Full text of DOI:10.1207/S15327558IJBM0702_2

INTERNATIONAL JOURNAL OF BEHAVIORAL MEDICINE, 7(2), 111–126
Copyright © 2000, Lawrence Erlbaum Associates, Inc.
Social Support and Stress:
Influences on Lipid Reactivity
Catherine M. Stoney and Montenique L. Finney
Social support protects individuals from cardiovascular morbidity and mortality, pos-
sibly by attenuating physiological stress responses. The presence of supportive indi-
viduals during acute psychological stress has been shown to influence cardiovascular
functioning, but in inconsistent directions. The purpose of this study was to test the
notion that the presence of a supportive friend during an acute stressor would influ-
ence lipid reactivity. Forty healthy women participated in the study. One half engaged
in a speech task while a friend was present; the other half participated in a speech task
without a friend present. The speech stressor elicited elevations in total cholesterol
and triglycerides under both conditions. Those with a friend present had greater total
cholesterol reactivity to stress, relative to those without a friend present. These find-
ings are similar to some studies in the cardiovascular literature, and may be due to
increased threat appraisal among those with a friend present.
Key words: lipids, reactivity, social support, affiliation, appraisal
The positive impact of social support on health has been well-documented (Cohen
&
Syme, 1985; Hazuda, 1994; Shumaker & Czajkowski, 1994). Many of these
studies have provided cross-sectional evidence for the impact of social ties on
all-cause morbidity and mortality. For example, the first prospective study on the
impact of social ties on mortality (Berkman & Syme, 1979) reported that individu-
Catherine M. Stoney and Montenique L. Finney, Department of Psychology, The Ohio State
University, Columbus, OH, USA.
This study was supported in part by National Institutes of Health grants HL–48363 and HL–48411,
and by MO1–RR0034 to The Ohio State University General Research Clinical Research Center.
We gratefully acknowledge the expert assistance of Karen Cook. The assays were performed in the
laboratory of Dr. Linda Bausserman of the Division of Cardiology in the School of Medicine at Brown
University.
Correspondence concerning this article should be sent to Catherine M. Stoney, Department of
Psychology, Ohio State University, 210 Townshend Hall, 1885 Neil Avenue, Columbus,
OH 43210–1222. E-mail: stoney.1@osu.edu

112
STONEY AND FINNEY
alswithfew contactswithfamilyandfriendswere 1.9to3.1times more likelytodie
within a 9-year follow-up period than were those individuals with more social con-
tacts. These early findings have been replicated in more recent prospective studies.
Although the particular elements of support that are protective may be somewhat
different for men and women, overall, data consistently indicate that both men and
women who are socially isolated have increased all-cause morbidity and mortality,
and those with strong support networks are particularly protected from early mor-
tality (Avlund, Damsgaard, & Holstein, 1998; Broadhead et al., 1983; Cohen,
1
988; Ruberman, Weinblatt, Goldberg, & Chaudhary, 1984).
Social ties are also specifically and inversely related to cardiovascular mortality
and morbidity (Orth-Gomér, 1994; Seeman & Syme, 1987), even after controlling
for established coronary heart disease (CHD) risk factors and health behaviors
(House, Landis, & Umberson, 1988; Orth-Gomér et al., 1998). As with the afore-
mentioned studies of all-cause mortality, most investigations of cardiovascu-
lar-specific morbidity and mortality have studied primarily men. Recently,
however, the cardioprotective effects of social support and social networks have
also been demonstrated for women (Orth-Gomér et al., 1998; Shye, Mullooly,
Freeborn, & Pope, 1995). Although the manner in which social support is defined
differs from study to study, taken together there is compelling evidence that as-
pects of social support are significant factors in determining cardiovascular risk
for both men and women.
The mechanisms by which social support affect cardiovascular health are still
unknown, but several viable hypotheses have been proposed (Knox &
Uvnäs-Moberg, 1998). For example, supportive individuals might encourage
CHD health-promoting behaviors and health-care seeking behaviors (Cohen,
1
988). Data from more recent investigations, however, strongly suggest that the
increased health-promoting behaviors among those with high levels of social sup-
port cannot fully account for the positive impact that social support has on cardio-
vascular disease (Cohen, Kaplan, & Manuck, 1994).
Currently, studies are testing the hypothesis that social support influences car-
diovascular health through the psychophysiologic consequences of social interac-
tions and social presence. For example, individuals who report high levels of
social support have lower blood pressure at rest and during psychological stress
(Knox, 1993). Similar findings have been reported in the context of ambulatory
blood pressure and social support among women (Linden, Chambers, Maurice, &
Lenz, 1993). Thus, it is possible that social integration affects cardiovascular
health by influencing blood pressure throughout the day.
To understand this relation more fully, subsequent investigations have used
the presence of supportive individuals during acute psychological stress as a labo-
ratory model for the physiological consequences of support. More specifically,
these studies have investigated whether the simple presence of a supportive indi-
vidual modifies cardiovascular responses to acute laboratory stressors (Allen,

