| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
REVIEWS |
From the Department of Psychiatry and Behavioral Science, Duke University Medical Center, Durham, North Carolina.
Address correspondence to Heather S. Lett, Psychology Service 116B, VA Palo Alto Health Care System, 3801 Miranda Ave., Palo Alto, CA 94304. E-mail: heather.lett{at}alumni.duke.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONCONCLUSIONSNOTESREFERENCES |
|---|
Methods: Articles for the primary review of social support as a risk factor were identified with MEDLINE (1966–2004) and PsychINFO (1872–2004). Reviews of bibliographies also were used to identify relevant articles.
Results: In general, evidence suggests that low social support confers a risk of 1.5 to 2.0 in both healthy populations and in patients with established CHD. However, there is substantial variability in the manner in which social support is conceptualized and measured. In addition, few studies have simultaneously compared differing types of support.
Conclusions: Although low levels of support are associated with increased risk for CHD events, it is not clear what types of support are most associated with clinical outcomes in healthy persons and CHD patients. The development of a consensus in the conceptualization and measurement of social support is needed to examine which types of support are most likely to be associated with adverse CHD outcomes. There also is little evidence that improving low social support reduces CHD events.
Key Words: social support • depression • coronary heart disease • physiological mechanisms
Abbreviations: AMI = acute myocardial infarction; ANS = autonomic nervous system; CHD = coronary heart disease; ENRICHD = Enhancing Recovery in Coronary Heart Disease; HPA = hypothalamic pituitary adrenal; SES = socioeconomic status; SNS = sympathetic nervous system.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONCONCLUSIONSNOTESREFERENCES |
|---|
In this review, we provide an overview of theories and conceptualizations of social support and summarize the evidence for a relationship between various types of social support and both the etiology and progression of CHD. Articles for the primary review of social support as a risk factor were identified with MEDLINE (1966–2004) and PsychINFO (1872–2004) searches using the terms prognosis, risk factors, social support, coronary disease, myocardial infarction, and coronary artery bypass. The searches were limited to studies using human subjects and available in English language. Reference sections of relevant articles also were used to identify additional studies that had not been identified by the database searches. Studies that measured social support at baseline, had longitudinal designs, and included "hard" endpoints like death or cardiac events were included in the primary review.
Theories of Social Support
Theoretical conceptualizations of social support have a relatively long history stemming from Durkheim’s (10) notion of attachment and regulation by society, Cooley’s (11) primary group, and Bowlby’s (12) theories of attachment. In the 1970s, 2 publications, by Cassel (13) and Cobb (14), described the disease-protective effect of social support (or "host resistance") and spawned an increased interest in social support and health. Cassel (13) and Cobb (14) observed that patients who were more socially connected appeared to have better prognosis in recovering from illness or in maintaining their health. Their conceptualization of social support emphasized feelings of belongingness. Alongside this perspective, an equally large literature arising out of epidemiological traditions also has grown and has focused more on the structure of the social network or more tangible types of support offered by the network (15).
Current Terminology
Although historically there has been little integration of differing theories of social support, there is increasing agreement among reviewers to structure social support within two broad domains: structural and functional support, or the structure of social ties and the support actually provided by that structure respectively (16–19).
Structural support refers to the size, type, density, and frequency of contact with the network of people surrounding an individual. Measures of the density of social support, frequency of interactions, the number of close contacts versus peripheral acquaintances, martial status, group or church membership, and geographic proximity describe varying structural patterns. Several limitations with this construct have been noted. Foremost among these criticisms is that merely describing the structure of relationships does not necessarily describe the nature of the relationships. That is, network measures may be problematic to the extent that they group together supportive and nonsupportive relationships or confuse quantity and quality.
Functional support is the support provided by the social structure. Functional measures of support are delineated by type, including instrumental (e.g., help getting tangible tasks done), financial (economic support), informational (providing needed information), appraisal (help evaluating a situation), and emotional support (e.g., providing emotional support, feelings of being loved). The label tangible is often used to describe types of support that are readily seen or quantified such as instrumental or financial support.
