| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Department of Psychology (M.C.D., C.M.), Arizona State University, Tempe, Arizona; and Department of Psychiatry (K.A.M.), University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania.
Address reprint requests to: Mary C. Davis, PhD, Department of Psychology, Box 871104, Arizona State University, Tempe, AZ 85287-1104. Email: mary.davis{at}asu.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
METHODS: Eighty undergraduate men and women were categorized as high or low in hostility on the basis of median splits of Cook-Medley Hostility Scale scores. Participants discussed a controversial topic with a confederate who disagreed with them, and hemodynamic responses were assessed with impedance cardiography.
RESULTS: High hostile individuals exhibited greater increases in diastolic blood pressure and total peripheral resistance and smaller increases in cardiac output during an interpersonal stressor than did low hostile individuals. Systolic blood pressure and heart rate increases were greater among high hostile relative to low hostile females and comparable among low and high hostile males. Affective responses and task perceptions were generally similar for high and low hostile participants, but the relationship between task perception and hemodynamic responses varied on the basis of hostility level.
CONCLUSIONS: These findings suggest that hostility in both men and women is associated with heightened vascular and dampened cardiac responsivity to interpersonal stress that is not deliberately anger provoking. Moreover, they indicate that the associations between task perception and hemodynamic responses vary between high and low hostile individuals.
Key Words: hostility • interpersonalstress • sex • cardiovascular reactivity.
Abbreviations: ANCOVA = analysis of covariance; ANOVA = analysis ofvariance; BP = blood pressure; CHD = coronary heart disease; CO = cardiac output; DBP = diastolic blood pressure; Ho= Cook-Medley Hostility Scale score; HR = heart rate; SBP =systolic blood pressure; TPR = total peripheral resistance.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Numerous empirical reports have documented that cynical hostility is positively related to heightened BP adjustments to behavioral challenge (eg, Refs. 9–11). However, hostile individuals are not overly reactive to all types of events; rather, they seem especially responsive to social situations (see Ref. 11 for a review). The types of social stressors selected for study have varied to some degree, but the majority have included anger provocation, often involving overt harassment of the participant by a stranger. The provocation typically elicits increased negative affect in both hostile and nonhostile individuals but heightened BP reactivity only in those who are hostile. This pattern of findings has held for both men (9, 10, 12) and women (12–14). In addition, high hostile men exhibit decreased forearm vascular resistance and increased blood flow in response to harassment (see Refs. 9; cf. Ref. 15), a pattern suggesting that heightened myocardial responsivity accounts for the BP elevations.
Less empirical attention has been directed toward understanding the relationship between hostility and cardiovascular arousal in stressful social contexts that are not deliberately provocative. Examining the physiological responses of hostile individuals to situations other than harassment is important because the majority of daily experiences are not marked by deliberate provocation, even among those who are hostile (eg, Ref. 16). In that regard, Smith et al. have demonstrated that hostility is related to elevated BP responses in men when they engage in conflictual role playing (17), arguing from an assigned position (18), and self-disclosure of personal information to a confederate (19). Because the strongest and most consistent physiological correlate of hostility was DBP, with no HR effects (17, 19), the authors postulated that the enhanced BP increases were attributable to increased peripheral resistance (19). Indeed, subsequent work demonstrated that men high in hostility who also reported that they expressed their anger outwardly tended to show smaller CO increases and larger TPR increases to a public speaking task than did their low hostile counterparts, with no differences based on hostility among men who did not express their anger outwardly or during an asocial task (20).
Thus, the limited data available suggest that hostile individuals respond to social stressors with enhanced BP responses, but the hemodynamic response patterns underlying BP elevations may depend on the presence of provocation. Anger-provoking interactions seem to enhance myocardial activity, whereas more ambiguous social interactions seem to enhance vascular responding. Williams et al. (21) were among the first to suggest that perception of the demands of the task is a key determinant of physiological response patterns and also proposed that two qualitatively distinct responses patterns are evident. The defense or "fight/flight" response is characterized by enhanced CO and occurs when an individual senses danger or the need for continuing mental effort, conditions likely to exist in social situations involving harassment. The vigilance response, on the other hand, is characterized by increased peripheral vasoconstriction and occurs when an individual feels uncertain and needs to be watchful of the environment, conditions likely to exist in ambiguous social interactions.
