Replicated item level factor analysis of the full MMPI

Indigenous Personality Measures

This article reviews the Chinese historical context of personality testing. Issues of interpretation and methodology related to initial adaptations of English-language personality tests are discussed. The deficiencies of the imposed-etic approach are addressed by the recent development of indigenous personality measures. Three large-scale indigenous personality inventories developed for the Chinese people are introduced and evaluated: Ko's Mental Health Questionnaire, Multi-Trait Personality Inventory, and Chinese Personality Assessment Inventory (CPAI). Research with indigenous Chinese instruments has provided relevant personality measures for use in local cultures. In addition, it has offered a means to examine the broader theoretical question of the universality and relevance of current Western personality theories, as illustrated by research with the CPAI.

MMPI-A Scale-Level Factor Structure: Replication in a Clinical Sample

The adolescent form of the Minnesota Multiphasic Personality Inventory (MMPI-A) Structural Summary was developed from the results of a scale-level factor analysis conducted by Archer, Belevich, and Elkins based on the MMPI-A normative sample. The present study examined the scale-level factor structure of the MMPI-A in a clinical sample of 358 adolescents receiving outpatient or inpatient psychiatric services. A Principal Factor Analysis was performed using the raw score intercorrelation matrix from the 69 scales and subscales of the MMPI-A. The procedure yielded nine factors that accounted for 75.6% of the total variance in scale and subscale raw scores. Results from this clinical sample indicated that seven of the eight dimensions that appear on the MMPI-A Structural Summary were replicated in terms of producing highly similar factor structure correlation coefficients with those reported by Archer, Belevich, et al. for a normal sample. The present findings support the use of the MMPI-A Structural Summary for the assessment of adolescents in clinical settings.

The Recaptured Scale Technique: A Method for Testing the Structural Robustness of Personality Scales

Tellegen and Waller advocated a complex and time-consuming scale construction method that they called "exploratory test construction." Scales that are constructed by this method-such as the Multidimensional Personality Questionnaire (MPQ)-are presumed to be more "psychologically coherent" and "robust" than scales constructed by other means. Using a novel procedure that we call the "recaptured scale technique," we tested this conjecture by conducting a megafactor analysis on data from the 411 adult participants of the Minnesota Study of Twins Reared Apart who completed the MPQ, the MMPI, and the CPI. We extracted and obliquely rotated 21 factors from a matrix of gender-corrected tetrachoric correlations for the 1,102 nonredundant items of the three omnibus inventories. Robustness of the 11 MPQ scales was assessed by the degree to which these factors recaptured the MPQ item groupings. Our results showed that nine factors were clearly recognizable as MPQ scales and two additional factors represented a bifurcation of an MPQ scale. A higher-order factor analysis of all 21 factor scales yielded five factors that clearly resembled the Big Five. Our results provide strong support for (a) the method of exploratory test construction, (b) the structural robustness of most MPQ scales, and

A questionnaire-wide association study of personality and mortality: the Vietnam Experience Study

OBJECTIVE

We examined the association between the Minnesota Multiphasic Personality Inventory (MMPI) and all-cause mortality in 4462 middle-aged Vietnam-era veterans.

METHODS

We split the study population into half-samples. In each half, we used proportional hazards (Cox) regression to test the 550 MMPI items' associations with mortality over 15years. In all participants, we subjected significant (p<.01) items in both halves to principal-components analysis (PCA). We used Cox regression to test whether these components predicted mortality when controlling for other predictors (demographics, cognitive ability, health behaviors, and mental/physical health).

RESULTS

Eighty-nine items were associated with mortality in both half-samples. PCA revealed Neuroticism/Negative Affectivity, Somatic Complaints, Psychotic/Paranoia, and Antisocial components, and a higher-order component, Personal Disturbance. Individually, Neuroticism/Negative Affectivity (HR=1.55; 95% CI=1.39, 1.72), Somatic Complaints (HR=1.66; 95% CI=1.52, 1.80), Psychotic/Paranoid (HR=1.44; 95% CI=1.32, 1.57), Antisocial (HR=1.79; 95% CI=1.59, 2.01), and Personal Disturbance (HR=1.74; 95% CI=1.58, 1.91) were associated with risk. Including covariates attenuated these associations (28.4 to 54.5%), though they were still significant. After entering Personal Disturbance into models with each component, Neuroticism/Negative Affectivity and Somatic Complaints were significant, although Neuroticism/Negative Affectivity's were now protective (HR=0.73; 95% CI=0.58, 0.92). When the four components were entered together with or without covariates, Somatic Complaints and Antisocial were significant risk factors.

