David G. Kleinbaum

John Henryism and blood pressure differences among black men. II. The role of occupational stressors

In this study, the effects of psychosocial job stressors on the resting blood pressure (BP) of 112 black male workers were examined. The subjects resided in a rural, poor, predominantly black community in eastern North Carolina. The job stressors included unemployment, lack of job security, lack of job success, the perception that wages earned were too low for the work performed (and inhibited anger about unfair wages), and the perception that being black had hindered chances for achieving job success. The effect-modifying influence of on-the-job social support, and John Henryism, on several of these relationships was also examined. For systolic blood pressure, a main effect was observed for job security, and an interaction effect was observed for employment status and time of day of interview. For diastolic blood pressure, significant interactions were observed for job success and John Henryism, and for job success and the perception that being black had hindered chances for achieving job success. These findings further clarify under what conditions John Henryism may be associated with higher BPs in this sample of black men. These findings also shed light on the emotional pathways through which selected job stressors may influence resting BPs in these men.

A generalization of the growth curve model which allows missing data

This study presents methods for estimating and testing hypotheses about linear functions of the unknown parameters in a generalization of the growth curve model which allows missing data. The estimators proposed are best asymptotically normal (BAN). A testing method for large samples is described which uses a test criterion given in general form by Wald. The asymptotic null distribution of the test statistic is a central chi-square variable. A BAN estimator of a linear vector function of the unknown parameters of the expectation model and consistent estimators of the variance-covariance parameters are required for computation.

Logistic regression analysis of epidemiologic data: theory and practice

The use of logistic regression analysis is widely applicable to epidemiologic studies concerned with quantifying an association between a study factor (i.e., an exposure variable) and a health outcome (i.e., disease status). This paper reviews the general characteristics of the logistic model and illustrates its use in epidemiologic inquiry. Particular emphasis is given to the control of extraneous variables in the context of follow-up and case-control studies. Techniques for both unconditional and conditional maximum likelihood estimation of the parameters in the logistic model are described and illustrated. A general analysis strategy is also presented which incorporates the assessment of both interaction and confounding in quantifying an exposure-disease association of interest.

Clinic Policy and Retention in Methadone Maintenance

Advantage was taken of a natural experiment to investigate the relationship of clinic objectives and procedures with retention in methadone maintenance. Assessment, maximum dose, and time in treatment data on 238 patients were summarized using Cox regression. While allowing for patient descriptors and maximum dose of methadone, those subjects assigned to a strongly abstinence-oriented program were 26% (95% C.I., 18 to 33%), 98% (66 to 137%), 213% (132 to 322%), and 393% (224 to 651%) more likely to leave treatment in the first, second, third, fourth 6 months of treatment, respectively, than those subjects assigned to a more laissez-faire program.

The Edgecombe County High Blood Pressure Control Program: I. Correlates of uncontrolled hypertension at baseline.

To guide the planning of a multifacetted hypertension control program in Edgecombe County, North Carolina, a baseline survey of a stratified (by township) random sample of 1,000 households was conducted. All adults (greater than or equal to 18 years) were interviewed and had their blood pressures (BP) measured. Five hundred thirty-nine individuals, 27 per cent of the survey population, had diastolic BP greater than or equal to 90 mm Hg or were receiving anti-hypertensive drug therapy. The 539 hypertensives were divided into seven subgroups reflecting successive stages in the control of hypertension based on the awareness, treatment, and control of their hypertension. Unaware hypertensives were further subdivided into three groups according to the recency of their last BP check, and those aware but untreated were subdivided by whether they had previously received treatment. The seven subgroups of hypertensives were compared, separately for women and men, with respect to sociodemographic characteristics, health behaviors, and health status. In general, the progression from undetected hypertension to treatment and control appeared to be associated with being older, female, and White. This progression was further associated with greater educational levels and higher family incomes among women and increasing self-reported morbidity among men. The implications for intervention of these and other described associations are discussed.