SOCIAL SUPPORT AND LIPID REACTIVITY
113
Blascovich, Tomaka, & Kelsey, 1991; Edens, Larkin, & Avel, 1992; Gerin,
Milner, Chawla, & Pickering, 1995; Kamarck, Annunziato, & Amateau, 1995;
Kamarck, Manuck, & Jennings, 1990; Kors, Linden, & Gerin, 1997; Lepore, 1995;
Sheffield & Carroll, 1994). Some of these investigations have found that the pres-
ence of a friend reduces cardiovascular responsivity to psychological challenges
(Gerin et al., 1995; Kamarck et al., 1990). For example, in one of the first investi-
gations of this type, Kamarck et al. (1990) examined female university students
while they performed a mental arithmetic and concept formation task alone or in
the presence of a friend. The friend was encouraged to be silently supportive and to
touch the participants on the wrist during the experiment. Results of this study in-
dicated that the presence of a supportive friend induced smaller increases in heart
rate (HR) and systolic blood pressure (SBP) during the math task, relative to those
who performed the tasks alone.
In contrast, other investigations have failed to find that the supportive pres-
ence of an individual reduces cardiovascular reactivity to stress (Allen et al.,
1
991; Sheffield & Carroll, 1994), and others have noted that the presence of a
supportive partner results in enhanced cardiovascular reactivity to stress (Edens
et al., 1992). In sum, although some investigations of cardiovascular functioning
during acute psychological stress find that the presence of a supportive individ-
ual can attenuate the response to stress, others do not. The discrepancies in this
literature are currently an active area of investigation.
The literature on the psychophysiological effects of this model of social sup-
port has primarily focused on blood pressure and HR reactivity to stress. How-
ever, other physiological variables are associated with risk for CHD and are also
associated with social support. With regard to risk factors, blood lipid concentra-
tions are well known to be consistent and potent physiological risk factors for
CHD (Lipid Research Clinics Program, 1984; Schaefer et al., 1994). With re-
gard to factors associated with social support, blood lipid concentrations also
have been cross-sectionally associated with social support. One investigation
demonstrated that a high degree of social integration among a large group of
healthy elderly individuals was associated with significantly lower levels of se-
rum cholesterol (Thomas, Goodwin, & Goodwin, 1985); these investigators pos-
tulate that social support may mitigate the negative physiological consequences
of stress. More recent studies have reported significantly lower high-density li-
poprotein–cholesterol (HDL–c) concentrations among elderly men living alone
(Gliksman, Lazarus, Wilson, & Leeder, 1995), and higher levels of the
atherogenic lipids were found in young men, but not young women, with low
levels of social support (Unden, Krakau, Hogbom, & Romanus-Egerborg, 1995).
Taken together, these data suggest that social support may exert part of its pro-
tection through its association with blood lipid concentrations. Thus, it may be
fruitful to expand the investigation of physiological parameters outside of the
traditional measures of cardiovascular reactivity.

114
STONEY AND FINNEY
The data linking social support and blood lipid concentrations are particularly
intriguing in light of the literature that has examined the impact of psychological
stress and blood lipid concentrations. Stress reliably increases the concentrations
of the atherogenic lipids in both men and women (Brindley, McCann, Niaura,
Stoney, & Suarez, 1993; McCann et al., 1996; Niaura, Stoney, & Herbert, 1992;
Stoney, Niaura, Bausserman, & Matacin, 1999). Although the etiologic signifi-
cance of such changes is not known, individuals at increased risk of cardiovascular
diseases have larger atherogenic lipid reactivity to acute stressors, relative to those
who are not at elevated risk (Stoney & Hughes, 1999). Because social support is
related in some studies to blood lipid concentrations as described earlier
(Gliksman et al., 1995; Thomas et al., 1985), an examination of the interaction of
psychological stress, blood lipids, and social support is warranted. The purpose of
this study was to examine the effects of an acute stressor and the influence of a sup-
portive presence on lipid reactivity in healthy young women.
METHOD
Participants
Forty healthy women with a mean age of 19.30 years (SD = 2.09, range = 18–29
years) participated in this study. Female participants were chosen because women
may be more responsive to the effects of social support than are men (Linden et al.,
1
993). The women were recruited through introductory psychology classes at The
Ohio State University and through campus-wide advertisements. To control for
possible effects of health status on the lipid parameters, only nonsmokers and
individuals who were not on medications were eligible to participate. In addition,
women who were pregnant, nursing, or taking oral contraceptives were excluded
from participating in this study because there is evidence to suggest that fluc-
tuations in reproductive hormones that occur in these situations influence lipid
concentrations (Davis & Matthews, 1990; Stoney, Matthews, McDonald, &
Johnson, 1988).
Each woman (Target) was asked to recruit a female friend (Friend) to accom-
pany them to the laboratory. We asked the participants to recruit same-sex friends
because such pairings reduce the possible variability associated with the interac-
tions of opposite-sex pairings (Kamarck et al., 1990). Each target–friend pair was
randomly assigned to one of two conditions: the “Alone” condition or the “Friend”
condition. In the Alone condition, friends were asked to remain in the hallway
while the target participant took part in the study. In the Friend condition, friends
were asked to directly participate in the study with the target participant.
All participants received monetary compensation for taking part in the study,
and target participants also received partial course credit for their involvement.