Furthermore, theorists carefully distinguish functional support that is actually received ("received functional support") from an individual’s subjective appraisal of their satisfaction with support or their perception that support would be available if needed ("perceived functional support"). Received functional support is limited by its overlap with distress, illness, or extent of need. Indeed, measures of received support are rarely used in the literature covered by this review for this reason. Instead, measures of perceived functional support are more commonly used. Perceived functional support has been conceptualized as the subjective appraisal of the degree of match between the amount and type of support needed and amount and type of support available, or the perception that support would be available if needed. As noted earlier, various domains of perceived and received functional support can be measured, including instrumental, financial, informational, appraisal, and emotional support. Throughout, perceived support is used as shorthand for perceived functional support. For example, "perceived emotional support" refers to the emotional domain of perceived functional support.
Emerging Theories
Theories of social support continue to evolve. Important contemporary lines of research include reconsidering basic assumptions about social support, developing theories of match between support and both situation-specific (e.g., type of stressor, stage of disease) and individual-difference (e.g., attachment style, degree of need for autonomy) variables, and improving measurement strategies. Many of these theories are not represented in the CHD literature, but they are presented here to suggest alternate interpretations of results and additions to future research.
Theorists and researchers have explored the possibility that individuals’ perceptions of the social support are determined not only by the social environment, but also by individual personality factors. The stability of these measures over time, their correlation with measures of personality, and relative lack of correlation with received functional support suggest this possibility (20–22).
Another area of research, summarized by Coyne and Bolger (7), suggests that negative aspects of interpersonal relationships such as abuse or hostility are more influential than positive aspects such as feelings of belongingness. Other work suggests that the most beneficial support is often not noticed by the recipient and is therefore unlikely to be reported on questionnaires. For example, a study of couples showed that partners who received supportive behaviors but did not report them (presumably because they did not notice them) had higher levels of well-being than partners who did report receiving support (23). Behavioral or peer-report measures that include items assessing negative aspects of relationships have been proposed to address some of these issues (24).
Theories of match propose that social support is more or less effective, depending on the degree of match with the demands of the situation and/or the preferences and needs of the support recipient. Temperament, personality, attachment style, and/or needs for autonomy, competence, and relatedness (25) may affect an individual’s tolerance or desire for intimacy, willingness to seek help from others, and perceptions of behaviors intended to be supportive (26). These considerations again highlight the need for behavioral or peer-report methods (7) or measures that allow a participant to indicate their unique perception of a given behavior intended to be supportive (24,27). Others have explored the optimal match between support and various stressful events or disease stage. For example, Antonucci and Johnson (22) have postulated that close relationships may protect against the onset of cardiovascular disease by buffering stress, but extended networks may be more important to patients with existing disease as they allow connection with others who have similar experiences and can offer practical advice.
Empirical Evidence for Lack of Social Support as a CHD Risk Factor
Prospective Studies With Patients Without Established CHD
Prospective studies of initially healthy individuals have provided evidence in support of social support as a predictor of the onset and progression of CHD. These studies are summarized in Table 1. Some have shown that low perceived emotional support is associated with increased CHD incidence (28,29). Several studies have shown an effect for structural support measures such as the number of social contacts, marital status, living arrangements, membership in clubs, or social activities. The effect of structural support has been shown for initial cardiac events (28,30) and mortality due to CHD (31–33). However, not all studies have found a prospective relationship between structural support and CHD (34,35). For example, in a study by Reed et al. (34), network support was not prospectively related to subsequent CHD incidence.
|
Of note are the studies by Orth-Gomer et al. (28) and House and colleagues (33), both of which directly compared a network measure and a perceived social support measure as predictors of CHD incidence. Orth-Gomer et al. (28) reported an effect for both perceived emotional support and the number of people met during a given week in a sample of 736 men. However, in a sample of 2754 men and women, House et al. (33) reported an effect for the number of social relationships and activities but did not find an effect for satisfaction with these activities.
