|
Article Excerpt
An abundance of social support research has accumulated over the past several decades (e.g., Barrera, 1986; Cohen & Willis, 1985; House, Umberson, & Landis, 1988; Sarason, Sarason, & Shearin, 1986; Schwarzer & Leppin, 1992; Vaux, 1988). The emergence of this research has been traced to the early work of Cassel (1976), Caplan (1974), and Cobb (1976), which identified social support as an important factor in the prevention of and susceptibility to illness and disease. Since this time, various disciplines have contributed to this body of literature including social psychology, health psychology, rehabilitation psychology, behavior medicine, and nursing. The fundamental premise driving these studies has been the widespread consensus that social support has a positive relationship to health and wellbeing. Specifically, researchers from an array of disciplines have found positive relationships between social support and a variety of outcomes typically associated with health, well-being, and coping (Auslander & Litwin, 1992; Pierce, Lakey, Sarason & Sarason, 1997; Schwarzer & Leppin, 1992). Further, social support has been found to be negatively related to stress (Russell & Cutrona, 1991), emotional distress (Dilworth-Anderson, Williams & Cooper, 1999), and mortality (Berkman & Syme, 1979), and shown to predispose persons to engage in health-promoting or in self-care behaviors (Hubbard, Muhlen, & Brown, 1984; Muhlenkamp & Sayles, 1986; Riffle, Yoho, & Sams, 1989).
Contemporary social support research has extended to a variety of health conditions (e.g., cancer, cardiovascular disorders, multiple sclerosis, AIDS), psychological disorders (e.g., depression, anxiety, severe mental illness, addiction), and other life stressors (e.g., divorce, caregiving), and has been studied with an array of health and mental health related outcomes (i.e., recovery time, improved condition, health maintenance, and relapse). Of particular interest are the studies investigating social support within the rehabilitation context. For years, scholars have consistently stressed the importance of familial and peer support in the rehabilitation process (Brady, Koch, & Griffith, 2003; Bray, 1978; Cook & Ferritor, 1984; Kelly & Lambert, 1992; McKenna & Power, 2000; Power & Dell Orto, 1986; Safilios-Rothschild, 1970; Story & Certo, 1996; Sutton, 1985; Wright, 1960). And, there is growing evidence of a relationship between social support and rehabilitation related outcomes. For example, research suggests that social support is related to adjustment to disability, quality of life, psychological well-being, employment, treatment compliance, and survival rates among persons with disabilities and chronic illness (Belgrave & Walker, 1991; Dunbar, Ford, & Hunt, 1998; Elliot, Herrick, Patti, Witty, Godshall, & Spruell, 1991; Elliott, Herrick, Witty, Godshall, & Spruell, 1992; Evers, Kraaimaat, Geenen, Jacobs, & Bijlsma, 2002; Hatchett, Friend, Symister, & Wadhwa, 1997; Herrick, Elliott, & Crow, 1994; Holosko, Huege, 1998; Kaplan, 1990; Kaplan & Hartwell, 1987; Leach, Frank, Bouman, & Farmer, 1994; McColl & Rosenthal, 1994; McShane & Karp, 1993; Orr & Aronson, 1990; Rintala, Young, Hart, Clearman, & Fuhrer, 1992; Schulz & Decker, 1985; Symister & Friend, 2003; Zea, Belgrave, Townsend, Jarama, & Banks, 1996).
While a number of studies have been conducted in the area of social support and rehabilitation, the evidence is ambiguous. The varied evidence base for social support is not unique to rehabilitation however; it is pervasive across a wide array of health related disciplines and contributes to the universal myth that social support is a unidimensional, global, positive force that works in some inexplicable way (Heller & Rook, 2001). Nonetheless, this area is of particular importance to our field as social support is integral to the ecological approach of rehabilitation, wherein characteristics of the individual and the environment are integrated to achieve optimal functioning and quality of life (Chronister, Johnson, & Berven, in press). In fact, federal mandates and funding initiatives have occurred to address this component. For example, the State-Federal Vocational Rehabilitation Program is mandated to incorporate significant others into rehabilitation planning if necessary to enable the client to achieve employment (Roessler, & Rubin, 1998). Within the area of psychiatric rehabilitation, federal initiatives since the 1970s have addressed the need provide a wide range of non-mental health issues including social support through the development of Community Support Programs (Turner & TenHoor, 1978). In 1986, the importance of improving the social support networks of clients with psychiatric disabilities was reaffirmed in an NIMH-funded study that cited the need for case managers to be knowledgeable about social support systems such as churches, social clubs, self-help groups, families, and neighborhood leaders and to be skilled in mobilizing community resources for clients (Friday, 1986).
These broad-based efforts to incorporate social support into rehabilitation planning and intervention are limited by a dearth of empirically validated social support interventions. The rehabilitation related studies that do offer empirical support for social support interventions do not typically make connections to the theoretical and conceptual underpinnings of social support in the literature (e.g., Brown & Hanis, 1995; McKay, Glasgow, Feil, Boles, & Barrera, 2002; McColl & Friedland, 1993; Mignone & Guidotti, 1999; Samarel, Tulman, & Fawcett, 2002; Sherman, DeVinney, & Sperling, 2004; Stewart et al. 2001), limiting the potential for replication and for use as evidence to support clinical decision making. For example, support groups appear to be the most widely used form of social support (Kessler, Mickelson, & Zhao, 1997), yet there is a dearth of evidence supporting this intervention or the processes that comprise this approach. According to Gottlieb (1988), social support interventions have "the status of a black box, appealing strongly to practitioners but leaving obscure the pathways to support afforded by different features of the natural and engineered social surrounding" (p. 530).