The literature is limited in at least three respects. First, the majority of investigations have included only measures of BP and HR in the study of ambiguous social contact. As a result, little is known about the hemodynamic response patterns underlying the stress-related BP increases among high and low hostile individuals during social stress that lacks deliberate provocation. A second concern is that the majority of research exploring the association between hostility and reactivity is based on male samples. Among the exceptions to this trend are two studies reporting that high hostile women showed greater BP increases during stress than low hostile women (13, 14). Several additional investigations have included both men and women, which permits evaluation of the extent to which the association between hostility and reactivity varies by sex. The results yielded by these studies have been inconsistent. Some suggest that hostility is associated with enhanced BP stress responses in both sexes (22, 23), in men but not women (12, 24, 25) or in neither sex (26, 27). Of note, no investigation has reported that the effects of hostility are more pronounced in women than in men. More comprehensive assessment of cardiovascular reactivity, including measures of TPR and CO, in a sample that includes both sexes would permit determination of the nature and extent of sex differences in the hostility-reactivity association. Finally, some findings point to the likely importance of task perception and affective arousal as determinants of the association between hostility and physiological responses to mild social stress (eg, Ref. 19), but these data are few and were obtained primarily from men.
The primary purpose of the present investigation was to examine the association between hostility and hemodynamic patterns underlying BP increases during a social stressor without deliberate anger provocation. Consistent with the only other investigation examining hemodynamic responses to social stress in a nonprovocative context (20), we hypothesized that individuals high in hostility would exhibit more marked increases in TPR and smaller increases in CO in response to the stress relative to low hostile individuals. In addition, we tested whether the relationship between hostility and physiological responses to interpersonal stress were similar for women and men. Finally, we explored the association between affective arousal and task perception, and hemodynamic response patterns.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Data from two female participants were discarded before analyses, one because of equipment failure and the second because of a marked cardiac arrhythmia that distorted the impedance waveform. Additionally, two male subjects did not complete the hostility measure. Therefore, statistical analyses were based on data from 20 participants in each sex by hostility group.
Psychological Measures
Hostility.
The Cook-Medley Hostility Scale (2) consists of 50 items that assess suspiciousness, resentment, and cynical mistrust. Each item is rated on a four-point scale (from "strongly agree" = 4 to "strongly disagree" = 1), resulting in a possible range of scores from 50 to 200. Scores on this measure (Ho scores) were normally distributed and were similar for males (mean = 118.4, SD = 14.0) and females (mean = 115.3, SD = 16.2) (t(78) = 0.90, p = .37). These values are akin to those reported in other studies of college undergraduates (eg, Ref. 23). Individuals were identified as high or low hostile by median splits, performed separately for men and women to maintain equal cell size. A 2 x 2 (sex by hostility group) analysis of variance did not yield a significant sex by hostility group interaction, indicating that males and females within each hostility group had similar Ho scores.
Affect and task perception.
A questionnaire completed immediately after the baseline rest period and again after the task asked participants to indicate the extent to which they felt anxious or angry. The questionnaire comprised three items assessing anger and four items assessing anxiety, each rated on a four-point scale (ranging from "not at all" = 1 to "very much so" = 4). Both the anger and anxiety items were internally consistent (Cronbach’s 
values = 0.74) and were summed to yield separate scores for anger and anxiety at both baseline and immediately after task completion.
After the task, participants completed a questionnaire assessing their perception of the task and of their own performance. Items were rated on a five-point scale (ranging from "not at all" = 1 to "very much so" = 5). Three items were summed to yield a score labeled "effort," which reflected the extent to which participants reported that they tried hard and found the task difficult and challenging (Cronbach’s = 0.61). An additional six items assessing the extent to which participants felt the task required them to be persuasive, competitive, dominant, and aggressive and to show a lack of both attunement and understanding were summed to generate a score termed "interpersonal control" (Cronbach’s = 0.87).
Procedure
To determine eligibility, individuals attended a screening session during which they provided information regarding their health and completed psychological measures, including the Cook-Medley Hostility Scale. They were subsequently contacted by telephone, told that they were being recruited to participate in a study examining heart functioning during speaking, and scheduled for a laboratory session. Participants were instructed to abstain from caffeine and exercise for 15 hours, and from medication and alcohol for 3 days, before their sessions. Female participants were scheduled to participate during the follicular phase of their menstrual cycles, between days 4 and 11, to control for the effects of menstrual cycle phase on hemodynamic stress responses (30).