SOCIAL SUPPORT AND LIPID REACTIVITY
115
Table 1 shows the relevant sample characteristics of the participants, by experimental
condition.
Physiological Measures
Blood lipid measures in this study included total cholesterol, HDL–c, and triglycer-
ides. These measures were assessed from blood samples drawn over a 2-min inter-
val at the end of the baseline, stressor, and recovery periods.
Total cholesterol was determined enzymatically on a Beckman CX4 analyzer
(
Allain, Poon, Chan, Richmond, & Fu, 1974). Triglycerides were first corrected
for free glycerol, and concentrations were then determined using Beckman reagent
Buccolo & David, 1973); all triglyceride values were log-transformed because of
(
nonnormal distributions. HDL–c was determined following sequential precipita-
tion with heparin–MnCl2 and dextran sulfate (Gidez, Miller, Gurnstein, Slagle, &
Eder, 1982). The laboratory responsible for assays participates in the national sur-
vey for clinical laboratories sponsored by the College of American Pathologists
and has participated in the Centers for Disease Control Lipid Standardization pro-
gram since 1977. The coefficients of variation for total cholesterol, triglycerides,
and HDL–c are 0.8, 1.5, and 2.0%, respectively.
Hematocrit and hemoglobin were measured in each blood sample to estimate
changes in plasma volume during the stressor (Dill & Costill, 1974). Analysis of
each of the lipid measures was accomplished after correction for stress-associated
plasma volume shifts (Patterson, Gottdiener, Hecht, Vargot, & Krantz, 1993;
Stoney & West, 1997).
TABLE 1
Sample Characteristics by Condition
Alone Conditiona
SD
.72
Frienda
Characteristic
M
M
SD
Age (years)
Height (in.)
Weight (lbs.)
Cook–Medley Hostility
ISEL total score
Tangible support
Appraisal support
Self-esteem support
Belonging support
18.53
64.92
140.26
20.3
34.6
9.1
19.95
63.74
138.72
19.1
33.3
9.0
8.4
8.3
7.8
2.63
6.67
12.96
7.8
6.2
1.6
1.9
1.8
1.8
5.37
11.49
7.9
4.0
1.6
9.1
1.1
8.3
2.0
8.2
1.2
Note. ISEL = Interpersonal Support Evaluation List.
a
n = 20.

116
STONEY AND FINNEY
Psychological Measures
In addition to testing the primary question of whether the social support manipula-
tion would influence lipid reactivity, we also wished to test two secondary ques-
tions. Some investigations have reported that hostility moderates the relation be-
tween social support and cardiovascular reactivity. We included a standard
measure of hostility to test a similar question with regard to lipid reactivity. The
only published investigations to date to investigate social support and lipid concen-
trations have examined self-report measures of social support. To replicate these
findings in a laboratory manipulation of social support, we included a measure of
perceived social support in this study.
The Cook–Medley Hostility Scale (Cook & Medley, 1954) is a 50-item
true–false scale derived from the Minnesota Multiphasic Personality Inventory
that has been associated with cardiovascular disease morbidity and mortality
(Barefoot, Dahlstrom, & Williams, 1983; Shekelle, Gale, Ostfeld, & Paul, 1983;
Williams et al., 1980). Responses to these 50 items are summed to yield a total hos-
tility score. Test–retest reliability of the Cook–Medley Hostility Scale is good and
ranges between .85 and .89 over 1–4 years (Shekelle et al., 1983).
The Interpersonal Support Evaluation List (ISEL; Cohen, Mermelstein,
Kamarck, & Hoberman, 1985) contains 40 true–false statements about the per-
ceived availability of social resources. We tested the relations between the overall
score of perceived social support, as well as the four subscales on this question-
naire. These subscales include tangible support, appraisal support, self-esteem
support, and belonging support (Cohen et al., 1985). The ISEL has high internal
consistency (alphas range from .77 to .86), and the test–retest reliability over a
4
-week interval is high (Cohen et al., 1985).
Stressor
A videotaped speech task was used in this study because it reliably elicits increases
in lipid parameters (Stoney et al., 1988). Target women were given 2 min to con-
struct and 3 min to deliver a speech about how they would respond to a hypothetical
situation. They were asked to suppose that after careful preparation for their best
friend’s wedding, a measurement mistake by the seamstress caused major faults in
their maid of honor dress just 1 week prior to the wedding. They were asked how
they would confront the store manager about the situation and asked to talk about
potential solutions to the problem. A video camera was placed approximately 5 feet
away, and target participants were asked to speak into the camera for the entire
3
-min period. Target participants were also informed that their speeches would be
evaluated for organization, clarity, and delivery. Individuals were prompted to con-
tinue speaking if they completed their speech before the end of the 3-min period.

SOCIAL SUPPORT AND LIPID REACTIVITY
117
Manipulation
The social support manipulation was based on the procedure used by Kamarck et al.
1990). For those participants randomized to the Friend condition, the friend re-
(
ceived support training during the time the target participant was being instru-
mented. Specifically, friends were instructed how to demonstrate support by lightly
touching the target participant on the wrist. This study involved the presence of a
friend and touching in the social support manipulation because these have been
shown to influence the effects of social presence on cardiovascular reactivity
(Christenfeld et al., 1997). The friends were specifically asked not to speak during
the protocol or to try and distract the target participant in any way.
For those participants who had been randomized to the Alone condition, the
friend remained in the hallway during the entire time that the target participant
completed the protocol.
Manipulation Checks
We employed four types of measures to check our stressor manipulation both with
and without the presence of a friend. First, we wished to assess the quality of the re-
lationship between the target participant and the recruited friend. We used a short
friendship assessment questionnaire that has been previously reported (Kamarck et
al., 1995) to assess overall perceived trustworthiness and loyalty of the friend who
was brought to the experimental session. Second, we wished to assess the impor-
tance of having a friend present during the experimental manipulation. We used
5
-point Likert-type scales to assess this during both tasks.
Third, we wished to assess how helpful and supportive the target participant felt
during the experiment. We used two, previously reported true–false questions
Kamarck et al., 1995) for this assessment. Finally, perceptions of the tasks them-
(
selves were assessed using 5-point Likert scales regarding perceived competency,
stress, confidence, amount of anger elicited, and task-difficulty.
Procedure
Target participants were asked to fast for 12 hr prior to coming to the laboratory,
and an interview confirmed that participants had complied with this requirement.
All participants were asked to have their same-sex friend accompany them to the
laboratory. Upon their arrival, the experimenter reviewed each of the respective
consent forms with the target participant and the friend. Once the consent forms
were read and signed by both participants, the target participant and friend were
separated; the target participant entered the laboratory where testing took place.

