Social Support 61“ndings have been reported by others (Gerin, Milnor,
Chawla, & Pickering, 1995; Lepore, Allen, & Evans, 1993;
Christenfeld et al., 1997; Gerin, Pieper, Levy, & Pickering,
1992).
Several studies have evaluated whether individual differ-
ence factors such as cynical hostility moderate the effects of
social support on cardiovascular reactivity. For example,
Lepore (1995) found that subjects with high scores on cynical
hostility did not have lowered cardiovascular activity when
provided with social support, whereas those subjects with
lower scores on cynical hostility did derive bene“t from
social support.
Because correlational studies are limited in terms of
causal inferences that can be made, intervention studies that
manipulate social support may provide more insight into the
relation between social support and cardiovascular function.
Only a small number of studies manipulate social support.
Overall, these studies found that social support interventions
result in reduced blood pressure when the participants under-
went a stressor challenge assessment postintervention. For
example, Sallis, Grossman, Pinski, Patterson, and Nader
(1987) randomly enrolled participants in a support education
group (support group), relaxation training, or a multicompo-
nent stress management intervention, and intervention or a
control group. Results indicated that support education and
relaxation training intervention resulted in smaller increases
in DBP from preintervention to follow-up, and lower DBP
levels during recovery from a cold pressure stress test,
compared with the multicomponent stress management inter-
vention. Among individuals at higher cardiovascular risk
(hypertensives), studies have consistently shown that inter-
ventions focusing on increasing positive support, particularly
support provided by family, result not only in short-term
decreases in DBP, but also in long-term effects on blood pres-
sure regulation (Levine et al., 1979). Indeed, a recent meta-
analysis of these intervention studies suggests that social
support manipulations can assist in the reduction of blood
pressure (Uchino et al., 1996).
Several studies have examined the link between
social support and cholesterol. Welin and colleagues (Welin,
Rosengren, & Wilhelmsen, 1996) found that low serum cho-
lesterol was associated with low social support in a study of
middle-aged men. This association was also found in a study
of healthy women, even after controlling for the effects of
other psychosocial factors including depression and recent
life events, and lifestyle factors (smoking, alcohol, obesity)
(Horsten, Wamala, Vingerhoets, & Orth-Gomer, 1997).
These “ndings are interesting from a clinical perspective, as
low lipid levels have been associated with increased mortal-
ity from violent causes (Muldoon et al., 1993).
Endocrine FunctionThe most commonly studied endocrine measures are the cate-
cholamines (e.g., norepinephrine [NE] and epinephrine [EPI])
and cortisol. Studies evaluating endocrine function are impor-
tant because of its association with the cardiovascular and
immune systems. Catecholamines play an important role in
cardiovascular regulation functions such as constriction of
arterial blood vessels. The association between endocrine
function and social support has not been well documented.
The majority of these studies have found an association be-
tween social support and catecholamine levels (e.g., Seeman,
Berkman, Blazer, & Rowe, 1994; Fleming, Baum, Gisriel, &
Gatchel, 1982). For example, Ely and Mostardi (1986) studied
331 men and found that high social support, de“ned as social
resources and marital status, was associated with lower NE
than low social support. However, studies of cortisol and sup-
port suggest that increasing social contact does not in”uence
cortisol levels. One study examined the association between
support from a stranger or partner and cortisol reactivity dur-
ing acute psychological stress (Kirschbaum, Klauer, Filipp, &
Hellhammer, 1995). The results indicated that men who re-
ceived support from their partners evidenced lower cortisol
levels than men who received stranger support or no support.
However, women evidenced a trend toward greater cortisol
response during the partner-supported conditions compared
with the other two conditions. Overall, the link between social
support and endocrine function has not been very consistently
con“rmed.Immune FunctionStudies linking social support to immune function indices
suggest that higher social support is associated with better
immune system function. Levy and colleagues (1990) exam-
ined the association between perceived emotional support
from spouse, family member, friends, doctors, and nurses and
the immune system function in women with breast cancer.
The results indicated that emotional support from spouse
and physician was associated with greater natural killer cell
activity (NKCA). Some studies have controlled for the in”u-
ence of other psychological factors, such as mood and stress-
ful life events, that could contribute to the association
between support and immune function. Baron and colleagues
(Baron, Cutrona, Hicklin, Russell, & Lubaroff, 1990) evalu-
ated the association between perceived support and immune
indexes among spouses of cancer patients and found that all
aspects of support were related to phytoheammagluttinin
(PHA) and NKCA, even after controlling for life events
and depression. Kiecolt-Glaser, Dura, and Speicher (1991)