Independent association between lower level of social support and higher coagulation activity before and after acute psychosocial stress.
ABSTRACT To investigate the relationship between social support and coagulation parameter reactivity to mental stress in men and to determine if norepinephrine is involved. Lower social support is associated with higher basal coagulation activity and greater norepinephrine stress reactivity, which in turn, is linked with hypercoagulability. However, it is not known if low social support interacts with stress to further increase coagulation reactivity or if norepinephrine affects this association. These findings may be important for determining if low social support influences thrombosis and possible acute coronary events in response to acute stress. We investigated the relationship between social support and coagulation parameter reactivity to mental stress in men and determined if norepinephrine is involved.
We measured perceived social support in 63 medication-free nonsmoking men (age (mean +/- standard error of the mean) = 36.7 +/- 1.7 years) who underwent an acute standardized psychosocial stress task combining public speaking and mental arithmetic in front of an audience. We measured plasma D-dimer, fibrinogen, clotting Factor VII activity (FVII:C), and plasma norepinephrine at rest as well as immediately after stress and 20 minutes after stress.
Independent of body mass index, mean arterial pressure, and age, lower social support was associated with higher D-dimer and fibrinogen levels at baseline (p < .012) and with greater increases in fibrinogen (beta = -0.36, p = .001; DeltaR(2) = .12), and D-dimer (beta = -0.21, p = .017; DeltaR(2) = .04), but not in FVII:C (p = .83) from baseline to 20 minutes after stress. General linear models revealed significant main effects of social support and stress on fibrinogen, D-dimer, and norepinephrine (p < .035). Controlling for norepinephrine did not change the significance of the reported associations between social support and the coagulation measures D-dimer and fibrinogen.
Our results suggest that lower social support is associated with greater coagulation activity before and after acute stress, which was unrelated to norepinephrine reactivity.
- SourceAvailable from: Roland Von Känel[Show abstract] [Hide abstract]
ABSTRACT: Psychosocial factors have been associated with both a prothrombotic state and an increased risk of venous thromboembolism (VTE). We examined the relation of depressive symptoms and social support with D-dimer, an integrative measure of enhanced coagulation activity, and several additional prothrombotic measures in patients with VTE. We studied 173 patients with a previous deep venous thrombosis and/or pulmonary embolism (mean age ± SD 45 ± 14 years, 55% men). Clinical and lab assessments took place ≥ 3 months after VTE and ≥ 1 month after discontinuation of oral anticoagulants. The patients rated depressive symptoms and social support by validated questionnaires. After adjusting for sociodemographic and medical covariates, interactions emerged between depressive symptoms and social support for D-dimer (p=0.012) and aPTT (p=0.002). As opposed to patients with high levels of social support, those with low levels of social support showed a direct association of depressive symptoms with D-dimer (r=0.19, p=0.014) and an inverse relationship with aPTT (r=-0.14, p<0.09). Depressive symptoms were associated with levels of thrombin-antithrombin complex (r=0.19, p=0.016). Greater social support was associated with prolonged aPTT (r=0.16, p=0.046). There were no significant associations of depressive symptoms and social support with D-dimer, fibrinogen, FII:C, FV:C, FVII:C, FVIII:C, FX:C, INR, and thrombin time. Depressive symptoms are associated with enhanced coagulation activity in patients with VTE, particularly so in those who perceive low levels of social support. Conversely, high levels of social support may contribute directly and through buffering the effect of depressive symptoms to attenuated clotting activity in VTE patients.Thrombosis Research 05/2012; 130(3):374-80. · 3.13 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: BACKGROUND: We investigated the psychometric properties of a short questionnaire for combined assessment of different perceived stress management skills in the general population and tested whether scores relate to physiological stress reactivity. METHODS: For psychometric evaluation, we determined the factor structure of the questionnaire and investigated its measurement invariance in the participant groups and over time in three different independent samples representing the general population (total N=332). Reliability was tested by estimating test-retest reliability, internal consistency, and item reliabilities. We examined convergent and criterion validity using selected criterion variables. For endocrine validation, 35 healthy non-smoking and medication-free men in a laboratory study and 35 male and female employees in a workplace study underwent an acute standardized psychosocial stress task. We assessed stress management skills and measured salivary cortisol before and several times up to 60min (workplace study) and 120min (laboratory study) after stress. Potential confounders were controlled. RESULTS: The factor structure of the questionnaire consists of five scales reflecting acceptably distinct stress management skills such as cognitive strategies, use of social support, relaxation strategies, anger regulation, and perception of bodily tension. This factor structure was stable across participant groups and over time. Internal consistencies, item reliabilities, and test-retest reliabilities met established statistical requirements. Convergent and criterion validity were also established. In both endocrine validation studies, higher stress management skills were independently associated with lower cortisol stress reactivity (p's<.029). CONCLUSIONS: Our findings suggest that the questionnaire has good psychometric properties and that it relates to subjective psychological and objective physiological stress indicators. Therefore, the instrument seems a suitable measure for differential assessment of stress management skills in the general population.Psychoneuroendocrinology 08/2012; · 5.14 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Objective Hypertension and an atherogenic lipid profile are known risk factors for coronary heart disease (CHD). Hypertensives show greater changes in atherogenic plasma lipids to acute stress than normotensives. In this study we investigated whether attribution of failure is associated with lipid stress reactivity in hypertensive compared with normotensive men. Methods 18 normotensive and 17 hypertensive men (mean ± SEM; 45 ± 2.2 years) underwent an acute standardized psychosocial stress task that can be viewed as a situation of experimentally induced failure. We assessed external-stable (ES), external-variable (EV), internal-stable (IS), and internal variable (IV) attribution of failure and psychological control variables (i.e. extent of depression and neuroticism). Moreover, total cholesterol (TC), low-density-lipoprotein cholesterol (LDL-C), and norepinephrine were measured immediately before and several times after stress. Results ES moderated TC- and LDL-C-stress reactivity in hypertensives as compared to normotensives (interaction mean arterial pressure [MAP]-by-ES for TC: F = 3.71, p = .015; for LDL-C: F = 3.61, p = .016). TC and LDL-C levels were highest in hypertensives with low ES immediately after stress (p ≤ .039). In contrast, hypertensives with high ES did not differ from normotensives in TC and LDL-C immediately after stress (p’s > .28). Controlling for norepinephrine, depression, and neuroticism in addition to age and BMI did not significantly change results. There were no significant associations between lipid baseline levels or aggregated lipid secretion and IS, IV, or EV (p’s > .23). Conclusion Our data suggest that ES may independently protect from elevated lipid stress reactivity in hypertensive individuals. ES thus might be a protective factor against CHD in hypertension.Journal of Psychosomatic Research. 01/2014;
