62 Coping and Social Support
conducted a prospective longitudinal study of caregivers to
individuals with Alzheimer•s disease. Caregivers were as-
sessed twice in a one-year time period. Both positive and un-
helpful support were assessed. Results indicated a buffering
effect for social support; caregivers low in positive support
evidenced greater negative changes in immune response
(Con A, PHA, and EBV) after controlling for age, income,
and depression. Similar “ndings among caregivers were re-
ported by Esterling and colleagues (Esterling, Kiecolt-Glaser,
Bodnar, & Glaser, 1994). Persson and colleagues (Persson,
Gullberg, Hanson, Moestrup, & Ostergren, 1994) reported
that low social participation, low satisfaction with social par-
ticipation, and low emotional support were associated with
CD4 counts compared with HIV-positive men without
AIDS who scored high on these social support measures. The
associations were stronger when age and length of time since
treatment were taken into account. Studies of older popula-
tions have consistently found an association between support
and immunity (e.g., Seeman et al., 1994; Thomas, Goodwin,
& Goodwin, 1985). Finally, Ward and colleagues (1999)
found an association between perceived adequacy of social
support and immune parameters associated with systemic
lupus erythmetosus activity (SLE). Greater SLE activity was
associated with less adequate social support. However, sev-
eral studies have not found an association between social sup-
port and immunological outcomes (Arnetz et al., 1987;
Kiecolt-Glaser et al., 1985; Perry, Fishman, Jacobsberg, &
Frances, 1992).
Social Support and Disease Recovery
Cardiac disease is the most studied disease when the role of so-
cial support is being considered. There is evidence that social
support in”uences recovery from cardiac events. Ostergren
et al. (1991) found that practical support predicted improve-
ment in physical working capacity among a small group of
40 persons admitted with “rst-time myocardial infarction
(MI). Yates (1995) interviewed a mixed group of patients post-
MI, coronary artery bypass grafting, and/or coronary artery
angioplasty. Emotional information provided during the re-
covery from spouse and health care provider, along with per-
ceived physical recovery were evaluated. Results indicated
that greater satisfaction with health care provider support
was associated with 1-year perceived physical recovery. This
study is limited because self-reported physical health was
measured using a single item perceived health measure, which
is quite subjective.
Hamalainen and colleagues (2000) reported a small asso-
ciation between support factors (de“ned as formal services),
semi-formal assistance, and informal social support (network
size, frequency of contacts, availability, and reciprocity in
relationships) and either functional capacity or working ca-
pacity (de“ned using a bicycle er gometer test as well as func-
tional limitations) among 147 MI and 159 coronary artery
bypass patients. High functional capacity at one year was as-
sociated with less assistance and emotional support in both
patient groups. It is dif“cult to infer causality from these “nd-
ings, as the need for assistance may be driven by poor func-
tional capacity. However, the authors also suggest that it is
possible that supportive family members may actually lead to
poorer health outcomes because family members overprotect
the patient during the recovery by reinforcing unhealthy
sedentary behaviors.
Social support has also been investigated as a factor pre-
dicting readmissions among patients with ischemic heart dis-
ease (MI, unstable angina, stable angina). Stewart, Hirth,
Klassen, Makrides, and Wolf (1997) did not “nd signi“cant
differences in total social support or support from different
network sources that predicted readmission among patients
with a history of multiple admissions.Disease Progression and MortalityAIDSA relatively large literature evaluates the association between
social support and human immunode“ciency virus (HIV)
progression in gay and bisexual men. Theorell and colleagues
(1995) evaluated the association between perceived support
and CD4 T-lymphocyte levels in HIV-infected hemophil-
iacs, and found that lower perceived support was associated
with greater declines in CD4 levels over a “ve-year period.
Patterson and colleagues (1996) followed a large group of
HIV-positive men over a “ve-year period, using measures of
CD4counts, symptomatology, AIDS diagnosis, and mortal-
ity as outcome variables. Social support was assessed as
received informational and emotional support, as well as net-
work size (number of social contacts). Results indicated that a
larger network size was actually associated with a shorter
symptom-free period among individuals who were symptom
free at baseline. After controlling for this interaction, higher
ratings of informational support predicted a longer time until
the onset of an AIDS-de“ning opportunistic infection. After
controlling for depressive symptoms, the size of the social
network was a predictor of mortality among individuals
with symptoms at baseline. Individuals with 15 persons
in their network had an 84% chance of remaining alive after
48 months, while those who listed only two people had a 44%
chance. Among participants that were symptomatic at base-
line, higher ratings of informational support predicted a
longer survival time after controlling for depressive symp-
toms and network size. Overall, support played a mixed role