In light of the weak evidence base for social support within the rehabilitation context, coupled with a health care climate driven by evidence based practice, this study will attempt to strengthen the evidence base of social support in rehabilitation by clarifying the effect of social support on rehabilitation outcomes using meta-analysis. Meta-analysis combines study findings to maximize power, and thus, provides answers to questions in rehabilitation by giving an overall estimation of the effectiveness from multiple studies (Wampold, 2001). This study has three specific aims: (a) to provide a review of theoretical and conceptual issues relevant to social support research; (b) to strengthen the evidence base for social support within rehabilitation using recta-analysis; and (c) to prompt researchers and practitioners to develop theoretical models of social support that can be empirically validated within the rehabilitation context and incorporated as 'best evidence' for clinical utilization.
Relevant Literature
The literature is robust with theoretical models and conceptual taxonomies of social support. The most widely recognized theoretical model is the stress buffering hypothesis, which suggests that social support works by moderating the effect of stress on health and adjustment, as evidenced by the demonstration of a statistical interaction effect between social support and the stressor (Cohen & Wills, 1985; Cobb, 1976). A number of studies have empirically validated this model by showing that stress has a more negative effect on health under conditions of low support than under conditions of high support (Andrews, Tennant, Hewson, & Schonell, 1978; Brown & Harris, 1978; Gore 1978; Pearlin, Lieberman, Menaghan, & Mullan, 1981 ; Turner, 1981). Another prominent model is the main-effect model, which posits that regardless of the level of stress, high levels of support promote well-being (Thoits, 1986). Evidence for this model derives from the demonstration of a statistical main effect of support with no interaction between stress and support (Cohen & Wills, 1985).
An array of other models are cited in the literature such as the support mobilization model, the support deterioration model, the perceived support model, the inequity theory and the esteem threat hypothesis (Aneshensel & Frerichs, 1982; Barrera, 1986; Dunbar, Ford, & Hunt, 1998; Thoits, 1986; Turner, 1981). These models differ primarily in how social support interacts with stress; for example, some models posit that stress initiates the mobilization of support, while other models suggest that stress inhibits seeking social support, and still other models propose that it is the degree of reciprocity in the relationship that determines the level of support. More contemporary models of social support seek to differentiate between the positive and negative elements of social relationships as well as the match between the type of support wanted and that received (Rook, 1992; Reynolds & Perrin, 2004). Indeed, support that is perceived helpful by one individual may be viewed as unsupportive by another (Brickman, 1982).
Conceptually, "social support" is the most widely used term, but its definition is multidimensional and lacks consensus, referring at times to structural and/or functional characteristics of social support, and alternatively to personal appraisals or received supportive behaviors (Veiel & Baumann, 1992). According to Vaux (1988), "no single and simple definition of social support is adequate because social support is a metaconstruct: a higher order theoretical construct comprised of several legitimate and distinguishable theoretical constructs" (p. 28). Researchers have attempted to clarify the definition of social support, with the most common conceptualizations encompassing two broad categories, structural and functional dimensions (Chak, 1996). The structural dimension focuses on the quantity (size, frequency of contacts) and characteristics (composition, density, homogeneity and multi-plexity) of a person's social network. This dimension is typically measured by the presence or absence of certain indices or a count measure (e.g., marital status, participation in community organizations, close friends, and relatives) as well as the number and frequency of contacts with specified social ties. This dimension has been criticized because of the assumption that all social interactions are supportive, wherein "the presence or absence of network is taken as proxy measures of supportiveness" (Wellman, 1981, p. 77). Not surprisingly, research shows this dimension to be a weak predictor of health and well-being (Knipscheer & Antonucci, 1990; Porritt, 1979; Seeman & Syme, 1987).
The functional dimension involves the type of supportive behavior and social exchange (Chak, 1996; Cohen & Wills, 1985; Schumaker & Brownell, 1984) and typically includes such functions as emotional (expressing affection and concern, listening, sharing a task), instrumental (tangible aid, financial and physical assistance), and informational support (advice, guidance, feedback; Barrera, 1981; Cohen & Wills, 1985; Cutrona & Russell, 1990; House & Kahn, 1985). From this perspective, social support has been defined as the "interactive process in which emotional, instrumental, or functional aid is obtained from the person's social network (Tolsdorf, 1976). Within this dimension, there have been several operational definitions of social support including received supportive behaviors, perceived satisfaction, and perceived availability of functions of social support. Received supportive behaviors is considered to be a more objective measure of social support, and is typically measured by inquiring about the types of support received during a specific period in the past, either in general or in a specific stressful situations (Dunkel-Schetter, 1984). Conversely, perceived availability of support targets one's subjective appraisal of supportive functions and denotes the amount of support that is expected to be available in case of need (Barrera, 1986; Sarason, Pierce, Shearin, Sarason, Waltz, & Pope, 1991; Vaux, 1988). Finally, perceived satisfaction involves appraisal of supportive functions received, and typically involves evaluating past received supportive behaviors.
Perceived appraisal of support is one of the most well researched operational definitions of social support, and the inverse relationship between perceived social support and psychological distress is well documented. The development of this definition has both methodological and theoretical roots. From a methodological perspective, this definition evolved from the recognition that data from subjective measures of social support were only minimally related to...
|