118
STONEY AND FINNEY
The target participant’s height and weight were obtained. Once this was com-
pleted, an indwelling catheter was inserted into the antecubical space of the
dominant arm.
For those participants randomized to the Friend condition, the following proce-
dure was followed. After instructions were administered to the friend, she was
brought into the laboratory, where the target participant was already instrumented,
and the experiment began. During the initial 30-min acclimation period, the target
participantwasaskedtorelaxinacomfortablechairandlistentorelaxingmusic.The
friend was instructed to begin showing active support as she had been instructed.
A 10-min baseline period followed the acclimation period, during which relax-
ing music was played. Two minutes before the end of baseline, a blood sample was
taken. Once the baseline period had been completed, instructions for the speech
task were given, and the video camera was turned on. Following a 2-min prepara-
tion period, participants were verbally prompted to begin speaking for 3 min. An-
other blood sample was taken 2 min before the end of the speech. Finally, target
participants completed a 10-min recovery period, during which time relaxing mu-
sic was again provided. A final blood sample was taken 2 min before the end of the
recovery period. Once the protocol was completed, the participants were asked to
complete the set of questionnaires and were then debriefed and excused.
For those participants randomized to the Alone condition, the target participant
completed the experiment while her friend remained in the hallway. Once the tar-
get participant had completed the experiment, the friend was brought in to the lab-
oratory for debriefing and to receive payment.
Data Reduction and Analysis
Demographic characteristics of the women in the Friend and Alone conditions
were compared with a series of one-way analyses of variance (ANOVAs) for each
of the relevant variables and for the baseline lipid parameters. Each of the variables
regarding perceptions of the stressor and support conditions was compared in a
similar manner. A series of one-way ANOVAs examining Time (Baseline vs.
Speech Stressor vs. Recovery) were performed for each of the lipid variables, to de-
termine which were significantly elevated during stress, relative to baseline.
Change scores were calculated by subtracting the baseline values from the corre-
sponding values during the stressor. A separate series of one-way ANOVAs on
change scores compared those in the Alone and Friend conditions on lipid reactiv-
ity. High- and low-hostile groups were formed by median splits on the total
Cook–Medley scores. Two (Condition: Alone vs. Friend) × 2 (Hostility: High vs.
Low) ANOVAs were performed on each of the lipid change scores, separately.
Finally, a series of one-way ANOVAs on the changes from baseline to stress com-
pared those scoring high and low on the ISEL on lipid reactivity.

SOCIAL SUPPORT AND LIPID REACTIVITY
119
Triglyceride values were log transformed prior to analyses; nontransformed
data are displayed in all tables for easier interpretation. All other parameters
were normally distributed. Post-hoc tests were accomplished using Tukey’s
HSD test statistic (Kirk, 1968). For all analyses involving a repeated factor with
more than two levels, the Greenhouse–Geiser epsilon correction was applied to
the error degrees of freedom.
RESULTS
Demographic Characteristics
The age of the participants in the two groups was slightly, but significantly, differ-
ent, F(1, 35) = 4.66, p = .04. Subsequent inspection of the means indicated that
women in the Alone condition (M = 18.5 years) were somewhat younger than were
women in the Friend condition (M = 19.9 years). However, women in the two
groups did not differ significantly with respect to height, F(1, 38) = 2.45, p = .13, or
weight, F(1,38) = .03, p = .86. There were no group differences in any of the
psychosocial variables measured (see Table 1).
Manipulation Checks
Women in the Friend and Alone conditions rated their friends equally important in
their lives. Thus, the overall perceived quality of the relationship prior to task expo-
sure was the same in both groups. As expected, participants in the Friend condition
rated the presence of their friend significantly more helpful during the session than
did participants in the Alone condition, F(1, 37) = 13.86, p = .001. Participants in
the Friend condition were also more likely to report that it is important to have a
friend present in order to do well on the task, relative to those in the Alone condi-
tion, F(1, 37) = 5.54, p = .024. The analysis of perceived stress ratings during the
tasks also indicated Group differences, but in the opposite direction. Specifically,
women in the Friend condition reported significantly more anger, F(1, 37) = 5.61, p
=
.023, and slightly more stress, F(1, 37) = 2.87, p = .09, during the speech task, rel-
ative to those in the Alone condition. No other task impressions or friendship-re-
lated variables differentiated the groups.
Baseline Measures
The one-way ANOVAs on baseline lipid values indicated that participants in both
groups had similar baseline concentrations of total cholesterol (all ps > .05).

