Psychosomatic Medicine
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zhang, J.
Right arrow Articles by Ward, K. D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhang, J.
Right arrow Articles by Ward, K. D.
Related Collections
Right arrow Endocrinology
Right arrow Personality
Right arrow Stress and Coping
Right arrow Other Cardiovascular Medicine
Psychosomatic Medicine 68:718-726 (2006)
© 2006 American Psychosomatic Society

ORIGINAL ARTICLES

Hostility and Urine Norepinephrine Interact to Predict Insulin Resistance: The VA Normative Aging Study

Jianping Zhang, MD, PhD, Raymond Niaura, PhD, Joshua R. Dyer, MS, Biing-Jiun Shen, PhD, John F. Todaro, PhD, Jeanne M. McCaffery, PhD, Avron Spiro, III, PhD and Kenneth D. Ward, PhD

From the Department of Psychology, Indiana University Purdue University Indianapolis (IUPUI), Indianapolis, Indiana (J.Z., J.R.D.); The Butler Hospital and Brown Medical School, Providence, Rhode Island (R.N.); the Department of Psychology, University of Miami, Miami, Florida (B.-J.S.); Centers for Behavioral & Preventive Medicine, The Miriam Hospital and Brown Medical School, Providence, Rhode Island (J.F.T., J.M.M.); Boston University School of Public Health and Massachusetts Veterans Epidemiology Research and Information Center (MAVERIC), VA Boston Healthcare System, Boston, Massachusetts (A.S.); and the Department of Health & Sport Sciences, and Center for Community Health, University of Memphis, Memphis, Tennessee (K.D.W.).

Address correspondence and reprint requests to Jianping Zhang, MD, PhD, Department of Psychiatry and Psychology, Cleveland Clinic Foundation, 9500 Euclid Ave, P57, Cleveland, OH 44195. E-mail: jpzhang2005{at}gmail.com.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 METHOD
 RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 
Objective: Previous research has produced mixed results pertaining to the association between hostility and insulin resistance. These inconsistent findings may be the result of a lack of studies examining potential moderators of this relationship and inconsistent measures of insulin resistance and/or hostility. We hypothesized that hostility may interact with circulating norepinephrine (NEPI) levels, indexed by 24-hour urine concentrations, to affect insulin resistance.

Methods: Six hundred forty-three men (mean age = 63.1 years) free of diabetic medications completed the Minnesota Multiphasic Personality Inventory and participated in a laboratory assessment. The Cook-Medley Hostility (Ho) and 24-hour urine NEPI were used to predict insulin resistance defined by the homeostatic model assessment (HOMA) index, 2-hour postchallenge glucose (PCGL), and insulin levels (PCIL) after controlling for nine common covariates.

Results: Multiple regression showed that the two-way interaction between Ho and NEPI significantly predicted HOMA and PCIL, but not PCGL, after controlling for covariates. Simple regression slopes of Ho on HOMA and PCIL were explored and indicated that, at higher levels of NEPI, higher Ho was associated with higher HOMA (ß = 0.14, p < .05). Ho was not a significant predictor of HOMA at mean and lower levels of NEPI. Similar results were obtained for PCIL, but not PCGL. Cynicism, but not other subscales of Ho, was similarly related to insulin resistance and NEPI.

Conclusion: Individuals with high stress and high hostility were more likely to have insulin resistance. It is important to study moderators in the relationship between hostility and insulin resistance.

Key Words: hostility • norepinephrine • insulin resistance • HOMA • metabolic syndrome

Abbreviations: CVD = cardiovascular disease; CHD = coronary heart disease; NIDDM = noninsulin-dependent diabetes mellitus; HTN = hypertension; HOMA = homeostatic model assessment approach; OGTT = oral glucose tolerance test; CMHOST = Cook-Medley hostility scale; Ho = full scale score of CMHOST; QUICKI = quantitative insulin-sensitivity check index; NEPI = norepinephrine; NAS = Normative Aging Study; MMPI = Minnesota Multiphasic Personality Inventory; BMI = body mass index; WHR = waist-to-hip ratio; FFQ = food frequency questionnaire; SD = standard deviation.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
METHOD
 RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 
The metabolic syndrome, also called "syndrome X" and "insulin resistance syndrome," is a cluster of pathophysiological features, including visceral obesity, insulin resistance, hyperglycemia, hypertension, and dyslipidemia, that increases the risk for both cardiovascular disease (CVD) and type II diabetes (noninsulin-dependent diabetes mellitus [NIDDM]) (1–4). Among these features, insulin resistance is believed to play a central role in the development of the metabolic syndrome (5–7). Although the pathogenesis of insulin resistance may be related to proliferation of fat tissue, genetic disposition, hormonal dysfunction, and other pathophysiological processes (8,9), psychosocial stress has also been shown to be an important factor (10). Given the fact that stress and certain psychosocial factors have been linked to the development of CVD and NIDDM (11,12), insulin resistance and the metabolic syndrome might play a mediating role in these relationships, although there is no direct evidence to support this hypothesis.

One psychosocial factor that has been intensively studied is hostility, a personality disposition characterized by chronic anger, cynicism, distrust, and antagonism (13). Hostility has been reported to be an independent risk factor for CVD as well as all-cause mortality in a number of prospective studies (3,14–16). However, previous research regarding the association between insulin resistance and hostility has been inconsistent. One issue is the diversity of measurement methods of insulin resistance across studies, potentially contributing to the inconsistency. Using a homeostatic model assessment approach (HOMA) (17), Surwit and colleagues found hostility was correlated with greater insulin resistance (18), but Philip and Facchini did not find such a significant relationship using the standard oral glucose tolerance test (OGTT) to define insulin resistance (19).

Other studies have used proxies such as fasting glucose levels, fasting insulin levels, and measures of obesity to assess insulin resistance. Hostility has been positively correlated with an increase in visceral adiposity and fasting insulin in a sample of American postmenopausal women (20) and average blood glucose measured by HbA1c in a sample of Japanese adult men (21). More recently, a large epidemiologic study of a Swedish sample (22) found a longitudinal association between cynical hostility and metabolic syndrome in which insulin resistance was a major component and measured by an insulin sensitivity check index (QUICK-I) (23). Using a similar strategy to define the metabolic syndrome, another study of 134 children aged 8 to 17 years found that children with higher scores of hostility were more likely to exhibit metabolic syndrome at a 3-year follow-up (24). Among the components of metabolic syndrome, insulin resistance, measured by the HOMA method, contributed most to the factor score.

