Journal of Gerontological Nursing


Depression and Social Support: Effective Treatments for Homebound Elderly Adults

Agnes Loughlin, RN, MSN


Detecting, assessing, and properly treating depression in homebound older adults can be tricky. Social support alone may not be the answer.


Detecting, assessing, and properly treating depression in homebound older adults can be tricky. Social support alone may not be the answer.

With the increase in the older adult population, the incidence of depression will likely increase, affecting quality of life, morbidity, and mortality. According to the National Institutes of Health's (NIH) consensus Development Conference on Diagnosis and Treatment of Depression in Late Life, depressive symptoms occur in approximately 15% of community residents older than 65 (NIH, 1991). The prevalence of depression in homebound elderly adults is estimated to be between 26% and 44% (Banerjee et al., 1996). Lack of social support, physical disability, and chronic illness are associated with depression (Banerjee et al., 1996). Because of isolation due to disability, an increase in single elderly adult households, and the lack of primary medical care provided directly in the home, homebound elderly adults are especially vulnerable to poor outcomes in cases of depression. Nurses' and medical providers' role in assessing for depression and the availability of social supports in the home is especially significant in managing health costs and maintaining quality of life in elderly individuals.

The purposes of the study were to identify the prevalence of depression in homebound elderly adults receiving home health services and to determine the relationship between depression and social support systems in this pilot study sample.


Despite having fewer physical, cognitive, and functional impairments, frail elderly adults living alone were found to have higher rates of psychological distress, including depression, than their counterparts who did not live alone (Burnette & Muí, 1994). Depressive symptoms were higher among elderly adults who were White, and those who had less education, lower incomes, decrements in health and functional status, and unmet social and personal needs (Burnette & Mui, 1994).

Blixen and Kippes (1999) examined the relationship among depression, social support, and quality of life in 50 elderly individuals with osteoarthritis. Mild to moderate depression was found in 40% of the sample, and severe symptoms were reported by only 6% of participants. Blixen and Kippes found that depression did not significandy vary with osteoarthritis severity and attributed this finding to large social support networks of family and friends. Elderly participants gave high satisfaction ratings to those providing them with social support (Blixen & Kippes, 1999). Only 13 of the participants with osteoarthritis reported living alone and relying on formal outside support. This study seems to support Burnette and Mui's (1994) finding that depression is more common among those living alone with little social support. Aneshensel & Stone (1982) stated that social support is an effective buffer in moderating depressive symptoms.

Holahan and Holahan (1987) also examined the relationship of social support and depression in older adults in a longitudinal analysis and, as predicted, found that social support showed a significant inverse relationship to depression (r = -.47). The authors formally linked social support to Bandura's (1982) framework of self-efficacy which interacts in the social milieu and leads to the initiation and maintenance of effective coping and social support behaviors.

Banerjee et al. (1996) studied the effect of a psychogeriatric team consisting of psychiatric nurses, occupational therapists, doctors, a social worker, and a psychologist on depression in frail homebound elderly adults in London, England. The prevalence and severity of depression was found to be greater among disabled elderly individuals receiving home care than among elderly individuals in the general population. At 6 months follow up, the team was significantly more effective than general practitioners as measured by elderly adults' recovery from depression (Banerjee et al., 1996). Interventions provided by the team included prescription of antidepressants, psychological interventions (e.g., bereavement counseling), and social interventions (e.g., referral to a day care center).

Rotenberg and Hamel (1988) found that depression was negatively related to having peers to talk to frequently, but positively correlated to having reciprocal intimate relationships. The 42 elderly adults in a senior citizen complex did not show reciprocity of self-disclosure or a willingness to discuss intimate topics (e.g., personal habits, deep feelings, fears, close relationships) with their peers. The researches suggest these findings indicate elderly individuals become depressed not because they lack frequent social interaction, but as a result of negative communications and complaints. Thus, increased social interaction with more peers was associated with decreased depression, but negative communication (e.g., complaints about health problems) was interpreted to increase depression.