CONCLUSIONS

Somatic Complaints and Personal Disturbance are associated with increased mortality risk. Other components' effects varied as a function of variables in the model.

Construct, concurrent and discriminant validity of Type D personality in the general population: associations with anxiety, depression, stress and cardiac output

The Type D personality, identified by high negative affectivity paired with high social inhibition, has been associated with a number of health-related outcomes in (mainly) cardiac populations. However, despite its prevalence in the health-related literature, how this personality construct fits within existing personality theory has not been directly tested. Using a sample of 134 healthy university students, this study examined the Type D personality in terms of two well-established personality traits; introversion and neuroticism. Construct, concurrent and discriminant validity of this personality type was established through examination of the associations between the Type D personality and psychometrically assessed anxiety, depression and stress, as well as measurement of resting cardiovascular function. Results showed that while the Type D personality was easily represented using alternative measures of both introversion and neuroticism, associations with anxiety, depression and stress were mainly accounted for by neuroticism. Conversely, however, associations with resting cardiac output were attributable to the negative affectivity-social inhibition synergy, explicit within the Type D construct. Consequently, both the construct and concurrent validity of this personality type were confirmed, with discriminant validity evident on examination of physiological indices of well-being.

Neuroticism, cognitive ability, and the metabolic syndrome: The Vietnam Experience Study

OBJECTIVES

The purpose of this study is to explore the association of neuroticism with the metabolic syndrome, separate components of the metabolic syndrome, and the number of components of metabolic syndrome an individual possesses. The purpose of this study is to examine also the extent to which any associations are accounted for by sociodemographic factors, health behaviors, and cognitive ability.

METHOD

Participants were 4208 men drawn from the Vietnam Experience Study. From military archives, and a later telephone interview and psychological and medical examination, sociodemographic, health behavior, cognitive ability, neuroticism, and health data were collected. Neuroticism and cognitive ability were assessed with standardized tests during the medical examination. Presence of the metabolic syndrome was based on body mass index, fasting blood glucose or a diagnosis of diabetes, high blood pressure or taking antihypertensive medication, high-density lipoprotein cholesterol, and triglyceride levels.

RESULTS

Neuroticism was positively associated with the occurrence of the metabolic syndrome and several of its components in both age-, and sociodemographic- and health behavior-adjusted analyses. Many associations were accounted for by individual difference in cognitive ability. Neuroticism was robustly associated with the number of components of the metabolic syndrome after adjustment.

CONCLUSIONS

Individuals with higher neuroticism scores had a higher prevalence of the metabolic syndrome and a larger number of its components. On the whole, differences in cognitive ability appeared to partially mediate the relationship between neuroticism and the metabolic syndrome.

Personality Assessment from the Nineteenth to the Early Twenty-First Century: Past Achievements and Contemporary Challenges

The historical basis of personality assessment that led to the development of today's approaches and applications is described. The modern era of personality assessment began in late nineteenth-century Europe. Early twentieth-century highlights included the development of projective techniques like the Rorschach and several early self-report inventories, culminating in the development of the most widely used measure, the Minnesota Multiphasic Personality Inventory (MMPI). The most recent 30-year period showed expansions into personnel screening; clinical assessment, including wide use in forensic settings; and therapeutic assessment. However, contemporary controversies are apparent with two of the most widely used measures, the Rorschach and the MMPI instruments. These controversies are described, including concerns about the Exner Comprehensive System for the Rorschach and the last five years of changes to the MMPI-2, including the introduction of the Restructured Clinical (RC) Scales, the adoption of the Fake Bad Scale (FBS) into the instrument, and the release of the MMPI-2 Restructured Form (MMPI-2-RF). Current challenges facing psychologists in personality assessment are highlighted.

Emotionally stable, intelligent men live longer: the Vietnam Experience Study cohort

OBJECTIVES

To determine whether neuroticism, cognitive ability, and their interaction predicted mortality and to test whether neuroticism or cognitive ability effects were mediated by socioeconomic status (SES), physical health, mental health, or health behaviors.

METHODS

Participants were 4200 men followed up for > 15 years. Participants took part in telephone interviews and medical and psychological evaluations. The neuroticism measure was based on the Minnesota Multiphasic Personality Inventory, which was administered during the psychological examination. Cognitive ability was measured via the Army General Technical Test given at induction and Wechsler Adult Intelligence Scale subtests administered during the psychological examination, approximately 17 years later. We used covariance structure modeling to analyze the data because it enabled us to conduct Cox proportional hazards analyses with latent variables and mediator variables.

