A comparison of African American and White American veteran MMPI-2 profiles.

The present study compared a matched sample of 180 African American and 180 White American veterans who completed the MMPI-2 as a part of their evaluation while receiving inpatient psychiatric treatment. Findings indicated no significant multivariate or univariate effects associated with race on the basic validity and clinical scales, a significant multivariate effect but no significant univariate effects associated with race across the supplementary scales. Overall the two groups had very similar mean profiles across the basic validity, clinical, and supplementary scales. Comparison of the two groups on the content scales yielded a significant multivariate and significant univariate effects with African Americans scoring higher on the FRS, BIZ, CYN, and ASP scales. Because the two groups differed in terms of frequency and type of drug abuse, follow-up 2 x 2 univariate analyses of variance were conducted for the FRS, BIZ, CYN, and ASP content scales comparing participants classified in terms of presence or absence of a primary or secondary drug abuse diagnosis by race. A significant main effect associated with drug abuse was obtained for ASP. Results are discussed and considered in light of earlier research on the MMPI-2 and race. Content scale differences are also discussed in terms of possible differences in worldview.

Comparison of two methods of MMPI-2 profile classification.

The present study investigated the extent of agreement of the highest scale method and the best-fit method in matching MMPI-2 profiles to database code-type profiles and considered profile characteristics that may relate to agreement or disagreement of code-type matches by these two methods. A sample of 519 MMPI-2 profiles that had been classified into database profile code types by these two methods was studied. Resulting code-type matches were classified into three groups: identical (30%), similar (39%), and different (31%), and the profile characteristics of profile elevation, dispersion, and profile code-type definition were studied. Profile code-type definition was significantly different across the three groups with identical and similar match profile groups showing greater profile code-type definition and the different group consisting of profiles that were less well-defined.

An ethnocultural study of posttraumatic stress disorder in African-American and white American Vietnam War veterans.

This study examined the intensity of Posttraumatic Stress Disorder (PTSD) symptoms as measured by the Mississippi Scale, the Keane (PK), and the PTSD (PS) Scales of the MMPI-2 in a sample of 34 African-American and 34 White American Vietnam War Veterans who sought treatment in a Specialized Inpatient PTSD Unit. The scores of the two groups on the Beck Depression Inventory and the clinical scales of the MMPI-2 were also compared. The ethnoculturally different sample was matched on intensity of combat exposure, marital status, employment status, age, and education. No significant differences on the measures of PTSD symptoms were noted and no significant differences were found on the Beck scale or the MMPI-2 clinical scales.

The MCMI-II and race.

In this study, we investigated MCMI-II profile differences in a sample of 65 Black and 164 White psychiatric inpatients. A multivariate analysis of variance (MANOVA) yielded a significant multivariate effect associated with race, with Black patients scoring significantly higher on the Histrionic, Narcissistic, Paranoid, Drug Dependent and Delusional Disorder scales. A second MANOVA was conducted on these 5 scales with a smaller sample of 46 Black and 46 White patients, who were matched for primary Axis I discharge diagnosis and matched for substance abuse comorbidity. This MANOVA did not yield a significant multivariate effect associated with race, and scale differences were attenuated.

Posttraumatic stress disorder and the MCMI-II.

This study investigated the MCMI-II profile characteristics of 39 veterans diagnosed with Posttraumatic Stress Disorder. Characteristics of the mean group profile were similar to prior findings reported in the literature on the MCMI and Posttraumatic Stress Disorder with highest mean elevations found on the Avoidant, Passive-Aggressive, Schizoid, and Antisocial basic personality scales, the Borderline and Schizotypal pathological personality scales, and with elevations on the Anxiety, Dysthymia, Alcohol Dependence, Drug Dependence, and Major Depression clinical syndrome scales. A multivariate analysis of variance comparing the group with Posttraumatic Stress Disorder with a non-PTSD comparison group of 39 on the basic personality, pathological personality, and the clinical syndrome scales of the MCMI-II was not statistically significant. Nonetheless, univariate analyses of variance comparing the two groups on the individual modifier scales and the individual personality and clinical syndrome scales of the MCMI-II using a Bonferroni adjusted probability indicated significant differences on the Desirability and Histrionic scales. Response-style bias as a possible factor in MCMI-II profiles for the group with Posttraumatic Stress Disorder is also discussed.

Post-traumatic stress disorder and the MMPI-2.