Krause (1986) identified four dimensions of social support: informational support, tangible help, emotional support, and integration (i.e., embeddedness of a person in a reciprocal network of shared obligations). Krause examined the effects on four stressors (i.e., bereavement, crime, network crises, financial problems) and found that all four dimensions of support buffered the affect of bereavement on depressive symptoms in 351 older adults. Emotional support was also found to buffer the affect of crime and legal stressors on emotional well being; and integration buffered the effects of network crises. However, none of the four types of support buffered the affect of financial stressors on depression (Krause, 1986).

In summary, current research indicates that depression in elderly individuals is a common problem which can be affected by social support. Social support is reported to be negatively correlated to depression or an effective intervention for depression in studies by Blixen and Kippes (1999), Banerjee et al. (1996) and Holahan and Holahan (1987). Rotenberg and Hamel (1988) indicated that depression in elderly individuals was negatively related to having a greater number of peers to communicate with, but positively related to exchanges of negative communication (e.g., complaints related to health and adverse events). Krause (1986) further delineated the effects of four types of social support on four kinds of stressors and depressive symptoms. Krause's findings indicate that different types of support are beneficial in different types of stressful situations.


Miller's (1992) framework for coping with chronic illness was used for this study. This framework focuses on overcoming powerlessness in chronic illness through achieving greater personal control which empowers the individual to function optimally in all aspects of life (Miller, 1992). Miller delineates seven sources of an individual's power to cope:

* Physical strength (physical reserve).

* Psychologic stamina and social support.

* Positive self-concept.

* Energy.

* Knowledge.

* Motivation.

* Belief system and hope.

This framework assumes that individuals with chronic illness have deficits in physical strength and energy, which are overcome by other power components that may need to be developed to prevent or overcome powerlessness. An example of drawing on resources to cope in the Miller model is an elderly adult with arthritis and depression who can no longer prepare his own meals. The older adult may rely on social supports such as friends or relatives to prepare meals, or on meals on wheels from a local organization. An elderly adult with depression, in addition to taking medications, may cope with and overcome illness by talking to family and friends, engaging in exercise, using humor, and using prayer and faith. Miller's framework can be used by nurses to aid elderly individuals in identifying resources to cope with chronic illness and depression.






Depression, the dependent variable, was measured by using the Geriatric Depression Scale (GDS) (Yesavage & Brink, 1983) long form with a sensitivity of 100% and specificity of 87% (Lyness et al., 1997). The GDS Yes or No format makes it easy to administer, and it is useful in detecting minor depression (Lyness et al., 1997).

Social support, the independent variable, was defined in terms of both quantity and quality of social interaction. Quantity of social support and interaction was defined as the number of formal social support services. Operationally, it was measured by having participants indicate the number of formal social support services they received, including:

* Meals on wheels.

* Homemaker.

* Home health nurse.

* Home health aide.

* Physical therapist.

* Occupational therapist.

* Adult day care.

* Support group.

* Transportation service.

* Case management.

Quality of social support was defined as perceived level of satisfaction with social interaction. It was operationally defined by the short form of the Social Support Questionnaire (SSQ-R) (Sarason et al., 1987) which measures perceived social support. The scale is composed of six, two-part items for identifying members of the person's support network (SSQ-N) and rating satisfaction from the network members (SSQ-S). The SSQ-N scoring is a possible 0 to 9 members of the support network per item, with a possible range of 0 to 54 individuals in the respondent's social support network. The SSQS measures support satisfaction on a 6point Likert-type scale. The range of SSQS scores is 6 to 36 with 1 being very dissatisfied and 6 being very satisfied. The long form SSQ has been used to measure social support in elderly individuals with osteoarthritis (Caifas, Kaplan, & Ingram, 1992). The reliability of the instrument is .94 or higher with an elderly population (Heitzman & Kaplan, 1988). The SSQR has similar internal consistency reliabilities and test-retest reliabilities as the 27-item SSQ (Sarason et al., 1987).

Demographic data was collected about the participants' age, gender, marital status, race, education, and diagnosis and current or prior use of antidepressants. Results were calculated as percents.


The study used a cross-sectional convenience sample of 25 home health patients. Eligible participants were 75 or older and homebound. Data was obtained in patients' homes.