RESULTS

Even after adjusting for age, ethnicity, and marital status, high neuroticism and low cognitive ability were independent mortality risk factors. A significant interaction indicated that participants high in neuroticism and low in cognitive ability were particularly at risk. In a second series of models, we examined whether education, income, seven physical health measures, two mental health measures, drinking, and smoking were related to mortality. SES and physical health variables attenuated the effect of cognitive ability but not that of neuroticism. A third series of models revealed that cognitive ability was related to mortality via its direct effects on income and health.

CONCLUSIONS

The effects of high neuroticism, low cognitive ability, and their interaction predict mortality. Cognitive ability effects are mediated by health, income, and education.

On the integration of personality assessment methods: the Rorschach and MMPI

Despite being the most studied and used personality assessment tools, data from the Rorschach and MMPI generally disagree (Archer & Krishnamurthy, 1993a, 1993b). Independence is proposed to result from at least 3 factors: (a) the methods tap unique levels of personality, (b) personality has a complex organization, and (c) response styles generate considerable method variance that must be considered in nomothetic research. These ideas led to 5 hypotheses, each of which received support. Rorschach and MMPI response styles are uncorrelated, although response styles are quite consistent within a method family. MMPI-2 and Rorschach constructs of dysphoria, psychosis, or wariness are uncorrelated when response styles are ignored. However, robust convergent validity is evident when patients have similar response styles on each method (e.g., for dysphoria, M r = .59) and dysphoria is expressed in opposing ways on each method when response styles are discordant (i.e., M r = -.54). Data from the latter analyses were correlated with genuine clinical phenomena and implications were discussed for clinical practice and research.

Domains and facets: hierarchical personality assessment using the revised NEO personality inventory

Personality traits are organized hierarchically, with narrow, specific traits combining to define broad, global factors. The Revised NEO Personality Inventory (NEO-PI-R; Costa & McCrae, 1992c) assesses personality at both levels, with six specific facet scales in each of five broad domains. This article describes conceptual issues in specifying facets of a domain and reports evidence on the validity of NEO-PI-R facet scales. Facet analysis-the interpretation of a scale in terms of the specific facets with which it correlates-is illustrated using alternative measures of the five-factor model and occupational scales. Finally, the hierarchical interpretation of personality profiles is discussed. Interpretation on the domain level yields a rapid understanding of the individual interpretation of specific facet scales gives a more detailed assessment.

Giving the Self a Voice in MMPI Self-report: Jerry Wiggins and the Content Scales

This article places one of Jerry Wiggins' contributions to the Minnesota Multiphasic Personality Inventory (MMPI), his content scales, in the twin contexts of past and recent research, and of Wiggins' own preoccupations growing out of the interpersonal point of view. It highlights Wiggins' perspective on the position of the person in the process of personality assessment, with special reference to the means by which the person's communication with the assessor may become accessible as a voice to be heard in the total pattern of assessment findings. Finally, one of the newer trends in personality assessment, one highly congenial to Wiggins' interpersonal outlook, is discussed with reference to the role that the MMPI-2 Content Scales may play in its future development.

Police corruption and psychological testing: a strategy for preemployment screening

The prediction, control, and prevention of police corruption represent pervasive and enduring problems. Researchers have suggested that intervention at the preemployment screening stage may be the best solution. However, investigators have acknowledged that existing assessment practices are flawed. This article proposes a strategy for the preemployment screening of law enforcement personnel. In particular, it examines the utility of the Inwald Personality Inventory and the Revised-NEO Personality Inventory in relation to assessing antisocial behavioral tendencies and conscientious personality traits, respectively, and argues that their combined use, appropriately administered in a testing situation, represents a reliable and valid predictor of good job performance. The article speculatively comments on this strategy for purposes of psychological testing, future research in the field, and law enforcement administration practices.

A quantitative review of the comprehensiveness of the five-factor model in relation to popular personality inventories

Reports of associations between the five-factor model (FFM) and the scales of popular personality inventories have generated controversy regarding the comprehensiveness of the FFM. The controversy is fueled by a preoccupation with capturing scale variance and differentiating between specific scales, whereas the focus should instead be on whether the FFM captures the common variance and the dimensions that exist in personality constructs. Analyses of published data revealed that the portions of scale variance captured by the FFM (mean = 38%) were substantial when evaluated in relation to the portions of common variance that exist in most personality inventories (mean = 50%). Furthermore, interbattery factor analyses indicated that the factor structures in most personality inventories can be closely replicated using data derived solely from scale associations with the FFM. Exceptions to this finding occurred for only 2 of 28 personality inventories. The findings support the comprehensiveness of the FFM.

