Understanding AIDS-risk behavior among adolescents in psychiatric care: links to psychopathology and peer relationships.

OBJECTIVE: Severely mentally ill youths are at elevated risk for human immunodeficiency virus infection, but little is known about acquired immunodeficiency syndrome (AIDS) risk behavior in adolescents who seek outpatient mental health services or about the links between psychiatric problems and particular high-risk behaviors. This pilot study used structural equation modeling to conduct a path analysis to explore the direct and indirect effects of adolescent psychopathology on risky sex, drug/alcohol use, and needle use. METHOD: Ethnically diverse youths (N = 86) and their caregivers who sought outpatient psychiatric services in Chicago completed questionnaires of adolescent psychopathology. Youths reported their relationship attitudes, peer influence, sexual behavior, and drug/alcohol use. RESULTS: Different AIDS-risk behaviors were associated with distinct forms of adolescent psychopathology (e.g., delinquency was linked to drug/alcohol use, whereas aggression was related to risky sexual behavior), and peer influence mediated these linkages. Some patterns were similar for caregiver- and adolescent-reported problems (e.g., peer influence mediated the relation between delinquency and drug/alcohol use), but others were different (e.g., caregiver-reported delinquency was associated with risky sex, whereas adolescent-reported delinquency was not). CONCLUSIONS: Findings underscore the complexity of factors (types of informants and dimensions of psychopathology) that underlie AIDS risk in troubled youths, and they offer specific directions for designing and implementing uniquely tailored AIDS prevention programs, for example, by targeting delinquent behavior and including high-risk peers and important family members in interventions.

Li+/Mg2+ competition at therapeutic intracellular Li+ levels in human neuroblastoma SH-SY5Y cells.

OBJECTIVES: One proposed mechanism of lithium action in the treatment of bipolar disorder is that Li+ competes with Mg2+ for Mg2+ binding sites within the cell. In this study, we investigated this competition at therapeutic intracellular Li+ levels in human neuroblastoma SH-SY5Y cells. METHODS: We used fluorescence spectroscopy and a Mg2+ indicator, furaptra, to investigate this competition in human neuroblastoma SH-SY5Y cells. Atomic absorption spectrophotometry was used for determination of the intracellular Li+ levels. RESULTS: The neuroblastoma cells, incubated in 15 mM or 30 mM Li+-containing buffer, showed a significant increase in free intracellular Mg2+ levels [using a positive linear within-groups contrast t-test, the 15 mM condition produced t(2) = 5.0, one-tailed p < 0.02, and the 30 mM Li+-incubation conditions gave t(2) = 9.2, one-tailed p < 0.006] but did not significantly increase over time in the Li+-free condition [t(2) = 0.1, one-tailed p > 0.96]. At the earlier times during the incubation (1 or 10 min for the 15 mM or 30 mM Li+-containing buffers), the intracellular Li+ concentrations were 0.6-2.5 mM, values which are comparable to those reached in the brain of Li+-treated patients. CONCLUSION: We demonstrated that competition between Li+ and Mg2+ can occur at therapeutic intracellular Li+ levels.

Statistical methodology: VIII. Using confirmatory factor analysis (CFA) in emergency medicine research.

How many underlying characteristics (or factors) does a set of survey questions measure? When subjects answer a set of self-report questions, is it more appropriate to analyze the questions individually, to pool responses to all of the questions to form one global score, or to combine subsets of related questions to define multiple underlying factors? Factor analysis is the statistical method of choice for answering such questions. When researchers have no idea beforehand about what factors may underlie a set of questions, they use exploratory factor analysis to infer the best explanatory model from observed data "after the fact." If, on the other hand, researchers have a hypothesis beforehand about the underlying factors, then they can use confirmatory factor analysis (CFA) to evaluate how well this model explains the observed data and to compare the model's goodness-of-fit with that of other competing models. This article describes the basic rules and building blocks of CFA: what it is, how it works, and how researchers can use it. The authors begin by placing CFA in the context of a common research application-namely, assessing quality of medical outcome using a patient satisfaction survey. They then explain, within this research context, how CFA is used to evaluate the explanatory power of a factor model and to decide which model or models best represent the data. The information that must be specified in the analysis to estimate a CFA model is highlighted, and the statistical assumptions and limitations of this analysis are noted. Analyzing the responses of 1,614 emergency medical patients to a commonly-used "patient satisfaction" questionnaire, the authors demonstrate how to: 1) compare competing factor-models to find the best-fitting model; 2) modify models to improve their goodness-of-fit; 3) test hypotheses about relationships among the underlying factors; 4) examine mean differences in "factor scores"; and 5) refine an existing instrument into a more streamlined form that has fewer questions and better conceptual and statistical precision than the original instrument. Finally, the role of CFA in developing new instruments is discussed.

