Gaps in Social Support Resources in Later Life: An Adaptational Challenge in Need of Further Research

Abstract

Gaps in social support resources in later life may arise when older adults lose social network members due to illness, death, or residential relocation. Gaps also may arise when social networks remain intact but are not well suited to meet older adults’ intensifying support needs, such as needs for extended or highly personal instrumental support. Significant gaps in support resources are likely to require adaptive responses by older adults. This discussion highlights theoretical perspectives and illustrates empirical findings regarding the nature and effectiveness of older adults’ responses to gaps in their social support resources. The literature examining these issues is relatively small and, as a result, is ripe for further development. Promising directions for future research are suggested.

The literature on social support is fascinating and important, but also somewhat overwhelming, which makes a special issue like this, designed to distill important themes and unresolved issues in the literature, especially valuable. I greatly appreciate the opportunity to offer some comments on the social support field, and I will do so from the perspective of later adulthood, focusing on life circumstances that may create gaps in older adults' social support resources. Understanding the nature and effectiveness of older adults' responses to gaps in their support resources is important in view of compelling evidence that social support contributes to health and well-being in later life (Seeman, 2000).

The sections that follow focus on two central questions: First, how do older adults derive needed forms of support and companionship following the loss or disruption of important social relationships? Second, how do older adults derive sustained instrumental support when it is needed and when their social networks lack normative sources of such support? These questions gain a sense of urgency from evidence of the continued brisk pace of population aging, rising rates of chronic illness across all age groups, and declining marriage and fertility rates -- trends that collectively raise concerns about the adequacy of the social support resources for future generations of older adults.

Currently, however, it is important to emphasize that older adults in the United States generally do not experience vast unmet needs for social support. Social network involvement often declines in later life (Lang & Carstensen, 1994; Shaw, Krause, Liang, & Bennett, 2007), but these declines appear to reflect older adults' preference to interact with their closest, most emotionally rewarding social network members networks (Carstensen, Fung, & Charles, 2003). Older adults winnow from their social networks individuals with whom interaction is less gratifying, while maintaining contact with their closest family members and friends (Lang & Carstensen, 1994). From this perspective, age-related declines in social activity reflect older adults’ selective involvement in meaningful, gratifying social ties (Carstensen et al., 2003).

This view of older adults as proactively managing their social lives is a valuable antidote to earlier views of older adults as passive victims of societal rejection or prone to social withdrawal (Carstensen, 1993). Yet without subscribing to such negative views, it would be an oversight to ignore the changing life circumstances that can cause the loss or disruption of older adults’ social relationships or that can reveal limitations of their intact social support resources. Such altered life circumstances may require a different repertoire of adaptive responses in order to preserve health and well-being in later life.

Deriving Support after the Loss or Disruption of Important Social Relationships

The loss or disruption of important social relationships can occur for many reasons in later life. Widowhood is a common experience, with nearly 60% of women and 22% of men likely to be widowed by their mid-70s (He, Sengupta, Velkoff, & DeBarros, 2005). The death of a spouse can affect other relationships, as contacts with in-laws and couples with whom the widowed person previously socialized often fade over time (Lamme, Dykstra, & Broese Van Groenou, 1996). Deaths of relatives and friends also become increasingly common as people age. In one study, 59% of men and 42% of women over the age of 85 reporting that a close friend had died in the past year (Johnson & Troll, 1992).

Life events other than the death of friends or family members can disrupt existing social network ties. Retirement reduces contact with former coworkers, a change often lamented by retirees when they discuss what they most miss from their former jobs. Incomes decline substantially after retirement, and the lack of discretionary funds for transportation, entertainment, and other costs associated with socializing can restrict opportunities for social contact. Residential relocation often disperses social network members geographically, reducing in-person support and companionship.

Turning to substitute sources of support

Many of the relationships that are lost or disrupted by such life circumstances will have served as key sources of needed support and companionship for older adults. Older adults may be motivated to turn to other relationships to obtain the missing support and companionship, if not immediately then after an initial period of adjustment (Arbuckle & de Vries, 1995).

There is no guarantee, however, that a substitute relationship will effectively compensate for the lost relationship (Rook & Schuster, 1996). The support derived from a substitute relationship may differ in important respects from that originally derived from the lost relationship. Emotional support provided by an adult child, for example, may be less readily accessible, less intimate, or less sensitively rendered than the emotional support once provided by a spouse. As a result, the emotional support provided by the adult child may not boost the older adult’s morale as much as the support formerly provided by the spouse. For this reason, some researchers have urged a distinction between substitution (the "replacement" of missing social ties) and compensation (or the extent to which substitute social ties benefit well-being; Rook & Schuster, 1996; Zettel & Rook, 2004). This distinction has sometimes been blurred in the literature, and substitute social ties have simply been assumed to compensate for lost relationships.