Interpretation of a full-information item-level factor analysis of the MMPI-2: normative sampling and nonpathognomonic descriptors

An exploratory item-level full-information factor analysis was performed on the normative sample for the MMPI-2 (Butcher, Dahlstrom, Graham, Tellegen, & Kaemmer, 1989). This method of factor analysis, developed by Schilling and Bock (Bock & Schilling, 1997) and based on item response theory, works directly with the response patterns and avoids the artifacts associated with phi coefficients and tetrachoric coefficients. Promax rotation of the factor solution organizes the clinical scale items into 10 factors that we labeled Distrust, Self-Doubt, Fitness, Serenity, Rebelliousness, Instrumentality, Irritability, Artistry, Sociability, and Self-Reliance. A comparison was made to the results of Johnson, Butcher, Null, and Johnson (1984), who performed a principal-component analysis on an item set of 550 items from the previous version of the MMPI (Hathaway & McKinley, 1943). Along with version changes and sampling differences, the essential differences between Johnson et al.'s results and ours may be attributed to differences between the Schilling and Bock method, which uses all information in the item responses, and the principal-component analysis, which uses the partial information contained in pairwise correlation coefficients. This study included 518 of the complete 567 items of the MMPI-2, versus Johnson et al.'s retention of 309 of the initially included 550 items of the previous MMPI. The full-information analysis retained all 518 initially included items and more evenly distributed the items over the 10 resulting factors, all sharply defined by their highest loading items and easy to interpret. Sampling effects and factor label considerations are discussed, along with recommendations for research that would validate the clinical utility of the implied scales for describing normal personality profiles. The full-information procedure provides for Bayes estimation of scores on these scales.

Conceptualizing and measuring global interpersonal mistrust-trust

Global interpersonal mistrust is conceptualized as a general mistrust of the motives of others in situations related to one's well-being: a general tendency to view others as mean, selfish, malevolent, or unreliable people who are, thus, not to be depended on to treat one well. The authors developed an 18-item unidimensional self-report inventory measuring interpersonal mistrust as a negative cognitive orientation toward others. The measure comprises items describing perceptions of specific hypothetical interpersonal situations rather than items asking respondents to describe their own general behavior. The measure was reliable and evidenced construct validity in a heterogeneous sample of Australians.

Scale 5 of the MMPI and MMPI-A: evidence of disparity

Scale 5 of the MMPI and MMPI-A was compared in a repeated measures design. Participants for the study were 43 adolescents classified as emotionally disturbed in a public school system and 17 inpatients at a residential treatment center. The MMPI Scale 5 mean score was substantially higher than that of the MMPI-A. The alternate-form reliability between Scale 5 of the two forms was surprisingly low, suggesting that the deletion of 16 items and rewording of 6 additional items changed the scale on the MMPI-A to an extent that may have significantly altered the underlying construct. The authors discuss factors that could be associated with the findings, including: (a) the diminished ability to express feminine interest on the MMPI-A, and (b) general changes in attitudes among adolescents over the 3 or more decades since the MMPI norms were developed.

Personality and coronary heart disease: the type-D scale-16 (DS16)

Clinical diagnoses of depression, self-reported negative emotions, and personality traits have been associated with both the development and progression of coronary heart disease (CHD). This article focusses on negative affectivity (the tendency to experience negative emotions) and social inhibition (the tendency to inhibit self-expression in social interaction) in CHD patients. Two independent samples of patients with CHD were included in this study. Both empirical and internal-structural criteria were used to devise a brief self-report measure comprising an eight-item negative affectivity and an eight-item social inhibition scale in Sample 1 (N = 400). These scales were internally consistent (alpha = .89 and .82), stable over time (three-month test-retest reliability = .78 and .87) and were validated against standard personality scales. CHD patients with a "distressed" personality (Type-D) report high levels of negative affectivity and social inhibition; accordingly, this self-report measure was termed Type-D Scale-16 (DS16). Previous research showed that Type-D was associated with cardiac events and incidence of cancer in patients with CHD. Type-D as measured by the DS16 was associated with depressive affect and symptoms, stress, poor self-esteem, dissatisfaction with life, and low positive affect in Sample 2 (N = 100) of the present article. It is concluded that research on CHD should focus on affective disorder, specific negative emotions, and global personality traits, and that the DS16 is a practical, sound research tool that can be used to assess Type-D.