Chi square was used to determine statistical associations between depression severity and perceived quality and quantity of social support Data was analyzed using Statistical Package for the Social Sciences (SPSS Inc., Chicago, IL) program for Windows (Redmond, WA) with levels of significance at the .05 level.


A sample of 25 home health patients were screened for depression using the long form of the GDS. Twenty (80%) participants were White, and 20% were AfricanAmerican. Nineteen participants were women and 6 were men. Participants ranged from age 75 to 98. Six of the participants were taking antidepressants; and three of these participants scored in the moderately depressed range. Depression did not increase with advancing age (Table 1).

Depression was significantly related to race (?2? = .003) with 55% of White participants and 40% of African- American participants reporting depressive symptoms. Depression was significantly related to gender (p = .009) with 67% of men and 47% of women reporting mild depression (Table 2). Depression was not significantly related to measures of formal social support, such as having meals on wheels, a home health nurse, or a home health aide (p = .162). Living alone and amount of education also were not significantly related to depressive symptoms in this sample. Using Pearson's correlation, depression was significantly negatively associated with increasing age, but positively associated with being a man, being unmarried, and needing formal social supports such as home health care and meals on wheels (Table 3).









This study implicates functional decline more than social support networks as affecting those with depressive symptoms. The measures of perceived level of social support in the six items in Sarason's Social Support Scale were tabulated by number of individuals. The number of actual individuals (range 1 to 8) providing social support was significantly related to depression (p = .028). Eightysix percent of those with more individuals (5 to 8) in their social support network scored moderately depressed. Satisfaction with individuals who provide social support was rated high by all participants and was not significantly related to depression (p = .102).

In this study, 52% of participants reported mild symptoms of depression, which supports Burnette and Mui (1994) who found 58.7% of elderly individuals reporting depressed feelings. Blixen and Kippes (1999) had similar findings, with 40% of their participants having mild depressive symptoms and 6% having severe symptoms. Banerjee et al. (1996) noted as much as 44% of their sample had depression. Participants in all of these studies were older than age 65 and had chronic medical conditions.


This study reinforces the complexity of assessing depression and social support because all participants rated their social support as satisfactory. For instance, a male participant with severe congestive heart failure was living with a supportive wife and had three daughters who visited regularly and provided support. However, he continued to have depressive symptoms caused by his disability and grief over the loss of an only son who had died of alcoholism 2 years prior. The participant's physician prescribed paroxetine HCl (Paxil) for a brief time, but the medication was discontinued because of side effects, and no other medications were substituted. Other participants also expressed grief over loss of functioning in spite of supportive families. The main source of support for most of the individuals was the immediate family or hired caregivers.

In previous studies, depressive symptoms were found to be significantly related to declines in medical and functional status (Blixen & Kippes, 1999; Burnette & Mui, 1994). This study supports previous research implicating functional decline more than lack of social support or social support satisfaction with depressive symptoms. This study, however, does not negate the efficacy of social support in overcoming depressive symptoms. The role of social support in coping with physical disability, illness, and depressive symptoms leaves unanswered questions requiring further study.

This study's non-random sample and small sample size limit the interpretation and generalizabiltiy of the findings. Although 52% of 25 participants reported signs of mild depressed in this small pilot study, a larger sample size would better assess the prevalence of depression and its relation of social support.


The results of this study and prior research have significant implications for interventions to help elderly individuals cope with depression and disability. In this study, chronic medical conditions limiting functional abilities were an influential predictor of depression in elderly individuals. Depression is reported to be influenced by a sense of control in life that includes satisfaction with every day life, choice, and self-confidence (Burnette & Mui, 1994). Unmet needs in personal and instrumental activities of daily living contributed to 11.9% of variance in depressive symptoms in Burnette and Mui's group. Home health nurses and geriatric practitioners are in a position to help homebound elderly adults cope with chronic disease by educating them in self-management skills. Miller's (1992) model also emphasizes that when power resources (e.g., knowledge, coping resources, problem solving attitude, sense of personal mastery, motivation) are decreased in chronically ill individuals, more nursing interventions are needed to help the person achieve a sense of control.