The majority of studies in this area show an effect for measures of structural support, and prior reviewers (16) have concluded that the majority of studies point to network measures as predictors of CHD onset. However, few studies included a measure of perceived functional support, so it is premature to conclude that this type of support is not related of the etiology of CHD until further studies are conducted. There is preliminary support for perceived emotional support in particular. In addition, although most evidence exists in support of structural support as a predictor of the initial onset of CHD, the specific dimensions of structural support that are most important remain unknown. For example, several studies reviewed here showed an effect for multidimensional measures of structural support (including the size, type, density, and frequency of contact with the network of people), and others showed an effect for specific domains ranging from involvement in activities to simply the number of relationships an individual has.
Prospective Studies With CHD Samples
Prospective studies with CHD samples address the role of varying types of social support in predicting health outcomes for patients with existing CHD. Studies of this kind that use "hard" endpoints (e.g., mortality, acute myocardial infarction [AMI]) were reviewed with attention to the specific type of social support that was predictive of outcome and are summarized in Table 2. Various domains of perceived functional support and structural support have been shown to predict mortality and cardiac morbidity in patients with CHD, conferring a 2- to 4-fold increased risk. For example, several studies of this kind have reported a main effect for varying domains of perceived functional support, including perceived emotional support (36,37), perceived marital quality (38,39), perceived tangible support (40), and perceived support phrased in general terms (e.g., perceived availability of needed support) (41,42). Berkman et al. (36) followed 194 AMI patients for 6 months and found that perceived emotional support was a predictor of mortality. Similarly, Gorkin et al. (41) followed 647 AMI patients and found that 1 item assessing the perceived availability of needed support was associated with a modest hazard ratio of 1.46 for subsequent mortality. On the other hand, Frasure Smith et al. (43) did not find a main effect for perceived social support in a sample of 887 post-AMI patients but rather found an interaction with depression, such that depressed patients with low perceived social support were at the greatest risk for mortality during the first year after AMI.
|
In addition to the evidence for perceived functional support, several studies of patients with existing CHD have reported an effect for structural support measured in such terms as network size (44–46), marital status (47–50), and participation in clubs or social/recreational activities (51–55). For example, in a study of 1368 CHD patients, Williams et al. (47) showed that individuals with CHD who were unmarried without a confidant had a relative risk of 3.34 compared with those who were either married or had a confidant, or both. Case et al. (49) reported a similar trend for post-AMI patients. Brummett et al. (44) followed 430 patients with documented CHD for an average of 4 years and found that low network size was associated with a hazard ratio of 2.43 for cardiac mortality. Jenkinson et al. (52) followed 1376 AMI patients for median of 3 years and found that lack of membership in a club or religious group or lack of contact with family and friends predicted mortality, with a modest hazard ratio of 1.49. By contrast, Berkman et al. (36) did not find an effect for social network size in predicting mortality for 194 post-AMI patients.
The few studies that have directly compared the effect of a variety of measures of network support with various domains of perceived functional support have provided mixed results (36,41,42,45,47,53,54). The results of some studies have shown that perceived functional support measures are more prognostic than structural measures (36,41,42). On the other hand, other studies have favored structural measure to perceived functional measures (45,47,53,54). For example, Gorkin et al. (41) found that a single item assessing the perceived availability of needed support was predictive of mortality for post-AMI patients but that participation in social or recreational activities was not. Horsten et al. (45) found that the size of social networks in post-AMI patients was predictive of cardiac events by 5 years but that the availability of emotional support from friends and family was not.
In contrast to the evidence presented in this review, several prior reviewers and theorists have concluded that perceived functional support, (particularly perceived emotional support) but not structural support, is related to CHD progression (3,16,56). For example, the work of Berkman and colleagues (36) has highlighted the role of structural support in predicting the initial incidence of CHD and the importance of perceived functional support in predicting outcomes for patients with existing CHD. In addition, there is a long tradition in psychology that emphasizes the role of perceived emotional support in particular (13,14). However, as noted above and in Table 2, there is in fact substantial evidence that various domains of both structural and perceived functional support are important predictors of prognosis for patients with existing CHD. Among structural measures, martial status and living with a partner have considerable support. However, multidimensional measures of structural support, simply the size of the social network, as well as extended connections and involvement in activities have been related to outcome. Furthermore, although there has been an emphasis theoretically on the importance of perceived emotional support, and evidence does indeed exist to support the role of perceived emotional support in prognosis, evidence also exists for the importance of perceived tangible support and multidimensional measures of perceived functional support. Thus, further research is needed to determine which types of functional and structural support are most related to outcome and therefore promising targets of intervention.