When participants arrived for a session, they found a same-sex confederate sitting quietly in a reception area. A female experimenter escorted the participant to the laboratory, briefly described the protocol, and obtained written informed consent. All participants were told that the confederate was undergoing an identical procedure in an adjacent laboratory. HR and impedance electrodes were positioned, a BP cuff was placed on the nondominant arm, and the participant was seated in a comfortable chair.
Physiological measures began with a 10-minute baseline period, during which participants relaxed while sitting quietly. After the prestress baseline period, participants completed the mood questionnaire and then were instructed that they would engage for 6 minutes in a videotaped discussion with the confederate regarding capital punishment. Participants were told that they were matched with the confederate because they held opposite positions on this topic and were randomly assigned either to persuade the confederate that their own view was correct or to empathize with the confederate’s view of capital punishment.
Participants were told that their effectiveness would be evaluated immediately after the task by the confederate and later by raters watching the videotape. They were instructed that the confederate knew he or she would be engaging in a discussion about capital punishment with another individual but did not know the purpose of the task. Confederates were trained to maintain a neutral tone and demeanor throughout the discussion while establishing a position on capital punishment opposite of the position held by the participant. They were explicitly instructed to avoid provoking the participant and to respond to comments made by the participant with calmly delivered statements regarding their own position. Confederates remained resolute in their position on capital punishment throughout the discussion.
After instructions were given to the participant, the confederate was escorted into the experimental room wearing band electrodes. A BP cuff was placed on the confederate’s arm, and bogus leads were attached to the electrodes. Directions regarding the discussion were given, and discussants were reminded that they would be videotaped for later evaluation. The task began with a 1.5-minute preparation period, during which the discussants were allowed to prepare for the subsequent discussion. After the preparation period, participants engaged in 4.5 minutes of structured talk with the confederate. The participant and confederate alternated speaking and listening for three 45-second periods, with participants initiating the discussion when signaled to begin the task. At the conclusion of the structured talk period, participants and confederates continued the discussion for 1.5 additional minutes with no constraints regarding taking turns.
After completing the discussion, the participant and the confederate filled out questionnaires regarding their mood and impressions of the task. The confederate was excused, and the participant relaxed for a final 10-minute period. After the rest period, the participant was debriefed, given a credit slip or $10, and excused.
Physiological Measures
SBP and DBP were measured with an IBS (model SD-700A) automated BP monitor using a standard inflatable BP cuff placed on the participant’s nondominant arm. The IBS assesses BP using the auscultatory method and is equipped to detect artifact caused by movement or poor cuff placement and correlates highly (r > 0.90) with BP measured with a mercury sphygmomanometer. Impedance cardiography, performed with the Minnesota Impedance Cardiograph (model 304B) in conjunction with a tetrapolar configuration,was used to measure cardiac performance. Disposable Mylar band electrodes (M6001, Contact Products, Inc., Dallas, TX) were applied with one voltage electrode encircling the base of the neck and one encircling the thorax over the tip of the xiphisternal junction. One current electrode was placed around the neck and one was placed around the thorax, at a minimum distance of 3 cm above and below the upper and lower voltage electrodes, respectively.
