Recognizing the importance of social support to the health of elderly persons, Crawford (1987) has recommended that nurses in clinical practice should routinely assess a client's social support needs and resources and intervene to promote the quality and availability of needed social support. Both the structure of the support network (network analysis) and the kinds and characteristics of support needed and received should be considered. Studies have shown that social support tends to be situation-specific; kinds, amounts, and sources of the support needed and provided differ for persons at different levels of wellness (Norbeck, 1981). To ensure that clinical assessments and interventions are based upon validated knowledge, nursing research is needed to investigate these aspects of social support in exemplary health and illness situations.
One purpose of this study was to describe, compare, and contrast the social support needs and resources of community-based older adults in two environments designed to provide social support and stimulation. The senior center represents a situation of relative health while the Geriatric Day Treatment Center (GDTC) exemplifies a situation where mental health impairment could influence the need for and provision of social support.
Since the Older Americans Act of 1965, there has been a proliferation of senior centers to the point where they have become essential resources in most communities. Participants in senior center activities are usually cognitively-intact older adults who have reasonably high levels of social skills. Activities are planned to promote social interactions among participants, thereby enlarging their social support networks. Studies employing nonpsychiatric samples have repeatedly demonstrated the positive impact of social support on a wide range of health-related factors.
The earliest investigations of the effects of social support emphasized the importance of having a confidant for coping successfully with the multiple losses experienced by older adults, especially with retirement and the loss of a spouse (Lowenthal, 1968). More recently, Minkler and associates (1983) found that the ability to give and seek advice and other forms of assistance showed a positive relationship with perceived health status among 678 older adults. Pynoos and associates (1984) reported that a neighborhood program which helped elders offer support to each other resulted in higher lifesatisfaction scores and greater social interaction among participants. In a longitudinal study oí 3,559 frail elders (Mor-Barak, 1991), the effects of traumatic life events were seen to be ameliorated by social support for as long as 12 to 18 months. Goodwin and associates (1991) found that among newly diagnosed cancer patients, those having adequate social support were more compliant with their treatment regimes - a finding they speculated may be important to additional years of survival.
GERIATRIC DAY TREATMENT CENTERS
GDTCs are rehabilitation-oriented programs where multiple strategies are used to facilitate ine reintegration of their participants into the community. The social support derived from interactions among participants, or with families and /or health care providers, is a vital factor in assisting the participants with their resocialization process. Participants in GDTCs are older adults who suffer from acute or chronic psychiatric disabilities. Many of the participante have diminished social skills and/or coping abilities, and some are cognitively impaired. For these reasons, they are not likely participants in the usual senior citizen programs and, thus, look to the GDTC for socialization and stimulation.
Research studies have also shown many beneficial effects of social support among samples of psychiatrically impaired individuals. Lowenthal and Haven (1968) found that although lifelong isolates were not more prone to hospitalization for mental illness, persons who had attempted to establish social relationships and failed were particularly vulnerable to the development of mental illness. Henderson and associates (1978) found that psychiatric outpatients had fewer friends, fewer attachment figures, and fewer contacts with others outside their homes when compared with controls. Patients felt that their principal attachment figures gave them inadequate support. Women who lacked an intimate confidant reported more severe psychiatric symptomatology than did their more adequately supported peers. In addition, the absence of casual, less intimate friends was also associated with higher physical and psychiatric symptom levels in a study reported by Miller and Ingham (1976). In a 6-month follow-up of women discharged after psychiatric hospitalization, social support was the one factor that differentiated those who were readmitted from their more successful counterparts. Lack of social support was also related to poor symptom control and difficult social adjustment (Goering, 1983). The sources of social support were found to be important in relieving the symptoms of depression among 1,174 elders (Dean, 1990). In descending order of effectiveness, the sources were spouse, friends, and adult children. In this study, low levels of support, rather man the absence of it, was more strongly associated with increased depression.
Norbeck's (198Ia) model of social support was the perspective that guided this study. Norbeck defines social support as a simple contact or presence of another during a stressful experience. However, she also asserts that social support is necessary to help individuals perform adequately on a daily basis, thereby contributing to well-being throughout the course of one's life.
Following the work of Kahn and Antonucci (1981), Norbeck identifies three dimensions or kinds of social support
* Affective support (for example, expressions of liking, love, or respect);
* Affirmational support (for example, having a confidant and receiving agreement with or endorsement of one's behaviors, perceptions, or views); and
* Aid as a type of social support (some kind of direct assistance, either symbolic or tangible, such as accompanying one on a visit to the doctor or providing financial support).
Norbeck (198Ia) also adopts the metaphorical term "convoy" from Kahn and Antonucci (1981). Each person can be thought of as moving through life surrounded by a set of significant others to whom that person is related through the giving and receiving of social support.
The need for social support, the actual support available, and its adequacy and influence on outcomes all arise from a combination of factors that Norbeck (198Ia) calls properties of the situation and properties of the person. In this study, properties of the person included age, sex, marital status, religion and ethnicity, living situation, perceived health status, and level of psychiatric symptomatology. Properties of the situation included type of setting, program goals and activities, types of staff, frequency and length of attendance, and the size and composition of support convoys. Since Norbeck's original work in developing and testing her model was done with a young adult sample, a second purpose of this study was to test the applicability of Norbeck's model to the investigation of social support pertaining to older adults. Findings are relevant to both clinical and research situations.
A convenience sample of 18 older adults who attended a senior center in a rural northeastern setting at least once a week comprised one group of study participants. Permission to recruit these subjects was obtained from the acting senior center director: Eighteen clients who attended a GDTC in a major teaching psychiatric facility at least once a week constituted the second group of study participants. Permission to invite clients to participate was obtained from the Research Review Committee of the institution housing the GDTC. Inclusion criteria for both groups required that participants must be at least 65 years of age, able to speak and understand English, and able to understand and respond to oral questions.
Each participant met individually with the researcher, who read the questions from each of the instruments and recorded the respondent's answers verbatim. Subjects were given a copy of each instrument for reference during the interviews. Prior to questioning, a consent form containing an assurance of confidentiality and the freedom to refuse participation without penalty was read aloud to each participant, questions were answered, and the form was signed. Because of the availability of previously collected data and the limited attention span of some of the GDTC clients, demographic information, psychiatric diagnoses, and the number of hospitalizations were collected from their hospital records. Permission to use these records was included in the consent form.
The Norbeck Social Support Questionnaire (NSSQ)
The NSSQ (Norbeck, 198Ib) operationalizes selected concepts of Norbeck's model of social support through nine items developed to measure the respondent's perceptions about the social support received from each convoy member. Participants are asked to list all of the significant persons in their lives (convoy members) and their relationship to each person. Respondents then rate on a scale of 1 (low) to 5 (high) their estimation of the extent of social support that each convoy member would provide in six hypothetical situations representing each type of support (affect, affirmation, and aid). Remaining items address the frequency of contacts and information about recent losses of important relationships.
Although Norbeck's model includes the need for support, the NSSQ does not measure the individual's perceptions of his or her need for each of the types of support. Therefore, the NSSQ was revised through the addition of a measure of need for this study. The six items measuring social support were reworded to ask respondents how important it was to them that they received the three types of social support represented by the hypothetical situations. The same rating scale of 1 (low) to 5 (high) was used. These items preceded the items that measured the sources and amounts of social support received. Cronbach's alpha was calculated at .97 for the instrument based upon data collected from both groups of elders.
The Brief Symptom Inventory (BSI)
Since mental status could have been a confounding variable for either group, an assessment of psychiatric symptomatology was included. The BSI, which measures the number and intensity of symptoms recently experienced in the nine primary dimensions of psychiatric symptomatology, has been widely used with older adults. Two exemplary items of the BSI include, "During the past week, how much were you distressed by 'feeling suddenly scared for no reason?' and by 'temper outbursts you couldn't control?'" Reliability data from several studies include Cronbach alphas ranging from .71 to .85 and test/retest values from .68 to .91. Construct validity was established using factor analysis resulting in values from .35 to .71 (Derogotis, 1983).
Health Status Rating
Ferraro (1980) presented evidence from a national sample showing that global measures of self-reported health status are significantly correlated with objective health status among elderly respondents. Therefore, the question, "How do you rate your health: excellent, good, fair, or poor?" was included. Data related to other properties of the person were obtained from demographic questions.
Properties of the Person
Table 1 presents the properties of the person for both of the groups studied. Only two properties of the person were found to be significantly different when comparing the two groups. The senior center group had completed more years of education than did the GDTC group [f(34) = 2.56, p^.05]. Although all of the participants were Caucasian, the religious affiliations of the two groups were significantly different. The majority of senior center members identified themselves as Protestant, while the majority of the GDTC group were Catholic [x2 (3, N = 36) = 9.53, p^.05]. There were no significant correlations between any of the properties of the person and the amount of social support needed or received by either group.
Psychiatric diagnoses among the GDTC members included 13 members with depression, 2 with dementia, 2 with schizophrenia, and 1 with bipolar disorder. Sixteen of the GDTC members had one or more psychiatric hospitalizations. However, there were no significant differences in current psychiatric symptomatology between the two groups as measured by the BSI scores [i(34) = 1.89,p>.05].
Need for Social Support
Scores for the kinds of social support needed show that both groups rated the need for affective support (love and respect) highest of the three types, with need for long- and shortterm tangible aid second, and the need for affirmation (a confidant and agreement and support of actions) third. In fact, many subjects made it clear that it mattered little to them whether convoy members agreed with or supported their beliefs or actions. There was much similarity between the two groups in terms of their need for the three kinds of social support, with t-tests showing no significant differences between them for affect [f(34) = 1.11, p>.05]; affirmation [i(34) = .10, p>.05]; or aid If(34) = .34,p>.05].
Table 2 presents the sources of social support identified by each group. In the senior center group, convoy size ranged from 4 to 21 persons identified as sources of social support. Family members were identified most frequently as convoy members. Only 10 of the 18 participants identified spouses, which was primarily a reflection of their widowed status. Friends were the secondmost frequent sources of social support. Eleven participants identified from 1 to 11 friends, but 7 participants failed to name any friends as convoy members. Neighbors were less frequently named, and, again, seven subjects failed to name a neighbor as a source of social support. Ten participants named clergy, and eight named health care providers as convoy members.
The GDTC participants described convoys ranging in size from 3 to 24 members. Family members were again the most frequently named. Although seven participants indicated that they lived with spouses, only four of them named their spouses as sources of support. Four GDTC participants named one friend as a convoy member, while four named more than one friend. However, 10 participants failed to name any friend as a convoy member. Similarly, 11 participants failed to name a neighbor in their convoys. Health care providers were named by 11 GDTC participants, 5 of whom identified more than one health care provider.
T-tests indicate that there were significant differences between the groups on the average number of convoy members, as well as on the composition of the convoys (Table 2).
Since all participants had multiple sources of social support, the support received exceeded the support needed in both groups. Senior center participants had larger convoys; therefore, they received more social support than did the GDTC participants in aoect [f(34) = 2.77, ps.05]; affirmation [t(34) = 2.67/ p.05]; and aid [t(34)=2.67, pas.05]. The average amount of the various types of support provided by each convoy member did not differ between the groups, with one important exception: Senior center members perceived the average amount of respect received from their convoy members to be significantly higher than that perceived by GDTC members [i(34) = 2.02,p^.05].
Properties of Persons lor Senior Center and Geriatric Day Treatment Center (GOTC) Grovps
While generalizability is limited by sample size and selection, several implications for nursing practice and research are suggested by the findings. Participants in both community-based programs with missions to promote socialization named multiple sources as members of their social support convoys. Both groups indicated that members of their respective programs were important sources of social support. This finding indicates that these programs strengthen and enhance the support systems of their members and thus are important referral resources for nurses engaged in working with older adults. While the senior center group identified family, friends, and clergy as their major sources of support, the GDTC participants named family and health care professionals as their major sources. This would suggest that the GDTC members may be quite dependent upon their program for much of their social support. It should be noted that members of both groups identified formal sources of support (health care providers and clergy) as convoy members. Cantor (1989) has stated that older persons turn to formal sources only when informal sources are unable to provide the required assistance. It may be that nurses and clergy need to focus more on helping older adults to develop alternative sources of social support among family, friends, and neighbors in addition to providing support directly.