120
STONEY AND FINNEY
Effects of Stress on Lipid Parameters
The mean values for each of the lipid and hematologic parameters during baseline,
stress, and recovery periods by condition are presented in Table 2. The ANOVAs
and subsequent post-hoc tests indicated that total cholesterol, F(2, 46) = 3.84, p =
.
029, and triglycerides, F(2, 46) = 4.50, p = .016, were significantly elevated during
the stress period, relative to baseline, ps < .05. There was no significant stressor
effect for HDL–c (p = .09).
Effects of the Presence of a Friend on Lipid Reactivity
Subsequent analyses compared women in the Alone and Friend conditions on the
changes in each of the lipid parameters during stress. Women in the Friend condi-
tion had higher total cholesterol reactivity during stress, F(1, 29) = 3.97, p = .056,
relative to women in the Alone condition (see Figure 1). No other significant condi-
tion main effects emerged in these analyses.
Hostility did not interact with the social support condition for any of the lipid
parameters, as reflected in the nonsignificant interaction terms for these analyses
(all ps > .20).
When comparing those scoring high and low on the ISEL and each of the
subscales, no group differences were apparent for lipid reactivity, all ps > .11.
TABLE 2
Means and Standard Deviations of Lipids During Baseline,
Stress, and Recovery by Condition
Baseline
Speech Task
Recovery
Condition
M
SD
M
SD
M
SD
Alone
Total cholesterol (mg/dl)
Triglycerides (mg/dl)
HDL–c (mg/dl)
Friend
145.60
53.50
43.80
29.52
22.02
8.94
147.69
53.67
44.14
30.90
20.38
8.23
146.21
50.61
43.48
31.55
22.78
8.26
Total cholesterol (mg/dl)
Triglycerides (mg/dl)
HDL–c (mg/dl)
145.80
49.47
50.33
29.96
25.92
10.21
153.10
50.80
52.33
30.77
25.95
12.31
150.45
48.62
52.97
27.69
26.74
12.65
Note. HDL–c = high-density lipoprotein–cholesterol.

SOCIAL SUPPORT AND LIPID REACTIVITY
121
FIGURE 1 Comparison of total cholesterol (mg/dl) reactivity during a speech stressor be-
tween participants in the Alone condition and those in the Friend condition.
DISCUSSION
Acute psychological stress was associated with reliable increases in the blood con-
centrations of total cholesterol and triglycerides in this sample of young, healthy
women. The magnitude of this increase is similar to what has been previously re-
ported (Niaura et al., 1992; Stoney et al., 1988) and was apparent after correction
for stress-induced plasma volume shifts. Although the physiological importance of
short-term lipid elevations is not yet known, indirect evidence suggests that they
may play a role in risk for CHD. For example, cross-sectional studies have demon-
strated larger magnitude elevations in the atherogenic lipids during acute stress
among healthy individuals at risk for the future development of CHD, relative to
those at lower risk (Davis & Matthews, 1990; Groover, Jernigan, & Martin, 1960;
Stoney & Hughes, 1999). The current data replicate previous findings that blood
lipid concentrations are transiently and mildly elevated during acute psychological
stress in healthy young individuals.
The main thrust of this study was to investigate whether the presence of a sup-
portive friend would alter lipid reactivity to an acute stressor. The results indicate
that the social manipulation induced significantly greater increases in total choles-
terol in the women with a friend present during the stressor, relative to the women
who were alone during the stressor. To our knowledge, no other published study
has tested lipid reactivity using this model. These findings suggest that the pres-
ence of a supportive friend during a challenging task actually increases physio-
logic responsivity to psychological stress. These findings are similar to the results
of some studies testing the effects of the presence of a supportive other on the car-
diovascular system (Lepore, Allen, & Evans, 1993; Snydersmith & Cacioppo,
1
992). For example, one investigation reported that the presence of a supportive
friend enhances cardiovascular and skin conductance reactivity to psychological

122
STONEY AND FINNEY
stressors (Allen et al., 1991). Another recent investigation reported increased car-
diovascular reactivity as a function of increased perceived social support (Roy,
Steptoe, & Kirschbaum, 1998). This latter study did not utilize the presence of a
friend as a model for social support, but the results call into question the ecological
validity of such a model.
One possible explanation for the increased lipid reactivity noted in this study is
that the presence of a supportive friend increases the threat appraisal by the partici-
pant. In his review of the cardiovascular literature reporting increased autonomic
reactivity to stress in the presence of a partner, Kamarck et al. (1995) noted that
each of these studies included conditions where partners viewed task performance
but did not provide verbal feedback regarding that performance. Thus, the absence
of a specific condition to reduce the threat associated with social evaluation may
have resulted in the enhanced cardiovascular reactivity. In this study, there was no
explicit, verbal condition by which the supportive friend could reduce appraisals
of threat. Additionally, there was a tendency for those with a friend present to re-
port more stress during the speech task. Thus, it is possible that participants in this
study appraised the presence of their friends as potential social threats, resulting in
enhanced lipid reactivity to stress. Future studies could explicitly test this notion
by manipulating the degree of social threat and evaluation.
The increased lipid reactivity in this study may be a function of the touch ma-
nipulation that we used. A previous study demonstrated that the presence of a
friend or other who showed support using touch resulted in larger HR and blood
pressure reactivity, relative to a condition not utilizing touch (Edens et al., 1992).
However, touch has also been reported to reduce subjective stress (Lynch,
Thomas, Paskewitz, Katcher, & Weir, 1977), and other studies of cardiovascular
reactivity have not demonstrated enhanced stress reactivity during a touch manip-
ulation (Kamarck et al., 1990). Future investigations including both a touch and
no-touch condition using this paradigm will be necessary to determine the extent
to which the touch manipulation is relevant to lipid responsivity to stressors.
Although the bulk of the cardiovascular reports are in contrast to the findings
reported here for lipid reactivity, there is considerable variability in the existing lit-
erature (Uchino, Cacioppo, & Kiecolt-Glaser, 1996). Although several investiga-
tions of cardiovascular reactivity have found that the presence of a friend or
stranger attenuates cardiovascular reactivity to stress (Gerin et al., 1995; Kamarck
et al., 1995; Kamarck et al., 1990), others have failed to find an effect (Sheffield &
Carroll, 1994; Spitzer, Llabre, Ironson, Gellman, & Schneiderman, 1992). When
social presence has been shown to attenuate reactivity, the finding does not appear
to be uniform for men and women or for all response systems measured
(Kirschbaum, Klauer, Filipp, & Hellhammer, 1995; Linden et al., 1993). For ex-
ample, the Linden et al. (Linden et al., 1993) study found that ambulatory SBP was
associated with self-reported high social support in women but not in men. No ef-
fects of social support were apparent for ambulatory diastolic blood pressure or