Moderating Factors
A priority for future research is to determine not only the types of social support that predict outcome but for whom and under what circumstances (2,3,8,9). Sociocultural factors such as gender, socioeconomic status (SES), and ethnicity, as well as psychological factors like personality and depression, are potential moderators of the effect of social support on CHD that deserve further investigation.
Sociocultural Factors
Sociocultural factors like gender, ethnicity, and SES are increasingly studied as potential moderators of the effect of social support. Various theories of social support suggest that factors associated with sociocultural status may moderate the relation between social support and CHD. For example, according to the stress buffering theory, individuals who face more stressful events are more likely to benefit from the buffering effects of social support (45). The added-value hypothesis suggests that social support may add more "value" to those who face increased stressors (58). Other theories suggest that those who emphasize communion or affiliation over agency or individualism are likely to be more reactive to the effects of social support (59). Few prospective studies in CHD populations using hard clinical endpoints exist at this time, but preliminary data are reviewed where available.
Early studies investigating gender-specific effects of social support suggested that the association between social support and CHD was stronger for men (31,33,48). However the most recent and adequately powered studies have either failed to show gender-specific effects (60,61) or have suggested that the relation is stronger for women, such that women are both more likely to benefit from positive support and suffer the ill effects of negative support (36,39,62–64).
There is a relative paucity of data investigating the role of ethnicity in moderating the effect of social support on CHD. Reviewers have emphasized the role social networks play in health behaviors for various ethnic minority populations (65). Overall, available empirical evidence has suggested that members of ethnic minority groups are effected equally or more by social support in terms of CHD outcomes. For example, Farmer et al. (66) compared the effect of social support on survival after AMI for Mexican Americans and non-Hispanic whites. Whereas the survival of non-Hispanic whites was not decreased by low levels of social support, low social support placed Mexican Americans at more than 3 times the risk for mortality over the follow-up period. Other research using surrogate endpoints has suggested that chronic racial discrimination may contribute to the development of cardiovascular disease (67,68).
SES
SES is another potential moderator of the effect of social support. Existing research has been mixed. Strogatz and James (69) reported that lower perceived tangible support, but not perceived emotional support, was associated with hypertension for low-income individuals, whereas there was no association for the remainder of the sample. By contrast, Vitaliano et al. (70) found that emotional support but not tangible support was associated with a composite measure of cardiovascular risk for low-income patients, but neither measure was associated with risk for patients with higher incomes. The use of varying surrogate markers as outcome measures may be responsible for the varying effects shown in these studies. Further research is needed to determine the dimensions of social support that affect particular surrogate endpoints, mediating mechanisms, and ultimately disease outcomes for patients of varying sociocultural backgrounds.
Depression
Low social support often coexists with depression, and both social support and depression have been implicated as important predictors of CHD onset and progression. Existing studies have suggested that it is important to consider the combined effect of depression and social support in predicting CHD prognosis. For example, Frasure-Smith and colleagues (43) found that high levels of perceived social support buffered the effects of depression on 1-year mortality in post-AMI patients, such that depressed patients with high levels of social support were not at increased risk relative to nondepressed patients. Likewise, Horsten et al. (45) reported a significant interaction between social integration (a measure of peripheral network size) and depression in predicting cardiac events and deaths in patients with CHD such that patients with elevated symptoms of depression who also lacked network support were at the greatest risk.