HR was detected using silver or silver chloride electrodes placed on the right arm and abdomen below the impedance electrodes, and a ground electrode was placed beside the navel. The electrocardiogram and impedance cardiogram were processed by using the Cardiac Output Program, an online computerized system developed by Bio-Impedance Technology (distributed by Instrumentation for Medicine, Inc., Greenwich, CT). A continuous sample of impedance waveforms was processed to generate an ensemble-averaged cardiac cycle for each 60-second unit of interest during baseline and recovery periods and each 45-second unit of interest during the task periods. The impedance cardiogram was used to derive measures of CO and HR. TPR was calculated by the Cardiac Output Program from a formula adapted from Guyton (31): 
|
|
Data Reduction
BP was monitored at minutes 5, 7, and 9 of the baseline and recovery periods, once during the 45-second preparation period, and once during each of the three 45-second speaking turns of the participant. During the task and rest periods throughout the session, HR and impedance measures were monitored continuously. Using the Cardiac Output Program ensemble averaging technique, HR and CO values were averaged across each of the last 5 minutes of the 10-minute baseline and recovery periods and across each 45-second period of each phase of the discussion task. The BP readings were matched with appropriate values for CO to derive estimates of TPR for the last 5 minutes of the baseline and recovery periods and during the discussion task. For the present study, we examined hemodynamic stress responses during the preparation and structured talk phases of the task, because these periods were highly similar across the persuasion and empathy task conditions. Furthermore, task condition was not included as a variable of interest in the current analyses.2
Data collected during the last 5 minutes of the baseline and recovery periods were used to compute prestress baseline and poststress recovery means, respectively, for each physiological variable. Mean values were computed for the task periods for each physiological variable by averaging values collected within the preparation and structured talk periods. Mean change from baseline during the task and recovery periods for all physiological measures were calculated by subtracting the mean level at prestress baseline from the mean level during the task and recovery periods (32).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
|
Blood pressure.
Analyses for SBP, covarying baseline, revealed a significant sex by hostility group by trial interaction (see Table 1). Inspection of the adjusted means revealed that the hostility effects were apparent only among men during the talk period. High hostile men showed smaller increases in SBP during the talk period than did their low hostile counterparts (mean values = 26.50 vs. 32.72 mm Hg, respectively; p < .05) but showed comparable changes during the preparation (means = 19.93 vs. 18.84 mm Hg, respectively) and recovery periods (mean = 4.37 vs. 4.44 mm Hg, respectively; p > .05). In contrast, high and low hostile women showed comparable changes during the preparation period (mean = 16.76 vs. 14.20 mm Hg, respectively), talk period (mean = 29.35 vs. 24.52 mm Hg, respectively), and recovery period (mean = 2.44 vs. 2.27 mm Hg, respectively) (p > .05).
Hostility group also affected DBP stress responses. The analyses yielded a significant main effect for hostility and a significant hostility by trial interaction. Follow-up comparisons of adjusted means revealed that high hostile relative to low hostile individuals exhibited greater DBP increases during the preparation and talk periods (p < .05) but not the recovery period (p > .05) (see Table 1 and Figure 1). Additionally, as reported previously (28), results revealed a main effect for sex (F(1,75) = 5.62, p = .02) and a significant sex by trial interaction (F(2,152) = 5.74, p = .004). Examination of adjusted means indicated that DBP reactivity was more pronounced in women relative to men during the preparation period (mean = 11.35 vs. 5.22 mm Hg, respectively; p < .05) but not during the talk (mean = 18.46 vs. 14.97 mm Hg, respectively) or recovery period (mean = 2.65 vs. 2.96 mm Hg, respectively) (p > .05).
|
|
|
No other effects involving sex or hostility emerged in the analyses.
Affect and Task Perception Measures
Sex by hostility ANOVAs of baseline anger and anxiety scores yielded no significant effects, indicating that groups reported low and comparable levels of anger and anxiety at rest. Anger and anxiety change scores served as dependent variables in subsequent 2 x 2 (sex by hostility) ANCOVAs, controlling for baseline anger or anxiety. Results yielded a significant interaction between sex and hostility for anxiety (F(1,75) = 5.43, p = .02). Post hoc comparisons within each sex revealed only that hostile women tended to report greater increases in anxiety than did nonhostile women (p < .07), with no difference between hostile and nonhostile men. In the current study, the effect size reflecting pre- to posttask change in anger was approximately 0.33. In contrast, studies comparing the anger levels of harassed vs. nonharassed groups yield much larger effect sizes, typically >1.0 (eg, 14). The small anger effect that we observed, particularly when compared with those observed in studies of harassment, suggests that the interaction task did not elicit substantial anger in participants. With regard to task perception, sex by hostility ANOVAs of posttask scores yielded no main effects or interaction terms. Thus, the extent to which participants felt angry, effortful, and a need to exert interpersonal control over the interaction did not vary by sex or hostility group.