Stress and coping among Russian and American early adolescents.

Self-reported stress, coping, and depression were compared between 270 Russian and 270 American early adolescents, 10 to 14 years of age. Russian and American adolescents reported equal levels of major life stress, but Russian adolescents reported greater levels of everyday life stress. Russian adolescents reported that they were less likely to use externalizing coping and more likely to use social support and problem-solving coping compared with American adolescents. Russian adolescents also reported that they were more depressed. However, LISREL path model analyses examined the relationships among these variables and showed that Russian and American adolescents coped with stress in similar ways. A buffering effect for social support on depression was found for both national groups.

Confirmatory P-technique analyses of therapist discourse: high-versus low-quality child therapy sessions.

From a sample of 35 child therapy sessions, 3 of the highest and 3 of the lowest quality sessions (HQS and LQS) were identified by independent ratings on the Loyola Child Psychotherapy Process Scales. Each of the 535 therapist-utterances in the HQS and the 372 therapist-utterances in the LQS were rated on 15 language interaction scales. Principal-components P-technique analyses were applied separately to the HQS and LQS, revealing 3 factors, Responsive Informing (RI), Initiatory Questioning (IQ), and Positive Regard (PR), that seemed to correspond across session types. Using confirmatory P-technique, a general measurement model with acceptable goodness-of-fit was determined for HQS. Two factors emerged: RI and IQ but not PR. Using a bootstrap procedure, random halves of the HQS were repeatedly sampled and compared via multigroup confirmatory factor analysis to create a reference distribution of the magnitudes of their difference chi-squares. Random resamples of HQS and LQS were then compared, resulting in a second reference distribution. Statistical tests and effect-size measures revealed replicable and large differences between sessions types.

Comparing five alternative factor-models of the Student Jenkins Activity Survey: separating the wheat from the chaff.

The Student Jenkins Activity Survey (SJAS; Glass, 1977) is the most popular instrument for measuring Type A behavior in college undergraduates, and a variety of different unidimensional and multidimensional models for scoring the SJAS have been employed in the past. This study compared five alternative factor-models developed for the short (21-item) form of the SJAS in terms of their overall goodness of fit, factorial invariance, and factor reliabilities, using the data of two independent samples of undergraduates. Indices of relative fit for the five models ranged from .64 to .94 (median = .85), but only one unidimensional model and one multidimensional model achieved a minimum satisfactory goodness of fit (.90 or greater) for both samples. Reconceptualizing these models within a broader nomological net sheds light on their meaning and highlights existing gaps in the measurement of multifaceted Type A behavior.

First-year residents' caring, medical knowledge, and clinical judgment in relation to laboratory utilization.

PURPOSE: To investigate first-year residents' levels of caring (concern for others' well-being), medical knowledge, and clinical judgment in relation to their levels of laboratory utilization. METHOD: Self-report questionnaires about caring, knowledge, and judgment were given in 1986-87 to 36 first-year residents in a three-year internal medicine residency program of the McGaw Medical Center of Northwestern University. Inpatient laboratory utilization data obtained from structured chart audits over a one-year period were used to construct comparable diagnosis- and severity-specific physician practice profiles, from which the residents received overall utilization scores for laboratory test charges. Statistical methods included Cronbach's alpha reliability coefficient and multiple regression analysis. RESULTS: The multiple regression analysis showed that medical knowledge was an independent predictor of increased laboratory utilization (standardized beta = .54, p < .04, partial R2 = .07); clinical judgment was an independent predictor of decreased utilization (standardized beta = -.53, p < .05, partial R2 = .06); and caring was unrelated to utilization (standardized beta = .15, ns, partial R2 = .01). CONCLUSION: The finding that clinical judgment was related to less laboratory utilization suggests that future research should investigate the decision-making concomitants of judgment to better understand its translation into resource utilization. It is possible that the relationship between medical knowledge and laboratory utilization is developmentally specific, and thus the knowledge of more experienced physicians, who would likely be more precise decision makers than first-year residents, may be related to decreased rather than increased utilization.