Personality
Match theories posit that social support can be beneficial, neutral, or even harmful, depending on the degree of match with the characteristics of the support recipient (26). Research by Hobfoll et al. (71) suggests that self-esteem is an important moderator of the effect of support. In one study, individuals who were low in self-esteem and close to family members tended to report the lowest satisfaction with support. The authors conjectured that these participants may have seen reliance on others as an assault to their already low self-esteem. Further, temperament or attachment style may limit an individual’s tolerance or desire for intimacy or help from others and influence their perceptions of behaviors intended to be supportive. There is evidence that individuals with high levels of cynical hostility benefit less from social support (72). Indeed, it is important to remember that not all people want or benefit from certain types of social support, nor are all people distressed if they lack social support (44).
Biobehavioral Mechanisms
There are a number of potential mechanisms by which social support may affect health, including lifestyle factors such as exercise, smoking, alcohol consumption, diet, and adherence to medical treatments; physiological factors related to autonomic nervous system (ANS) dysfunction such as cardiovascular reactivity; and psychosocial factors like self-efficacy, coping ability, negative affect, and depression.
Health Behaviors
There is evidence that improved social support is associated with greater adherence to medical recommendations and other health-promoting behaviors. For example, Hartel et al. (73) reported cross-sectional evidence that individuals with more social contacts and ties were less likely to smoke. Doherty et al. (74) found that wives’ reports of the social support they provide to their husbands were significantly related to their husbands’ adherence to medical recommendations. In another study of cardiovascular patients, one of the factors most predictive of smoking cessation was the degree of disapproval expressed by a member of the patient’s social network (75). Oka et al. (76) found that support specific to exercise received from friends and family members was an important predictor of adherence to an exercise program for cardiovascular disease patients. Overall, the most consistent findings in this area have been provided by network measures of support. This often has been explained under the rubric of "social control," in which the mere presence of other person implicitly demands more socially acceptable healthy behaviors (7).
Physiological Mechanisms
A review by Knox and Uvnas-Moberg (77) examined evidence that ANS and hypothalamic pituitary adrenal (HPA) axis dysregulation may link social support and CHD (78). For example, ANS dysregulation has been implicated in CHD. Hyperactivity of the sympathetic nervous system (SNS) has been linked to hypertension (79) and other risk factors for CHD such as decreased heart rate variability, decreased vagal tone, and reduced heart rate recovery (80–82). SNS activation may elicit coronary vessel constriction, resulting in myocardial ischemia (83,84).
Low social support may also be linked to autonomic dysregulation (2,77) Uchino et al. (85) conducted a comprehensive review of the literature on the association between various domains of social support and the autonomic dysregulation, as indicated by cardiovascular reactivity and neuroendocrine measures. The authors cited a small but consistent cross-sectional correlation between social support and blood pressure. A meta-analysis of experimental reactivity studies revealed a particularly robust trend, such that low social support was related to greater reactivity. In a prototypical experimental reactivity study, participants were asked to perform a stressful speech task with or without support from a confederate. Systolic blood pressure, diastolic blood pressure, and heart rate were measured throughout. All participants showed increases in blood pressure and heart rate during the task, but this effect was moderated by the availability of support. That is, individuals who were offered support showed comparatively less reactivity (86). In a narrative review of cross-sectional, experimental, and prospective observational studies of the relation of social support and neuroendocrine measures, the authors reported an inconsistent but promising effect such that lower levels of social support were associated with increased cortisol and catecholamine levels.
In a similar manner, evidence links HPA axis dysregulation to both CHD and low social support. HPA axis dysregulation is related to many cardiovascular disease risk factors such as obesity, hypercholesterolemia, hypertriglyceridemia, hypertension, and elevated heart rate (87,88). Two studies cited by Knox and Uvnas-Moberg (77) have linked social isolation in rats (89) and monkeys (90) to HPA dysregulation.
There is evidence that platelet activity, vascular endothelial functioning, and inflammation play important roles in the etiology of CHD. Emerging evidence suggests that these factors are in turn linked to stress and psychological disorders such as depression (91). These mechanisms are promising targets for future research aiming to uncover links between social support and CHD (78). In addition, Knox and Uvnas-Moberg (78) suggest that the release of oxytocin in response to social support and physical touch may buffer the effects of ANS and HPA axis dysregulation.
