Review: Lack of social support reduces survival after first myocardial infarction
ACP J Club. 1994 Nov-Dec;121:80. doi:10.7326/ACPJC-1994-121-3-080
Bucher HC. Social support and prognosis following first myocardial infarction. J Gen Intern Med. 1994 Jul;9:409-17.
To examine the contribution of psychosocial factors, particularly social support, to the prognosis of patients after a first myocardial infarction (MI).
MEDLINE was searched using the terms coronary heart disease, myocardial infarction, psychosocial factors, social support, and social isolation.
Studies were selected if an inception cohort with a complete report of follow-up was identified; objective outcome criteria were considered; and adjustment of confounding factors was reported.
Data on measurement of social support; outcomes (cardiac death, all-cause mortality, sudden death, and reinfarction); and the relative risk (RR) for the outcome were extracted from both cohort studies and intervention studies.
Of 523 articles reviewed, 9 cohort studies evaluating 11 675 patients after their first MI and 2 randomized controlled trials of psychological support interventions in 818 patients met the inclusion criteria. 7 of the 9 cohort studies showed an association between lack of social ties and increased mortality from coronary heart disease. 4 of the 7 studies included details about the social networks measured (size, presence of emotional support, and number of contacts). The 3 remaining studies measured social support as living either alone or with someone. An increased risk for cardiac death with lower social support was shown in 4 of the 7 studies. Individual studies showed that cardiac death was associated with living alone (RR 1.8); lack of emotional support (RR 2.9, 95% CI 1.2 to 6.9); network scope, contact frequency, and size (RR 1.47, CI 1.17 to 1.85; 1.42, CI 1.12 to 1.81; and 1.31 CI 1.09 to 1.56; respectively); and lack of a spouse and close confidants (RR 3.34, CI 1.84 to 6.20). 1 study showed an association between a high level of stress and lack of a close confidant with sudden death (RR 5.62), another showed a trend toward an association between total cardiac events and living alone (RR 1.58, CI 0.91 to 2.74), and a third showed an association between total mortality and lack of spouse (P for both men and women < 0.025). 1 trial of social support intervention showed decreased sudden death rates (P = 0.01) and cardiac death rates (P = 0.02) and a trend toward lower reinfarction rates; the other trial showed reduced reinfarction rates (P = 0.028) and a trend toward lower sudden death (P = 0.08) and cardiac death (P = 0.21).
Lack of social support is a risk factor for poor outcome in patients who have a first myocardial infarction.
Source of funding: Not stated.
For article reprint: Dr. H.C. Bucher, Medizinische Universitäts-Poliklinik, Kantonsspital Basel, CH-4031 Basel, Switzerland. FAX 41-61-265-43-00.
Common sense suggests that patients with no one to turn to might be worse off after a heart attack than those with good social supports. The review by Bucher examined this hypothesis and found a reasonably consistent association between less social support and higher cardiac mortality and morbidity after a first MI. 7 of 9 cohort studies found a moderately strong association (RRs ranged from 1.5 to 5.6) and acceptable precision (most CIs excluded 1.0). The 2 experimental studies supported a relation between social support interventions and improved outcome after MI.
Some unanswered questions remain about the evidence. The search strategy was specified for this systematic review, but no unpublished sources were identified. Although "adequate adjustment" for confounding was used as one of the specific criteria for study selection, the details of how, and how well, each of the primary studies accomplished this important task were not presented. Selected strengths and weaknesses of some of the studies were presented, but consistent application of these critical appraisal criteria would have been helpful. As the author points out, the science of measuring social supports is not yet fully developed or validated, making interpretation of these results difficult. Nevertheless, this high-quality review yields important information for clinicians. However, because so many biological factors are also associated with poor outcome after MI (1, 2), it would be useful to know the relative contribution of specific biological and psychosocial variables in predicting poor outcomes.
How does this evidence bear on clinical practice? Data obtained by clinicians about their patients' families, friends, and colleagues may help to classify patients into those at greater or lesser risk for cardiac morbidity and mortality.
This review may prompt clinicians to encourage post-infarction patients who have social supports to accept their help. Whether to intervene when a patient lacks social support and how to select the best method of support are challenges that await future randomized controlled trials.
Scott Richardson, MD
University of Rochester School of Medicine and DentistryRochester, New York, USA