A cross-validation of MMPI-2 substance abuse scales

Spanish Versions of the MMPI-2 and PAI: An Investigation of Concurrent Validity With Hispanic Patients

Delivery of mental health services to Hispanic populations requires that assessment methods address the linguistic needs of their clientele. Toward this end, Spanish translations of two recent multiscale inventories (i.e., Minnesota Multiphasic Personality Inventory-2, MMPI, and Personality Assessment Inventory, PAI) were published in order to meet these needs. Unfortunately, the linguistic equivalence and criterion-based validation of these Spanish versions were not addressed prior to publication. We sought to take the first step in remedying this oversight by administering the MMPI-2 and PAI to 105 Hispanic patients for whom Spanish is the sole or preferred language. We employed a revised Spanish version of the Diagnostic Interview Schedule (DIS) as a criterion measure. By focusing on four common disorders (i.e., major depression, schizophrenia, anxiety disorders, and alcohol dependence), we examined the usefulness of codetypes and clinical elevations for establishing these disorders. Overall, we found moderate hit rates for the MMPI-2 ( N = .60, range from .46-.69) and moderate to high hit rates for the PAI ( M = .72, range from .57-.87). Correlations of selected scales with DIS symptoms varied widely and were generally in the low to moderate range. Finally, exploratory data suggested few within-minority differences on selected MMPI-2 and PAI scales when DIS symptoms were employed as covariates.

Liver Transplantation and Abuse of Drugs and Alcohol: A Correlation Between Scales of the MMPI-2

BACKGROUND

Clinical practice requires an accurate psychological assessment of subjects with clinical history of alcohol abuse and/or substance abuse (abuse history [AH]) for therapeutic choice. This study aims to identify significant correlations between the Minnesota Multiphasic Personality Inventory (MMPI)-2 scales in patients awaiting liver transplantation.

METHODS

We evaluated a personality questionnaire containing MMPI-2 scales in the sample of 308 patients (81.8% males and 18.2% females) awaiting liver transplantation. The AH group composed 44.49% of patients and in the abuse free (AF) group, 55.51%. Scales were compared using Shapiro-Wilk test and Mann-Whitney U test. Interrelationships were examined using Spearman's correlation.

RESULTS

This analysis found 27 scales of the MMPI-2 that were statistically different between 2 groups (AF and AH). In the AH group, we found a significant correlation between the following pairs of scales: Schizophrenia Scale (Sc) with the Addictions Potential Scale, Social Introversion scale (Si) with the Psychopathic Deviate scale (Pd), and Social Discomfort scale with Pd; the ES scale was negatively correlated with the Sc and Si scales. This interim study showed that the understanding of these indicators is crucial both for the assessment accuracy and for a prediction of the degree of therapy compliance after the transplantation.

CONCLUSIONS

The scales of the MMPI-2 indicated a marked tendency to emotional rigidity, a lack of self-esteem and susceptibility judgment. Social introversion and social discomfort trends lead to impulsive behavior and deviant actions that combine poorly with good compliance with treatment.

Comparison of personality characteristics in Parkinson disease patients with and without impulse control disorders and in healthy volunteers

OBJECTIVE

We aimed to assess personality characteristics in patients with Parkinson disease (PD) with and without impulse control disorders (ICD).

METHODS

We tested patients and controls with the Minnesota Multiphasic Personality Inventory-2 (MMPI-2) scales that have expected high sensitivity to apparent addictive behavior. We recorded mean disease duration and mean levodopa dose in the PD groups.

RESULTS

Of the 46 PD patients, 13 had ICD: hypersexuality, binge eating, or dopamine dysregulation. The PD patients with ICD had a longer duration of disease (11 vs. 5 y) and were taking higher doses of levodopa (900 vs. 500 mg/d). They scored above the pathologic threshold in 4 domains of the MMPI-2 Clinical Scales and in 8 Clinical Subscales and Content Scales. The most significant abnormality was Alienation-Self and Others.

CONCLUSIONS

ICDs in the general population have similarities to disorders of substance addiction. In PD patients, some personality profiles could play a role in development of ICDs or dopamine dysregulation syndrome. The MMPI-2 may be a useful test for PD patients in general, and for detecting ICD in particular.

Differences in the personality profile of medication-overuse headache sufferers and drug addict patients: a comparative study using MMPI-2

BACKGROUND

Medication-overuse headache (MOH) refers to headache attributed to excessive use of acute medications. The role of personality needs studies to explain the shifting from drug use to drug abuse. The main aim of this study is to study personality, according to Minnesota Multiphasic Personality Inventory, comparing MOH, episodic headache, substance addicts (SA) vs healthy controls.