Relationship to Depression
The association between social support and depression is particularly relevant as depression itself is known to be an important predictor of outcome for CHD patients (2,91). Longitudinal studies with population samples have supported the hypothesis that low social support primarily acts as a risk factor for the subsequent development or worsening of depression. This effect has been shown for varying types of social support, including perceived emotional support and network size (92–96). On the other hand, there has been only limited support for the hypothesis that depression prospectively predicts decreased levels of social support (95). There is mixed evidence that a third factor, such as a personality trait or genetic factor, may directly predict both lower social support and depression (93,95,97).
Similar trends have been shown in CHD samples. There has been extensive documentation of the cross-sectional relationship between depression and perceived social support (44,97–104) and some evidence for an association between depression and network size (45,103). Longitudinal studies have provided evidence that there is a causal relationship between perceived social support and depression in cardiovascular disease populations. For the most part, these studies only have addressed the hypothesis that low perceived social support predicts depression. These studies have suggested that low perceived social support prospectively confers a risk for the development or worsening of depression in cardiovascular disease populations by 1 month to 4 years (42,98,99,102, 105,106). Although there is evidence for the role of network size in predicting depression, the evidence for perceived social support is more robust, which is consistent with theories that suggest that perceptions of social support are more likely than social network measures to be related to such factors as depressogenic cognitions, negative affect, or personality.
Interventions
There is a large body of research assessing the effectiveness of social support interventions. In a recent comprehensive review, Hogan et al. (107) assessed 100 such studies. Interventions differ by support provider (peer, existing family or friends, professional), format (group or individual), and content (e.g., skills training, discussion, support provision). Most studies have not used low social support as a selection criterion but have selected participants undergoing a stressful event, such as a recent AMI. Due to considerable methodological limitations in a large number of studies, including failure to formally assess social support, Hogan et al. (107) were unable to compare the relative effectiveness of different types of interventions. However, overall 83% of the studies showed some benefit in one or more domains, including well-being, social support, and clinical health outcomes.
An important example of a social support intervention for CHD patients is the M-HART trial which was designed to replicate the promising results of the Ischemic Heart Disease Life Stress Monitoring Program (108–112), with some design improvements such as the inclusion of women and a larger sample size (113). Patients were randomized to 12 months of a psychosocial nursing intervention or usual care. The intervention consisted of monthly telephone monitoring (administration of the General Health Questionnaire) by an untrained person. Patients who reported high levels of stress were visited in their homes by a nurse. Nurses were not given specific instructions on how to help the patient but reported providing support and assurance. Twelve-month follow-up data indicated that, in contrast to the prior trial, which reported significant reductions in cardiac mortality and morbidity for the men who received the intervention, the M-HART intervention did not protect against reinfarction or cardiac or all-cause mortality. The effect on social support was not reported, but the authors did report a lack of significant reductions in depression or anxiety as a result of the intervention. In addition there was a statistically nonsignificant trend (p = .051) suggesting that the intervention may have actually increased all-cause mortality in women. This surprising finding illustrates the importance of including a representative sample in trials and reminds us that interventions can have unforeseen effects.
In the recently completed Enhancing Recovery in Coronary Heart Disease (ENRICHD) (114) patients with a recent AMI were screened for depression or low perceived social support; 2481 patients with depression, low perceived social support, or both were randomly assigned to usual care or the psychosocial intervention, which consisted of 6 sessions to 6 months of individual cognitive behavioral therapy, group therapy, and treatment with antidepressant medication for patients with severe or persistent depression. Patients with low social support were helped with skill deficits, cognitive factors related to low levels of social support, and outreach. Initial results of the ENRICHD trial indicate that the intervention did lead to statistically significant but small improvements in low social support compared with usual care (the treatment group’s social support scores changed by 27% and the usual care group’s scores changed scores by 18%). However, the intervention was not effective in reducing rates of mortality or reinfarction.