It would be methodologically important to investigate which measures of insulin resistance are associated with hostility in a single sample. Inspecting the formulas (17), QUICKI (=1/[log(I0)+log(G0)], where I0 and G0 are fasting insulin and glucose concentrations, respectively) is basically the same as HOMA (=G0 * I0/22.5) with an inversed log transformation. Studies have shown that HOMA was correlated well with the hyperinsulinemic euglycemic clamp derived insulin resistance, the gold standard measure of insulin resistance (17), with correlation coefficients of 0.73 to 0.82 in two studies (25,26). Other proxies of insulin resistance we proposed to test were 2-hour postchallenge glucose levels (PCGL) and insulin levels (PCIN). These measures are routinely used in clinical practice, but have not been frequently examined in terms of their relationships with psychological factors in the literature. It is worthy to note that various measures and proxies of insulin resistance are physiologically and computationally related. Both HOMA and QUICKI are based on fasting glucose and insulin levels. PCGL is part of the standard OGTT protocol routinely used in clinical practice to diagnose diabetes. Although PCIN is usually not part of OGTT, it can be added to obtain information on how insulin has responded to the glucose load. PCGL and PCIN were moderately correlated with the hyperinsulinemic euglycemic clamp-derived insulin resistance, 0.58 and 0.50, in two separate studies (27,28). Glycosalated hemoglobin A1C (HbA1c) assesses long-term glucose control and is often used in gauging diabetic treatment. It is not recommended for initial diagnosis of diabetes and not considered as a good measure of insulin resistance. The hyperinsulinemic-euglycemic clamp technique is regarded as the gold standard measure of insulin resistance, but it requires constant intravenous infusion of insulin while monitoring glucose levels; therefore, technically complex and expensive (17). Hence, the present study used three insulin resistance measures: HOMA, PCGL, and PCIN.

It is also important to examine potential moderators of the relationship between hostility and insulin resistance. For example, Surwit et al. (18) found that the correlation between hostility and HOMA was significant among women, but not among men, and that ethnicity might also moderate the association. Another potentially important moderator of the influence of hostility on insulin resistance that has received little attention is psychosocial stress. The stress moderation model (29) speculates that hostile individuals, when faced with stressful situations, may be more likely to perceive stress and have an exaggerated stress response. An earlier study (30) found higher fasting insulin levels in a group of elderly individuals with high hostility and high daily hassles and higher fasting glucose levels in individuals with high hostility and under chronic stress (i.e., taking care of a spouse with dementia) compared with other groups. However, this study did not directly measure insulin resistance. Hence, to advance previous research, it is necessary to examine the interaction effect of hostility with stress on measures of insulin resistance. In the present study, we propose to use a 24-hour urinary norepinephrine (NEPI) level as an index of stress. Previous research has shown that the rate of urinary catecholamine excretion averaged over a 24-hour period may be an appropriately sensitive measure of real life stress (31–33) and that urinary NEPI is thought to be an index of sympathetic activity integrated over time (32–35). This may be advantageous because it is a more objective measure of stress and may be free of subjective bias as suffered by many self-report measures of stress (36).

Another issue in this area of research is that different studies have used somewhat different measures of hostility. For example, some studies used the full scale scores of the Cook-Medley Hostility scale (CMHOST) (37) to test its association with insulin resistance (19,20), but other studies used a hostility scale as part of the Profile of Mood States (21). In addition, subscales of CMHOST such as cynicism scales derived by Barefoot et al. (38) or by Costa et al. (39) have frequently been used in studies of insulin resistance (24). Surwit et al. (18) assessed hostility with a 27-item scale from the CMHOST, including cynicism, hostile affect, and aggressiveness (38). In contrast, Nelson et al. (22) measured cynical hostility using seven items from the CMHOST based on a factor analysis. In an attempt to resolve the previous inconsistencies in the literature, the present study conducted exploratory analysis on five different subscales of the CMHOST in terms of their associations with insulin resistance.

In summary, it seems that the inconsistencies in the existing literature examining associations among hostility and insulin resistance may be the result of a number of issues, including differences among measures of insulin resistance, differences among measures of hostility, and failure to consider important moderators such as race, gender, and stress. Although it is difficult to resolve these issues in a single study, the present study attempted to investigate how different measures of hostility were related to different measures of insulin resistance while at the same time examining the moderation effect of acute stress. Given the nature of the sample, we were not able to study the moderation effects of gender and race, like in Surwit et al. (18).

The present study used a dataset from the Normative Aging Study (NAS), which is a longitudinal study designed to examine biomedical and psychosocial changes involved in the normal aging process. Previous works have been published using this dataset to study issues related to metabolic syndrome. Niaura et al. (40) examined how hostility was related to metabolic syndrome, which included blood pressure, serum lipids, and fasting glucose and insulin levels. A path analysis showed that the association between hostility and metabolic syndrome might be mediated by body mass index (BMI) and waist/hip ratio (WHR). However, they did not explore how hostility was related to indices of insulin resistance and potential moderators. In another study, Shen et al. (4) conducted a confirmatory factor analysis to show that metabolic syndrome was primarily represented by insulin resistance and obesity and, to a lesser extent, by lipids and blood pressure. The study did not examine whether hostility was related to the common factor of metabolic syndrome and potential moderators. The present study was designed to ask questions that were unanswered by the previous findings. The primary goal was to test the hypothesis that hostility (measured by the full-scale CMHOST) is related to insulin resistance (measured by HOMA) and that this relationship may be moderated by 24-hour urinary NEPI levels. There were two secondary goals with data analyses of exploratory in nature. We examined which of other measures of insulin resistance (i.e., 2-hour PCGL and 2-hour PCIN) demonstrated associations with hostility. We then explored whether certain subscales of CMHOST were significantly associated with insulin resistance.


   METHOD
TOP
 ABSTRACT
 INTRODUCTION
 METHOD
RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 
Participants
The present study used a subsample of the NAS cohort. The NAS consists of 2280 men living in the greater Boston area who at enrollment (1961–1970) were 21 to 80 years of age, predominantly white (approximately 98%), had a high school education, and were free of chronic medical conditions such as cancer, CVD, or diabetes mellitus. Volunteers participated in regular examinations every 3 (age 52 or older) or 5 (younger than age 52) years. Since 1986, however, all men have been examined at 3-year intervals. Additional details of the sample and admission criteria have been described elsewhere (41).