A measurement model for the Type A Self-Rating Inventory.

Compared to other questionnaire measures of Type A behavior, the Type A Self-Rating Inventory (TASRI) possesses particularly strong face validity. This study sought to develop a formal measurement model for the TASRI using a sample of 352 male and 479 female undergraduate psychology students. Assessed via structural equation modeling, an oblique two-factor (Hard-Driving, Extroverted) model explained over 90% of the common variance in responses to a subset of 13 of the original 28 TASRI items for men, women, and the pooled data. As hypothesized, relative to women, men had a greater mean score on the Hard-Driving factor and a lower mean score on the Extroverted factor. However, the magnitude of these gender differences is small. When reconceptualized in the context of work on the Big Five factor model, the Hard-Driving and Extroverted factors were found to reflect related elements of positive sociability and negative power, respectively.

The factor structure and cross-sectional distributional properties of the Beth Israel/UCLA Functional Status Questionnaire.

The functional status of ambulatory patients is an important correlate of their emotional and physical well-being and is subject to influence by the delivery of health care. First, the present study obtains a profile on the FSQ for a large sample of young, healthy adults (N = 508 college undergraduates), for use as an "empirical standard" against which to evaluate the effect of disease or injury on an individual's functional status. Second, cross-sectional analyses contrast distributional features of responses to the FSQ between (a) the present younger, healthy sample and (b) a previously published, older ambulatory patient sample. These analyses suggest that the FSQ has acceptable discriminant validity for intergroup comparisons. Finally, confirmatory structural analyses suggest that the theoretical six-factor model hypothesized to underlie responses to the FSQ has a modest, yet acceptable, goodness of fit to actual data.

An analysis of decision making in discharge planning.

A repeated-measures design was used to examine medical professionals' discharge planning strategies. Physicians, residents, nurses, and social workers were presented with 16 hypothetical case scenarios and asked to: (1) rate the appropriateness of four discharge options (nursing home, community nursing, adult day, and outpatient clinic care), and (2) select the most appropriate discharge plan for each case. Four within-group variables were included in the scenarios: physical impairment, caregiver availability, follow-up required, and patient compliance. Decisions were greatly influenced by caregiver availability. When a caregiver was available, respondents preferred community-based options (i.e., community nursing care or outpatient clinic); if the case involved complications (i.e., severe physical impairment, heavy follow-up, noncompliant patient), they considered community nursing care more appropriate than outpatient clinic. When a caregiver was unavailable, respondents preferred institution-based options (i.e., nursing home or adult daycare); if there were complications, they considered nursing home more appropriate than adult daycare.

A path analysis of medical school and residency performance. Implications for housestaff selection.

The selection and subsequent performance of 212 internal medicine residents was examined by factor analysis and path analysis. A three-factor solution accounted for most of the variance among the nine selection variables. These three factors, labeled Board Scores, Faculty Evaluations, and Academic Distinction, were then combined with in-training residency performance evaluations and composite scores on the ABIM certifying examination to produce a comprehensive path model of house staff selection and performance. The Academic Distinction factor emerged as the strongest predictor of residency performance,; the Faculty Evaluations factor was also a significant component of the model. Standardized test scores correlated poorly with clinical performance. The data suggest that increased attention to the content of letters of reference could substantially improve their predictive validity. Other means of reporting subjective evaluations may also be needed to increase the stature of non-cognitive attributes in house staff selection decisions.