This study compared the MMPI-2 profiles of 27 veterans diagnosed with post-traumatic stress disorder with a non-PTSD comparison group of 27 veteran patients receiving inpatient treatment for other mental disorders. Three multivariate analyses of variance were conducted comparing the two groups on the 10 traditional clinical scales, the 12 supplemental scales and the 15 new content scales on the MMPI-2. The PTSD group obtained a mean profile with peak elevations on the F validity scale and on clinical Scales 2 (D) and 8 (Sc). The multivariate analysis of variance comparing the PTSD and non-PTSD groups across the 10 traditional clinical scales was not significant. The multivariate analyses of variance comparing the two groups on the 12 supplemental scales and the 15 content scales were significant. Significant univariate supplemental scale differences were found on the Keane PTSD scale (PK) and the Post-Traumatic Stress Disorder (PS) scale with the PTSD group scoring higher on PK and PS. Significant univariate content scale differences were found for the Anger (ANG) scale with the PTSD group scoring higher. A cut-off score of 28 on the PK scale correctly classified 76% of the overall sample, 67% of the PTSD group and 85% of the non-PTSD-comparison group.

MMPI-2 negative treatment indicators scale and irregular discharge.

A sample of 18 psychiatric inpatients who had completed the MMPI-2 and subsequently received an irregular discharge from inpatient treatment were compared on the MMPI-2 Negative Treatment Indicators scale (TRT) with a random sample of 18 inpatients who received a regular discharge from inpatient care. Analysis showed no significant difference between the two groups on the Negative Treatment Indicators scale. The two groups did differ in K scale elevation. The possible need to interpret the Negative Treatment Indicators scale in the context of K scale elevation is discussed.

Inpatient PTSD treatment: a study of pretreatment measures, treatment dropout, and therapist ratings of response to treatment.

Pretreatment measures including demographic variables, adjustment index variables and psychological testing variables were studied in relationship to treatment dropout and therapist ratings of overall response to treatment among PTSD veterans in an inpatient PTSD program. Analysis comparing a group of fourteen veterans who dropped out of treatment early and a random sample of fourteen who successfully completed treatment showed no significant differences. Analysis comparing a group of 35 veterans who received the highest therapist ratings on response to treatment with a group of 35 veterans receiving the lowest ratings on response to treatment also showed no significant differences. Analysis of subgroups of patients who had completed the Millon Clinical Multiaxial Inventory (MCMI) and received high versus low therapist ratings showed one significant difference on the hypomania scale. Overall findings on the MMPI and MCMI appeared similar to other investigations of PTSD.

F Scale elevation and PTSD MMPI profiles.

MMPI profiles for 87 PTSD veteran inpatients were classified and studied according to MMPI F Scale elevation. Mean MMPI profiles and frequency of high two-point code types were studied for different levels of F Scale elevation. Similar mean profile configurations were found for subgroups with F > or = 70 with Scales 2 and 8 appearing as the two highest clinical scales. For F < 70 the configuration was different in that Scale 8 was not one of the two highest scales. The 2-8/8-2 high two-point code was the modal high two-point code for the total sample but the relative frequency of high two-point codes did vary somewhat within and across levels of F Scale elevation.

Predicting Wechsler Adult Intelligence Scale-Revised (WAIS-R) IQ scores from the Luria-Nebraska Neuropsychological Battery (Form I).

The study examined a modification of two sets of previously published LNNB-derived WAIS IQ estimates in order to estimate WAIS-R IQs in a sample of neuropsychiatric patients. Findings suggested that the modification will allow practicing clinicians to estimate WAIS-R IQs with a level of accuracy similar to that found in previous research. The level of predictive accuracy for WAIS-R FSIQ and VIQ was somewhat higher than would be expected on the basis of previous research, which suggests that the modified McKay, Golden, Moses, Fishburne, and Wisniewski (1981) formulae for estimating IQ may be more accurate than previously thought. The correlation matrix of WAIS-R subtests and LNNB Form I clinical and summary scales also is described.

Confidence intervals for the MMPI-2.

The confidence intervals for the Minnesota Multiphasic Personality Inventory (MMPI-2) clinical scales were investigated. Based on the clinical scale reliabilities published in the MMPI-2 manual, estimated true scores, standard errors of measurement for estimated true scores, and 95% confidence intervals centered around estimated true scores were calculated at 5-point MMPI-2 T-score intervals. The relationships between obtained T-scores, estimated true T-scores, scale reliabilities, and confidence intervals are discussed. The possible role of error measurement in defining scale high point and code types is noted.

A comparison of MMPI-2 and MMPI T-scores for men and women.

The original MMPI linear T-score norms are compared with the new linear and uniform T-score norms for the MMPI-2 across basic validity and clinical scales for both men and women. The comparison indicated more pronounced normative changes for certain scales and somewhat more pronounced changes for men than for women. Relationships between MMPI-2 and MMPI T-scores across scales are complex; a few MMPI-2 scales show enhancement, others show attenuation across the full range of the scale, and other scales show attenuation in the middle and lower ranges of the scale with enhancement in the upper ranges. Possible implications and effects of the new MMPI-2 T-score norms on original MMPI scale high-point elevation and high-point codes are discussed.