The Convergent Validity of MMPI and Rorschach Scales: An Extension Using Profile Scores to Define Response and Character Styles on Both Methods and a Reexamination of Simple Rorschach Response

Past research indicated the convergence of Rorschach and MMPI scales may be a function of (a) simple Rorschach response frequency (R) or (b) complex response-character styles on both methods. In this study, new criteria were developed for defining the second assumption using F and K from the MMPI and R and Lambda from the Rorschach. Although substantially different from the factor criteria used previously (KS = .45 and .30), the new criteria still produced the expected pattern of correlations among MMPI and Rorschach scales. Averaged across 17 constructs, the new criteria produced strong validity coefficients for patients with similar styles (M composite r = .50), though they were less effective for patients with discordant styles (M composite r = -.27). It was also demonstrated that R by itself does not moderate convergent validity. Rather, statistical modeling with two sets of 300 random samples (a) demonstrated the prior findings related to R were the result of sampling error and (b) supported the general hypothesis that Rorschach and MMPI scales correlate to the extent response-character styles correlate. Implications are considered.

The Antisocial Personality Questionnaire: An Inventory for Assessing Personality Deviation in Offender Populations1 * An earlier version of this paper was presented at the Fifth European Conference on Law and Psychology, Budapest, Hungary, August 1995

Summary: The development is described of the Antisocial Personality Questionnaire (APQ), a short multitrait, self-report inventory that measures intrapersonal and interpersonal dispositions of relevance to antisocial populations. Scales were generated through factor analysis of an item pool adapted from the MMPI, Buss-Durkee Hostility Inventory, and a self-report scale of Psychopathy, using samples of male mentally disordered offenders (N = 499) and male volunteer nonoffenders (N = 238). Eight factors extracted were identified as Self-Control, Self-Esteem, Avoidance, Paranoid Suspicion, Resentment, Aggression, Deviance, and Extraversion. Short scales constructed to measure these have satisfactory reliability (α), and correlations with measures of personality disorder, observer ratings of interpersonal style, and criminal career data support their construct validity. Scale intercorrelations yield two higher-order dimensions of hostile impulsivity and social withdrawal that reflect orientations towards others and the self, respectively. The APQ provides comprehensive coverage of the deviant traits implicated in personality disorder and antisocial behavior, and appears to tap three of the Big Five personality dimensions (Neuroticism, Extraversion and Agreeableness). The questionnaire has satisfactory psychometric properties and can aid research and intervention with offenders.

MMPI and MMPI-2 scores on the Cook-Medley Hostility Scale

Raw-score means of the Cook-Medley Hostility scale are smaller in the Minnesota Multiphasic Personality Inventory-2 (MMPI-2) normative group than in the MMPI normative group, and T scores are larger. The results contrast with those for the clinical scales but are consistent with results for some other scales. Reasons for the different patterns of MMPI and MMPI-2 scores are discussed.

Colon cancer: personality factors predictive of onset and stage of presentation

This study examined premorbid personality correlates of colon cancer and stage of presentation of colon cancer to health care providers. Sixty-one male veterans who completed the MMPI between 1947 and 1975 and were then diagnosed with colon cancer between 1977 and 1988 were matched with control patients. A 21-factor solution of the MMPI [1] was used to seek potential personality differences between colon cancer cases and their controls in terms of presence of colon cancer and stage of presentation for this disease. A stepwise conditional regression analysis found significant differences between the colon cancer and control groups on the Aggressive Hostility variable (p < 0.018). A multivariate analysis of variance conducted across the stages of colon cancer presentation found that patients who presented later on for colon cancer had higher Phobia scores (p < 0.05). Religious Fundamentalism was also related to presentation (p < 0.05), but in a nonlinear manner. Discussion is related to previous findings regarding the relationship between personality and development of cancer, as well as to implications for patient screening.