To be included in the present study, NAS participants were required to meet the following criteria: a) completion of the Minnesota Multiphasic Personality Inventory (MMPI) in 1986; b) participation in a subsequent comprehensive physiological and laboratory assessment; and c) not taking diabetic medications at the time of laboratory assessment. Of the original NAS sample of 2280, 1081 provided a valid MMPI assessment and participated in the laboratory visit. The average time from MMPI to laboratory assessment was 2.44 years (±1.25 years). Among these men, 238 were excluded because of missing information on glucose, insulin, or 24-hour urine sample. Another 14 men were excluded as a result of taking one or more diabetic medications at the time of laboratory assessment. An additional 186 men were excluded because their serum creatinine levels were either too high (>1.5 mg/dL) or missing as a result of concerns of their impaired renal function, which could potentially affect urine NEPI levels. These resulted in a final sample of 643 men for the present study.

Procedures
Participants completed the MMPI assessment by mail. On the night before their scheduled laboratory visit, participants were instructed to refrain from eating or drinking after midnight and avoid smoking after 8:00 pm. The physiological assessment included blood pressure measurement, blood work, an anthropometric evaluation, and assessment of health behaviors by standardized questionnaires. The first blood sample was drawn at 8:00 am to measure fasting insulin and glucose. Another blood sample was taken 2 hours after the participant orally consumed a 100-g glucose load to obtain postchallenge insulin and glucose concentrations. Twenty-four-hour urine samples were collected at home by each participant and returned at the examination. Each participant completed a questionnaire that elicited information on urine collection times, missed collections, spilled urine, and medication use. It should be noted that there was a time lapse, approximately 2.44 ± 1.25 years, between the MMPI assessment (of which hostility scores were derived) and the laboratory testing (of which insulin resistance measures and 24-hour urine norepinephrine were obtained). This study has been approved by the Institutional Review Board and the Human Studies Subcommittee of the Research and Development Committee at the Boston VA Medical Center.

Measures
Fasting and Postchallenge Serum Insulin and Glucose
The two blood samples were analyzed to obtain fasting and postchallenge insulin and glucose values. Serum glucose concentration was measured in duplicate on an autoanalyzer by the hexokinase method (42). Serum insulin concentration was determined by a solid phase [125I]-radioimmunoassay (Diagnostic Products Corp., Los Angeles, CA). The intraassay and interassay coefficients of variation for insulin were 3% to 5% and 5% to 7%, respectively.

Index of Insulin Resistance
Two indices of insulin resistance were calculated based on each participant's fasting glucose and insulin levels. The HOMA was computed as: HOMA = glucose * insulin/22.5 (43). Because of its skewed distribution, a log transformation was performed and the transformed data (Ln-HOMA) were used in subsequent analyses. The QUICKI was computed as: QUICKI = 1/[log(insulin) + log(glucose)] (23). Note that Ln-HOMA and QUICKI are perfectly correlated at –1.0 because the former assesses insulin resistance and the latter measures insulin sensitivity. Therefore, Ln-HOMA, not QUICKI, was used as an index of insulin resistance in subsequent data analysis. The PCGL and PCIN were also used as indicators of insulin resistance and/or glucose intolerance.

Urinary Norepinephrine
Urinary NEPI levels were measured by high-performance liquid chromatography with electrochemical detection according to the method of Smedes et al. (44) as modified by MacDonald and Lake (45). The intraassay coefficient of variation for urine samples (corrected for recovery) was 4% to 6% and the interassay coefficient of variation was 6% to 7%. NEPI levels were adjusted for 24-hour urine volume and creatinine levels before being used in data analysis.

Body Mass Index and Waist-to-Hip Ratio
A series of anthropometric measurements were made with participants dressed in underwear and socks only. Measurements included height, measured against a wall chart to the nearest 0.25 cm; weight, measured on a balance beam scale to the nearest 0.45 kg; waist circumference, measured in centimeters at the level of the umbilicus; and hip circumference, measured in centimeters at the level of greatest protrusion of the buttocks. The indices were calculated from these measurements as follows: BMI = weight (kg)/height (m)2 and WHR = waist circumference/hip circumference. The BMI was a standard measure of obesity, whereas the WHR is more reflective of ventral obesity. Both variables were used as covariates in data analysis because previous studies have shown their association with metabolic syndrome in the NAS data (4,40).

Hypertension
Hypertension (HTN) was defined on the basis of diagnosis by the study physicians (board-certified internists), review of medical records, or antihypertensive medication use (46).

Health Behaviors
Behavioral risk factors assessed included alcohol and tobacco consumption, physical activity levels, and diet. Dietary data were obtained by means of a semiquantitative food frequency questionnaire (FFQ) (47), which was mailed to each participant and completed before the examination. The FFQ lists food items with serving sizes and elicits information on frequency of intake during the past year. Nutrient scores were computed by multiplying the frequency of intake by the nutrient content of the food items. Macronutrients examined in the present analyses were total caloric intake (kcal/day) and alcohol drinks per year. Information was also obtained on number of cigarettes currently smoked per week. Physical activity was assessed on a scale derived by Paffenbarger et al. (48). Responses to questions about the number of flights of stairs climbed per day, walking pace, and frequency of various sports activities were used to derive a continuous physical activity variable that assessed total kilocalories expended per week.

Hostility
Hostility was measured with the CMHOST scale (37) taken from the MMPI Form AX (49), which included items from both the MMPI and the MMPI-2. It is generally believed that the Ho scale measures an individual's cynicism, distrust, resentment, and trait anger (38,50). The full CMHOST scale score (Ho) was calculated based on all 50 items in the scale. Five subscale scores were calculated to measure different components of hostility, including four subscales described by Barefoot et al. (38), i.e., cynicism (13 items), hostile attribution (12 items), hostile affect (5 items), and aggressive responding (9 items) and another cynicism scale based on Costa et al. (24 items) (39).

Demographic Risk Factors
Age in years was assessed at the time of MMPI assessment. Education was divided into four categories: less than high school, high school graduate (including attainment of a general education diploma), some college or college graduate (2 years of technical school or 4 years of college), and some postgraduate or postgraduate education.