Internal Analyses
We were interested in evaluating whether task perception or affective arousal was associated with different response patterns in hostile and nonhostile individuals. Individuals were first categorized separately by hostility group as high or low in effort, need for interpersonal control, or change in negative affect based on median splits of these scores. Next, a series of 2 x 2 x 2 x 2 (sex by hostility group by perception group by trial, ie, preparation or talk) repeated-measures ANOVAs was conducted, with TPR and CO change scores serving as dependent variables.
Task perception.
Analyses examining effort ratings yielded significant main effects of effort for both CO and TPR (F(1,72) > 4.83, p < .03) and significant interactions between hostility and effort (F(1,72) > 5.76, p < .02). Follow-up comparisons indicated that among low hostile individuals, those high in effort showed enhanced CO and dampened TPR responses during the task relative to those low in effort (p < .05). In contrast, effort ratings were unrelated to CO and TPR responses among high hostile individuals.
With regard to need for interpersonal control, findings revealed a significant interaction between hostility and control perception for TPR (F(1,72) = 4.94, p = .03). Post hoc comparisons showed that among hostile individuals, those who perceived a high need to control the confederate exhibited greater TPR increases relative to those who perceived a low need for interpersonal control (p < .05). Among low hostile individuals, those high in control perception showed dampened TPR responses relative to those low in control perception (p < .05). No other main effects or interaction terms involving control perception emerged in the analyses.
Affective arousal.
None of the main effects or interaction terms involving anger or anxiety groups achieved significance, indicating that the relationships between affect arousal and CO and TPR responses were similar for high and low hostility groups.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
What can account for the heightened vascular stress responses of hostile relative to nonhostile individuals? Some investigators have posited that the relationship between hostility and reactivity is mediated by arousal of negative affect (10). For example, high levels of hostility are associated with more self-reported anger after conflictual social interactions (15) and frustrating cognitive tasks (22). Moreover, arousal of negative affect among hostile men (9, 10) and women (14) is positively related to BP and HR responses during harassment, a relationship that is not evident among low hostile individuals. However, in the current study and others including nonprovocative social stress (eg, Refs. 17), hostile and nonhostile individuals generally reported comparable levels of low anger and moderate anxiety in response to the social task and showed comparable associations between physiological and affective arousal. Thus, negative affect arousal may be likely to account for the hostility-reactivity association only in social contexts involving extreme provocation.
The current findings suggest that in less provocative, more ambiguous social contexts, subtle perceptions of the situational demands may determine responsivity. Hostile and nonhostile participants perceived the interaction with an unchangeable confederate similarly in terms of the degree of effort and the amount of interpersonal control required to perform well. However, the associations between task perception and physiological responding varied by hostility level. Greater perceived effort was related to heightened myocardial responsivity in low hostile individuals, and a greater perceived need to control was related to heightened vascular responsivity in high hostile individuals. Stated differently, nonhostile individuals appear to be more physiologically responsive to effort perceptions, and hostile individuals to control perceptions, during mild interpersonal stress. Thus, hostile individuals did not exhibit a propensity to perceive that they needed to control the interaction but did show a propensity to respond to a control perception with heightened vascular responsivity relative to their nonhostile counterparts. Christensen and Smith (19) reported compatible findings in their study of self-disclosure in men. High and low hostile men did not differ in how open they intended to be with a confederate regarding a stressful event they had experienced; however, the association between intent to be open and BP reactivity was positive for high hostile and negative for low hostile men (19). Together, these findings suggest that even when high and low hostile individuals report that their perceptions of and intentions to behave in uncertain social circumstances are similar, their perceptions and intentions are differentially related to physiological responses. Future research that includes more detailed assessment and/or manipulation of participants’ perception of social stressors and evaluation of neuroendocrine parameters may elaborate the extent to which the processes linking perception and physiological reactivity vary by hostility.
The present findings also showed that hostility was related to stress-related changes in cardiovascular measures in both sexes. Hostility was associated with enhanced DBP in both men and women, consistent with earlier findings (eg, Refs. 22). Moreover, hostile men and women showed elevated TPR and dampened CO changes underlying the BP increases relative to their nonhostile counterparts. To our knowledge, this is the first study to document that the hemodynamic responses underlying BP increases to social stress are generally comparable among hostile men and women.
