The convergent and discriminant validity of the MCMI as a measure of the DSM-III personality disorders

Presurgical Psychological Assessments as Correlates of Effectiveness of Spinal Cord Stimulation for Chronic Pain Reduction

BACKGROUND

Spinal Cord Stimulator (SCS) is a surgically implanted device for patients with certain types of chronic pain. While some studies show the value of psychological screening of potential SCS candidates, no consensus exists. This single-site study analyzed the association of SCS success with psychological assessments (e.g., Millon Behavioral Medicine Diagnostic), beliefs regarding SCS efficacy, self-reported pain and quality of life (QOL) among patients approved for SCS.

METHODS

Potential SCS candidates (N = 200) were contacted 3-7 years after initial psychological and medical clearance for SCS; 59 consented to a structured telephone interview. Thirty-four of the 59 had received a SCS; 25 had not received a SCS. Of the 34 that had received a SCS, 22 were approved by routine psychological evaluation while 12 went through in-depth psychological testing.

RESULTS

The majority of respondents (62%) reported effective pain reduction, and 64% of SCS recipients reported improved QOL. Younger patients reported higher pre-implantation pain scores, and participants with higher levels of pain preimplantation were more likely say they would undergo the procedure again. Finally, persons reporting preoperative alcohol problems were more likely to report lower levels of post-SCS pain.

CONCLUSION

Predictors of pain relief and QOL following SCS may depend on expectations of the device and on individuals' interpretation of pain or psychosocial health.

Evolution of the Millon Clinical Multiaxial Inventory

Dr. Theodore Millon (1928-2014) was a primary architect for the personality disorders in the DSM-III, a structure that has endured into the DSM-5. His 1969 book, Modern Psychopathology, created an elegant framework into which the well-known personality prototypes could be fitted and understood. His theoretical work soon led into the creation of several psychological inventories, most notably the Millon Clinical Multiaxial Inventory (MCMI). The MCMI, now in preparation for its 4th major edition, has been a very popular instrument among clinicians. This article explores the history of the MCMI's development from its origins, through 2 distinct theoretical phases, and to its current status as the MCMI-IV is finalized.

Childhood Emotional Neglect Related to Posttraumatic Stress Disorder Symptoms and Body Mass Index in Adult Women

The relationships among the severity of childhood abuse and neglect, posttraumatic stress disorder (PTSD), and adult obesity were investigated. 207 women ( M age = 26.5 yr., SD = 6.7) completed the Childhood Trauma Questionnaire, Millon Clinician Multiaxial Inventory, and a demographic questionnaire. Analyses of variance indicated that women who reported moderate-to-extreme emotional neglect ( n = 71) had significantly higher PTSD scores and increased BMI compared to women who reported low emotional neglect ( n = 84). Women who reported severe sexual or emotional abuse also had higher PTSD scores, but no relationship was found with BMI when other factors were controlled. Although PTSD scores and self-reported severity of childhood emotional neglect were strongly correlated ( r206 = .61, p < .001), PTSD was not found to be a mediating factor in obesity in women who reported childhood emotional neglect, although depression was.

Narcissism at the crossroads: phenotypic description of pathological narcissism across clinical theory, social/personality psychology, and psychiatric diagnosis

This review documents two themes of emphasis found in phenotypic descriptions of pathological narcissism across clinical theory, social/personality psychology, and psychiatric diagnosis. Clinical theories of narcissism spanning 35 years consistently describe variations in the expression of pathological narcissism that emphasize either grandiosity or vulnerable affects and self-states. Recent research in social/personality psychology examining the structure of narcissistic personality traits consistently finds two broad factors representing Grandiosity-Exhibitionism and Vulnerability-Sensitivity-Depletion respectively. However, the majority of psychiatric criteria for narcissistic personality disorder (NPD) in the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV; American Psychiatric Association, 1994) emphasize expressions of grandiosity. By placing most of the diagnostic emphasis on overt grandiosity, DSM NPD has been limited by poor discriminant validity, modest levels of temporal stability, and the lowest prevalence rate on Axis II. Despite converging support for two phenotypic themes associated with pathological narcissism, psychiatric diagnosis and social/personality psychology research often focus only on grandiosity in the assessment of narcissism. In contrast, clinical theory struggles with a proliferation of labels describing these broad phenotypic variations. We conclude that the construct of pathological narcissism is at a crossroads and provide recommendations for diagnostic assessment, clinical conceptualization, and future research that could lead to a more integrated understanding of narcissistic personality and narcissistic personality pathology.