Data Analysis
Descriptive statistics of all variables were inspected for nonnormality and outliers. Log transformation was computed for skewed variables, including fasting and postchallenge insulin, NEPI, and HOMA. Hierarchical linear regressions were conducted for each of the three measures of insulin resistance: HOMA, PCGL, and PCIN, to examine the interaction between hostility and 24-hour urinary NEPI, controlling for age, education, smoking, alcohol use, caloric intake, BMI, WHR, HTN status, and physical activity. These covariates were selected based on their possible associations with insulin resistance and hostility shown in previous studies (40). Nonsignificant covariates were kept in the final model because model estimates more accurately reflect population values when conceptually important covariates are retained even if they are not statistically significant (51). Outliers with standardized residuals greater than 3.0 were dropped from the analysis. No more than three outliers were actually excluded in each regression analysis. Some variables such as postchallenge glucose and insulin levels had more missing data than others, but the amount of missing data was less than 5% of the total sample data. In each regression analysis, covariates were entered in the first step, and the main effects of hostility and NEPI were entered at the second step. All variables were centered before entering in regression analysis. The interaction term between hostility and NEPI, which was a product of the centered hostility and centered NEPI, was entered in the third step. If the interaction effect was significant, simple regression slopes were explored following the recommendations of Cohen and colleagues (52) to investigate how NEPI might moderate the relationship of hostility with insulin resistance. Specifically, simple slopes of hostility were calculated when NEPI took the value of the mean, one standard deviation (SD) above the mean, and one SD below the mean. These represent how hostility was related to insulin resistance at different levels of NEPI (i.e., average, higher, or lower levels). The computation and significance testing used a web-based R program at www.unc.edu/~preacher/interact/mlr2.htm created by Kristopher J. Preacher.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 METHOD
 RESULTS
DISCUSSION
 NOTES
 REFERENCES
 
Sample Characteristics
Table 1 presents demographic, behavioral, psychological, and physiological characteristics of the sample at the time of the MMPI administration in 1986. Study participants were 643 men with an average age of 60.60 years (SD = 7.55) ranging from 42 to 87. Demographically, almost two thirds of the sample had at least some college education. Notably, the average BMI (26.46 ± 3.43 kg/m2) is suggestive of a relatively overweight sample. Approximately 10.3% of the sample smoked cigarettes. An average person in the sample had more than one drink of alcohol per day. Table 2 shows the zero-order correlation coefficients between different measures of hostility and some of the major covariates, insulin resistance measures, and NEPI. Table 3 shows the intercorrelations among hostility measures and among insulin resistance measures. The six different measures of hostility were moderately to highly correlated with each other, with rs ranging from 0.37 to 0.94. The three measures of insulin resistance, HOMA, PCIN, and PCGL, were moderately correlated with each other, ranging from 0.39 to 0.55. From Table 1, Ho was not correlated with either PCIN or PCGL (rs = 0.01 and 0.07, respectively, ps > .05). However, Ho was significantly correlated with HOMA (r = 0.10, p < .05). Note that these are zero-order correlations. The following analyses will examine the relationship between Ho and measures of insulin resistance controlling for common covariates that are known to be associated with insulin resistance.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Descriptive Statistics for the Variables in the Study (N = 643)

 

View this table:
[in this window]
[in a new window]
  		TABLE 2. Zero-Order Correlation Coefficients Between Hostility Measures and Physiological Measures (N = 643)

 

View this table:
[in this window]
[in a new window]
  		TABLE 3. Zero-Order Correlation Coefficients Among Hostility Measures and Among Insulin Resistance Measures (N = 643)

 

Ho x Norepinephrine Interaction on Homeostatic Model Assessment Approach
Hierarchical multiple regression showed that covariates explained approximately 25% of total variance in HOMA (F [9, 623] = 20.26, p < .001). Significant covariates included WHR (ß = 0.17, p < .001), BMI (ß = 0.38, p < .001), and HTN (ß = 0.10, p < .01). The main effects of Ho and NEPI were not significant, but the interaction of Ho and NEPI was significant, explaining an additional 1% of variance in HOMA (F [1, 620] = 3.89, p < .05). Simple regression slopes of Ho were then explored using one SD above the mean, the mean, and one SD below the mean for NEPI. When NEPI was at one SD above the mean (i.e., higher level), the standardized regression coefficient for Ho was ß = 0.14 (t [620] = 2.35, p < .05). When NEPI was at the mean (i.e., medium level), ß for Ho was 0.06 (t [620] = 1.34, p > .10). When NEPI was at one SD below the mean (i.e., lower level), ß for Ho was –0.02 (t [620] = –0.30, p > .20). Figure 1 shows the differential regression slopes of Ho at lower, medium, and higher levels of NEPI. When individuals had higher levels of NEPI, higher Ho was associated with worse insulin resistance measured by HOMA, but this was not the case when individuals had either medium or lower levels of NEPI.


Figure 111
View larger version (10K):
[in this window]
[in a new window]
  		Figure 1. A plot of simple regression slopes used to interpret the hostility x norepinephrine interaction on homeostatic model assessment approach (z-score).

 

Ho x Norepinephrine Interaction on Postchallenge Insulin Level
The results of hierarchical multiple regression on PCIN were similar to those of HOMA. Covariates explained approximately 22% of total variance in PCIN (F [9, 586] = 15.53, p < .001). Significant covariates included WHR (ß = 0.20, p < .001), BMI (ß = 0.32, p < .001), and HTN (ß = 0.09, p < .05). The main effects of Ho and NEPI were not significant, but the interaction of Ho and NEPI was significant, explaining an additional 1% of variance in PCIN (F [1, 583] = 4.71, p < .05). Simple regression slopes for Ho were then explored. When NEPI was at one SD above the mean (i.e., higher level), the standardized regression coefficient for Ho was ß = 0.12 (t [583] = 2.48, p < .05). When NEPI was at the mean (i.e., medium level), ß for Ho was 0.05 (t [583] = 1.46, p > .10). When NEPI was at one SD below the mean (i.e., lower level), ß for Ho was –0.03 (t [583] = –0.49, p > .20). Figure 2 shows the differential regression slopes of Ho at lower, medium, and higher levels of NEPI, similar to those of Figure 1. When individuals had higher levels of NEPI, higher Ho was associated with worse insulin resistance measured by PCIN, but this was not the case when individuals had either medium or lower levels of NEPI.