Comparison of Mean Convoy Compositions at Senior Center and Geriatric Day Treatment Center (GDTC) Members
In almost every community, a wide array of educational and support group programs are offered by nurses and other providers to older adults. Social networking and the enhancement of social support among participants should be primary goals of such programs.
Both groups perceived that the social support they received was greater than that needed on all three parameters (affect, affirmation, and aid). Additionally, both groups perceived that approximately the same amount of support was supplied by their individual convoy members. These findings are in opposition to the literature that describes the social support provided to psychiatric outpatients by their attachment figures as inadequate (Henderson, 1978) and merit further study. However, by virtue of having more extensive convoys, the senior center participants received more total support than did the GDTC participants. Given the losses of significant relationships that accompany aging and their dependence upon their program for social support, it could be that the GDTC participants are at greater risk for having inadequate support in the future. This is of particular concern considering the vulnerability to crisis in this group. It seems essential then that programs like the GDTC include mechanisms for promoting ongoing social ties among participants and with other sources of social support in the community. Such community sources might be mutual support groups, such as the one described by Pynoos and associates (1984), church groups, programs involving both young persons and peers, and programs where participants can make useful contributions as volunteers in hospitals, senior centers and day care facilities.
IMPLICATIONS FOR NURSING
It might be expected that among persons who are often dependent upon others for transportation, and even for activities of daily living, tangible aid would have represented the area of social support most needed. Instead, both groups indicated love and respect as the areas of greatest need. This is a very meaningful finding for all nurses who work with older adults to consider. While scores for both groups indicated that these needs were met, it is important to note that the two groups differed significantly on their perceptions of the amount of respect they received. Senior center members received significantly more respect from their individual convoy members than did the GDTC members. While this finding may well be a function of the outlook on life prevalent among individuals with chronic mental health problems, it suggests that GDTC staff need to be conscious of their participants' needs for respect during interactions and to communicate this need to family members. Even well-meaning family and staff often deny these older adults needed respect by infantilizing and patronizing them and promoting feelings of dependence. Nurses are in an excellent position to identify and discourage such behavior, offer more positive alternatives, and model behaviors and attitudes that demonstrate respect for and promote dignity among older adults through their own interactions.
The use of Norbeck's model and questionnaire produced a very comprehensive and revealing picture of the social support convoys of older adults, as well as detailed information about the kinds and amounts of social support they felt they received. However, in this study, the questionnaire had to be expanded to measure the amount of support needed, since absence of support could indicate absence of need, as well as meaning no available support (Thoits, 1982). The items measuring affirmation as the convoy members' support and agreement with the participant's beliefs and actions may be inappropriate for persons at this developmental level. Such ideas seem more important to the adolescents and young adults with whom Norbeck did her original work.
Two aspects of social support that have particular relevance for older adults have been explored since Norbeck published these materials and are therefore addressed neither by her model nor the NSSQ. The first of these is reciprocity (Lubben, 1988). Social exchange theorists have argued that being able to give support is important to the strength and durability of support networks. In particular, the lack of reciprocal support relationships can enhance feelings of dependence and reduced selfesteem, for which the elderly are particularly at risk. The second area of concern is that of negative support. Negative dimensions of social support include such behaviors as provoking feelings of anger and conflict, taking advantage of the individual, breaking promises (Rook, 1984), criticizing behavior, and upsetting the recipient (Fiore, 1983). In their studies with older adults, Finch and associates (1989) demonstrated that negative and positive social support are two distinctly different domains of experience. They advise that efforts to reduce distress and enhance well-being must include interventions to reduce negative social support, as well as to promote positive support.
While the findings of this study have provided some tentative suggestions for nursing practice, further investigation is needed before definitive conclusions can be drawn about the effectiveness of communitybased socialization programs for older adults in general. In addition, several unexpected findings of this study suggest areas for future research studies. For example, it was not anticipated that affective support would have been the major type of social support needed by participants. It was also surprising to find that almost 40% of senior center members failed to name one friend among their convoys. It was a revelation to learn that some respondents derived support from deceased family members and friends. In addition to the many nursing research questions these findings generate, they also illustrate the clinical importance of careful assessment of social support cited by Crawford (1987) as an essential nursing action when working with older adults.