Interpersonal aspects of personality disorders in opioid-dependent patients: the convergence of the ICL-R and the SIDP-IV

This study aims to establish the convergence of the empirically based Diagnostic Statistical Manual (DSM)- and theory-based interpersonal behavioral approaches to personality, in opioid-dependent patients (n = 110), with the use of the Structured Interview for DSM-IV Personality [Pfohl et al.: Structured Interview for DSM-IV PD. Iowa City, UICM, 1995] and the Interpersonal Checklist-Revised (ICL-R) [DeJong and VandenBrink: ICL-R. St. Oedenrode/NL, Novadic, 1998]. As hypothesized, based on prior research, we found the two approaches to be complementary rather than interchangeable. However, some overlap was found between the SIDP-IV dimensions and the ICL-R, mainly with rebellious/distrustful, reserved/silent and masochistic/self-effacing styles. Results indicate that drug dependence in itself is not a predictor of interpersonal style, while personality pathology is. Patients with a personality disorder (PD) perceive themselves as hostile and submissive, while patients without a PD view themselves as friendly and controlling. The SIDP-IV seems informative in classifying PDs, in addition guidelines for behavioral change, in addicted patients, were provided based on the ICL-R.

Millon Clinical Multiaxial Inventory-III subtypes of opioid dependence: validity and matching to behavioral therapies

The concurrent and predictive validity of 2 different methods of Millon Clinical Multiaxial Inventory-III subtyping (protocol sorting, cluster analysis) was evaluated in 125 recently detoxified opioid-dependent outpatients in a 12-week randomized clinical trial. Participants received naltrexone and relapse prevention group counseling and were assigned to 1 of 3 intervention conditions: (a) no-incentive vouchers, (b) incentive vouchers alone, or (c) incentive vouchers plus relationship counseling. Affective disturbance was the most common Axis I protocol-sorted subtype (66%), antisocial-narcissistic was the most common Axis II subtype (46%), and cluster analysis suggested that a 2-cluster solution (high vs. low psychiatric severity) was optimal. Predictive validity analyses indicated less symptom improvement for the higher problem subtypes, and patient treatment matching analyses indicated that some subtypes had better outcomes in the no-incentive voucher conditions.

The incremental validity of passive-aggressive personality symptoms rivals or exceeds that of other personality symptoms in suicidal outpatients

We examined the incremental validity of passive-aggressive and other personality symptoms, by determining personality syndromes' ability to account for unique variance in distress and impairment as indexed by several Axis I-related symptoms. Two hundred fifty young adults, referred for suicidal behavior or ideation, completed the Millon Clinical Multiaxial Inventory (Millon, 1983) as well as several other symptom measures. The personality syndromes varied considerably in their ability to account for unique variance in Axis I-related symptoms. Several syndromes, including schizoid, schizotypal, and compulsive symptoms, showed relatively low incremental validity, whereas the incremental validity of passive-aggressive symptoms exceeded all others. The criterion of incremental validity represents an inclusive and persuasive means to evaluate personality disorder validity. On this criterion, passive-aggressive symptoms display high validity, and passive-aggressive personality disorder, in some form, may deserve inclusion on Axis II of future editions of Diagnostic and Statistical Manual of Mental Disorders (American Psychiatric Association).

Authority conflicts and personality disorders

We investigated personality disorders (PDs) of young men with chronic authority conflicts using the Millon Clinical Multiaxial Inventory-I (MCMI-I; Millon, 1983). PDs occurred considerably more often in the young men with chronic authority conflicts (62%) compared with a clinical control group of young men with acute authority conflicts (39%). The prevalence of PDs in a normal control group was considerably lower (11%). The passive-aggressive PD is over-represented among young men with chronic authority conflicts; there is hardly any antisocial PD. Millon's conceptualization of these disorders could have influenced these results. The high prevalence of passive-aggressive PDs complicates a good psychotherapeutic outcome in young men with chronic authority conflicts. Our results underscore the clinical utility of the controversial PAPD diagnosis.