Figure 211
View larger version (11K):
[in this window]
[in a new window]
  		Figure 2. A plot of simple regression slopes used to interpret the hostility x norepinephrine interaction on postchallenge insulin (z-score).

 

Ho x Norepinephrine Interaction on Postchallenge Glucose Level
Hierarchical multiple regression showed that covariates explained approximately 11% of total variance in PCGL (F [9, 586] = 7.25, p < .001). Significant covariates included WHR (ß = 0.10, p < .05), BMI (ß = 0.21, p < .001), age (ß = 0.14, p < .01), and HTN (ß = 0.15, p < .01). Neither the main effects of Ho or NEPI, nor the interaction of Ho and NEPI, was significant (all ps > .05). Therefore, simple regression slopes of Ho were not examined. Ho was not associated with PCGL, and the interaction of Ho and NEPI was not significant.

Other Measures of Hostility x Norepinephrine Interaction on Insulin Resistance Measures
As an exploratory analysis, we also examined which subscales/components of CMHOST were associated with insulin resistance in the same pattern as described previously for the full scale score. Four subscales from Barefoot et al. (38), including cynicism, hostile attribution, hostile affect, and aggressive responding and another version of cynicism from Costa et al. (39) were used. The analyses followed the same scheme as described previously. Each hostility subscale was used to predict each of the three insulin resistance measures. Therefore, an additional of 15 (5 x 3) hierarchical multiple regressions were conducted. As a result of limitation of space, a detailed report of the results of all 15 analyses is not possible. Instead, a summary of the results is shown in Table 4. None of the main effects of hostility measures were significant after controlling for covariates, although the zero-order correlations were significant between Ho and HOMA, between cynicism (Barefoot et al.) and PCIN, and between the aggressive responding subscale and PCIN. Similar to the Ho x NEPI interaction, the cynicism (Costa et al.) x NEPI interaction was also significant for HOMA and PCIN. The cynicism (Barefoot et al.) x NEPI interaction was significant for PCIN, but only marginally significant for HOMA. None of the interaction effects were significant for PCGL. The other three subscales, i.e., hostile attribution, hostile affect, and aggressive responding, did not significantly interact with NEPI in predicting any of the three insulin resistance measures. For the significant interaction effects between cynicism scales and NEPI, the same approach was followed to calculate the simple slopes of cynicism at higher, medium, and lower levels of NEPI. The patterns of the simple slopes were similar to Figures 1 and 2.


View this table:
[in this window]
[in a new window]
  		TABLE 4. Summary of the Main Effects of All Hostility Measures and Their Interaction Effects With Norepinephrine in Predicting Insulin Resistance Measures

 


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 METHOD
 RESULTS
 DISCUSSION
NOTES
 REFERENCES
 
The results reported here partially support the proposed hypotheses. Ho was positively associated with insulin resistance measured by HOMA, especially when participants had higher levels of 24-hour urinary NEPI, an indicator of sympathetic arousal and stress. Ho was positively associated with insulin resistance measured by PCIN only when participants had higher levels of 24-hour urinary NEPI. In contrast, Ho and PCGL levels were not related regardless of NEPI levels. It appears that Ho was differentially related to different measures of insulin resistance and that HOMA was especially sensitive to the relationship with Ho. The subsequent exploratory analysis showed that cynicism, but not other components of hostility, had essentially the same relationship with insulin resistance measures.

The finding that Ho was positively related to HOMA is consistent with other studies (18,24) in different populations. Although previous studies have found moderation effects of sex and race on the relationship between Ho and HOMA, to our knowledge, the present study is the first to report the moderation effect of 24-hour urinary NEPI, a putative stress indicator. These results are also consistent with an earlier study demonstrating that daily hassles and chronic stress moderated the relationship between hostility and fasting glucose and insulin levels (30), although this study did not measure insulin resistance. Therefore, higher hostility, combined with higher levels of sympathetic arousal, may be associated with increased insulin resistance, which may place individuals at higher risks of developing either NIDDM, coronary heart disease (CHD), or both. Based on the stress moderation model (29), when in a stressful environment, hostile individuals tend to be more likely to perceive threats, more likely to provoke hostile responses from other people as a result of their own interpersonal styles, and more likely to have exaggerated physiological responses to stress. With the release of stress hormones such as catecholamines and cortisol, the body is more likely to mobilize and release hepatic glucose into the bloodstream so that different organ systems can use the extra energy supply to cope with stress. Catecholamines, indicative of sympathetic arousal, can suppress the secretion of insulin from pancreatic beta cells (31,53). Cortisol can suppress the effect of insulin in regulating glucose (54), which leads to the phenomenon of higher glucose levels combined with higher insulin levels. Cortisol can also redistribute body fat to central locations and downregulate insulin receptors in adipose tissues, which over time may be associated with the development of NIDDM (10,53).

We found that Ho interacted with NEPI to significantly predict PCIN, but not PCGL. The main effect of Ho on either measure was not significant. The lack of a main effect of Ho is consistent with an earlier study (19), which did not find a significant relationship among Ho, perceived stress, and plasma insulin levels after a standard OGTT. However, these investigators did not examine interaction effects between Ho and stress on insulin resistance. The OGTT is a standard method to assess insulin resistance and diagnose NIDDM in clinical practice (55), but it has not been used extensively to examine its association with psychosocial factors. The present study suggests that post-OGTT challenge glucose is not sensitive enough of a measure of insulin resistance to demonstrate a significant association with Ho. In contrast, postchallenge insulin levels, as well as the HOMA index, do appear to be sensitive enough to detect an association with Ho, although potential moderators such as stress should be considered.

Another interesting finding in the present study was that cynicism, as measured by either the Barefoot et al. version or Costa et al. version, showed similar relationships with NEPI and insulin resistance measures. Other components of hostility such as hostile attribution, hostile affect, and aggressive responding were not associated with insulin resistance. This is not surprising because previous studies have shown that cynicism is the key component of hostility predicting higher risks of mortality (15,38,56), worsened insulin resistance (22,24), increased abdominal fat distribution (57), and poor health behavior (58). Another reason that other subscales of CMHOST was not related to insulin resistance in the same fashion as the cynicism and the full-scale Ho is that they had relatively low reliability. The alpha coefficients in the study were 0.63 for hostile attribution, 0.54 for hostile affect, and 0.52 for aggressive responding. In contrast, the alpha coefficients were 0.74 for cynicism (Barefoot et al.) and 0.83 for cynicism (Costa et al.). Low reliability of a measure tends to attenuate its correlation with other measures.