Different forms of social network substitution can be distinguished, with potentially distinctive effects on older adults’ well-being. Substitution can involve the formation of a new social relationship. For example, widows may befriend other widows to whom they can turn for support and companionship (Morgan, Carder, & Neal, 1997). Older adults also may attempt to rekindle dormant social relationships (such as relationships with friends, neighbors, or family members) in which contact has dwindled because of such factors as time conflicts or competing role demands. For example, a widow may renew a relationship with a cousin who was seen only infrequently preceding the death of her spouse, but with whom she now has more frequent or more intimate contact. When such dormant relationships are renewed, they may help to provide missing support and companionship. Consistent with this idea, sibling ties often assume increased importance in later life, compared to midlife, a shift that is especially evident among widows (Cicirelli, 1995). Another form of substitution involves turning to currently active social network members to provide some of the support formerly provided by the spouse. Widowed individuals often turn to their adult children for this purpose (Connidis & Davies, 1992). In this sense, existing social network members assume expanded support roles, taking over some roles previously performed by a network member who is no longer present.

These examples are not meant to imply that long-term relationships can be readily replaced (or perhaps ever replaced) by other social ties (Stroebe, Stroebe, Abakoumkin, & Schut, 1996). The shared memories, world views, and well-honed ways of relating that have evolved over decades cannot be replicated. Nonetheless, older adults typically do seek to reorganize their social lives following a network loss (e.g., Stylianos & Vachon, 1993), and the formation of new social ties, rekindling of dormant ties, or expansion of support functions in existing ties may bolster psychological well-being.

Evidence suggests that older adults do turn to substitute sources of support following the loss of a major social relationship, such as a spouse (Connidis & Davies, 1992; Lamme et al., 1996). For example, all three forms of social network substitution discussed above were documented in a study of the support resources of older widowed women, with the duration of widowhood influencing the particular kind of substitution that occurred (Zettel & Rook, 2004). The most recently widowed women turned to existing and dormant relationships, whereas those who had been widowed longer turned not only to these relationships but also to new relationships. Whether substitute social ties effectively compensate for lost or disrupted social ties in later life is a separate question, as noted earlier.

Deriving benefits from substitute social ties

Studies that distinguish formally between substitution and compensation are scarce, but they suggest that turning to substitute ties may be easier than deriving psychological benefits from such ties. In the study of widowed women just described, the substitute social relationships that were formed did not appear to confer compensatory benefits, as gauged in terms of reduced loneliness or depression (Zettel & Rook, 2004). In fact, greater substitution was related to worse, rather than better, psychological health over a one-year period. Methodological limitations of the study (e.g., a short follow-up period or insensitive outcome measures) may have been partly responsible for the lack of evidence for compensation. Still, other research using different measures, samples, and study designs also has failed to find evidence of compensatory benefits of substitute social ties. Van Baarsen (2002) found, for example, that increased social support from network members during the early years of widowhood did not alleviate the widowed individuals’ loneliness.

Given the sparse research that has distinguished between social network substitution and compensation, it would be premature to conclude that substitute ties do not compensate for lost or disrupted ties. It is useful, nonetheless, to think about possible reasons why deriving compensatory benefits from substitute ties in later life may be difficult. Relatively few contexts exist in later life that are particularly conducive to forging new social relationships (Rook, 1991), and "linking persons" who can facilitate relationship formation (e.g. fellow widowed individuals who can provide links to similar others) may not always be present. Without such structural supports for relationship formation, forging new ties after having lost one or more important social ties may be effortful and fraught with risks, including the risks of being rebuffed or, alternatively, being unable to meet the new person's expectations for support or for the frequency or intimacy of interaction (Rook, 1991). If forging new social relationships is particularly effortful or risky for vulnerable older adults (those who have experienced recent losses), the costs may detract from the benefits of the new relationships. It is also possible that the need to forge new social ties in later life is at odds with the age-related preference to interact selectively with familiar, close social network members (Carstensen et al., 2003), discussed earlier. This discrepancy between current needs and age-related shift in social goals and motivations may create emotional distress in its own right (Zettel & Rook, 2004). Newly formed social ties or those that have undergone relatively recent changes also need time to stabilize and solidify (Blieszner & Adams, 1992). Greater benefits of new social ties might emerge in studies with longer time frames than those of existing studies.