Psychopathology, treatment completion and 5 years outcome. A prospective study of drug abusers

Two hundred Norwegian drug abusers who consecutively applied for treatment in a Phoenix-House-type of therapeutic community, were interviewed in intake and personally followed up, on average, 5 years after (response rate 79%). The year prior to follow-up, 20% had no/light use of substances and 56% heavy abuse, including the deceased (12.4%). Personality disorders and psychopathology at application did not directly predict either substance abuse problems nor level of social functioning, but were mainly related to the risk of death. Treatment completion was positively related to the 5-year outcome of social functioning, but not to substance abuse problems. Our findings indicate the need for more systematic training in how to control the use of substances in the Phoenix House model. One should also offer a more longstanding outpatient programme of social skills training, aiming at following these clients for years.

Comparative validity of MMPI-2 and MCMI-II personality disorder classifications

Minnesota Multiphasic Personality Inventory-2 (MMPI-2) overlapping and nonoverlapping scales were demonstrated to perform comparably to their original MMPI forms. They were then evaluated for convergent and discriminant validity with the Million Clinical Multiaxial Inventory-II (MCMI-II) personality disorder scales. The MMPI-2 and MCMI-II personality disorder scales demonstrated convergent and discriminant coefficients similar to their original forms. However, the MMPI-2 personality scales classified significantly more of the sample as Dramatic, whereas the MCMI-II diagnosed more of the sample as Anxious. Furthermore, single-scale and 2-point code type classification rates were quite low, indicating that at the level of the individual, the personality disorder scales are not measuring comparable constructs. Hence, each instrument is providing similar and unique information, justifying their continued use together for the purpose of diagnosing personality disorders.

Personality disorder correspondence among the MMPI, MBHI, and MCMI

MMPI, MBHI, and MCMI personality disorder scales were analyzed for convergent and discriminant validity. Friedman's ANOVA indicated that there were no significant differences among the sample's averaged scale scores. Further analyses of these data, however, demonstrated that the Millon instruments classified significantly more of the sample as personality disordered when compared to Morey's MMPI personality disorder scales. In addition, codetype correspondence among the three instruments was only 4 to 6%. When the instruments were analyzed in a pair-wise fashion, codetype correspondence increased to approximately 10 to 20%. These data indicate that these personality disorder scales do not demonstrate construct equivalence, particularly at the level of the individual profile.

Diagnosing Personality Disorders: An Examination of the MMPI-2 and MCMI-II

Personality disorders are highly prevalent in clinical populations and affect outcomes across all forms of intervention. This investigation examined the diagnostic efficiency of two widely used, self-report measures of personality disorder (MMPI-2; Butcher, Dahlstrom, Graham, Tellegen, & Kaemmer, 1989; MCMI-II; Millon, 1987), as compared to a structured interview (SCID-II; Spitzer et al., 1987) diagnosis. The measures were administered to 150 residential and outpatient volunteer subjects. Persons with primary organic or psychotic-spectrum disorders were excluded from participation. Results were variable across disorders measured, with low to moderate levels of diagnostic agreement observed. The MCMI-II appears to be a more sensitive measure, whereas the MMPI-2 is more specific. The two self-report measures demonstrated greater convergence with each other than with the interview measure. Both the MMPI-2 and MCMI-II were more accurate at identifying the absence of a given disorder. Although overall diagnostic powers exist at acceptable levels. the results suggest that diagnoses generated by self-report versus interview are not interchangeable.

A comparison of the differential diagnostic efficiency of three personality disorder inventories

This study was a comparison of the relative diagnostic efficiency of the Millon Clinical Multiaxial Inventory-II, the MMPI Personality Disorder Scales, and the Personality Disorder Questionnaire-Revised. The Structured Clinical Interview for DSM-III-R, Axis II was used as the criterion measure. The operating characteristics of all three objective instruments revealed a high rate of false-positive diagnoses, but negative test results were generally valid. Each instrument was found to possess clinical utility in its ability to improve diagnostic efficiency over base rate predictions for most disorders. It is suggested that many of the psychometric limitations revealed in this study reflect the limitations of DSM-III-R, Axis II.