There are both advantages and limitations of the present study. We examined three different measures of insulin resistance, HOMA, postchallenge glucose, and insulin levels, and compared their sensitivity in terms of their associations with Ho. We followed a theoretical stress moderation model to study the effect of Ho as opposed to the simple main effect of Ho examined in many previous studies. We also explored which components of hostility were more important in relating to insulin resistance. In addition, the present study had a large sample size, which yielded more robust results compared with studies with smaller samples. Finally, we used a relatively more objective indicator of psychosocial stress (24-hour urinary samples of NEPI), which is exempt from self-report bias and the confounding effects of personality variables. Despite these advantages, several limitations should also be discussed. First, the nature of the sample (i.e., older males, primarily white, normal health status at baseline) may have obscured relationships between hostility and insulin resistance that may have been evident in a more general and representative population sample. Therefore, generalizability of our results to younger men, women, and ethnic minority populations is limited. Second, we did not include measures of cortisol, which may have assisted with the interpretation of pathways of how hostility is associated with insulin resistance. Third, hostility measures were derived at a different time point than the insulin resistance and 24-hour urine NEPI measures. Although hostility is a relatively stable personality trait over time, changes in hostility may be predictive of progression of certain pathophysiological processes (59). Fourth, the interactive prediction between NEPI and hostility on insulin resistance (e.g., HOMA) was statistically significant but not robust. Further work refining this relationship both conceptually and methodologically should seek to increase the amount of variance in insulin resistance accounted for by hostility and levels of stress. Finally, all three measure of insulin resistance are proxy measures. We were not able to perform the gold standard assessment of insulin resistance, the hyperinsulinemic euglycemic clamp, which is technically complex (17). The use of the clamp technique would have made the association between hostility and insulin resistance more definitive. More research is also needed to study the postchallenge glucose and insulin levels in association with psychosocial variables given their importance in clinical practice.

In summary, the present study found that the relationship between Ho and insulin resistance was moderated by 24-hour urinary NEPI. Higher Ho combined with higher NEPI was associated with greater insulin resistance, especially when measured by HOMA and postchallenge insulin levels. It appears that cynicism was the critical component of hostility in terms of its association with insulin resistance. Future research is needed to replicate these findings in other populations such as women, younger individuals, and individuals from diverse racial and ethnic status. Future research should also use more direct measures of insulin resistance such as the clamp technique to strengthen the findings and measures of adrenocortical activities through cortisol to further elucidate psychophysiological pathways from hostility to insulin resistance. If it is true that hostility combined with high stress reactivity leads to insulin resistance, and subsequently NIDDM and CHD, intervention programs could be designed to help individuals predisposed to hostile responding to better cope with stress. Potential protective factors should also be studied, such as social support (60), sense of coherence (61), and life satisfaction (62).

The VA Normative Aging Study (NAS) is supported by the Cooperative Studies Program/ERIC, U.S. Department of Veterans Affairs, and is a research component of the Massachusetts Veterans Epidemiology Research and Information Center (MAVERIC). Some of the data analyzed in this project were obtained with support provided by NIH grants HL37871 and AG02287 and by the Research Service of the U.S. Department of Veterans Affairs. Part of the data analysis and preparation of the manuscript is supported by a Purdue Research Foundation Summer Faculty Fellowship awarded to the first author.


   NOTES
TOP
 ABSTRACT
 INTRODUCTION
 METHOD
 RESULTS
 DISCUSSION
 NOTES
REFERENCES
 

Received for publication May 17, 2005; revision received April 27, 2006.

DOI:10.1097/01.psy.0000228343.89466.11


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 METHOD
 RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 