SOCIAL SUPPORT AND LIPID REACTIVITY
123
HR for either men or women. Other studies suggest that the physiological effects
of this type of manipulation may be dependent on individual differences in psy-
chological variables. For example, one study showed that the presence of a sup-
portive friend resulted in diminished blood pressure reactivity only among
individuals low in cynicism; no effects of the presence of a friend were apparent
among those high in cynicism (Lepore, 1995). In this study of women only, we did
not find any evidence for an interaction between the presence of a friend and hos-
tility. However, there may be other important individual differences in psycholog-
ical factors, particularly for women, that may interact to result in differential lipid
reactivity to stressors.
One potential limitation of this study is that the baseline physiological measures
were assessed after the manipulation (presence or absence of the friend) occurred.
Although this was necessary for pragmatic reasons, it does not appear to explain the
reactivitydifferencesnoted,becausetherewerenobaselinedifferencesbetweenthe
two groups for any of the physiological measures assessed. Thus, it appears that the
presence of the friend had effects on lipids during the stressor only but not during
baseline.Anotherpotentiallimitationofthisstudyconcernstheexclusiveinvestiga-
tionofyoung, healthy, educatedwomen. Giventhegenderdifferencesincardiovas-
cular disease progression and in some investigations of social support and lipid
concentrations,studiesinourlaboratoryarefurtherinvestigatingtheextenttowhich
the current findings are generalizable to men and other populations. Several previ-
ous investigations have reported a significant and negative association between so-
cial support and blood lipid concentrations at rest (Gliksman et al., 1995; Thomas et
al., 1985). However, two of the three studies were investigating a healthy elderly
population. In contrast, a recent investigation of older African Americans reported
oppositefindings(Waldstein,Toth,&Poehlman,1998).Theseinvestigatorsreported
thatthosewithmoresocialsupporthadhighertotalcholesterolandlowdensitylipo-
protein–cholesterol (LDL–c) concentrations, relative to those with lower levels of
social support. Investigation of similar relationships among a healthy young adult
population has resulted in finding either no association for social support and blood
concentrations of cholesterol (Knox, Jacobs, Chesney, Raczynski, & McGreath,
1
996;Terborg, Hibbard, &Glasgow, 1995)oronlylimitedsupportforapositiveas-
sociation(Undenetal., 1995). Inthisstudy, wefoundnoevidenceforanassociation
between overall social support, as measured by the ISEL, and lipids either at rest or
duringthestressor. Nopreviousstudyhasinvestigatedtheimpactofsocialpresence
on lipids at rest and during stress. This study suggests that the presence of a friend
may, in some circumstances, actually increase cholesterol stress responses.
REFERENCES
Allain, C. C., Poon, L. S., Chan, C. S. G., Richmond, W., & Fu, P. C. (1974). Enzymatic determination of
total serum cholesterol. Clinical Chemistry, 20, 470–475.

124
STONEY AND FINNEY
Allen, K. M., Blascovich, J., Tomaka, J., & Kelsey, R. M. (1991). Presence of human friends and pet
dogs as moderators of autonomic responses to stress in women. Journal of Personality and Social
Psychology, 61, 582–589.
Avlund, K., Damsgaard, M. T., & Holstein, B. E. (1998). Social relations and mortality. An eleven year
follow-up study of 70-year-old men and women in Denmark. Social Science and Medicine, 47,
6
35–643.
Barefoot, J. C., Dahlstrom, W. B., & Williams, R. B. (1983). Hostility, CHD incidence and total mortal-
ity: A 25-year follow-up study of 255 physicians. Psychosomatic Medicine, 45, 59–63.
Berkman, L. F., & Syme, S. L. (1979). Social networks, host resistance, and mortality: A nine-year fol-
low-up study of Alameda County residents. American Journal of Epidemiology, 109, 186–204.
Brindley, D. N., McCann, B. S., Niaura, R., Stoney, C. M., & Suarez, E. C. (1993). Stress and lipoprotein
metabolism: Modulators and mechanisms. Metabolism, 42(9, Suppl. 1), 3–15.
Broadhead, W. E., Kaplan, B. H., James, S. A., Wagner, E. H., Schoenbach, V. J., Grimson, R., Heyden,
S., Tibblin, G., & Gehlbach, S. H. (1983). The epidemiological evidence for a relationship between
social support and health. American Journal of Epidemiology, 117, 521–537.
Buccolo, G., & David, H. (1973). Quantitative determination of serum triglycerides by the use of en-
zymes. Clinical Chemistry, 19, 476–482.
Christenfeld, N., Gerin, W., Linden, W., Sanders, M., Mathur, J., Deich, J. D., & Pickering, T. G. (1997).
Social support effects on cardiovascular reactivity: Is a stranger as effective as a friend? Psychoso-
matic Medicine, 59, 388–398.
Cohen, S. (1988). Psychosocial models of the role of social support in the etiology of physical disease.
Health Psychology, 7, 269–297.
Cohen, S., Kaplan, J. R., & Manuck, S. B. (1994). Social support and coronary heart disease: Underlying
psychological and biological mechanisms. In S. A. Shumaker & S. M. Czajkowski (Eds.), Social
support and cardiovascular disease (pp. 195–221). New York: Plenum .
Cohen, S., Mermelstein, R., Kamarck, T., & Hoberman, H. M. (1985). Measuring the functional compo-
nents of social support. In I. G. Sarason & B. R. Sarason (Eds.), Social support: Theory, research
and applications (pp. 73–94). The Hague, Netherlands: Martinus Nijhoff.
Cohen, S., & Syme, S. L. (Eds.). (1985). Social support and health. Orlando, FL: Academic.
Cook, W. W., & Medley, D. M. (1954). Proposed hostility and pharisaic-virtue scales for the MMPI.
Journal of Applied Psychology, 38, 414–418.
Davis, M. C., & Matthews, K. A. (1990). Cigarette smoking and oral contraceptive use influence
women’s lipid, lipoprotein, and cardiovascular responses during stress. Health Psychology, 9,
7
17–736.
Dill, D. B., & Costill, D. L. (1974). Calculation of percentage changes in volumes of blood, plasma, and
red cells in dehydration. Journal of Applied Physiology, 37, 247–248.
Edens, J. L., Larkin, K. T., & Avel, J. L. (1992). The effect of social support and physical touch on car-
diovascular reactions to mental stress. Journal of Psychosomatic Research, 36, 371–382.
Gerin, W., Milner, D., Chawla, S., & Pickering, T. G. (1995). Social support as a moderator of cardio-
vascular reactivity in women: A test of the direct effects and buffering hypotheses. Psychosomatic
Medicine, 57, 16–22.
Gidez, L. I., Miller, G. J., Gurnstein, M., Slagle, S., & Eder, H. A. (1982). Separation and quantification
of subclasses of human plasma high density lipoproteins by a simple precipitation procedure. Jour-
nal of Lipid Research, 23, 1206–1223.
Gliksman, M. D., Lazarus, R., Wilson, A., & Leeder, S. R. (1995). Social support, marital status and liv-
ing arrangement correlates of cardiovascular disease risk factors in the elderly. Social Science and
Medicine, 40(6), 811–814.
Groover, M. E., Jr., Jernigan, J. A., & Martin, C. D. (1960). Variations in serum lipid concentration and
clinical coronary disease. American Journal of the Medical Sciences, February, 53/133–59/139.