Independent Association Between Lower Level of Social Support and Higher
Coagulation Activity Before and After Acute Psychosocial Stress
PETRA H. WIRTZ, PHD, LAURA S. REDWINE, PHD, ULRIKE EHLERT, PHD, AND ROLAND VON KA ¨NEL, MD
Objective: To investigate the relationship between social support and coagulation parameter reactivity to mental stress in men and
to determine if norepinephrine is involved. Lower social support is associated with higher basal coagulation activity and greater
norepinephrine stress reactivity, which in turn, is linked with hypercoagulability. However, it is not known if low social support interacts
with stress to further increase coagulation reactivity or if norepinephrine affects this association. These findings may be important for
determining if low social support influences thrombosis and possible acute coronary events in response to acute stress. We investigated
the relationship between social support and coagulation parameter reactivity to mental stress in men and determined if norepinephrine is
involved. Methods: We measured perceived social support in 63 medication-free nonsmoking men (age (mean ? standard error of the
mean) ? 36.7 ? 1.7 years) who underwent an acute standardized psychosocial stress task combining public speaking and mental
arithmetic in front of an audience. We measured plasma D-dimer, fibrinogen, clotting Factor VII activity (FVII:C), and plasma
norepinephrine at rest as well as immediately after stress and 20 minutes after stress. Results: Independent of body mass index, mean
arterial pressure, and age, lower social support was associated with higher D-dimer and fibrinogen levels at baseline (p ? .012) and with
greater increases in fibrinogen (? ? ?0.36, p ? .001; ?R2? .12), and D-dimer (? ? ?0.21, p ? .017; ?R2? .04), but not in FVII:C
(p ? .83) from baseline to 20 minutes after stress. General linear models revealed significant main effects of social support and stress on
fibrinogen, D-dimer, and norepinephrine (p ? .035). Controlling for norepinephrine did not change the significance of the reported
associations between social support and the coagulation measures D-dimer and fibrinogen. Conclusions: Our results suggest that lower
social support is associated with greater coagulation activity before and after acute stress, which was unrelated to norepinephrine
reactivity. Key words: social support, coagulation, clotting factor VII:C, D-dimer, fibrinogen, psychological stress.
CVD ? cardiovascular disease; ACS ? acute coronary syndromes;
BMI ? body mass index; MAP ? mean arterial pressure; BP ?
blood pressure; SD ? standard deviation; AUC ? area under the
curve; TSST ? Trier Social Stress Test; FVII:C ? clotting Factor
VII activity; FVIII:C ? clotting Factor VIII activity.
on one or more of the following three classes: information
leading the person to believe that he or she (1) is cared for and
loved (2), esteemed and valued (3), and belongs to a network
of communication and mutual obligation (1,2). In the litera-
ture, measures of perceived social support are more commonly
used than measures of received social support (3). Perceived
social support is conceptualized as the subjective appraisal of
the degree of match between the amount and type of support
needed and the amount and type of support available, or the
perception that support would be available if needed (3).
Accumulating evidence suggests that social support is asso-
prospectively increases the risk of coronary artery disease (CAD)
(3–6) with a relative risk two- to three-fold and a robust effect
observed even after controlling for conventional (e.g., age, body
ocial support is a context-specific advocative interpersonal
process centered on the reciprocal exchange of information
mass index (BMI), and high blood pressure (BP)), social, and
behavioral cardiovascular risk factors (4–7).
Previous research addressed the biological pathways through
which low social support increases cardiovascular risk or, alter-
natively, high social support enhances health. Based on the ef-
fects of social support on physiological processes implicated in
cardiovascular disease, the so-called “social support-reactivity
hypothesis” has been proposed. This view posits that social
support maintains cardiovascular health by reducing psychobio-
logical reactivity to stressors, thereby acting as a “stress buffer-
social support attenuates psychological and physiological stress
responses as indexed by cardiovascular, autonomic, and hypo-
thalamo-pituitary-adrenal (HPA) responses (8,10–13). In hyper-
tensive and normotensive middle-aged men, greater levels of
perceived social support were associated with reduced cat-
echolamines reactivity to acute psychosocial stress (14). Notably,
the accumulation of physiological hyperreactivity to stress
throughout a life span that may occur in individuals with low
social support has been proposed to enhance CAD risk (15–18).
Coagulation and fibrinolysis are implicated as stress-reac-
tive physiological systems important in the development of
CAD and acute coronary syndromes (ACS) (19). Patients
whose ACS had been triggered by intense emotions showed
greater platelet activation in response to psychological stress
compared with patients without emotion-triggering ACS (20).
In healthy individuals, acute mental stress activates both the
coagulation and the fibrinolysis components of hemostasis to
result in net hypercoagulability (19,21). Acute psychological
stress-induced increases in several hemostatic parameters,
such as clotting Factor VII activity (FVII:C), clotting Factor
VIII activity (FVIII:C) and FXII:C, thrombin-antithrombin
complex, fibrinogen, von Willebrand Factor, and D-dimer
levels, have been observed (19).
In contrast to stress effects, higher social support may
reduce coagulation activity. Greater social support as assessed
From the Department of Clinical Psychology and Psychotherapy (P.H.W.,
U.E.), University of Zurich, Switzerland; Department of Medicine (L.S.R.),
Cardiology Branch, University of California, San Diego, California; and the
Department of General Internal Medicine (R.v.K.), Bern University Hospital,
Inselspital, and University of Bern, Bern, Switzerland.
Address correspondence and reprint requests to Petra H. Wirtz, Department
of Clinical Psychology and Psychotherapy, University of Zurich, Binzmu ¨-
hlestrasse 14, Box 26, CH-8050 Zurich, Switzerland. E-mail: p.wirtz@
Received for publication November 19, 2007; revision received July 3,
Supported by research Grants 56233203 and 56233204 from the University
of Zurich (P.H.W.) and by a research grant from the University of Bern
Copyright © 2009 by the American Psychosomatic Society
Psychosomatic Medicine 71:30–37 (2009)
via social integration, social isolation, and social network is
associated with lower amount of fibrinogen (22–27) even
when controlling for cardiovascular risk factors (22,24,26,27).