  1. Timar O, Sestier F, Levy E. Metabolic syndrome X: a review. Can J Cardiol 2000;16:779–89.[Medline]
  2. Vitaliano PP, Scanlan JM, Zhang J, Savage MV, Hirsch IB. A path model of chronic stress, the metabolic syndrome, and coronary heart disease. Psychosom Med 2002;64:418–35.[Abstract/Free Full Text]
  3. Niaura R, Todaro JF, Stroud L, Spiro A III, Ward KD, Weiss S. Hostility, the metabolic syndrome, and incident coronary heart disease. Health Psychol 2002;21:588–93.[CrossRef][Medline]
  4. Shen B-J, Todaro JF, Niaura R, McCaffery JM, Zhang J, Spiro A III, Ward KD. Are metabolic risk factors one unified syndrome? Modeling the structure of the metabolic syndrome X. Am J Epidemiol 2003;157:701–11.[Abstract/Free Full Text]
  5. Reaven GM, Lithell H, Landsberg L. Hypertension and associated metabolic abnormalities—the role of insulin resistance and the sympathoadrenal system. N Engl J Med 1996;334:374–82.[Free Full Text]
  6. Reaven GM. Insulin resistance, cardiovascular disease, and the metabolic syndrome: how well do the emperor's clothes fit? Diabetes Care 2004;27:1011–2.[Free Full Text]
  7. Scanlan JM, Vitaliano PP, Zhang J, Borson S, Dyck D. Insulin: the heart of the metabolic syndrome? The Fifth International Congress of Behavioral Medicine; Copenhagen, Denmark; 1998.
  8. Frayne KN. Adipose tissue and the insulin resistance syndrome. Proc Nutr Soc 2001;60:375–80.[Medline]
  9. Smith U. Impaired (‘diabetic') insulin signaling and action occur in fat cells long before glucose intolerance—is insulin resistance initiated in adipose tissue? Int J Obes Relat Metab Disord 2002;26:897–904.[CrossRef][Medline]
  10. Hjemdahl P. Stress and the metabolic syndrome: an interesting but enigmatic association. Circulation 2002;106:2634–6.[Free Full Text]
  11. Rozanski A, Blumenthal JA, Kaplan J. Impact of psychological factors on the pathogenesis of cardiovascular disease and implications for therapy. Circulation 1999;99:2192–217.[Abstract/Free Full Text]
  12. Surwit RS, van Tilburg MA, Zucker N, McCaskill CC, Parekh P, Feinglos MN, Edwards CL, Williams P, Lane JD. Stress management improves long-term glycemic control in type 2 diabetes. Diabetes Care 2002;25:30–4.[Abstract/Free Full Text]
  13. Smith TW. Concepts and methods in the study of anger, hostility, and health. In: Siegman AW, Smith TW, eds. Anger, Hostility and the Heart. Hillsdale, NJ: Lawrence Erlbaum Associates; 1994:23–42.
  14. Barefoot JC, Dahlstrom WG, Williams RB. CHD incidence and total mortality: a 25 year follow-up study of 255 physicians. Psychosom Med 1983;42:59–64.
  15. Miller TQ, Smith TW, Turner CW, Guijarro ML, Hallet AJ. A meta-analytic review of research on hostility and physical health. Psychol Bull 1996;119:322–48.[CrossRef][Medline]
  16. Todaro JF, Con A, Niaura R, Spiro A, Ward K, Roytberg A. Combined effects of the metabolic syndrome and hostility on the incidence of CHD: the Normative Aging Study. Am J Cardiol. In press.
  17. Radziuk J. Insulin sensitivity and its measurement: structural commonalities among the methods. J Clin Endocrinol Metab 2000;85:4426–33.[Abstract/Free Full Text]
  18. Surwit RS, Williams RB, Siegler IC, Lane JD, Helms M, Applegate KL, Zucker N, Feinglos MN, McCaskill CM, Barefoot JC. Hostility, race, and glucose metabolism in nondiabetic individuals. Diabetes Care 2002;25:835–9.[Abstract/Free Full Text]
  19. Philip P, Facchini F. Self-perceived stress, hostility, and insulin resistance: lack of association. Diabetes Care 1995;18:1198–9.[Medline]
  20. Raikkonen K, Matthews KA, Kuller LH, Reiber C, Bunker CH. Anger, hostility, and visceral adipose tissue in healthy postmenopausal women. Metabolism 1999;48:1146–51.[CrossRef][Medline]
  21. Kawakami N, Akari S, Ohtsu H, Hayashi T, Masumoto T, Yokohama K. Effects of mood states, smoking and urinary catecholamine excretion on hemoglobin A1c in male Japanese workers. Ind Health 1995;33:153–62.[Medline]
  22. Nelson TL, Palmer RF, Pederson NL. The metabolic syndrome mediates the relationship between cynical hostility and cardiovascular disease. Exp Aging Res 2004;30:163–77.[CrossRef][Medline]
  23. Katz A, Nambi SS, Mather K, Baron AD, Follmann DA, Sullivan G, Quon MJ. Quantitative insulin sensitivity check index: a simple, accurate method for assessing insulin sensitivity in humans. J Clin Endocrinol Metab 2000;85:2402–10.[Abstract/Free Full Text]
  24. Raikkonen K, Matthews KA, Salomon K. Hostility predicts metabolic syndrome risk factors in children and adolescents. Health Psychol 2003;22:279–86.[CrossRef][Medline]
  25. Bonora E, Targher G, Alberiche M, Bonadonna RC, Saggiani F, Zenere MB, Monauni T, Muggeo M. Homeostasis model assessment closely mirrors the glucose clamp technique in the assessment of insulin sensitivity: studies in subjects with various degrees of glucose tolerance and insulin sensitivity. Diabetes Care 2000;23:57–63.[Abstract]
  26. Emoto M, Nishizawa Y, Maekawa K, Hiura Y, Kanda H, Kawagishi T, Shoji T, Okuno Y, Morii H. Homeostasis model assessment as a clinical index of insulin resistance in type 2 diabetic patients treated with sulfonylureas. Diabetes Care 1999;22:818–22.[Abstract/Free Full Text]
  27. Mari A, Pacini G, Brazzale AR, Ahren B. Comparative evaluation of simple insulin sensitivity methods based on the oral glucose tolerance test. Diabetologia 2005;48:748–51.[CrossRef][Medline]
  28. Leonetti F, Iacobellis G, Zappaterreno A, Ribaudo MC, Tiberti C, Vecci E, Di Mario U. Insulin sensitivity assessment in uncomplicated obese women: comparison of indices from fasting and oral glucose load with euglycemic hyperinsulinemic clamp. Nutr Metab Cardiovasc Dis 2004;14:366–72.[CrossRef][Medline]
  29. Smith TW, Gallo LC. Personality traits as risk factors for physical illness. In: Baum A, Revenson TA, Singer JE, eds. Handbook of Health Psychology. Mahwah, NJ: Lawrence Erlbaum Associates; 2001:139–74.
  30. Vitaliano PP, Scanlan JM, Krenz C, Fujimoto W. Insulin and glucose: relationships with hassles, anger, and hostility in nondiabetic older adults. Psychosom Med 1996;58:489–99.[Abstract/Free Full Text]
  31. Ward KD, Sparrow D, Landsberg L, Young JB, Vokonas PS, Weiss ST. Influence of insulin, sympathetic nervous system activity, and obesity on blood pressure: the Normative Aging Study. J Hypertens 1996;14:301–8.[Medline]
  32. Dimsdale JE, Ziegler MG. What do plasma and urinary measures of catecholamine tell us about human responses to stressors? Circulation 1991;83:II36–42.
  33. Steptoe A. The assessment of sympathetic nervous function in human stress research. J Psychosom Res 1987;31:141–52.[CrossRef][Medline]