SOCIAL SUPPORT AND LIPID REACTIVITY
125
Hazuda, H. P. (1994). A critical evaluation of U.S. epidemiological evidence and ethnic variation. In S.
A. Shumaker & S. M. Czajkowski (Eds.), Social support and cardiovascular disease (p. 360). New
York: Plenum.
House, J. S., Landis, K. R., & Umberson, D. (1988). Social relationships and health. Science, 241,
5
40–545.
Kamarck, T. W., Annunziato, B., & Amateau, L. M. (1995). Affiliation moderates the effects of social
threat on stress-related cardiovascular responses: Boundary conditions for a laboratory model of so-
cial support. Psychosomatic Medicine, 57, 183–194.
Kamarck, T. W., Manuck, S. B., & Jennings, J. R. (1990). Social support reduces cardiovascular reactiv-
ity to psychological challenge: A laboratory model. Psychosomatic Medicine, 52, 42–58.
Kirk, R. E. (1968). Experimentaldesign:Procedures for thebehavioralsciences. Belmont, CA:Brooks/Cole.
Kirschbaum, C., Klauer, T., Filipp, S.-H., & Hellhammer, D. H. (1995). Sex-specific effects of social
support on cortisol and subjective responses to acute psychological stress. Psychosomatic Medicine,
5
7, 23–31.
Knox, S. S. (1993). Perception of social support and blood pressure in young men. Perceptual and Mo-
tor Skills, 77, 132–134.
Knox, S. S., Jacobs, D. R., Chesney, M. A., Raczynski, J., & McGreath, H. (1996). Psychosocial factors
and plasma lipids in Black and White young adults: The Coronary Artery Risk Development in
Young Adults Study data. Psychosomatic Medicine, 58, 365–373.
Knox, S. S., & Uvnäs-Moberg, K. (1998). Social isolation and cardiovascular disease: An
atherosclerotic pathway? Psychoneuroendocrinology, 23, 877–890.
Kors, D., Linden, W., & Gerin, W. (1997). Evaluation interferes with social support: Effects on cardio-
vascular stress reactivity. Journal of Social and Clinical Psychology, 16(1), 1–23.
Lepore, S. J. (1995). Cynicism, social support, and cardiovascular reactivity. Health Psychology, 14(3),
2
10–216.
Lepore, S. J., Allen, K. A. M., & Evans, G. W. (1993). Social support lowers cardiovascular reactivity to
an acute stressor. Psychosomatic Medicine, 55, 518–524.
Linden, W., Chambers, L., Maurice, J., & Lenz, J. W. (1993). Sex differences in social support, self-de-
ception, hostility, and ambulatory cardiovascular activity. Health Psychology, 12, 376–380.
Lipid Research Clinics Program. (1984). The lipid research clinics coronary primary prevention trial re-
sults. I. Reduction in incidence of coronary heart disease. Journal of American Medical Association,
2
51, 351–364.
Lynch, J. J., Thomas, S. A., Paskewitz, D. A., Katcher, A. H., & Weir, L. O. (1977). Human contact and
cardiac arrhythmia in a coronary care unit. Psychosomatic Medicine, 39, 188–192.
McCann, B. S., Benjamin, A. H., Wilkinson, C. W., Carter, J., Retzlaff, B. M., Russo, J., & Knopp, R. H.
(1996). Variations in plasma lipid concentrations during examination stress. International Journal
of Behavioral Medicine, 3, 251–265.
Niaura, R., Stoney, C. M., & Herbert, R. N. (1992). Lipids in psychological research: The last decade.
Biological Psychology, 34, 1–43.
Orth-Gomér, K. (1994). International epidemiological evidence for a relationship between social sup-
port and cardiovascular disease. In S. A. Shumaker & S. M. Czajkowski (Eds.), Social support and
cardiovascular disease (pp. 97–117). New York: Plenum.
Orth-Gomér, K., Horsten, M., Wamala, S. P., Mittleman, M. A., Kirkeeide, R., Svanes, B., Rydén, L., &
Schenck-Gustafsson, K. (1998). Social relations and extent and severity of coronary artery disease:
The Stockholm Female Coronary Risk Study. European Heart Journal, 19, 1648–1656.
Patterson, S. M., Gottdiener, J. S., Hecht, G., Vargot, S., & Krantz, D. S. (1993). Effects of acute mental
stressonserumlipids:Mediatingeffectsofplasmavolume. PsychosomaticMedicine, 55, 525–532.
Roy, M. P., Steptoe, A., & Kirschbaum, C. (1998). Life events and social support as moderators of indi-
vidual differences in cardiovascular and cortisol reactivity. Journal of Personality and Social Psy-
chology, 75, 1273–1281.