These possible explanations address the limited compensatory benefits of social network substitution that involves the formation of the new social relationships. Some of these factors also may have a role in the limited benefits of social network substitution that involves renewing dormant relationships or expanding the support functions of existing relationships, but other factors may be important, too. For example, social support may be expanded in ways that increase its quantity but not its quality, thereby limiting its contributions to an older person's life. Johnson and Catalano (1983) noted that, unlike spouses and adult children, more distant relatives often provide perfunctory support to older adults. In addition, much support in later life must be provided over long periods of time, and network members who are called upon to assume expanded support roles of indefinite duration may experience distress that, in turn, detracts from the older person’s own emotional health.

Finally, it is important to recognize that older adults’ needs may change over time after experiencing a major relationship loss (Arbuckle & de Vries, 1995). The processes of substitution and compensation, therefore, are not necessarily directed toward a static set of social needs; rather, they may occur in concert with other processes that involve redefining social goals and aspirations (Zettel & Rook, 2004).

Redefining social goals and aspirations

Psychological well-being following a relationship loss may sometimes be preserved not by augmenting or realigning existing social ties but, rather, by modifying social aspirations to accommodate existing realities. This process has been referred to as an accommodative strategy of coping (Brandstadter & Renner, 1990) or a form of secondary control (Heckhausen & Schulz, 1995). A modification of goals and aspirations may be stimulated, in part, by a lack of success in efforts to attain a goal (Brandstadter & Renner, 1990; Heckhausen & Schulz, 1995). For example, lonely older adults who have tried unsuccessfully to find new companions may be compelled at some point to scale back on their aspirations for companionship.

Weiss (1974) has questioned whether such a redefinition of social needs or aspirations can preserve psychological well-being when key forms of social support are missing. He suggested that lonely individuals rarely succeed in alleviating the distress of loneliness by attempting to define away their needs for social contact. Baumeister and Leary (1995) similarly argued that the need for strong, stable social bonds is universal and fundamental to human functioning, and that efforts to deny or disavow this need are unlikely to be effective.

Other research, however, suggests that as people age their willingness to adjust personal goals as a means of coping with losses and obstacles increases (Brandstadter & Renner, 1990; Heckhausen & Schulz,, 1995). Studies of age differences in goals and aspirations for the future have found that older adults more often emphasize preservation of the status quo and prevention of declines (particularly health declines), rather than the achievement of positive changes in their lives (Cross & Markus, 1991). Such age group differences are consistent with the notion that psychological well-being may be maintained in later life, in part, by adjusting aspirations to mesh with objective conditions.

Evidence of the adjustment of social goals emerged in a study by Dykstra (1995), who found that adults who had been widowed longer developed more favorable beliefs about being single and less favorable beliefs about being married. A shift in relationship aspirations may have occurred over time, as part of the process of adapting to widowhood. This apparent shift appeared to be adaptive, moreover, as it was related to lower levels of loneliness.

A related possibility is that the criteria, rather than the desire, for particular kinds of relationships change following the loss of major social ties. Adults over the age of 85 in one study responded to social network deficiencies by broadening their criteria for friendship (Johnson & Troll, 1994). Their more inclusive friendship criteria allowed casual acquaintances and even employees (e.g., home health care aides) with whom social exchanges were quite circumscribed and unlikely to grow more intimate to be regarded as friends.

Summary

The disruption or loss of important social relationships in later life has the potential to create deficiencies in older adults’ access to needed forms of support. Such deficiencies may prompt a variety of compensatory responses, some directed toward the social environment and others directed toward internal states. Compensatory responses and outcomes are related, but nonetheless distinct, phenomena, and compensatory responses may or may not yield beneficial outcomes (Dixon & Backman, 1992; Rook & Shuster, 1996).

Deriving Support from Non-normative Sources

The discussion thus far has emphasized challenges that may arise in later life when older adults seek to derive support and companionship from alternative sources following the loss or disruption of important relationships. Challenges also can arise when existing social network ties remain intact but needs emerge for forms of support not previously provided by the network. Deriving sustained or intensive instrumental support, in particular, can present challenges when the social network does not include normative sources of such support.

Some social support theorists, such as Weiss (1974), argue that social relationships tend to become specialized in the functions they perform and that relationships cannot easily perform interchangeable functions. Patterns of relationship specialization are not idiosyncratic, but, rather, reflect norms about who should provide particular kinds of support. In addition, the level of comfort experienced when support is derived from particular sources influences patterns of relationship specialization (Litwak, 1985). In later life, for example, family members typically function as sources of instrumental and emotional support and, to a lesser extent, companionship; friends, in contrast, typically function as sources of emotional support and companionship but not instrumental support (Rook & Schuster, 1996). Personal care (e.g., bathing, grooming, preparing meals), in particular, is most comfortably and effectively provided by spouses or family members (Litwak, 1985).

Obtaining intensive or highly personal instrumental support may be difficult when the social network does not include normative sources of such support (Rook & Schuster, 1996). If so, then older adults who lack spouses or (geographically proximal) adult children may experience gaps in the support resources provided by their social networks (Rook & Zettel, 2005). The number of people who are remaining single (DePaulo & Morris, 2005) and childless (Abman & Martinez, 2006) has risen markedly in recent decades in many developed nations. These demographic changes raise the possibility that future cohorts of older adults may lack the customary sources of instrumental support that have proven to be valuable in helping to forestall declines in older adults' health or ability to live in their own homes.

Yet perhaps greater interchangeability exists among different categories of support providers than is implied by theories of relationship specialization. Cross-domain buffering theories, for example, suggest that the social support provided in one relationship domain (e.g. friends) can make up for problems or support deficits that exist in other relationship domains (e.g., family relationships; Lepore, 1992; Manne et al., 2003). Similarly, Cantor’s (1979) hierarchical-compensatory model of social support suggests that that older adults’ preferences for support follow a sequential hierarchy, with spouses preferred as the primary source of support support, followed -- in order -- by children, other family members, and friends when spouses are not available. Empirical evidence indicates that adult children and other kin do provide support to older adults who lack a spouse (e.g., Connidis & Davies, 1992; Pinquart, 2003). Less evidence speaks to the question of how well non-kin (e.g., friends, neighbors) can substitute for kin in meeting intensive support needs that may arise in later life.

Existing empirical evidence is too sparse to draw conclusions about which of these different theoretical models best describes the support transactions that occur (or do not occur) when intensifying support needs lead older adults and their social network members to contemplate whether to negotiate unfamiliar support roles. Future work aimed at testing these different theoretical models also may need to consider the possibility that older adults who have lived for many years with social networks containing few kin may have developed robust capacities for self-reliance or strategies for obtaining instrumental support from formal sources (e.g., home health agencies).

Summary

Deriving social support in later life may prove to be challenging when older adults develop new needs for support, particularly instrumental support, and when their social networks do not include categories of network members who normatively provide such support. If the relationship specialization perspective is correct, social networks comprised largely of non-kin may function adequately to provide emotional support and companionship and to meet limited needs for instrumental support, but they may be less able to meet extensive needs for instrumental support. On the other hand, if the cross-domain buffering or hierarchical-compensatory perspectives are correct, then the composition of the social network may matter less in meeting intensified needs for support when they emerge. Research is needed to test these competing perspectives and to develop projections about the support needs and resources of future cohorts of older adults, particularly in view of the demographic trends noted earlier.

Conclusion

The focus of this discussion on gaps in social support resources that may arise in later life should not be taken to mean that older adults currently experience widespread limitations of their support resources. On the contrary, declines in social network size occur routinely with few negative consequences, presumably because such declines reflect older adults' proactive winnowing of peripheral relationships from their social networks (Carstensen et al., 2003). Yet it is common for older adults to experience the loss or disruption of their social networks due to others' illness, death, or residential relocation. For some older adults, such changes will create gaps in their social support resources that require an adaptive response. Even when such social network disruptions do not occur, some older adults may experience gaps in their support resources as their needs for support intensify in ways that are not well aligned with the support-providing capacities or past practices of their social networks. These circumstances, too, require adaptive responses.

Understanding how people seek to remedy significant gaps in their social support resources is an important topic that is ripe for further theoretical and empirical development. This topic has relevance, moreover, not only to later life but also to other life stages and contexts. Children or adolescents who are isolated or rejected at school (East & Rook, 1992; Milevsky, 2005), young or middle-aged adults who experience marital separation and divorce (Wrosch & Heckhausen, 1999), and people whose support networks are upended by natural disasters (Norris & Kaniasty, 2004) all face challenges that include the need to forge or reconstitute key supportive relationships, or, alternatively, to find satisfactory ways to move forward in their lives without such relationships. Identifying commonalities and differences across these different life contexts would help to build a richer base of knowledge regarding the nature and effectiveness of individuals’ responses to gaps in their social support resources.