  34. Young JB, Rosa RM, Landsberg L. Dissociation of sympathetic nervous system and adrenal medullary responses. Am J Physiol 1984;247:E35–40.
  35. Daly PA, Landsberg L. Hypertension in obesity and NIDDM: role of insulin and sympathetic nervous system. Diabetes Care 1991;14:240–8.[Abstract]
  36. Watson D, Clark LA. Negative affectivity: the disposition to experience aversive emotional states. Psychol Bull 1984;96:465–90.[CrossRef][Medline]
  37. Cook WW, Medley DM. Proposed hostility and pharisaic-virtue scales from the MMPI. J Appl Psychol 1954;38:414–8.[CrossRef]
  38. Barefoot JC, Dodge KA, Peterson BL, Dahlstrom WG, Williams RB. The Cook-Medley hostility scale: item content and ability to predict survival. Psychosom Med 1989;51:46–57.[Abstract/Free Full Text]
  39. Costa PT Jr, Zonderman AB, McCrae RR, Williams RB Jr. Cynicism and paranoid alienation in the Cook and Medley HO scale. Psychosom Med 1986;48:283–5.[Abstract/Free Full Text]
  40. Niaura R, Banks SM, Ward KD, Stoney CM, Spiro A III, Aldwin CM, Landsberg L, Weiss ST. Hostility and the metabolic syndrome in older males: the normative aging study. Psychosom Med 2000;62:7–16.[Abstract/Free Full Text]
  41. Bosse R, Ekedert DJ, Silbert JE. The Vetrans Administration Normative Aging Study. In: Mednick SA, Finello KM, eds. Handbook of Longitudinal Research, vol 2. Teenage and Adult Cohorts, vol 2. New York: Preager; 1984:273–93.
  42. Leon LP, Chu DK, Stasiw RP. New, More Specific Methods for the SMA 12/60 Multichannel Biochemical Analyzer: Advances in Automated Analysis. Tarrytown, NY: Mediad; 1977.
  43. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28:412–9.[CrossRef][Medline]
  44. Smedes F, Kraak JC, Poppe H. Simple and fast solvent extraction system for selective and quantitative isolation of adrenaline, noradrenaline, and dopamine from plasma and urine. J Chromatogr 1982;231:25–39.[Medline]
  45. MacDonald IA, Lake DM. An improved technique for extracting catecholamines from body fluids. J Neurosci Methods 1985;13:239–48.[CrossRef][Medline]
  46. Zhang J, Niaura R, Todaro JF, McCaffery JM, Shen BJ, Spiro A 3rd, Ward KD. Suppressed hostility predicted hypertension incidence among middle-aged men: the Normative Aging Study. J Behav Med 2005:1–12.
  47. Willett WC, Sampson L, Stampfer MJ, Rosner B, Bain C, Witschi J. Reproducibility and validity of a semi-quantitative food frequency questionnaire. Am J Epidemiol 1985;122:51–65.[Abstract/Free Full Text]
  48. Paffenbarger RS Jr, Hyde RT, Wing AL, Hsieh CC. Physical activity, all-cause mortality, and longevity of college alumni. N Engl J Med 1986;314:605–13.[Abstract]
  49. Butcher JN, Dahlstrom WG, Grahm JR, Tellegen A, Kaemmer B. MMPI-2: Minnesota Multiphasic Personality Inverntory-2. Manual for Administration and Scoring. Minneapolis: University of Minnesota; 1989.
  50. Barefoot JC. Developments in the measurement of hostility. In: Friedman HS, ed. Hostility, Coping, and Health. Washington, DC: American Psychological Association; 1992:43–66.
  51. Steyerberg EW, Eijkemans MJ, Harrell FE Jr, Habbema JD. Prognostic modeling with logistic regression analysis: in search of a sensible strategy in small data sets. Med Decis Making 2001;21:45–56.[Abstract/Free Full Text]
  52. Cohen J, Cohen P, West SG, Aiken LS. Applied Multiple Regression Analysis for the Behavioral Sciences. Mahwah, NJ: Lawrence Erlbaum Associates; 2003.
  53. Brunner EJ, Hemingway H, Walker BR, Page M, Clarke P, Juneja M, Shipley MJ, Kumari M, Andrew R, Seckl JR, Papadopoulos A, Checkley S, Rumley A, Lowe GDO, Stansfeld SA, Marmot MG. Adrenocortical, autonomic, and inflammatory causes of the metabolic syndrome: nested case-control study. Circulation 2002;106:2659–65.[Abstract/Free Full Text]
  54. Rosmond R. Role of stress in the pathogenesis of the metabolic syndrome. Psychoneuroendocrinology 2005;30:1–10.[CrossRef][Medline]
  55. American Diabetes Association. Standards of medical care in diabetes. Diabetes Care 2005;28:S4–36.[Free Full Text]
  56. Almada SJ, Zonderman AB, Shekelle RB, Dyer AR, Daviglus ML, Costa PT Jr, Stamler J. Neuroticism and cynicism and risk of death in middle-aged men: the Western Electric Study. Psychosom Med 1991;53:165–75.[Abstract/Free Full Text]
  57. Nelson TL, Palmer RF, Pedersen NL, Miles TP. Psychological and behavioral predictors of body fat distribution: age and gender effects. Obes Res 1999;7:199–207.[Medline]
  58. Leiker M, Hailey BJ. A link between hostility and disease: poor health habits? Behav Med 1988;14:129–33.[Medline]
  59. Siegler IC, Costa PT, Brummett BH, Helms MJ, Barefoot JC, Williams RB, Dahlstrom WG, Kaplan BH, Vitaliano PP, Nichaman MZ, Day RS, Rimer BK. Patterns of change in hostility from college to midlife in the UNC Alumni Heart Study predict high-risk status. Psychosom Med 2003;65:738–45.[Abstract/Free Full Text]
  60. Vitaliano PP, Scanlan JM, Zhang J, Savage MV, Barefoot JC, Brummett B, Siegler I. Are the salutogenic effects of social supports modified by income? A test of an ‘added value hypothesis.' Health Psychol 2001;20:155–65.[CrossRef][Medline]
  61. Zhang J, Vitaliano PP, Lutgendorf SK, Scanlan JM, Savage MV. Sense of coherence buffers relationships of chronic stress with fasting glucose levels. J Behav Med 2001;24:33–55.[CrossRef][Medline]
  62. Lyubomirsky S, King L, Diener E. The benefits of frequent positive affect: does happiness lead to success? Psychol Bull. In press.



This article has been cited by other articles:


Home page
 Psychosom. Med.Home page
A. Georgiades, J. D. Lane, S. H. Boyle, B. H. Brummett, J. C. Barefoot, C. M. Kuhn, M. N. Feinglos, R. B. Williams, R. Merwin, S. Minda, et al.
Hostility and Fasting Glucose in African American Women
Psychosom Med, July 1, 2009; 71(6): 642 - 645.
[Abstract] [Full Text] [PDF]

Home page
 Psychosom. Med.Home page
R. S. Surwit, J. D. Lane, D. S. Millington, H. Zhang, M. N. Feinglos, S. Minda, R. Merwin, C. M. Kuhn, R. C. Boston, and A. Georgiades
Hostility and Minimal Model of Glucose Kinetics in African American Women
Psychosom Med, July 1, 2009; 71(6): 646 - 651.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zhang, J.
Right arrow Articles by Ward, K. D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhang, J.
Right arrow Articles by Ward, K. D.
Related Collections
Right arrow Endocrinology
Right arrow Personality
Right arrow Stress and Coping
Right arrow Other Cardiovascular Medicine

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS