Risk factors for depression among the oldest-old in urban congregate housing: Contribution of grief

Abstract

The objective of this study was to research grief and depression in oldest-old people living in urban congregate housing by examining the various types of grief that they experience by using a stress process model for depression among the oldest old. This study used a convenience sample of the 128 participants aged 80 and older living in congregate housing. We found that the two types of grief most significantly associated with depressive symptoms were grieving about relocating, and loss of health. Our findings demonstrate the need to explore a variety of grief which influences mental health among oldest-old.

Keywords

congregate housing depressive symptoms grief loss oldest-old

Introduction

Although the definition of the oldest-old group varies from study to study, the oldest-old group (age >75–85 years) is currently the fastest growing population in many countries, including China, the United States, India, Japan, Germany, and Russia and expected to increase 151 percent between 2005 and 2030 globally (National Institute on Aging, 2007). Such a population trend implies that there will be an increase in need for formal and informal support by this age group. Accordingly, the number of older adults living in retirement communities, such as congregate housing ‘a shared living environment designed to enhance older adults’ independent functioning’ including independent living senior apartments or assisted living, will also increase (Adams and Roberts, 2010: 474). Independent senior apartments are private apartments that include one or more supportive services such as meals, housekeeping, transportation, social activities, and laundry (Adams and Roberts, 2010). Assisted living facilities provide support for older adults with services that include not only meals, laundry, and housekeeping, but also medication reminders, and assistance with Activities of Daily Living (ADLs) and Instrumental Activities of Daily Living (IADLs) and medical care (Adams and Roberts, 2010; Zimmerman et al., 2006). Older adults living in congregate housing are more likely to have a smaller social network, be older, female, live alone, be unmarried, and display higher rates of depressive symptoms compared to community-dwelling older adults (Adams and Roberts, 2010). The prevalence of depressive symptoms increases throughout late adulthood and becomes the highest in the oldest-old group (aged > 80) (Glass et al., 1997). Moreover, in old age, people may experience grief more often compared to other groups in response to negative life events common in old age, such as bereavement, declining health, and relocation. Previous research has found relationships between these negative life events and depression (Meller et al., 1997). Great attention has been paid to understanding stress, coping, and mental health among older adults, but few studies have focused on those living in congregate housing, despite the fact that this group is at higher risk of depression. The present study examines the types of grief experienced by the oldest-old living in congregate housing and explores the factors associated with depressive symptoms in this vulnerable group.

Oldest-old and depression

Depression is one of the most common emotional problems among older adults and is characterized by dysphoric mood, or loss of interest or pleasure in life and free time activities (Bulut, 2009). Depressive symptoms include poor appetite, insomnia or hypersomnia, loss of energy, fatigue or tiredness, feelings of excessive guilt, inability to concentrate or think, and suicidal thoughts (Bulut, 2009; Ryan and Shea, 1996). Depression is not only a risk factor for mortality (Glass et al., 1997; Kaplan and Reynolds, 1988) but also for functional impairment (Hybels et al., 2001), decreased quality of life, and subjective distress among older adults (Lavretsky and Kumar, 2002). Between 11 percent and 40 percent of community-dwelling older adults report depressive symptoms, with an average of around 20 percent (Bulut, 2009). Approximately 43 percent of institutionalized older adults have been diagnosed with depression (Bulut, 2009; Reeker, 1997). Furthermore, oldest-older adults living in congregate housing tend to be at higher risk of developing depressive symptoms and major depression than community-dwelling older adults (Adams and Roberts, 2010). Despite the large number of studies on depression in older adults, relatively little is known about depression among the oldest-old group and even less about those living in congregate housing.

Oldest-old and grief

Negative stressful life events including bereavement, onset of significant health problem, and relocation are significant risk factors of depressive symptoms among older adults (De Beurs et al., 2001; Glass et al., 1997). However, the specific effects of negative life events on depression are inconsistent among the oldest-old group. For example, Jeon and Dunkle’s (2009) longitudinal study of 193 oldest-old people (aged >85) living independently found that trajectories of negative life events were not significantly associated with depression among the oldest-old group. However, in a longitudinal study by Meller and colleagues (1997) with 402 people older than 85 years living in a community, negative life events (i.e. death of partners, illness of others, and moving) were risk factors for depression. Since there have been so few studies of the influences of life events in the oldest-old group, these different findings have yet to be reconciled.

Other risk factors for depression and oldest-old

Socio-demographic

Although the findings on the relationship between age and the prevalence of depression are inconsistent, several studies have revealed the highest prevalence of depression was found in the oldest-old group in some studies (Blazer et al., 1991; Jeon and Dunkle, 2009; Pàlsson et al., 2001; Valvanne et al., 1996). For example, in the longitudinal study by Jeon and Dunkle (2009), the oldest-old (aged >85) group was likely to report increased depressive symptoms with age. However, other studies have found a decline in the propensity for depression with increasing age (aged >65 years) (Mojtabai and Olfson, 2004) or no significant relationship with increasing age (aged >65 to >85 years) (Meller et al., 1997; Minicuci et al., 2002). Education level is also associated with depression. In a review of 122 studies of the prevalence and predictors of depression in populations of older adults, Djernes (2006) suggested that older adults (aged >65) with lower education tended to report more depressive disorders and depressive symptoms (Gostynski et al., 2002; McCall et al., 2002). This has also been found in the oldest-old group (Beekman et al., 1995).

Physical and cognitive function

Both physical and cognitive impairment are risk factors in older adults for the onset or persistence of depressive symptoms and depressive disorders (Djernes, 2006). This also holds true for the oldest-old group (aged >75–90), for which several studies have found that physical (Forsell et al., 1998; Meller et al., 1997; Païvarintä et al., 1999) and cognitive functional impairment (Forsell et al., 1998) predict depression.

Loneliness

Recent studies have supported the significance of loneliness in later life (Adams et al., 2004a; Kwag et al., 2011; Pinquart and Sorensen, 2001). For instance, Kwag and colleagues (2011) reported that perceived stress had direct and indirect effects on loneliness among community-dwelling older adults (aged >65). Adams and colleagues (2004a) demonstrated that loneliness is a potential risk factor for depressive symptoms among older adults in congregate housing (aged 60–90). In particular, loneliness is more prevalent in the oldest-old group (aged >80) than the middle-aged group (Pinquart and Sorenson, 2001), and is among risk factors for the onset of depressive symptoms among the oldest-old (Meller et al., 1997; Roberts et al., 1997).

Conceptual model and research questions

Few studies have specifically focused on the relationships between stressor and outcome in this age group (Jeon and Dunkle, 2009). Therefore, in this study, one of the most widely used stress process models in sociology, psychology, gerontology, nursing and public health was applied to conceptualize the process that occurs for oldest-old individuals in congregate housing (Thoits, 2006). The stress process model proposed by Pearlin and colleagues (1981) includes four domains: stressors, resources, outcomes and the contextual or background information. This process model focuses on direct effects of various types of stressors, and resources for coping on outcome. Also the model proposes the indirect effects (e.g. mediating effect) in the stress process by illustrating how individuals may reduce the potential negative outcomes (e.g. depression) to stressful situations by controlling stress in their lives through mechanisms such as social support. This study included stressors such as physical and cognitive impairment and grief from various recent losses; resources such as social support, here represented negatively through emotional and social loneliness; with depressive symptoms as an outcome.

Given the gaps in our knowledge of the risk factors for depression in the oldest-old group, particularly the role of grief from common stressful life events, this study aims to address the following research questions:

Research question 1: What kinds of grief do the oldest-old living in urban congregate housing experience?

Research question 2: What are the relationships among presence of grief/various types of grief and depressive symptoms in the oldest-old living in urban congregate housing?

Research question 3: What are the relationships among stressors (physical and cognitive function, grief), resources (social and emotional loneliness), and depressive symptoms in the oldest-old living in urban congregate housing?

Methods

Sample

This study used a convenience sample of the 128 participants who were aged 80 and older living in one of six Continuing Care Retirement Facilities in northeast Ohio, out of a total population of approximately 1500 residents. Data were collected through a self-administered survey form and follow-up phone interview with the participants’ permission. To recruit participants, signs were posted in prominent locations, ﬂyers were distributed in residents’ open mailboxes, and labeled boxes of blank survey packets were placed near the residents’ mailbox area. Self-addressed stamped return envelopes were included in the survey packets so that participants could return them by mail. During the follow-up interviews participants were asked additional screening questions and two open-ended questions about any concerns they had and how they cope with aging in a congregate living setting. Trained research assistants with experience in gerontological social work or psychology conducted the phone interviews. The participants who completed both the survey and the follow-up telephone interview were provided with a $20 gift card.