However, the role of social support as a buffer of acute stress
for hemostasis is not straightforward. In a small sample of 27
healthy middle-aged men, we recently did not find significant
associations between social support as assessed by the Ger-
man Social Support Questionnaire (F-SozU) and stress-in-
duced changes in coagulation parameters (28). However, in
that study, due to the small sample size, we did not control for
age, BMI, and BP. In contrast, a study by Steptoe and co-
workers assessed associations between social support mea-
sures and fibrinogen stress reactivity (27) and found higher
fibrinogen levels in socially isolated participants compared
with nonisolated participants before and after acute mild psy-
chological stress (27).
Catecholamines play a role in stress-induced hemostatic
changes because catecholamine infusion elicited hypercoagu-
lability in healthy persons (29). In addition, we recently found
associations between epinephrine and FVIII:C and between
norepinephrine and fibrinogen under resting conditions, inde-
pendent of age, BMI, and mean arterial pressure (MAP) in
middle-aged hypertensive and normotensive men (30). More-
over, in the same study, stress-induced changes in D-dimer were
predicted by norepinephrine stress change (30). In another study,
we found that the stress-induced increase in norepinephrine sig-
unknown whether lower social support is associated with greater
catecholamine stress reactivity, which in turn could induce hy-
percoagulability. More specifically, the role of norepinephrine in
stress-induced D-dimer changes in those with low social support
has not previously been examined.
The aim of this study was to extend previous research
investigating whether perceived social support is associated
with several coagulation parameter levels at rest and through-
out a potent acute psychosocial stressor in a sizeable group of
medication-free nonsmoking men. We hypothesized that
lower social support is associated with higher baseline
levels and greater increases in response to stress in the
coagulation measures fibrinogen and FVII:C, and the hy-
percoagulability marker D-dimer. Moreover, we examined
whether such associations were moderated or mediated by
The study is part of a project studying the biological response in general
and coagulation activation in particular to acute psychosocial stress in men as
previously described (32,33). Whereas our previous report addressed the
acute stress-induced increases of the coagulation parameters D-dimer and
fibrinogen in relationship to age (33), in the present study, we specifically
examined the relationship of social support to acute stress-induced coagula-
tion changes. In the present study, we also tested whether the relationship
between social support and coagulation activation to stress would be modu-
lated by norepinephrine.
The Ethics Committee of the State of Zurich, Switzerland, formally
approved the research protocol. Of the original 66 subjects, three persons had
to be excluded because of incomplete blood samples for coagulation param-
eters. The final study sample consisted of 63 subjects who provided their
written informed consent. The study was conducted between April 2004 and
August 2005. We intentionally recruited nonsmoking men between the ages
of 20 and 65 years who were in excellent physical and mental health
confirmed by an extensive health questionnaire (34) and telephone interview.
Specific exclusion criteria were obtained by the subjects’ self-report and were
as follows: regular strenuous exercise; alcohol and illicit drug abuse; any heart
disease, varicosis or thrombotic diseases; elevated blood sugar and diabetes;
elevated cholesterol, liver, and renal diseases; chronic obstructive pulmonary
disease; allergies and atopic diathesis, rheumatic diseases; and current infec-
tious diseases. In addition, participants were included only if they reported
taking no prescribed and/or over-the-counter medication, either regularly or
occasionally, and if their BP was in the normotensive or moderately hyper-
tensive range (systolic BP ?160 mm Hg and diastolic BP ?100 mm Hg).
When the personal or medication history was not conclusive, the subject’s
primary care physician was contacted for verification.
Assessment of Social Support
The first part of the Berlin Social Support Scale consists of 17 items
assessing perceived social support (PSS), support seeking (SS), and need for
support (35). Previous literature on social support and cardiovascular disease
as well as our own previous findings of greater catecholamine stress reactivity
with increasing PSS led us to use the 8-item PSS subscale composed of the
two subscales “emotional support” and “instrumental support” in the present
study (3,14). Using a 4-point rating scale ranging from 1 (completely wrong)
to 4 (completely right), we asked the participants whether they agree with
certain statements on their perception of social support (e.g., “there are people
who help me if I need help”; “there are people cheering me up when I am
sad”). The PSS renders scores between 1 (minimum score) and 4 (maximum
score). Higher scores mean higher PSS. Cronbach’s ? (n ? 437) is 0.83 for
the PSS subscale (35).
Subjects were tested between 2 PM and 4 PM. They abstained from
physical exercise, alcohol, and caffeinated beverages for at least 24 hours
before testing. We used the Trier Social Stress Test (TSST) combining a
5-minute preparation phase followed by a 5-minute mock job interview, and
5-minute mental arithmetic task in front of an audience (36). The TSST
evokes reliable physiological responses across different biological systems,
including coagulation factors also investigated in the present study (37).
During recovery, the subjects remained seated in a quiet room for 40 minutes.
Blood for coagulation and norepinephrine measures was obtained imme-
diately before stress, immediately after stress, and 20 minutes after stress. BP
was measured immediately before and 40 minutes after stress by sphygmo-
manometry (Omron 773, Omron Healthcare Europe B.V., Hoofddorp, Neth-
erlands) and MAP was calculated by the formula (2/3 ? mean diastolic BP) ?
(1/3 mean systolic BP).
Venous blood was drawn through an indwelling forearm catheter into
polypropylene tubes containing 3.8% sodium citrate and centrifuged at 2000
g for 20 minutes at 4°C. The obtained plasma sample was immediately
aliquoted in polypropylene tubes and frozen at ?80°C. All analyses of
coagulation factors used the BCS Coagulation Analyzer (Dade Behring,
Liederbach, Germany). Determination of FVII:C used standard coagulometric
methods, factor-deficient standard human plasma, and reagents (Dade Be-
hring); plasma fibrinogen was determined using a modified Clauss method
(Multifibren U, Dade Behring). Plasma D-dimer was measured by means of
an enzyme-linked immunosorbent assay (Asserachrom Stago, Asnie `res,
France). Inter- and intra-assay coefficients of variation were ?10% for all
Blood samples for measurement of plasma norepinephrine were drawn
into ethylenediaminetetraacetic acid-coated monovettes (Sarstedt, Numbre-
cht, Germany), and immediately centrifuged for 10 minutes at 2000 g; plasma
was stored at ?80°C until analysis. Plasma norepinephrine was determined by
means of high-performance liquid chromatography and electrochemical de-
SOCIAL SUPPORT, COAGULATION ACTIVITY, AND STRESS
31 Psychosomatic Medicine 71:30–37 (2009)
tection after liquid-liquid extraction (38,39). The limit of detection was 10
pg/ml. Inter- and intra-assay variance was ?5%. To reduce error variance
caused by imprecision of the intra-assay, all samples from one subject were
analyzed in the same run.
Data were analyzed using SPSS (version 13.0) statistical software package
(SPSS Inc., Chicago IL). All tests were two-tailed with level of significance
of p ? .05 and level of statistical trends of p ? .10. Using the trapezoid
formula, we calculated areas under the total response curves, expressed as area
under the measured time points with respect to ground for all coagulation
measures and norepinephrine (40). Before statistical analyses, all data were tested
for normality using the Kolmogorov-Smirnov test. Coagulation values and co-
agulation areas under the curve (AUCs) were logarithmically transformed to
achieve normal distributions. For clarity, we provide untransformed data.
To assess the associations between social support and coagulation activity
at baseline and after stress, we first calculated linear regression analyses with
the respective coagulation measure as the dependent variable and social
support score as a continuous independent variable. We used coagulation
baseline measures as the dependent variables to assess associations between
social support and coagulation activity at rest. We employed AUC measures
of the coagulation parameters to assess the associations between social
support and stress-induced coagulation changes. In light of previously re-
ported associations between BMI, MAP, and age with coagulation parameters
at rest and in response to stress (30,33,41), we controlled for BMI, MAP, and
age. We entered these parameters as predictors in all analyses. All indepen-
dent variables were simultaneously forced into the regression equations. The
optimal total sample size to predict stress reactivity in coagulation parameters
was n ? 59 for detecting a medium-to-large effect size of f2? 0.25 in
multiple regression analyses with a power of 0.80 using four predictors.
Second, we further tested regression results by performing general linear
models with repeated measures for each coagulation parameter as the depen-
dent variable and with social support as the continuous independent variable.
In these analyses, we again controlled for BMI, MAP, and age. For illustrative
purposes, we categorized the study group based on their social support scores
into four groups of subjects with increasing levels of social support termed as
follows: lowest (2.75–3.38, n ? 11), lower (3.50–3.63, n ? 16), higher
(3.75–3.88, n ? 25), and highest social support scores (4–4, n ? 11).
To test for associations between social support and norepinephrine stress
reactivity, we used general linear modeling with norepinephrine levels as
repeated dependent factor and levels of social support as the continuous
independent factor, at the same time controlling for age, BMI, and MAP. To
assess whether associations between social support and coagulation activity at
baseline and after stress are related to norepinephrine levels, we again calcu-
lated regression analyses. As dependent variables, we entered those coagula-
tion measures that were significantly associated with social support in the
linear model. As independent variables, we simultaneously entered social
support and norepinephrine to test for a mediation effect of norepinephrine.
To test for a moderating effect of norepinephrine, we simultaneously entered
social support, norepinephrine, as well as their interaction term (social sup-
port ? norepinephrine) as independent variables. We always controlled for
age, BMI, and MAP. Interaction terms were computed on Z-transformed data
rendering means of 0 and standard deviations (SD) of 1. All regression
analyses including interaction terms were performed on Z-transformed data.
We calculated Pearson’s product-moment correlations to assess the associa-
tions between social support, age, BMI, and MAP.
Social support scores of our 63 study participants ranged
between 2.75 and 4 (mean ? SD ? 3.66 ? 0.32). The mean
age was 37 ? 13.7 years, the mean BMI was 24.8 ? 3.04, and
MAP was 94.7 ? 10.62 mm Hg. Age, MAP, and BMI were
intercorrelated (p ? .001) but they did not correlate with
social support (p ? .21).
Relationship Between Social Support and Coagulation
Activity at Rest and in Response to Stress
Social Support and Coagulation Activity at Rest
D-Dimer at Rest
Lower social support scores were significantly associated
with higher D-dimer levels at rest (? ? ?0.22, p ? .012;
?R2? .05) independent of MAP (p ? .11), BMI (? ? 0.47,
p ? .001; ?R2? .15), and age (? ? 0.51, p ? .001; ?R2?
0.17) with the total model explaining 60% of the variance in
D-dimer resting levels.
Fibrinogen at Rest
Higher fibrinogen at rest was significantly associated with
lower social support (? ? ?0.35, p ? .001; ?R2? .11)
independent of MAP (p ? .90), BMI (? ? 0.42, p ? .001;
?R2? .12), and age (? ? 0.33, p ? .004; ?R2? .08) with
the total model explaining 52% of the variance in fibrinogen
FVII:C at Rest
Social support was not significantly associated with FVII:C
at rest (p ? .96).
Social Support and Coagulation Levels Between Rest
and 20 Minutes After Stress
Lower social support was associated with higher D-dimer
AUC (? ? ?0.21, p ? .017; ?R2? 0.04) independent of
MAP (p ? .07), BMI (? ? 0.42, p ? .001; ?R2? .13), and
age (? ? 0.56, p ? .001; ?R2? .21). The whole model
explained 59% of the variance in D-dimer AUC.
Independent of MAP (p ? .95), BMI (? ? 0.38, p ? .001;
?R2? .10), and age (? ? 0.34, p ? .005; ?R2? .08), lower
social support was associated with higher fibrinogen AUC
(? ? ?0.36, p ? .001; ?R2? 0.12). The whole model
explained 48% of the observed variance in fibrinogen AUC.
There were no associations between social support and
AUC of FVII:C (p ? .83).
General Linear Models
Across all subjects, the TSST elicited significant increases
in D-dimer (p ? .009, f ? 0.34), fibrinogen (p ? .012, f ?
0.32), and FVII:C (p ? .046, f ? 0.32) from baseline to
immediately after stress at the same time controlling for age,
BMI, and MAP. To validate the results from the above re-
gression analyses, we applied general linear models with
repeated measures of coagulation factors as dependent vari-
ables and social support as a continuous independent variable.
After controlling for BMI, MAP, and age, there were signif-
icant main effects for social support for repeated D-dimer and
fibrinogen levels (D-dimer: F(1,58) ? 6.0, p ? .018, f ? 0.31,
P. H. WIRTZ et al.
32Psychosomatic Medicine 71:30–37 (2009)
Figure 1A; fibrinogen: F(1,58) ? 13.8, p ? .001, f ? 0.47,
Figure 1B) whereas there were no significant interactions
between stress and social support (p ? .58). Neither main
effects nor interaction effects were observed in terms of re-
peated measurements of FVII:C (p ? .24).
For illustrative purposes, Figure 1 (panels A and B) shows
coagulation responses to the TSST in four groups of subjects
with increasing levels of social support (lowest: 2.75–3.38,
n ? 11; lower: 3.50–3.63, n ? 16; higher: 3.75–3.88, n ? 25),
and highest social support scores (4–4, n ? 11) based on
quartiles (note that subject numbers differ in the four groups
because individual PSS values were not balanced above and
below the particular cut-off values).
Norepinephrine, Social Support,
and Coagulation Parameters
Social Support and Norepinephrine Stress Reactivity
The TSST stimulated significant increases in norepineph-
rine levels (F(1/61) ? 94.9, p ? .001). General linear mod-
eling with norepinephrine as repeated dependent variable and
social support as continuous independent variable revealed
that higher social support was associated with lower norepi-
nephrine levels before and after stress (main effect: F(1/57) ?
4.65, p ? .035, f ? 0.27, Figure 2), at the same time control-
ling for age, BMI, and MAP. There was no interaction be-
tween social support and stress (p ? .33).
Norepinephrine and Associations Between Social
Support and Coagulation Parameters
Table 1 depicts regression results for associations between
norepinephrine, social support, and coagulation parameters at
rest and in response to stress.
Regression analyses after controlling for BMI, MAP, and
age revealed that resting norepinephrine levels were not sig-
nificantly related to any of the coagulation parameters (p ?
.60), and was determined to not be a significant factor in the
associations between social support and resting levels of D-
dimer and fibrinogen (Table 1). This suggests that resting
norepinephrine levels do not mediate the observed associa-
tions between social support and resting levels of D-dimer and
fibrinogen. Additionally, entering baseline norepinephrine,
levels, social support, plus the norepinephrine-by-social sup-
(2.75–3.38, n ? 11), lower (3.50–3.63, n ? 16), higher (3.75–3.88, n ? 25),
and highest social support (SS) scores (4–4, n ? 11). (A, B). Values are
mean ? standard error of the mean. Stress reactivity of D-dimer (A) and
fibrinogen (B) across all 63 subjects. We applied general linear models with
repeated measures of coagulation factors as dependent variables and social
support as continuous independent variable at the same time controlling for
body mass index, mean arterial pressure, and age in all analyses. The main
effect of social support was significant in terms of D-dimer F(1,58) ? 6.0,
p ? .018, f ? 0.31, (A) and fibrinogen (F(1,58) ? 13.8, p ? .001, f ? 0.47,
B). The figure shows these data for illustrative purposes by four categories of
SS scores. TSST ? Trier Social Stress Test.
Prothrombotic factor levels in four groups of subjects with lowest
(2.75–3.38, n ? 11), lower (3.50–3.63, n ? 16), higher (3.75–3.88, n ? 25),
and highest social support (SS) scores (4–4, n ? 11). Values are mean ?
standard error of the mean. Stress reactivity of norepinephrine across all 63
subjects. We applied general linear models with repeated measures of nor-
epinephrine as dependent variables and social support as continuous indepen-
dent variable at the same time controlling for body mass index, mean arterial
pressure, and age. Higher social support was associated with lower norepi-
nephrine levels before and after stress (main effect: F(1,57) ? 4.65, p ? .035,
f ? 0.27). Figure 2 shows these data for illustrative purposes by four
categories of SS scores. TSST ? Trier Social Stress Test.
Norepinephrine levels in four groups of subjects with lowest
SOCIAL SUPPORT, COAGULATION ACTIVITY, AND STRESS
33Psychosomatic Medicine 71:30–37 (2009)
port interaction into the regression equation revealed that
resting norepinephrine levels did not moderate the relation-
ship between social support and both D-dimer and fibrin-
ogen (p’s ? .69).
To determine if norepinephrine secretion mediates between
social support and stress reactivity of coagulation parameters
D-dimer and fibrinogen, regression analyses were recalculated
with norepinephrine AUC entered as an independent variable
(Table 1). To test for a moderation effect of norepinephrine,
its interaction with social support was additionally entered as
an independent variable. Controlling for norepinephrine AUC
did not significantly affect the observed associations between
social support and AUC of D-dimer and fibrinogen. Thus,
neither norepinephrine AUC alone nor its interaction with
social support were significantly associated with any of the
coagulation parameter AUCs (p ? .50). In sum, norepineph-
rine does not seem to be a mediator or a moderator of social
support effects on coagulation activity.
Notably, results of the PSS subscales “instrumental sup-
port” and “emotional support” were similar to the combined
PSS scale in all analyses (data not shown).
Research suggested that poor social support is associated
with health risks (2), such as CAD (3–6). The mechanisms
involved are not known, although overactivation of hemosta-
sis might be one contributing factor. The main objective of the
present study was to investigate whether social support is
associated with elevated coagulation factor levels at rest and
throughout an acute stress process. The coagulation factors
fibrinogen and D-dimer were examined at baseline, represent-
ing resting activity, and also for the AUC (i.e., immediately
before stress, immediately after stress, and 20 minutes after
stress), representing the total stress-response course. The re-
sults of our study suggest that low social support is associated
with increased fibrinogen and D-dimer levels at rest, and that
acute stress and low social support are independently associ-
ated with an increased AUC. This may indicate that individuals
with reduced social support run an even higher atherothrom-
botic risk during acute stress. Specifically, effects sizes of
stress-induced coagulation increases suggest clinical impor-
tance. A prothrombotic state of the blood promotes athero-
sclerosis development and, after rupture of an atherosclerotic
plaque, thrombotic occlusion of a coronary artery leading to
an ACS (42,43).
Even though FVII:C was responsive to the TSST, we did
not find a significant association between social support and
FVII:C both at baseline and throughout the stress period.