However, appropriate clinical assessments and interventions require that nurse theorists and researchers develop and test models and instruments that accurately reflect the complexity and multiple dimensions of the social support construct. Models and instruments that would be most helpful in nursing practice should include attention to health and illness situations; the appropriateness of developmental levels; sources of support; the kinds of support needed and received; negative, as well as positive, support; and the reciprocal aspects of support.
- Cantor, M.H. Social care: Family community support systems. Annals, AAPSS 1989; 503:99-112.
- Crawford, G. Support networks and health related change in the elderly: Theory based nursing strategies. Family & Community Health 1987; 102):30-48.
- Dean, A., Kolody, B., Wood, P. Effects of social support from various sources on depression in elderly persons. / Health Soc Behav 1990; 31:148-161.
- Derogóos, L.R., Melisaratis, N. The brief symptom inventory: An introductory report. Psychol Med 1983; 13:595-605.
- Ferraio, K.F. Self-rating of social support networks on the health of the elderly. / Health Soc Behav 1980; 21:377-383.
- Finch, J.F., Okun, M.A., Barrera, M., Zautra, J., Reich, J.W. Positive and negative social ties among older adults: Measurement models and the prediction of psychological distress and well-being. Am J Community Psychol 1989; 17:585-605.
- Fiore, J., Decker, J., Coppel, D.B. Social network interactions: A buffer or a stress? Am J Community Psychol 1983; 11:423-439.
- Goering, P., Wasylenki, D., Lancee, W., Freeman, S. Social support and post hospital outcome for depressed women. Can ] Psychiatry 1983; 28:612-617.
- Goodwin, J.S., Hunt, W.C., Samet, J-M. A population-based study of the functional status and social support networks of elderly patients diagnosed with cancer. Arch Intern Mea 1991; 151:366-70.
- Henderson, S., Duncan-Jones, P., McAuley, H., Ritchie, K. The patient's primary group. Br J Psychiatry 1978; 132:74-86.
- Kahn, R.L., Antonucci, T.C. Convoys over the life course: Attachment, roles, and social support. In PB. Baltes, QC. Brim (Eds.), Life-span development and behavior. New York: Academic Press, 1981.
- Kaplan, B.H., Cassel, J.C., Gore, S. Social support and health. Med Care 1977; 15(5):47.
- Lowenthal, M.F., Haven, C. interaction and adaptation: Intimacy as a critical variable. American Sociological Review 1968; 33:20-30.
- Lubben, J.E. Assessing social networks among elderly populations. Family & Community Health 1988; ll(3):42-52.
- Miller, P.M., Ingham, J.G. Friends, confidants and symptoms. SOCKJ/ Psychology 1976; 11:5158.
- Minkler, M.A., Salariano, W.A., Longhauser, C. Supportive exchange: An exploration of the relationships between social contacts and perceived health status in the elderly. Archives of Gerontology and Geriatrics 1983; 2:211-220
- Mor-Barak, M.E., Miller, L.S., Syme, L.S. Social networks, life events, and the health of the poor, frail elderly: A longitudinal study of the buffering versus the direct effect. Family and Community Health 1991; 14(2):1-13.
- Norbeck, J.S. Social support A model for clinical research and application. ANS 198Ia; 3(4):43-59.
- Norbeck, J.S., Lindsey, A.M., Carrieri, V.L. The development of an instrument to measure social support. Nurs Res 1981b; 30:264-269.
- Pynoos, ]., Hade-Kaplan B., Reisher, D. Intergenerational neighborhood networks: A basis for aiding the frail elderly. Gerontologist 1984; 24:233-237.
- Rook, K.S. The negative side of social interactions: Impact on psychological well-being. J Pers Soc Psycho! 1984; 46:1097-1108.
- Ryan, M.C., Austin, A.G. Social supports and social networks in the aged. Image: Journal of Nursing Scholarship 1989; 21:176-180.
- Thoits, P.A. Conceptual, methodological, and theoretical problems in studying social support as a buffer against life stress. / Health Soc Behav 1982; 23:145-159.
Properties of Persons lor Senior Center and Geriatric Day Treatment Center (GOTC) Grovps
Comparison of Mean Convoy Compositions at Senior Center and Geriatric Day Treatment Center (GDTC) Members