section III. BPD symptoms include a variety of maladaptive thoughts and behaviors, and it is important to determine if the Section III trait operationalization for BPD captures these behavioral symptoms, as well as shows similar associations as the traditional Section II version with external criteria. For this purpose, we used a sample of 285 undergraduate students and conducted correlation and regression analyses to delineate the associations between Section III BPD traits and conceptually relevant external criteria. A Section III Total score was meaningfully associated with all criteria. Moreover, externalizing psychopathology tended to be most highly associated with disinhibitory Section III BPD traits, whereas internalizing psychopathology tended to have its strongest unique associations with traits reflective of negative affectivity. These results provide support for the construct validity of the trait profile for BPD in DSM–5 Section III.