A comparison of MMPI profile types with corresponding estimated MMPI-2 profiles.

Changes in mean elevation, dispersion, overall configuration, and code type of well-defined MMPI profiles were examined after transformation to estimated MMPI-2 profiles. A total of 34 MMPI profiles from both Gilberstadt and Duker (1965) and Marks, Seeman, and Haller (1974) were analyzed using contemporary MMPI-2 T-scores. Results yielded a reduction in mean elevation and both increases and decreases in scatter about the mean upon transformation to MMPI-2 norms. The total configuration of linear MMPI and estimated MMPI-2 T-score profiles correlated highly with each other and manifested a similar pattern of correlation with the total configuration of Skinner and Jackson's (1978) three modal MMPI types.

A comparison of MMPI profile types across standard and contemporary norms.

Changes in mean elevation, dispersion, overall configuration, and code type of well-defined MMPI profiles derived from diagnostic groups with known clinical correlates were examined after transformation using contemporary norms developed by Colligan, Osborne, Swenson, and Offord (1983). A total of 34 MMPI profiles from both Gilberstadt and Duker (1965) and Marks, Seeman, and Haller (1974) were analyzed using contemporary normalized T scores. Results yielded the expected reduction in mean elevation and dispersion about the mean upon transformation to contemporary norms. Linear and contemporary normalized T-score profiles correlated highly with each other and manifested a similar pattern of correlation with Skinner and Jackson's (1978) three modal MMPI types. Similarity in the code types of 94% of the profiles was found. However, the substantial number of order changes in the scales that comprise the code type evident upon transformation highlights the caution and care necessary when one is interpreting MMPI profiles based on contemporary norms. Continued use of original norms together with contemporary norms appears indicated until the full implications of profile and code type differences upon transformation to contemporary norms are understood.

Psychological test findings on pathological gamblers in treatment.

A sample of 23 veterans entering the Pathological Gambling Treatment Program at the Veterans Administration Medical Center in Miami, Florida, were administered a battery of tests including the following: Wechsler Adult Intelligence Scale (WAIS), Minnesota Multiphasic Personality Inventory (MMPI), Edwards Personal Preference Schedule (EPPS), and the Personal Orientation Inventory (POI). Findings on the intelligence testing indicated that on the average the pathological gamblers were of bright normal intelligence with a mean WAIS Full Scale IQ of 116.78, mean Verbal IQ of 119.30, and a mean Performance IQ of 111.30. Group profile results on the MMPI showed significant mean clinical scale elevations on two of the nine clinical scales: Scale 2, Depression; and Scale 4, Psychopathic Deviate. Group profile results on the EPPS indicated relatively high scores on the variables of achievement, exhibition, autonomy, dominance, and heterosexuality; and relatively low scores on the variables of deference, order, and endurance. Group profile results on the POI were essentially within normal limits with a tendency toward time incompetency noted, suggesting difficulty living fully in the present here and now.

Demographic and clinical characteristics as predictors of readmission: a one-year follow-up.

Follows up an earlier investigation in which demographic and clinical characteristics of psychiatric patients were used to predict readmission within 3 months of discharge. In the initial study, stepwise multiple regression analysis identified six variables as the optimal set of predictors for readmission within 3 months of discharge: type of discharge, number of prior psychiatric hospitalizations, race, suicide attempt within 1 month of admission, subjective report of depression upon admission, and occupational level (R = .452). In the present study the same sample was followed up at 1 year after discharge, and demographic and clinical variables were used to predict readmission within 1 year of discharge. Stepwise multiple regression analysis identified three variables as the optimal set of predictors for readmission within 1 year of discharge: past history of suicidal behavior, subjective report of depression upon admission, and number of prior psychiatric hospitalizations. Changes in predictors as a function of length of follow-up period are considered, and implications of the findings for identifying high-risk readmission candidates are discussed.

Demographic and clinical characteristics as predictors of length of hospitalization and readmission.

The present study investigated demographic and clinical characteristics of psychiatric patients in relation to the two criterion variables of length of hospitalization and readmission within 3 months of discharge. Stepwise multiple regression analysis identified five variables as the optimal set of predictors for length of hospitalization: age, history of commitment, number of prior psychiatric hospitalizations, recent employment history, and past history of suicidal behavior (R = .451). Regression analysis also identified six variables as the optimal set of predictors for readmission within 3 months of discharge: type of discharge, number of prior psychiatric hospitalizations, race, suicide attempt within 1 month prior to admission, subjective report of depression upon admission, and occupational level (R = .452). Implications of the findings for identifying short-term treatment candidates and factors related to readmission are discussed.