A note on measurement issues in type A research: let's not throw out the baby with the bath water.

It has been suggested that researchers rely exclusively on the structured interview (SI) to assess Type A behavior instead of using objective self-report measures, because the SI is the only prospectively validated instrument currently available. This article considers the costs and benefits of relying solely on the SI to measure Type A behavior. Although using only the SI would assure that researchers measure, in a relatively unobtrusive fashion, actual Type A behaviors known to be predictive of heart disease, it would dramatically increase research costs, impede longitudinal studies of changes in Type A behavior, reduce the validity of statistical conclusions, restrict the convergent and discriminant validity of the Type A construct, and ultimately inhibit our ability to improve accuracy in predicting heart disease. A set of recommendations is proposed for improving the quality of measurement in Type A research.

Type A behaviour and time urgency: perception of time adjectives.

Relations between time perception and Type A behaviour were investigated. Extreme Type A (n = 26) and B (n = 28) male college undergraduates rated the appropriateness of 158 adjectives as descriptors of time. Adjectives were independently rated using seven-point Likert 'speed' and 'energy' scales. Analyses showed that Type As rate adjectives low in speed and energy as less descriptive of time than Type Bs and vice versa.

Intimacy motivation and subjective mental health in a nationwide sample.

Over 1,200 adults in a representative nationwide sample were administered the Thematic Apperception Test (TAT) and an unstructured interview which produced 25 measures of subjective mental health and ultimately six factor scores: unhappiness, lack of gratification, strain, feelings of vulnerability, lack of self-confidence, and uncertainty. The TATs were scored for intimacy motivation (McAdams, 1980)--a recurrent preference or readiness for experiences of warm, close, and communicative interaction with others. Controlling for age, education, and their interaction, high intimacy motivation in women was associated with greater happiness and gratification, whereas in men it was associated with lack of strain and lack of uncertainty. Women high in intimacy motivation who were living alone reported lower levels of gratification and more uncertainty in their lives than other women. Intimacy motivation also declined significantly over the life span for women, but not for men. Finally, demographic effects on intimacy motivation were examined. Controlling for age, education, and their interaction, professional men (e.g., doctors, lawyers, and teachers) scored higher on intimacy motivation than did men in other occupational categories, while among women the occupational groups with highest intimacy motivation were service workers and craftspersons.

Dimensions of social insecurity and their relation to coronary-prone behaviour in college undergraduates.

Research suggests that Type A behaviour and 'social insecurity' (SI) may represent independent additive task factors for coronary disease. Research examining SI and Type A behaviour, however, has not rigorously evaluated the structure of the SI scale. The present study examined the structure of a previously used eight-item SI scale, and its relation to a questionnaire measure of Type A (the student Jenkins Activity Survey), for a sample of 504 undergraduates. Exploratory and confirmatory factor analyses of responses to the SI items revealed a three-factor solution was most appropriate for Type As, and a two-factor solution for Type Bs. For both As and Bs, the first factor reflected insecurity regarding one's own capabilities and worth (self-specific insecurity), and the second factor reflected insecurity specific to a group or social context (group-specific insecurity). A third factor emerging only for Type As reflected insecurity regarding others' capabilities (other-directed insecurity). Semipartial correlations revealed Type A total score was marginally negatively related to group-specific insecurity and marginally positively related to other-directed insecurity, corroborating previous angiographic and prospective medical research. The hard-driving/competitive subscale was significantly negatively related to self-specific insecurity, supporting evidence regarding Type A and interpersonal dominance, competitiveness, and aggression.

Comparing the long and short forms of the student version of the Jenkins Activity Survey.

The 44-item (long-form) student Jenkins Activity Survey (SJAS), used in locating subjects on the Type A/B continuum, requires an average of 8 min more to complete than the short form (containing only the 21 scored items). This study was conducted to determine whether the long and short forms provide comparable score distributions and factor structures for independent random samples of undergraduates (N = 1248). Analyses revealed that the short form provides factors that are more independent than those of the long form, provides comparable factor structures for A's and B's, and provides comparable distributions of SJAS scores.