Measures

Depression

Depressive symptoms were measured using the Geriatric Depression Scale (GDS; Brink et al., 1982) which has demonstrated adequate reliability and validity in over 300 studies (Adams and Roberts, 2010). This scale consists of 30 items with yes/no responses. The original scale was constructed by adding 30 items on depressive symptoms (Yesavage et al., 1983). Although there is no clear consensus on cut-off scores, the original developers of the scale recommend that 11 points and above indicates moderate to severe depression with a sensitivity of 84 percent and a specificity of 95 percent (Yesavage et al., 1983). Others recommend a cut-off of 14 points (Lyons et al., 1989), a 12-point cut-off (Lavretsky and Kumar, 2002), and a nine-point cut-off for the oldest-old residents (aged >85) of retirement communities (Watson et al., 2004). Adams et al. (2004b) proposed a six factor structure for the GDS including a nine-item Dysphoric Mood (i.e. downhearted and blue), six-item Withdrawal-Apathy-Vigor (WAV, i.e. prefer to day home), four-item Worry (i.e. afraid something bad may happen), four-item Cognitive Impairment (i.e. more problem with memory than most), four-item Hopelessness (i.e. feel situation is hopeless), and three-item Agitation (i.e. restless and fidgety). Cronbach’s alpha of the GDS scale in the current study was 0.852 and of each subscale was 0.756 (Dysphoric Mood), 0.587 (WAV), 0.620 (Worry), 0.598 (Cognitive Impairment), 0.515 (Hopelessness), and 0.349 (Agitation). The Agitation subscale was excluded because of the low level of reliability (Cronbach’s alpha), and in the current study both a summed score and the five reliable subscales were used.

Physical and cognitive function

The levels of physical function in everyday tasks were measured by the Katz Index of Activities of Daily Living (ADLs; Katz et al., 1963) and the Older Americans Resources Scale for Instrumental Activities of Daily Living (IADLs; Duke University Center for the Study of Aging and Human Development, 1978). The six-item ADL scale used a three-anchor response (0 = unable, 1 = some help, 2 = able) with the total range from 0 to 12. Higher scores indicate better functioning. Cronbach’s alpha was 0.930. The seven-item IADL scale was rated on a three-point scale based on responses about capability to perform a task (0 = unable, 1 = some help, 2 = without help). The total score ranges from 0 to 14 with a higher score indicating greater functionality. Cronbach’s alpha was 0.808. The level of cognitive function was assessed by using the Telephone Interview for the Cognitive Screening – Modified (TICS-M; Welsh et al., 1993), a cognitive screening measure modeled after the Mini-Mental State Exam (MMSE; Folstein et al., 1975). This scale was designed for use in a telephone interview to test functioning in several domains such as concentration, orientation, memory, naming, comprehension, and abstraction. It consists of 21 items with a maximum score of 50 points, with lower scores reﬂecting higher cognitive impairment. Cronbach’s alpha was 0.647.

Loneliness

Loneliness was measured by the De Jong-Gierveld Loneliness Scale (De Jong-Gierveld and Van Tilberg, 1999), which consists of 11 dichotomous items. The authors identified two subscales of Social (five items, i.e. someone to talk to about problems) and Emotional Loneliness (six items, i.e. sense of emptiness). The scores for Social Loneliness range from 0 to 5 and for Emotional Loneliness from 0 to 6, and higher scores on both subscales indicate more loneliness. Cronbach’s alpha for Emotional Loneliness was 0.797 and Social Loneliness was 0.823.

Grief

We also included a yes/no item written for this study: ‘I am grieving over an important loss’. If they answered yes, participants were asked to indicate the source of their grief from a list of specific types of losses including death of their spouse, death of a family member or friend, death or loss of a pet, moving house, and loss of health. Respondents could check all types of grief that applied to them.

Demographic variables

Demographic variables included age in years, and education was measured with the following three categories: high school, beyond high school and beyond college.

Statistical analysis plan

To address the research questions, descriptive analyses of demographic and study variables were conducted. Bi-variate association was used to examine correlations among possible predictors to test for multicollinearity. Experiencing grief was included as a dichotomous variable as a predictor in six hierarchical multiple regression analyses on depression symptoms after controlling for demographic variables. And finally, various types of grief were included in six hierarchical multiple regression analyses on depressive symptoms (one total score and five subscales of depression) after controlling for demographic variables.

Results

Description of the sample population

As shown in Table 1, the average age of the participants was 86.56 years. The majority of participants were female (79.7%) and White (90.6%). Over two-thirds of the participants were widowed. The participants predominantly lived alone in an apartment, or in a unit within the congregate facility. More than three-quarters of the participants had some college education or more advanced education. Most participants (84%) in this study were strongly and moderately satisfied with their income adequacy. Of the 113 respondents who reported religion, 40.6 percent were Protestant and 30.5 percent were Jewish.

Table 1.

Descriptive statistics of socio-demographic and key variables.

Variables	% or M (SD)
Gender	Male = 20.3%
	Female = 79.7%
Race	African American = 7.8%
	White = 90.6%
Marital status	Single = 7.8%
	Widowed = 16.4%
	Living with a partner = .8%
	Separated/divorced = 5.5%
	Widowed = 69.5%
Income	Can’t make ends meet = .8%
	Just managed to get by = 14.1%
	Have enough money with a little extra = 45.3%
	Money is not a problem = 39.1%
Religion	Catholic = 10.2%
	Protestant = 40.6%
	Jewish = 30.5%
	No religious preference = 1.6%
Age	86.56 (4.12)
Years of education	8th grade or less = 4.7%
	Some High School = 5.5%
	High School graduate = 18.8%
	Some college = 16.4%
	College graduate = 17.2%
	Post-graduate work = 11.7%
	Advanced degree = 25%
GDS total score	6.52 (4.92)
GDS Worry	.57 (.93)
GDS Hopelessness	.55 (.85)
GDS Cognitive Impairment	1.11 (1.21)
GDS Dysphoric Mood	1.13 (1.72)
GDS WAV	2.66 (1.62)
ADLs	11.48 (2.30)
IADLs	11.48 (2.30)
Emotional Loneliness	1.84 (1.91)
Social Loneliness	1.85 (1.85)
Grief (yes/no)	No = 88 (68.9%)
	Yes = 40 (31.1% )

One-fourth of the sample reported moderate to severe depression based on Watson and colleagues’ nine-point cut-off for the oldest-old residents (aged >85) of retirement communities. The mean depression score for the sample was 6.52 (SD = 4.92), with a range from 0 to 23: the means of subscales of depressive symptoms were − 0.57 (SD = .93) (Worry), .55 (SD = .85) (Hopelessness), 1.11 (SD = 1.21) (Cognitive Impairment), 1.13 (SD = 1.72) (Dysphoric Mood). In terms of IADL, the most needed instrumental help was housework (63%). Respondents also needed some help with going to places outside of walking distance (40%), grocery shopping (35.9%), meal preparation (32.8%), handling finances (15.6%), taking medications (6%), and telephone use (3.1%). The ADL measure revealed that most participants did not need help with activities of daily living, with only 5.5 percent needing assistance in bathing, 2.3 percent in control of bladder or bowels, and 1.6 percent each in dressing, using the bathroom, transferring from chair to bed or vice versa, and eating. The TICS-M score ranged from 14 to 44, and 28 individuals (21.9%) were identified as having mild cognitive impairment. On average the participants reported low levels of emotional and social loneliness.

In order to examine what kinds of grief the oldest-old living in urban congregate housing experience, a descriptive analysis was performed. In this study, the majority of the sample was not currently grieving a recent loss. In this study, 40 (31.1%) participants reported grieving a recent loss, approximately three-quarters of which reported one type of grief, 3 percent reported two types, 0.8 percent reported three types, and 4.7 percent reported four types of grief. Of those who were grieving, 85 percent reported a loss of spouse, family, and friends, 10 percent were grieving moving, and 12 percent reported a loss of health.

Estimation of the predictors for depressive symptoms

We next examined the relationship among stressors including grief, social and emotional loneliness and depressive symptoms. Table 2 summarizes the OLS regression results with the categorical variable of grief as a predictor of the total score of depressive symptoms and scores of the five subscales (e.g. Worry, Hopelessness, Cognitive Impairment, Dysphoric Mood, Withdrawal-Apathy-Vigor) of depressive symptoms. Higher levels of emotional and social loneliness, and grieving a recent loss were significantly associated with more depressive symptoms. Grieving a recent loss was significantly related to being more worried. Higher levels of emotional and social loneliness were significantly associated with feeling hopeless. Lower level of TICS-M was related to experiencing more cognitive symptoms. Lower level of TICS-M, higher levels of emotional and social loneliness and grieving a recent loss were significant predictors of Dysphoric Mood. Lower levels of IADL and higher levels of social loneliness were significantly associated with WAV. ADLs, age, and education were not significant predictors of either the total score of depressive symptoms or subscales of depressive symptoms.

Table 2.

Multiple regression model for grief on depressive symptoms.