Personality style codetype concordance between the MCMI and MBHI

Personality scale profiles of Millon's Clinical Multiaxial Inventory and Behavioral Health Inventory were analyzed for codetype concordance using a sample of psychiatric patients with medical symptoms (N = 91). Results indicated that at least half of the sample would be diagnosed with a personality disorder by each instrument. Approximately 95% of the sample produced at least one elevated scale. The tests were not significantly different with respect to their classification rates of pathology or number of scales elevated and demonstrated significant and positive correlations between complementary scales. Nonetheless, codetype correspondence was found to be 29.67% for single scales and 20.88% for two-point codetypes. Correlational data, factor analysis, and paired mean score comparisons were used to analyze these findings.

The relationship between MCMI personality scales and clinician-generated DSM-III-R personality disorder diagnoses

This study examined the relationship between elevations on the personality scales of the Millon Clinical Multiaxial Inventory (MCMI) and clinician-generated Diagnostic and Statistical Manual of Mental Disorders (3rd ed., rev. [DSM-III-R]; American Psychiatric Association, 1987) diagnoses for 101 psychiatric patients at a VA medical center/psychiatric hospital in the Southeastern United States. Personality disorder diagnoses were made by employing a personality symptom checklist that consisted of all the verbatim criteria for personality disorders contained in the DSM-III-R. Clinicians who completed the checklists were required to have had at least 5 hr of direct contact with the patients who completed the MCMI. The results indicated that only the Schiozotypal scale of the MCMI was related to its respective DSM-III-R personality disorder in the simple correlation. An examination of the diagnostic efficiency statistics for each of the MCMI personality disorder scales revealed overall low sensitivity, poor specificity, poor positive predictive power, and low diagnostic power, which suggests that the MCMI may have only limited utility in identifying personality disorders.

Validation of the MCMI-I Borderline Personality Disorder Scale with a well-defined criterion sample

This study examined the MCMI-I BPD scale's accuracy in assigning borderline personality disorder as a primary diagnosis. Clinicians with particular expertise in Axis II pathology diagnosed borderline patients in two groups in this sample. Results indicate that the MCMI-I BPD scale has very limited utility as a screening instrument for individual diagnoses when a well-defined borderline sample is-used as a criterion. This study further suggests that at present only clinicians with particular expertise in the diagnosis of personality disorders can assess the degree and type of Axis II pathology present in a given patient. Because most structured interview and self-report validation studies have not included well-defined criterion groups, diagnostic validity of particular measures generally has not been established beyond the level of concordance.

MCMI-II diagnosis of borderline personality disorder: base rates versus prototypic items

The Millon Clinical Multiaxial Inventory-II (MCMI-II) profiles of 26 psychiatric inpatients diagnosed as having borderline personality disorders were compared with profiles of 42 patients with no personality disorders. The borderline group scored significantly higher on the following scales: Disclosure (X), Debasement (Z), Passive-Aggressive (8A), Self-Defeating (8B), Borderline (C), and Major Depression (CC). Differences approaching significance were also found on substance abuse measures: the Alcohol Dependence (B) scale and Drug Dependence (T) scale. These findings are consistent with criteria established in the Diagnostic and Statistical Manual of Mental Disorders (3rd ed., rev. [DSM-III-R]; American Psychiatric Association, 1987) and the results of other studies utilizing the MCMI-II. In addition, diagnostic efficiency of Scale C was assessed at various cutoff points defined by either base rate (BR) scores or the number of prototypic items endorsed. The greatest efficiency was found when a cutoff of seven or more prototypic items was utilized, with nearly 80% of the patients correctly classified. Results are discussed in terms of their relevance for further research.