Clotting factor FVII belongs to the extrinsic system of blood
coagulation, whereas fibrinogen is the precursor of fibrin
further down in the coagulation cascade after intrinsic and
extrinsic coagulation pathways have merged (21). Therefore,
it could be that social support is more strongly related to
factors involved in the intrinsic pathway of blood coagulation,
which reasoning we did, however, not address in the present
study. After clot formation, the fibrinolysis system degrades
fibrin, whereby fibrin degradation products such as D-dimer
are formed. Because it indicates activation of the entire coag-
ulation system (i.e., the step of fibrin formation) and also the
fibrinolysis system (i.e., the step of fibrin dissolution), D-
dimer is termed a hypercoagulability marker (44). Hyperco-
agulability markers are more sensitive to change than are
individual hemostatic factors (45), providing one explanation
for why D-dimer emerged as a significant correlate of social
support whereas FVII:C did not. Our findings correspond to
previous studies that social support scores were negatively
associated with resting fibrinogen levels (19,21) now extend-
ing these findings to D-dimer. These may be important find-
ings in relationship to health. Elevated fibrinogen levels are
associated with cardiovascular disease risk (46), and are re-
lated to both atherogenesis and thrombogenesis. For each SD
increase in fibrinogen above the mean, it was suggested that
there is an 84% increase in the 5-year risk of ischemic heart
disease (47). In addition, increased D-dimer levels are asso-
ciated with greater risk of myocardial infarction (48), cere-
brovascular events (49), and peripheral arterial disease (50).
Acute mental stress activates both the coagulation and the
fibrinolysis components of hemostasis to result in net hyper-
coagulability (19,21). Increased social support has been sug-
gested to act as a buffer to attenuate physiological stress
responses, as indexed by cardiovascular, autonomic, and HPA
Between Social Support and Coagulation Parameters Controlling for
Hierarchical Regression Analyses for Associations
D-dimer at baseline
NEPI at baseline
Fibrinogen at baseline
NEPI at baseline
BMI ? body mass index; MAP ? mean arterial blood pressure; NEPI ?
norepinephrine; AUC ? area under the curve; SS ? social support score.
P. H. WIRTZ et al.
34Psychosomatic Medicine 71:30–37 (2009)
responses (8,10–13). However, our present findings did not
suggest an interaction between social support and stress for
hemostatic measures, but social support and stress are inde-
pendent factors and therefore may be additive. Higher social
support predicted lower D-dimer and fibrinogen AUC from
baseline to 20 minutes after the stress task. These findings
support a previous study showing that higher fibrinogen levels
were found in socially isolated participants both before and
after a color-word and a mirror-tracing task (27). The present
study extends those results to include D-dimer, which is
involved with both coagulation and fibrinolysis. Furthermore,
an interpersonal stressor, the TSST, was used in the present
research, which expands the range of psychological stress
that is found to generate such hemostatic responses. Cat-
echolamines play a role in stress-induced hemostatic changes
(30), but their role in social support-related changes in coag-
ulation parameters is unclear. To examine if norepinephrine
mediates or moderates the association between social support
and hemostatic changes at rest and in response to stress,
fibrinogen and D-dimer were measured at baseline and
throughout the stress task (baseline through 20 minutes post
task). Resting norepinephrine levels and the interaction of
norepinephrine baseline levels with social support did not
predict levels of D-dimer and fibrinogen, suggesting that
norepinephrine is not associated with hemostatic reactions to
stress in relationship to social support.
Future studies are needed to examine additional pathways
that may mediate social support and health, besides the sym-
pathetic reactivity hypothesis, antistress hormones, inflamma-
tory responses, and sympatho-vagal balance may play roles as
alternative mediators. For example, oxytocin was found to
have potential antistress effects and when combined with
social support, there was a reduced cortisol response along
with increased calmness and decreased anxiety during the
TSST (13). In turn, cortisol changes during stress are associ-
ated with total fibrin formation (51), suggesting a potential
alternate pathway in which social support may buffer stress-
induced hemostatic activity and ultimately health.
Social networks were associated with reduced inflamma-
tory marker interleukin-6 responses in male Framingham
Heart Study participants (52), and reactivity in interleukin-6
and D-dimer to stress were positively correlated (51). Addi-
tionally, in patients with coronary heart disease (CHD), ele-
vated social support is associated with greater probability of
sympatho-vagal balance (i.e., reduced heart rate, MAP, car-
diac index, and increased high frequency heart rate variability)
during recovery from mental stress (53). In contrast, social iso-
lation was associated with decreased heart rate variability (54)
suggesting a reduction in parasympathetic or vagal tone. In
women with CHD, we previously observed elevated levels of
fibrinogen in association with reduced vagal cardiac control (55).
There are also hypotheses about social relationships sug-
gesting that they exhibit main effects on health but do not
necessarily interact with stress (56). Lack of social support
due to social isolation had a hazard ratio (HR) of 1.75 for
mortality in patients with heart failure (57). Compared with
heart failure patients reporting a high social network, hospital
readmission was more frequent among those who had mod-
erate social networks (HR ? 1.87; 95% Confidence Interval
(CI) ? 1.06–3.29; p ? .05) and low social networks (HR ?
1.98; 95% CI ? 1.07–3.68; p ? .05) (58). Similarly, loneli-
ness was significantly associated with elevated systolic BP,
but not so with differences in autonomic or endocrine reactivity
to stress (59). Comparisons of persons reporting more types of
contacts (parents, children, family members, and friends) with
those reporting fewer types yielded age- and gender-adjusted
HRs of 0.73 (95% CI ? 0.64–0.82) for mortality and HRs of
0.75 (95% CI ? 0.61–0.91) for CHD (60).
In summary, social support is associated with various fac-
tors that might also affect hemostasis and that warrant an
investigation as potential pathways linking social support and
hemostasis and ultimately health. Alternatively, social rela-
tionships may function as a main effect on health. As is often
the case, both hypotheses are likely true depending on the
situation and factors being measured.
Our study has several strengths, which include recruitment
of a sizable group of apparently healthy and unmedicated
subjects across a range of perceived social support levels in a
natural unselected setting. However, the study has also its
limitations. First, its cross-sectional nature does not allow us
to interpret the direction of the social support-coagulation-
stress link, although we feel it is unlikely that coagulation
influences social support. Second, our findings were obtained
in a sample of apparently healthy men with BP in the normo-
tensive and mildly hypertensive range and may not be gener-
alized to individuals with more severe hypertension and
women. Third, we restricted our analyses to three coagulation
parameters. Fourth, the significance of low scores on PSS
remains to be elucidated and future studies should include
assessments of a person’s social network.
In conclusion, social support seems to be important for coag-
ulation parameter levels, such as fibrinogen and D-dimer, which
may affect thrombosis and atherosclerotic disease. Lower social
support was associated with higher baseline levels of fibrinogen
related to an increase of these hemostatic factors throughout the
stress response period. Stress-related alterations in hemostasis
might act as a potential trigger for the onset of ACS (19).