Item-level and scale-level factor structures of the MMPI-A

The Minnesota Multiphasic Personality Inventory-Adolescent (MMPI-A) is a 478-item test that represents a substantial revision of the original form of the MMPI. This investigation sought to identify the item-level factor structure of the MMPI-A and also examined the scale-level factor structure of the 69 scales and subscales of this instrument. The study utilized the 1,620 normal adolescents (805 boys and 815 girls) of the normative sample for the MMPI-A. These adolescents ranged in age from 14 to 18, inclusive, with a mean age of 15.54 for boys and 15.60 for girls. A principal factor analysis of item-level responses resulted in extraction of 14 factors that were subjected to promax (oblique) rotation procedures. These 14 factors incorporated 81% of the total MMPI-A item pool and accounted for 44% of the total item-level response variance. For the scale-level analysis, 8 factors were selected for extraction and submitted to promax rotation procedures. These eight factors accounted for a total of 93.5% of the total variance in MMPI-A scale and subscale raw scores. Item-level results were discussed in terms of areas of congruence and dissimilarities from previously reported MMPI factor analyses in adolescent and adult samples, and scale-level factor results were presented in terms of clinical implications for the interpretation of MMPI-A scales and subscales.

The Rorschach's Factor Structure: A Contemporary Investigation and Historical Review

The Rorschach Inkblot Test was factor analyzed to assess for a two-dimensional structure that was expected to reflect the traditional interpretation of many scores, as well as two dimensions that have become a basic paradigm for studying self-reported personality and mood. In my sample (N = 268), Comprehensive System scores had a four-dimensional structure. These factors replicated many earlier findings but were independent of self-report measures. Two of the Rorschach dimensions were strongly defined by response frequency (R) and reflect the powerful impact that R has on determining scores and their intercorrelations. A third dimension was comprised of form-dominant shading determinants and partially corresponded to a hypothesized factor of dysphoria. The fourth dimension was comprised of holistic nonform-dominant color and shading determinants and appears to reflect Shapiro's (1965) description of the hysterical mode of cognition. This study suggests it is necessary to expand our conceptualization of the Rorschach to account for its reproducible factors.

MMPI factor analytic findings for adolescents: item- and scale-level factor structures

Although numerous studies have examined the factor structures of the Minnesota Multiphasic Personality Inventory (MMPI) in investigations of adults, very little attention has been directed to this issue in adolescent samples. In this study, we investigated the item-level and scale-level factor structures using a sample of 1,762 adolescents receiving psychiatric services at the time of their MMPI assessment. Results identified 17 factors related to item responses and 5 factors based on scale-level data. Findings are discussed in relation to results obtained in adult populations and the potential utility of factor analytically derived content scales for adolescent MMPI interpretation.

Replicated item level factor structure of the MMPI: racial and sexual differences

Item-level factor studies of the MMPI date back to the work of Comrey (1957). Surprisingly, there are no extant studies of the items that clinicians use most in the course of daily MMPI interpretive reporting, namely, the 399 x 399 matrix that represents all of the items that comprise the traditional clinical and validity scales. Furthermore, there are no prior studies with adequate Ns that have examined the replicability of MMPI factor structure via available factor comparison techniques (Harman, 1976). In this study, 20,000 MMPIs were factored by the principal components method, followed by Varimax rotations of 6 through 25 factors. The coefficient of congruence was used to compare the factor structures of randomly divided subsamples, as well as males vs. females and Blacks vs. Whites. Differences in factor structure were found, and suggestions are made with regard to the significance of these findings in clinical and applied settings.

Hostility, coronary heart disease, and total mortality: a 33-year follow-up study of university students

Hostility as measured by the Cook-Medley Hostility (HO) Scale on the Minnesota Multiphasic Personality Inventory has been suggested as a risk factor for coronary heart disease (CHD) and total mortality. This study tested the HO-CHD hypothesis in a sample of 1399 men who entered the University of Minnesota in 1953 and, as part of freshman orientation, completed the MMPI. Current health status was ascertained for 94% of the sample through telephone interviews 33 years later. Higher HO scores did not predict CHD mortality, CHD morbidity, or total mortality either before or after adjustment for baseline risk factors. Among the plausible explanations for these results are that (1) hostility is not a risk factor in all populations, (2) the HO scale at age 19 does not assess a stable psychological characteristic, or (3) the HO scale is not an adequate measure of hostility.

Refining the type A hypothesis: emergence of the hostility complex

Recent negative research findings have raised questions regarding the robustness of the Type A hypothesis. A growing body of evidence suggests that not all aspects of the global Type A behavior pattern are pathogenic, but only those concerned with hostility and anger. Biologic mechanisms responsible for increased risk of coronary disease in persons with high levels of hostility and anger appear to involve increased cardiovascular and neuroendocrine responses to behavioral challenge, and, possibly, inadequate parasympathetic antagonism of sympathetic nervous system effects. Future research should concentrate on further refining our understanding of the hostility complex and its pathophysiologic mechanisms.