Although some investigations have noted hostility effects in men but not women (eg, Refs. 12, 24), no study that has included both sexes has reported that reactivity is enhanced among hostile women but not among hostile men. Together, the data in this literature point to the possibility that hostile women respond with exaggerated physiological reactivity to a narrower range of social stimuli than do hostile men. For example, hostile individuals of both sexes respond excessively to stressful interpersonal interactions with a stranger (eg, Refs. 13), but only hostile men show enhanced responses to challenging discussions with their spouses (25). Thus, hostile women may make finer distinctions than do hostile men between the threats posed by stimulus situations, distinctions reflected in their more variable cardiovascular arousal. Recent data gleaned from ambulatory assessments point to this possibility. Hostility was associated with higher levels of DBP in men but not women when examining BP levels assessed throughout the entire work or school day (34, 35). However, hostility was related to elevations in SBP in both men and women during periods of social interaction (34). The possibility that the sexes differ in terms of the range of stimuli that elicits reactivity remains speculative, however, until additional research comparing response patterns in hostile men and women broadens to include a variety of social interactions with strangers, acquaintances, and familiar others.
Caution is warranted in interpreting the present findings, particularly with regard to generalizability. We examined a sample of white undergraduates who were originally recruited because they endorsed either an instrumental or expressive gender role orientation. Although instrumentality and expressivity scores in this sample were not extreme and were unrelated to hostility, the current results may not generalize to other ethnic, racial, age, or gender role groups. Additionally, because we included only a highly structured social interaction task involving a same-sex stranger in the laboratory, we cannot comment on the extent to which the patterns we observed are also apparent in routine daily interactions in naturalistic settings. Finally, our hypotheses regarding the impact of nonprovocative interpersonal stress on hemodynamic responses of hostile individuals were theoretically driven, but a stronger test of the hypotheses would have involved experimental manipulation of the social stressor. Still, the results of this study suggest that high levels of hostility are associated with enhanced vascular and dampened cardiac reactivity to nonprovocative social stress in both men and women and that experiencing a high need to control the interaction exacerbates the responding of hostile but not nonhostile individuals. The findings not only highlight the utility of social psychophysiological methods but also point to the need for further elaboration of the salient features of social context as a means of refining our understanding of the hostility-reactivity association.
|
NOTES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
2(1) = 0.205, p = .66). The same nonsignificant associations were apparent when these analyses were repeated separately by sex. 
The persuasion and empathy versions of the task were included in the original study to examine gender-relevant stress. Approximately half of high and low hostile individuals participated in each task condition. However, because task condition was not expected to interact with hostility group to predict reactivity, and because individuals were not randomly assigned to task condition on the basis of hostility group, task condition was not evaluated in the primary analyses. When we repeated the analyses, including task condition as an independent variable, no effects involving condition emerged, and the pattern of findings reported above remains intact.
Received for publication February 25, 1999.
Revision received July 13, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  L. M. Christian and C. M. Stoney Social Support Versus Social Evaluation: Unique Effects on Vascular and Myocardial Response Patterns Psychosom Med, November 1, 2006; 68(6): 914 - 921. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Carrere, D. Yoshimoto, A. Mittmann, E. M. Woodin, A. Tabares, J. Ullman, C. Swanson, and M. Hawkins The Roles of Marriage and Anger Dysregulation in Biobehavioral Stress Responses Biol Res Nurs, July 1, 2005; 7(1): 30 - 43. [Abstract] [PDF]
|
|
|
|
|
|
S. H. Boyle, R. B. Williams, D. B. Mark, B. H. Brummett, I. C. Siegler, M. J. Helms, and J. C. Barefoot Hostility as a Predictor of Survival in Patients With Coronary Artery Disease Psychosom Med, September 1, 2004; 66(5): 629 - 632. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Christensen, R. Lund, M. T. Damsgaard, B. E. Holstein, S. Ditlevsen, F. Diderichsen, P. Due, L. Iversen, and J. Lynch Cynical Hostility, Socioeconomic Position, Health Behaviors, and Symptom Load: A Cross-Sectional Analysis in a Danish Population-Based Study Psychosom Med, July 1, 2004; 66(4): 572 - 577. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. S. Thomas, R. A. Nelesen, and J. E. Dimsdale Relationships Between Hostility, Anger Expression, and Blood Pressure Dipping in an Ethnically Diverse Sample Psychosom Med, May 1, 2004; 66(3): 298 - 304. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