METHODS

Eighty-two MOH patients (mean age 44.5; 20 M, 62 F) and 35 episodic headache (mean age 40.2; 8 M, 27 F), were compared to 37 SA (mean age 32.5; 29 M, 8 F) and 37 healthy controls (mean age: 32.49; 20 M, 17 F). International Classification of Headache Disorders 2nd Edition criteria were employed. Chi-square test, Kruskal-Wallis test, and post hoc comparisons were used for statistics.

RESULTS

MOH patients scored higher on Hypochondriasis, Depression (only females), Hysteria (only females) (P < .000). MOH did not show higher scores than episodic headache or healthy controls in dependency scales, while SA did.

CONCLUSION

The data obtained show that MOH and SA do not share common personality characteristics linked to dependence. Although further studies are needed to understand if such a difference is related to instrumental characteristics or to yet undiscovered psychobiological characteristics of MOH patients; however, we hypothesize that the detected difference may rely on the fact that drug dependence in the 2 groups is promoted by entirely different needs: pleasure seeking in the SA group, pain avoidance in the MOH group.

An investigation of Goodman's addictive disorder criteria in eating disorders

The aim of this study was to examine how far Goodman's addictive disorder criteria were met by individuals with eating disorders according to subtypes. The study provided a cross-sectional comparison among three samples of eating disorders [restricting anorexia nervosa (R-AN), N = 68; purging anorexia nervosa (P-AN), N = 42; and bulimia nervosa (BN), N = 66], a sample of substance-related disorders (SRDs, N = 48) and a sample of matched controls (N = 201). Diagnosis of addictive disorder was made following Goodman's criteria. Addictive personality traits were assessed with the Addiction Potential Scale of the Minnesota Multiphasic Personality Inventory--2 and with the Zuckerman's Sensation Seeking Scale. Results showed that individuals with BN met Goodman's addictive disorder criteria in the same proportion as drug-addicted individuals (65% vs 60%, p = NS). They both showed higher rates than R-AN individuals (35%; R-AN versus BN: F = 11.9, p < 0.001 and R-AN versus SRD: F = 7.16, p < 0.01). Although BN and SRD showed higher rates of addictive disorders compared with P-AN, differences were not significant. Scores on the Sensation Seeking and on the Addictive Potential scales paralleled the distribution of addictive disorders, with individuals with BN and with P-AN showing higher levels than individuals with R-AN. Results showed that a subgroup of individuals with an eating disorder experiences their disorder as an addiction and may deserve specific therapeutic attention.

2010 Bruno Klopfer Distinguished Contribution Award. Some considerations for enhancing psychological assessment

The field of psychological assessment appears to be bogged down in recent decades on refining existing measures and techniques to the detriment of real change. This article attempts to highlight some of the changes that need to take place in psychological assessment in the next few decades if real progress is to be achieved: (a) developing guidelines for psychological tests and measures; (b) establishing explicit criteria for training; (c) establishing guidelines for competencies; (d) recognizing the impact of marketing; (e) embracing electronic technology; and (f) implementing computer adaptive testing.

Screening, Brief Intervention, and Referral to Treatment (SBIRT): toward a public health approach to the management of substance abuse

Screening, Brief Intervention, and Referral to Treatment (SBIRT) is a comprehensive and integrated approach to the delivery of early intervention and treatment services through universal screening for persons with substance use disorders and those at risk. This paper describes research on the components of SBIRT conducted during the past 25 years, including the development of screening tests, clinical trials of brief interventions and implementation research. Beginning in the 1980s, concerted efforts were made in the US and at the World Health Organization to provide an evidence base for alcohol screening and brief intervention in primary health care settings. With the development of reliable and accurate screening tests for alcohol, more than a hundred clinical trials were conducted to evaluate the efficacy and cost effectiveness of alcohol screening and brief intervention in primary care, emergency departments and trauma centers. With the accumulation of positive evidence, implementation research on alcohol SBI was begun in the 1990s, followed by trials of similar methods for other substances (e.g., illicit drugs, tobacco, prescription drugs) and by national demonstration programs in the US and other countries. The results of these efforts demonstrate the cumulative benefit of translational research on health care delivery systems and substance abuse policy. That SBIRT yields short-term improvements in individuals' health is irrefutable; long-term effects on population health have not yet been demonstrated, but simulation models suggest that the benefits could be substantial.

Screening and interventions for alcohol and drug problems in medical settings: what works?