Nursing interventions must be adapted to individuals' specific deficiencies as well as to their specific coping styles and strengths. Community and family resources must be activated in response to the needs of those with depression. For example, a hospital in Chicago has a team of a physician, nurses, and social workers assessing and monitoring depression in homebound patients with chronic illness. Organizations such as Little Brothers-Friends of the Elderly provide social support and services to frail elderly adults.

Sloss et al. (2000) rank 21 medical conditions among elderly individuals as needing quality improvement based on data such as prevalence, affect on health and quality of life, and effectiveness of interventions. Depression ranked second behind pharmacologic management in prevalence and affect on health and quality of life, further reinforcing the need to treat depression. Geriatric nurses will be increasingly involved in assessing and managing depressive illness.

The increasing numbers of elderly adults with depression and the lack of physicians willing to visit and treat homebound elderly adults will further reinforce the role of the geriatric nurse or nurse practitioner. Effective treatment requires a team approach involving home health nurses, social workers, families, primary care providers, and older adult patients.


  • Aneshensel, CS., & Stone, J.D. (1982). Stress and depression: A test of the buffering model of social support. Archives of General Psychiatry, 39, 1392-1396.
  • Bandura, A. (1982). Self-efficacy mechanism in human agency. American Psychologist, 37(2), 122-147.
  • Banerjee, S., Shamash, K., Macdonald, A., & Mann, A. (19%). Randomized controlled trial of effect of intervention by psychogeriatric team on depression in frail elderly people at home. British Medical Journal, 313(7064), 1058-1061.
  • Blixen, C, & Kippes, C. (1999). Depression, social support, and quality of life in older adults with osteoarthritis. Image: Journal of Nursing Scholarship, 31(3), 221-226.
  • Burnette, D., & Mui, A. (1994). Determinants of self-reported depressive symptoms by frail elderly persons living alone. Journal of Gerontological Social Work, 22(1/2), 1-19.
  • Calfas, K.J., Kaplan, R.M., & Ingram, R.E. (1992). One year evaluation of cognitivebehavioural intervention in osteoarthritis. Arthritis Care and Research, 5(4), 202-209.
  • Heitzman, CA. & Kaplan, R.M. (1988). Assessment of mediods for measuring social support. Journal of Health Psychology, 7(1), 75-109.
  • Holahan, C, & Holahan, C (1987). Self efficacy, social support, and depression in aging: A longitudinal analysis. Journal of Gerontology, 42(1), 65-68.
  • Krause, N. (1986). Social support, stress, and well-being among older adults. Journal of Gerontology, 41(4), 512-519.
  • Lyness, J., Noel, T.K., Cox, C, King, D., Conwell, Y., & Caine, E.D. (1997). Screening for depression in elderly primary care patients: A comparison of the Center for Epidemiologic Studies - Depression scale and the geriatric depression scale. Archives of Internal Medicine, 157, 449-454.
  • Miller, J.E (1992) Coping with chronic illness overcoming powerlessness. Philadelphia, PA: EA. Davis Co.
  • National Institutes of Health. (1991, November 4-6). Diagnosis and treatment of depression in late life. NIH Consensus Statement, 9(3), 1-27.
  • Rotenberg, K, & Hamel, J. (1988). Social interaction and depression in elderly individuals. International Journal of Aging & Human Development, 27(4), 305-319.
  • Sarason, I.G., Sarason, B.R., Shearin, E.N., & Pierce, CR. (1987). A brief measure of social support: Practical and theoretical implications. Journal of Social and Personal Rehtionships, 4, 497-510.
  • Sloss, E.M., Solomon, D.H., Shekelle, PC, Young, R.T., Saliba, D., MacLean, CH., Rubenstein, L.Z., Schnelle, J.F., Kamberg, CJ., & Wenger, N.S. (2000). Selecting target conditions for quality of care improvement in vulnerable older adults. Journal of the American Geriatric Society, 48(4), 363-369.
  • Yesavage, J., & Brink, T. (1983). Development and validation of a geriatric screening scale: a preliminary report. Journal of Psychiatric Research, 17, 85-95.








Sign up to receive

Journal E-contents