Thoits et al. (115) tested an intervention provided by a single peer. Men awaiting coronary artery bypass graft surgery were randomized to usual care or visits by a peer who had recently undergone the procedure. At first glance, the intervention appeared to be ineffective, as patients receiving the intervention did not differ in adjustment to surgery. However, the authors noted that many patients spontaneously sought support from fellow patients. Reanalysis revealed that those receiving the intervention and those who spontaneously sought help fared better after surgery than those who received neither type of support.
Many social support groups bring together peers who have undergone similar stressors in an effort to minimize stigma, decrease isolation, and allow friendships to form. These groups are very common, particularly for stigmatizing diseases or disorders such as acquired immunodeficiency syndrome, cancer, or addictive disorders but are less common for diseases like CHD (107). Typically, peer-support interventions are facilitated by a leader and include discussion and support among members and may include skills training as well. Work by Humphreys and colleagues (116) suggests that improved network support is an important function of these groups. Another important and often-overlooked aspect of peer groups is that they allow participants to both receive and give assistance. Indeed, reciprocity has been associated with greater improvements in well-being across various populations (117), a finding that is consistent with theories about the importance of preserving a sense of self-esteem and competence.
Overall, social support interventions are hypothesized to work by creating new relationships, by allowing patients to distance themselves from harmful personal relationships (7), or by teaching skills necessary to develop more beneficial personal relationships. Social support interventions, particularly those with a cognitive focus, may also work by improving perceived social support (107,118). To date, however, the types of social support that can be meaningfully altered and the therapeutic methods that are most effective are largely unknown (86,117,119–122).
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONCONCLUSIONSNOTESREFERENCES |
|---|
Research thus far has provided a piecemeal approximation of the specific dimensions of support that are most important at various disease stages and for differing patient profiles. Simply including a comprehensive battery of social support instruments and adequate demographic representation to existing research designs will clarify the dimensions of impaired social support that are associated with increased risk and the variables that may moderate this effect. Further work is needed to clarify the behavioral, psychological, and physiological mechanisms underlying the relation between social support and outcomes. Including emerging theories from social psychological literature, which encourage the reconsideration of basic assumptions about social support, the development theories of match between support and both situation-specific (e.g., type of stressor, stage of disease) and individual-difference (e.g., attachment style, degree of need for autonomy) variables will also allow us to design the most effective interventions with the aim of improving both disease prognosis and quality of life.
|
NOTES |
|---|
|
TOPABSTRACTINTRODUCTIONCONCLUSIONSNOTESREFERENCES |
|---|
This research was supported in part by grants HL 69672, MH 49679, and HC 55142 from the National Institutes of Health, Bethesda, MD. This work was conducted as part of the first author’s doctoral dissertation for her PhD in clinical psychology at Duke University.
DOI:10.1097/01.psy.0000188393.73571.0a
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONCONCLUSIONSNOTESREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. Ikeda, H. Iso, I. Kawachi, K. Yamagishi, M. Inoue, S. Tsugane, and for the JPHC Study Group Social Support and Stroke and Coronary Heart Disease: The JPHC Study Cohorts II Stroke, March 1, 2008; 39(3): 768 - 775. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Ohira, A. V. D. Roux, R. J. Prineas, M. A. Kizilbash, M. R. Carnethon, and A. R. Folsom Associations of Psychosocial Factors With Heart Rate and Its Short-Term Variability: Multi-Ethnic Study of Atherosclerosis Psychosom Med, February 1, 2008; 70(2): 141 - 146. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. J. Rodriguez, M. M. Burg, J. Meng, T. G. Pickering, Z. Jin, R. L. Sacco, B. Boden-Albala, S. Homma, and M. R. Di Tullio Effect of Social Support on Nocturnal Blood Pressure Dipping Psychosom Med, January 1, 2008; 70(1): 7 - 12. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Huisman, F. van Lenthe, and J. Mackenbach The predictive ability of self-assessed health for mortality in different educational groups Int. J. Epidemiol., December 1, 2007; 36(6): 1207 - 1213. [Abstract] [Full Text] [PDF]
|
|