This article summarizes current knowledge about the accuracy of screening tests and the efficacy of interventions for substance use disorders in different medical settings (including trauma centers) where the practitioners are not specialists in the management of substance use disorders. In the first section, we introduce basic screening approaches for psychoactive substance use disorders and issues of natural history, risk factors, and populations at risk. Next, we review recent scientific research on the development of screening tests and the evaluation of early intervention services for persons at risk. We conclude that reliable and valid screening tests are available to detect alcohol use disorders but that further work is needed before routine screening for drug use disorders is warranted. We found strong evidence to support the effectiveness of brief interventions in managing at-risk drinkers; however, the evidence is only suggestive for drug use disorders. Finally, we explore the implications of the findings for developing a public health approach to early intervention, particularly as it relates to the unique needs of trauma centers.

Assessing contemporary substance abusers with the MMPI MAC Andrews Alcoholism Scale: a review

We reviewed 71 United States-based MacAndrew Alcoholism Scale (MAC), as revised (MAC-R) studies totaling almost 32,000 Ss, with adolescent and adult substance abusers, from studies published since the last MAC reviews (1989) through 2001. Results suggest that the MAC, and to some extent, the MAC-R, significantly correlates with measures of alcohol and substance abuse in both male and female adolescents and adults, across a diverse spectrum of the use-abuse continuum. Nonclinical groups (100%) scored below the clinical ranges on the MAC/MAC-R, while 79% of adolescent substance abusing groups scored > R 23, indicative of problems with substance abuse. Persons who abused alcohol, drugs, and polydrugs had mean MAC/MAC-R scores > 23, which ranged from 77% to 100% of the cases. The MAC/MAC-R does well in discriminating persons who abuse substances compared to nonclinical, nonabusing groups, but appears to lose diagnostic efficiency with psychiatric patients, and especially with medical patients with seizure disorders. Using R > 25 seems to improve diagnostic accuracy with this population. Meaning of false positives and false negatives were explored and discussed.

Alcohol and anxiety: subtle and obvious attributes of abuse in adults with social anxiety disorder and panic disorder

Previous research has found a relation between social anxiety disorder and alcoholism but recent work found no differences in drinking levels among socially anxious individuals, dysthymics, and normal controls. Using a more sophisticated measure of substance abuse may further explicate the relation between social anxiety and drinking. We examined aspects of substance abuse in treatment-seeking individuals with social anxiety disorder or panic disorder (psychiatric control group) as well as nondisordered individuals (normal control group). We used the Substance Abuse Subtle Screening Inventory-3 because it includes both face valid and subtle items to control for social desirability. Contrary to the hypotheses, there were few obvious or subtle aspects of substance abuse significantly greater for individuals with social anxiety disorder than those with panic disorder or normal controls. Implications for understanding the social anxiety-alcohol relationship, assessment of substance abuse in socially anxious populations, and the construct of social anxiety-are discussed.

Diagnostic accuracy and factor structure of the AAS and APS scales of the MMPI-2

Receiver operating characteristics analysis and sensitivity analysis were used to compare the diagnostic accuracy of the Addiction Acknowledgement Scale (AAS; Weed, Butcher, McKenna, & Ben-Porath, 1992) and Addiction Potential Scale (APS; Weed et al., 1992) of the Minnesota Multiphasic Personality Inventory-2 (Butcher, Dahlstrom, Graham, Tellegen, & Kaemmer, 1989) to that of the MacAndrew Alcoholism Scale-Revised (Butcher et al, 1989), the CAGE (Mayfield, McLeod, & Hall, 1974), and Svanum's scale (Svanum & McGrew, 1995) in a sample of 338 university students. The AAS was the most accurate of these 5 scales at identifying current alcohol dependence (as measured by a structured diagnostic interview) and appears to offer considerable promise as an alcohol screening instrument. In contrast, the APS performed the most poorly of the 5 scales being evaluated, yielding results that would be of minimal clinical utility. Factor analysis yielded a 2-factor solution for the AAS (Acknowledgement of Alcohol/Drug Problems; Positive Alcohol Expectancies) and a 4-factor solution for the APS (Satisfaction with Self; Cynicism/Pessimism; Impulsivity; Risk-Taking).

Korsakoff's syndrome on the MMPI-2

BACKGROUND

This article describes an exploratory study of MMPI-2 administration to 20 patients with Korsakoff's disease, all admitted to the Korsakoff unit of a major psychiatric hospital in the Netherlands.

METHODS

We compared their MMPI-2 basic profiles with those of an inpatient alcohol-dependent non-Korsakoff group. Attention was also given to content scales and some selected supplementary scales. Furthermore, we explored the differences between MMPI-2 profiles of Korsakoff patients with and without insight into oneself and one's disease.