Although social support did not seem to affect the magnitude of
the coagulation parameter responses to stress, low social support
may have added risk because of its association with greater
overall coagulation parameter levels. Furthermore, social support
does not seem to drive stress-associated changes in norepineph-
rine. Therefore, future studies are needed to examine the mech-
anisms that are involved in the relationship between social sup-
port and hemostasis factors in order to develop interventions that
can reduce hemostatic risk factors of CHD related to low levels
of social support.
1. Cobb S. Presidential Address—1976. Social support as a moderator of
life stress. Psychosom Med 1976;38:300–14.
SOCIAL SUPPORT, COAGULATION ACTIVITY, AND STRESS
35Psychosomatic Medicine 71:30–37 (2009)
2. McEwen BS. Physiology and neurobiology of stress and adaptation:
central role of the brain. Physiol Rev 2007;87:873–904.
3. Lett HS, Blumenthal JA, Babyak MA, Strauman TJ, Robins C, Sherwood
A. Social support and coronary heart disease: epidemiologic evidence and
implications for treatment. Psychosom Med 2005;67:869–78.
4. Strike PC, Steptoe A. Psychosocial factors in the development of coro-
nary artery disease. Prog Cardiovasc Dis 2004;46:337–47.
5. Kuper H, Marmot M, Hemingway H. Systematic review of prospective
cohort studies of psychosocial factors in the etiology and prognosis of
coronary heart disease. Semin Vasc Med 2002;2:267–314.
6. Rozanski A, Blumenthal JA, Kaplan J. Impact of psychological factors on
the pathogenesis of cardiovascular disease and implications for therapy.
7. Andre-Petersson L, Engstrom G, Hedblad B, Janzon L, Rosvall M. Social
support at work and the risk of myocardial infarction and stroke in
women and men. Soc Sci Med 2007;64:830–41.
8. Christenfeld N, Gerin W. Social support and cardiovascular reactivity.
Biomed Pharmacother 2000;54:251–7.
9. Lepore SJ. Problems and prospects for the social support-reactivity
hypothesis. Ann Behav Med 1998;20:257–69.
10. Nausheen B, Gidron Y, Gregg A, Tissarchondou HS, Peveler R. Lone-
liness, social support and cardiovascular reactivity to laboratory stress.
11. Seeman TE, McEwen BS. Impact of social environment characteristics
on neuroendocrine regulation. Psychosom Med 1996;58:459–71.
12. Kirschbaum C, Klauer T, Filipp SH, Hellhammer DH. Sex-specific
effects of social support on cortisol and subjective responses to acute
psychological stress. Psychosom Med 1995;57:23–31.
13. Heinrichs M, Baumgartner T, Kirschbaum C, Ehlert U. Social support
and oxytocin interact to suppress cortisol and subjective responses to
psychosocial stress. Biol Psychiatry 2003;54:1389–98.
Ehlert U. Low social support and poor emotional regulation are associ-
ated with increased stress hormone reactivity to mental stress in systemic
hypertension. J Clin Endocrinol Metab 2006;91:3857–65.
15. Steptoe A, Willemsen G. Psychophysiological responsivity in coronary
heart disease. In: Stansfeld SA, Marmot MG, editors. Stress and the
Heart. London: BMJ Books; 2002:168–180.
16. Linden W, Gerin W, Davidson K. Cardiovascular reactivity: status quo
and a research agenda for the new millennium. Psychosom Med 2003;
17. McEwen BS. Protective and damaging effects of stress mediators. N Engl
J Med 1998;338:171–9.
18. Matthews KA, Katholi CR, McCreath H, Whooley MA, Williams DR,
Zhu S, Markovitz JH. Blood pressure reactivity to psychological stress
predicts hypertension in the CARDIA study. Circulation 2004;110:74–8.
19. Thrall G, Lane D, Carroll D, Lip GY. A systematic review of the effects
of acute psychological stress and physical activity on haemorheology,
coagulation, fibrinolysis and platelet reactivity: Implications for the
pathogenesis of acute coronary syndromes. Thromb Res 2007;120:
20. Strike PC, Perkins-Porras L, Whitehead DL, McEwan J, Steptoe A.
Triggering of acute coronary syndromes by physical exertion and anger:
clinical and sociodemographic characteristics. Heart 2006;92:1035–40.
21. von Kanel R, Mills PJ, Fainman C, Dimsdale JE. Effects of psychological
stress and psychiatric disorders on blood coagulation and fibrinolysis: a
biobehavioral pathway to coronary artery disease? Psychosom Med 2001;
22. Davis MC, Swan PD. Association of negative and positive social ties with
fibrinogen levels in young women. Health Psychol 1999;18:131–9.
23. Rosengren A, Wilhelmsen L, Welin L, Tsipogianni A, Teger-Nilsson AC,
Wedel H. Social influences and cardiovascular risk factors as determi-
nants of plasma fibrinogen concentration in a general population sample
of middle aged men. BMJ 1990;300:634–8.
24. Orth-Gomer K, Rosengren A, Wilhelmsen L. Lack of social support and
incidence of coronary heart disease in middle-aged Swedish men. Psy-
chosom Med 1993;55:37–43.
25. Helminen A, Rankinen T, Vaisanen S, Rauramaa R. Social network in
relation to plasma fibrinogen. J Biosoc Sci 1997;29:129–39.
26. Loucks EB, Berkman LF, Gruenewald TL, Seeman TE. Social integration
is associated with fibrinogen concentration in elderly men. Psychosom
27. Steptoe A, Kunz-Ebrecht S, Owen N, Feldman PJ, Rumley A, Lowe GD,
Marmot M. Influence of socioeconomic status and job control on plasma
fibrinogen responses to acute mental stress. Psychosom Med 2003;65:
28. von Kanel R, Kudielka BM, Preckel D, Hanebuth D, Herrmann-Lingen
C, Frey K, Fischer JE. Opposite effect of negative and positive affect on
stress procoagulant reactivity. Physiol Behav 2005;86:61–8.
29. Preckel D, Von Kanel R. Regulation of hemostasis by the sympathetic
nervous system: any contribution to coronary artery disease? Heart Drug
30. Wirtz PH, Ehlert U, Emini L, Rudisuli K, Groessbauer S, Mausbach BT,
von Kanel R. The role of stress hormones in the relationship between
resting blood pressure and coagulation activity. J Hypertens 2006;24:
31. von Kanel R, Mills PJ, Ziegler MG, Dimsdale JE. Effect of beta2-
adrenergic receptor functioning and increased norepinephrine on the
hypercoagulable state with mental stress. Am Heart J 2002;144:68–72.