Psychopathology and substance abuse as predictors of program completion in a therapeutic community for drug abusers: a prospective study

A total of 144 substance abusers entering consecutively a hierarchical therapeutic community, Phoenix House, Oslo, were followed prospectively 18 months through the entire program. Thirty percent completed the 1-year inpatient phase and 20% completed the total program. Using a structured interview and 3 self-report instruments, the Millon Clinical Multiaxial Inventory, Basic Character Inventory and Symptom Checklist-90, significant differences between dropouts and completers were found in 3 areas of pre-admission characteristics: gender, substance abuse and psychopathology. Completers were characterized by a greater percentage of females, a higher frequency of amphetamine use and a lower frequency of alcohol use and a higher frequency of histrionic and a lower frequency of schizotypal traits than dropouts. The study indicates the importance of paying closer attention to these factors during treatment.

Factor structure of the MCMI among personality disordered outpatients and in other populations

This study examined the factor structure of the Millon Clinical Multiaxial Inventory (MCMI) for a sample (N = 60), which consisted largely of personality disordered outpatients. Results are integrated with previous studies to identify the underlying MCMI dimensions across differing psychiatric populations. Three principal axis Varimax rotated factors emerged from the current study: affectively charged general maladjustment; impulsive acting-out style; and psychotically tinged isolation. Comparison with results of previous studies that involved four diverse samples indicated a large amount of stability for scales that load on the first factor. Factor two shows much less consistency across the five studies, but frequent reappearance of the drug abuse and other acting-out scales suggests a general dimension that reflects acting-out substance abuse. Across studies the third factor shows even more inconsistency, although there is some evidence for a social detachment pattern.

Personality disorders among depressed outpatients as identified by the MCMI

Personality disorders frequently are seen among depressed psychiatric inpatients. In a group of 73 depressed psychiatric outpatients, this study identified 85% whose response to the Millon Clinical Multiaxial Inventory (MCMI) suggested a personality disorder. On retest 12 weeks later, only 64% manifested a response style consistent with the DSM-III personality disorders. These findings are discussed in relation to methodology, characteristics of the MCMI, and the response style of the depressed patient, which may lead to an overidentification of personality disorders on a self-report measure. Related aspects of the diagnostic criteria for personality disorders also are discussed.

Personality disorder assessment for psychiatric inpatients

The standardised criteria for DSM-III personality disorders encouraged the development of numerous multidimensional instruments to make the diagnosis of such disorders more objective and reliable compared with clinical judgement. Yet, there is no published research on the concordance between these instruments when used with psychiatric inpatients. Two such measures, the Millon Clinical Multiaxial Inventory (MCMI) and the Structured Interview for DSM-III Personality (SIDP), were chosen for this study. The MCMI identified a significantly greater number of personality disorders than the SIDP for each subject. Despite overall high interrater reliability with the SIDP, low agreement was found between the two instruments for most of the 11 DSM-III Axis II (personality disorders) categories. Different normative populations on which the instruments were developed, method variance, and theoretical differences between Millon and DSM-III, were all likely to contribute to the low concordance. Systematic research of possible confounding factors is suggested.

MMPI personality disorder scales and the MCMI: concurrent validity

A set of DSM-III personality disorder scales recently were developed for the MMPI. The purpose of the present investigation was to provide data on the concurrent validity of these scales. Scores from both the overlapping and nonoverlapping MMPI personality disorder scales were compared against the criterion of the personality disorder scales of the MCMI in 47 psychiatric patients. In general, the MMPI personality scales correlated significantly with the corresponding MCMI scales. However, the antisocial (ANT), compulsive (CPS), and paranoid (PAR) scales of the MMPI failed to achieve adequate concurrent validity. For the ANT scale, this probably reflects conceptual differences between the Millon (1981) typology that underlies the MCMI and the DSM-III. The PAR scale appears to show some bias toward the assessment of paranoid and suspicious ideation, rather than paranoid personality characteristics.

An update of the Millon Clinical Multiaxial Inventory (MCMI)

Reviews, commentary, and recent research on the MCMI are examined. There has been controversy over the impact on research of delayed availability of hand scoring materials, the discrepancies between Millon's diagnostic conceptualizations and DSM-III, and MCMI construction idiosyncrasies. MCMI use for clinical diagnosis continues to be suspect, but clinicians now have an augmented awareness of DSM limitations as a result of this version of the Millon instrument.