RESULTS

Compared with both the Dutch norms and the scale means of the non-Korsakoff alcoholics, Korsakoff patients showed an extraordinary flat profile. Illness insight appeared to be related to the levels of the various scales.

DISCUSSION

The results show the potential usefulness of the MMPI-2 in the differential diagnosis of chronic alcoholism and Korsakoff's disease.

Adherence to substance abuse treatment: clinical utility of two MMPI-2 scales

In this study, we examined the ability of the MMPI-2 (Butcher, Dahlstrom, Graham, Tellegen, & Kaemmer, 1989) Addiction Acknowledgment scale (AAS; Weed, Butcher, McKenna, & Ben-Porath, 1992) and Negative Treatment Indicators scale (TRT; Butcher, Graham, Williams, & Ben-Porath, 1990) to predict adherence to and outcomes from substance abuse treatment. There was no evidence that the AAS was related to treatment adherence or outcome in our sample. However, results did reveal a significant positive relation between scores on the TRT scale and readmission to the hospital. Further analyses identified an optimal score for use in similar clinical populations and settings, and characteristics of high and low scorers. Compared to low scorers, high TRT scorers were more likely to not return for treatment after an initial screening interview. If they did return for treatment, high TRT scorers were more likely to experience fewer treatment days and to be rated as having lower motivation, poorer participation, and poorer comprehension of program materials. These findings provide promising initial evidence of the utility of the TRT scale for identifying patients who may be at a high risk for unsuccessful substance abuse treatment.

Use of the MMPI-A to detect substance abuse in a juvenile correctional setting

In this study, we examined the ability of the MMPI-A (Butcher et al., 1992) to detect substance abuse problems in a juvenile correctional setting. Specifically, we evaluated the Alcohol/Drug Problem Acknowledgment scale (ACK; Weed, Butcher, & Williams, 1994), the Alcohol/Drug Problem Proneness scale (PRO; Weed et al., 1994), and the MacAndrew Alcoholism Scale-Revised (MAC-R; Butcher et al., 1992) in the prediction of substance abuse. In addition, the incremental validity of ACK in comparison to PRO was evaluated, as was the incremental validity of PRO in comparison to ACK. The sample consisted of 123 boys and girls from ajuvenile correctional facility in Northeastern Ohio. Results indicate that ACK and PRO, but not MAC-R, were related to interviewer ratings of substance abuse. Results point to the superiority of ACK over PRO in substance abuse identification.

Efficacy of the Substance Abuse Subtle Screening Inventory-3 (SASSI-3) in identifying substance dependence disorders in clinical settings

This article presents analyses on psychometric properties of a recent revision of the Substance Abuse Subtle Screening Inventory (SASSI-3). Participants were 772 individuals from a range of clinical settings who were independently diagnosed according to the Diagnostic and Statistical Manual of Mental Disorders (3rd ed., rev. [DSM-III-R]; American Psychiatric Association [APA], 1987) or DSM-IV (APA, 1994) criteria regarding the presence or absence of a substance-dependence disorder. A development sample (n = 391) was used to formulate SASSI-3 decision rules that yielded 95% agreement with clinical diagnoses of substance dependence. Correspondence with the criterion variable was shown to be 97% in a cross-validation sample (n = 381) from the same clinical settings. Convergent relations were demonstrated with a variety of other indexes of substance misuse, including clinical assessments of substance abuse history, alcohol and drug-related arrests, self-reported use of illicit drugs, and other instruments designed to screen for substance misuse. Further, logistic regression analyses indicated no significant differences in the overall accuracy of the SASSI-3 as a function of respondents' demographic characteristics or clinical ratings of respondents' level of adjustment and functioning (Global Assessment of Functioning ratings; APA, 1987, 1994).

The Common Alcohol Logistic-Revised scale (CAL-R): a revised alcoholism scale for the MMPI and MMPI-2

The Common Alcohol Logistic-Revised (CAL-R) scale was developed for use with MMPI or MMPI-2 with medical patients. It was derived from the Common Alcohol Logistic (CAL) scale and developed by (1) dropping the six CAL items deleted from the MMPI during construction of the MMPI-2 and recomputing item weights (using logistic regression); (2) calculating norms; and (3) repeating the validation procedure used in developing CAL. We used the same criterion group (1,221 alcoholics) and same contrast sample (7,621 nonalcoholic medical patients) used for CAL. Comparison of receiver operating characteristic curves, positive predictive value, and negative predictive value for CAL and CAL-R indicates that the latter has the same favorable ability to screen medical patients for alcoholism as the former.