126
STONEY AND FINNEY
Ruberman, W., Weinblatt, E., Goldberg, J. D., & Chaudhary, B. S. (1984). Psychosocial influences on
mortality after myocardial infarction. New England Journal of Medicine, 311, 552–559.
Schaefer, E. J., Lamon-Fava, S., Ordovas, J. M., Cohn, S. D., Schaefer, M. M., Castelli, W. P., & Wilson,
P. W. F. (1994). Factors associated with low and elevated plasma high density lipoprotein choles-
terol and apolipoprotein A-1 levels in the Framingham offspring study. Journal of Lipid Research,
3
5, 871–882.
Seeman, T. E., & Syme, L. (1987). Social networks and coronary artery disease: A comparison of the
structure and function of social relations as predictors of disease. Psychosomatic Medicine, 49,
3
41–354.
Sheffield, D., & Carroll, D. (1994). Social support and cardiovascular reactions to active laboratory
stressors. Psychology and Health, 9, 305–316.
Shekelle, R. B., Gale, M., Ostfeld, A. M., & Paul, P. (1983). Hostility, risk of coronary disease, and mor-
tality. Psychosomatic Medicine, 45, 109–114.
Shumaker, S. A., & Czajkowski, S. M. (Eds.). (1994). Social support and cardiovascular disease. New
York: Press.
Shye, D., Mullooly, J. P., Freeborn, D. K., & Pope, C. R. (1995). Gender differences in the relationship
between social network support and mortality: A longitudinal study of an elderly cohort. Social Sci-
ence and Medicine, 41, 935–947.
Snydersmith, M. A., & Cacioppo, J. T. (1992). Parsing complex social factors to determine component
effects: I. Autonomic activity and reactivity as a function of human association. Journal of Social
and Clinical Psychology, 11, 263–278.
Spitzer, S. B., Llabre, M. M., Ironson, G., Gellman, M. I., & Schneiderman, N. (1992). The influence of
social situations on ambulatory blood pressure. Psychosomatic Medicine, 54, 79–86.
Stoney, C. M., & Hughes, J. W. (1999). Lipid reactivity among men with a parental history of myocar-
dial infarction. Psychophysiology, 36, 1–7.
Stoney, C. M., Matthews, K. A., McDonald, R. H., & Johnson, C. A. (1988). Sex differences in lipid, li-
poprotein, cardiovascular, and neuroendocrine responses to acute stress. Psychophysiology, 25,
6
45–656.
Stoney, C. M., Niaura, R., Bausserman, L., & Matacin, M. (1999). Lipid reactivity to stress: I. Compari-
son of chronic and acute stress responses in middle-aged airline pilots. Health Psychology, 18,
2
41–250.
Stoney, C. M., & West, S. (1997). Lipids, personality, and stress: Mechanisms and modulators. In M.
Hillbrand & R. T. Spitz (Eds.), Lipids and human behavior (pp. 47–66). Washington, DC: American
Psychological Association.
Terborg, J. R., Hibbard, J., & Glasgow, R. E. (1995). Behavior change at the worksite: Does social sup-
port make a difference? American Journal of Health Promotion, 10(2), 125–131.
Thomas, P. D., Goodwin, J. M., & Goodwin, J. S. (1985). Effect of social support on stress-related
changes in cholesterol level, uric acid level, and immune function in an elderly sample. American
Journal of Psychiatry, 142, 6–8.
Uchino, B., N., Cacioppo, J. T., & Kiecolt-Glaser, J. K. (1996). The relationship between social support
and physiological processes: A review with emphasis on underlying mechanisms and implications
for health. Psychological Bulletin, 119, 488–531.
Unden, A.-L., Krakau, I., Hogbom, M., & Romanus-Egerborg, I. (1995). Psychosocial and behavioral fac-
tors associated with serum lipids in university students. Social Science and Medicine, 41, 915–922.
Waldstein, S. R., Toth, M. J., & Poehlman, E. T. (1998, March). Social support and lipoprotein lipids in
older African-Americans: A negative association. Paper presented at the meeting of the Society of
Behavioral Medicine, New Orleans, LA.
Williams, R. B., Jr., Hanley, T. L., Lee, K. L., Kong, Y. H., Blumenthal, J. A., & Whalen, R. E. (1980).
Type A behavior, hostility, and coronary atherosclerosis. Psychosomatic Medicine, 42, 539–549.