The Edgecombe County (NC) High Blood Pressure Control Program: II. Barriers to the use of medical care among hypertensives.

As the initial step in a five-year project to improve control of high blood pressure in Edgecombe County, North Carolina, a survey was conducted in 1980 to determine the prevalence of hypertension and to identify factors which might constitute barriers to the use of medical care by hypertensives. This report summarizes the findings for the 539 hypertensives identified through the baseline survey. In general, Black hypertensives reported more access problems than Whites. Within race, however, males and females differed very little on selected measures of potential access to medical care. Among women, lower scores on potential access were strongly associated with being untreated, whereas for men, concerns about the safety of anti-hypertensive drug therapy were associated with being unaware. On a summary measure of the actual use of medical care in response to symptoms, both male and female treated hypertensives scored higher than their untreated counterparts. The implications of these and other findings for community-based blood pressure control activities are discussed.

The Edgecombe County High Blood Pressure Control Program: III. Social support, social stressors, and treatment dropout.

In a hypertension prevalence survey of a stratified random sample of 1,000 households, 2,030 adults (aged 18 years and over) were interviewed and information on psychosocial variables collected. Among 359 hypertensives, there was a consistent relationship between indicators of difficulty in the social environment and dropout from treatment in women. Compared to those who remained in treatment, women who dropped out can be characterized as having less social support on the job, having less perceived spouse approval (if married), having a lower level of perceived access to supportive resources, and being more likely to report feeling pushed most or all of the time if they are homemakers. Relationships between indicators of social support and dropout from treatment in men were found only with support on the job, and for White men, with perceived friend approval.

Testing Linear Hypotheses in Generalized Multivariate Linear Models

This study presents a general large sample procedure for resting linear hypotheses in multivariate linear models for which observations say be missing and/or for which different design matrices correspond Co different response variates. The theoretical foundation of this method is due to Wald. The asymptotic null distribution of the test statistic is a chi square variable. The test statistic is a quadratic form constructed from a BAN estimator of a linear function of the unknown parameters in the expectation model and a consistent estimator of the variance-covariance matrix. For the special case of the standard multivariate analysis of variance models, the Wald statistic is exactly equivalent to Hotellings Trace Criterion.

A method for selecting criteria to evaluate medical care.

This study tests a questionnaire method for eliciting process criteria for medical care appraisal. The questionnaire was sent to national samples of family physicians, pediatricians, and pediatricians specializing in infectious diseases asking their opinions about various clinical actions in 125 clinical situations concerning respiratory infection in infants. Five hundred twenty-four (54%) physicians returned completed questionnaires. Questionnaire responses favored the performance of a majority of actions and opposed very few. Opinions concerning individual actions, particularly diagnostic tests and treatments, varied widely depending upon the clinical situation presented. A second questionnaire sent one year later indicated that the opinions expressed in the first questionnaire remained stable over time, especially if the initial opinion favored performance of the action. Comparison of the questionnaire responses and medical records of a group of practitioners demonstrated that only 55% of actions favored in a practitioners questionnaire appeared in his records. Although the questionnaire method appears to be a feasible, specific, and reliable means of identifying clinical opinion, there remains considerable discordance between opinion as expressed in the questionnaire and recorded clinical practice.

On testing hypotheses concerning standardized mortality ratios

Abstract This paper describes tests of hypotheses concerning the equality of any k standardized mortality ratios (or, equivalently, of any k indirect adjusted rates) from p (≥ k ) populations. The distinction is made between the situation where the standard population is chosen independently of the p populations and one where the standard is formed by pooling all p populations. Different test procedures are required for these two situations. When the pooled standard is used, the appropriate test procedure is applicable only when k = p . Experimental evidence is given showing that when the pooled standard is used both test procedures lead to the same conclusion concerning the hypothesis for the case k = p . The recommendation is made, therefore, to use the “incorrect” test procedure when the standard is a pooled one even when k p .