32. Wirtz PH, Siegrist J, Rimmele U, Ehlert U. Higher overcommitment to
work is associated with lower norepinephrine secretion before and after
acute psychosocial stress in men. Psychoneuroendocrinology 2008;33:
33. Wirtz PH, Redwine LS, Baertschi C, Spillmann M, Ehlert U, von Kanel
R. Coagulation activity before and after acute psychosocial stress in-
creases with age. Psychosom Med 2008;70:476–81.
34. von Kanel R, Kudielka BM, Abd-el-Razik A, Gander ML, Frey K,
Fischer JE. Relationship between overnight neuroendocrine activity and
morning haemostasis in working men. Clin Sci (Lond) 2004;107:89–95.
35. Schulz U, Schwarzer R. Social support and coping with illness: the Berlin
Social Support Scales (BSSS). Diagnostica 2003;49:73–82.
36. Kirschbaum C, Pirke KM, Hellhammer DH. The “Trier Social Stress
Test”—a tool for investigating psychobiological stress responses in a
laboratory setting. Neuropsychobiology 1993;28:76–81.
37. von Kanel R, Preckel D, Zgraggen L, Mischler K, Kudielka BM, Haeberli
A, Fischer JE. The effect of natural habituation on coagulation responses
to acute mental stress and recovery in men. Thromb Haemost 2004;92:
38. Ehrenreich H, Schuck J, Stender N, Pilz J, Gefeller O, Schilling L, Poser
W, Kaw S. Endocrine and hemodynamic effects of stress versus systemic
CRF in alcoholics during early and medium term abstinence. Alcohol
Clin Exp Res 1997;21:1285–93.
39. 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.
40. Pruessner JC, Kirschbaum C, Meinlschmid G, Hellhammer DH. Two
formulas for computation of the area under the curve represent measures
of total hormone concentration versus time-dependent change. Psycho-
41. Wirtz PH, Ehlert U, Emini L, Rudisuli K, Groessbauer S, von Kanel R.
Procoagulant stress reactivity and recovery in apparently healthy men
with systolic and diastolic hypertension. J Psychosom Res 2007;63:51–8.
42. Falk E, Fernandez-Ortiz A. Role of thrombosis in atherosclerosis and its
complications. Am J Cardiol 1995;75:3B–11B.
43. Libby P. The molecular mechanisms of the thrombotic complications of
atherosclerosis. J Intern Med 2008;263:517–27.
44. von Kanel R, Dimsdale JE. Fibrin D-dimer: a marker of psychosocial
distress and its implications for research in stress-related coronary artery
disease. Clin Cardiol 2003;26:164–8.
45. Bauer KA, Rosenberg RD. The pathophysiology of the prethrombotic
state in humans: insights gained from studies using markers of hemostatic
system activation. Blood 1987;70:343–50.
46. Koenig W. Fibrin(ogen) in cardiovascular disease: an update. Thromb
47. Heinrich J, Balleisen L, Schulte H, Assmann G, van de Loo J. Fibrinogen
and factor VII in the prediction of coronary risk. Results from the
PROCAM study in healthy men. Arterioscler Thromb 1994;14:54–9.
48. Ridker PM, Hennekens CH, Cerskus A, Stampfer MJ. Plasma concen-
tration of cross-linked fibrin degradation product (D-dimer) and the risk
of future myocardial infarction among apparently healthy men. Circula-
49. Smith FB, Lee AJ, Fowkes FG, Price JF, Rumley A, Lowe GD. Hemo-
static factors as predictors of ischemic heart disease and stroke in the
Edinburgh artery study. Arterioscler Thromb Vasc Biol 1997;17:3321–5.
50. Fowkes FG, Lowe GD, Housley E, Rattray A, Rumley A, Elton RA,
MacGregor IR, Dawes J. Cross-linked fibrin degradation products, pro-
gression of peripheral arterial disease, and risk of coronary heart disease.
P. H. WIRTZ et al.
36 Psychosomatic Medicine 71:30–37 (2009)
51. von Kanel R, Kudielka BM, Hanebuth D, Preckel D, Fischer JE. Different
contribution of interleukin-6 and cortisol activity to total plasma fibrin
concentration and to acute mental stress-induced fibrin formation. Clin
Sci (Lond) 2005;109:61–7.
52. Loucks EB, Sullivan LM, D’Agostino RB Sr, Larson MG, Berkman LF,
Benjamin EJ. Social networks and inflammatory markers in the Framing-
ham heart study. J Biosoc Sci 2006;38:835–42.
53. Lache B, Meyer T, Herrmann-Lingen C. Social support predicts hemo-
dynamic recovery from mental stress in patients with implanted defibril-
lators. J Psychosom Res 2007;63:515–23.
54. Horsten M, Mittleman MA, Wamala SP, Schenck-Gustafsson K, Orth-
Gomer K. Social relations and the metabolic syndrome in middle-aged
Swedish women. J Cardiovasc Risk 1999;6:391–7.
55. von Ka ¨nel R, Orth-Gome ´r K. Autonomic function and prothrombotic
activity in women after an acute coronary event. J Womens Health
56. Cohen S. Keynote presentation at the Eight International Congress of
Behavioral Medicine: the Pittsburgh common cold studies: psychosocial
predictors of susceptibility to respiratory infectious illness. Int J Behav
57. Friedmann E, Thomas SA, Liu F, Morton PG, Chapa D, Gottlieb SS,
Sudden Cardiac Death in Heart Failure Trial Investigators. Relation-
ship of depression, anxiety, and social isolation to chronic heart
failure outpatient mortality. Am Heart J 2006;152:940.e1–8.
58. Rodriguez-Artalejo F, Guallar-Castillon P, Herrera MC, Otero CM,
Chiva MO, Ochoa CC, Banegas JR, Pascual CR. Social network as a
predictor of hospital readmission and mortality among older patients with
heart failure. J Card Fail 2006;12:621–7.
59. Hawkley LC, Masi CM, Berry JD, Cacioppo JT. Loneliness is a unique
predictor of age-related differences in systolic blood pressure. Psychol
60. Barefoot JC, Gronbaek M, Jensen G, Schnohr P, Prescott E. Social
network diversity and risks of ischemic heart disease and total mortality:
findings from the Copenhagen city heart study. Am J Epidemiol 2005;
SOCIAL SUPPORT, COAGULATION ACTIVITY, AND STRESS
37Psychosomatic Medicine 71:30–37 (2009)