Independent variables	GDS total β (SE)	GDS subscales
Independent variables	GDS total β (SE)	Worry β (SE)	Hopelessness β (SE)	Cognitive Impairment β (SE)	Dysphoric Mood β (SE)	WAV β (SE)
Age	.053 (.088)	.096 (.021)	.065 (.016)	.082 (.026)	−.050 (.030)	.077 (.037)
Education	−.060 (.647)	−.030 (.155)	−.112 (.119)	.063 (.193)	−.104 (.222)	.008 (.269)
IADL	−.123 (.162)	.012 (.039)	−.005 (.030)	−.174 (.048)	.082 (.056)	−.280 (.068) **
ADL	.004 (.427)	.154 (.102)	−.260 (.079)	−.073 (.127)	.075 (.146)	.012 (.178)
TICS-M	−.108 (.067)	−.040 (.016)	.027 (.012)	−.244 (.020) *	−.183 (.023) *	.075 (.028)
Grief (Yes/no)	.199 (.775) *	.281 (.186) **	.134 (.142)	.132 (.230)	.259 (.265) **	−.038 (.322)
Emotional Loneliness	.350 (.240) ***	.203 (.058)	.265 (.044) *	.221 (.071)	.337 (.082) **	.180 (.100)
Social Loneliness	.267 (.235) **	.170 (.056)	.215 (.043) *	−.038 (.070)	.242 (.081) *	.256 (.098) *
R ²	.449	.245	.367	.200	.439	.242
F	9.898***	3.941	7.021	3.025**	9.501***	3.875**

p < .05; **p < .01; ***p < .001.

Note: WAV = Withdrawal-Apathy-Vigor.

Table 3 summarizes the OLS regression results of six models with types of grief as predictors of both the total score of depressive symptoms and the five subscales of depressive symptoms. Our results indicated that the grief related to moving, and higher levels of emotional and social loneliness were significant predictors of a higher level of depressive symptoms. Grief related to moving and a higher level of emotional loneliness were associated with greater sense of worry. Having more problems with daily activities, grieving the loss of health, and feeling more emotional and social loneliness were significant predictors of greater sense of hopelessness. Lower scores on the TICS-M, grieving the loss of health and a higher level of emotional loneliness were significant predictors of more cognitively-related depressive symptoms. Lower scores on the TICS-M and higher levels of emotional and social loneliness were associated with Dysphoric Mood-related depressive symptoms. Lower levels of IADL and higher levels of social loneliness were significant predictors of more WAV-related depressive symptoms.

Table 3.

Multiple regression model for types of grief on depressive symptoms.

Independent variables	GDS total β (SE)	GDS subscales
Independent variables	GDS total β (SE)	Worry β (SE)	Hopelessness β (SE)	Cognitive Impairment β (SE)	Dysphoric Mood β (SE)	WAV β (SE)
Age	.098 (.091)	.168 (.021)	.089 (.017)	.132 (.027)	−.031 (.032)	.098 (.038)
Education	−.046 (.654)	.003 (.152)	−.105 (.119)	.073 (.194)	−.093 (.230)	.010 (.271)
IADL	−.107 (.167)	.036 (.039)	.026 (.030)	−.162 (.049)	.081 (.059)	−.269 (.069)**
ADL	.000 (.437)	.163 (.101)	−.266 (.080)**	−.089 (.130)	.078 (.154)	.003 (.181)
TICS-M	−.089 (.068)	−.020 (.016)	.068 (.012)	−.232 (.020)*	−.174 (.024)*	.081 (.028)
Types of grief Loss of spouse/friend/family	.037 (.920)	.116 (.213)	−.066 (.167)	−.046 (.272)	.159 (.323)	−.097 (.381)
Moving	.206 (1.415)*	.359 (.328)**	.073 (.258)	.220 (.418)	.086 (.497)	.087 (.586)
Loss of health	−.028 (1.448)	−.095 (.336)	.155 (.264)*	.071 (.418)*	−.022 (.508)	−.014 (.600)
Emotional Loneliness	.375 (.240)***	.226 (.056)*	.305 (.044)**	.236 (.071)*	.373 (.084)***	.174 (.099)
Social Loneliness	.251 (.239)**	.141 (.055)	.214 (.043)*	−.058 (.071)	.240 (.084)*	.246 (.099)*
R ²	.452	.298	.383	.213	.415	.251
F	7.851***	4.026***	5.905***	2.575**	6.729***	3.175***

p < .05; **p < .01; ***p < .001.

Note: WAV = Withdrawal-Apathy-Vigor.

Discussion

The objective of this study was to research grief and depression in oldest-old (aged >80 years) people living in urban congregate housing by examining the various types of grief that they experience. In order to do so, we utilized a stress process model for depression among the oldest-old in congregate housing. This study provides evidence that oldest-old who were grieving a recent loss were likely to report higher levels of depressive symptoms. We found that the two types of grief most significantly associated with depressive symptoms were grieving about relocating, which was predictive of a higher total GDS score and the Worry subscale on the GDS, and loss of health, which was predictive of the Hopelessness and Cognitive Impairment subscales on the GDS.

This study makes three significant contributions. First, results here confirm that grief due to recent losses plays a significant role in the depressive symptoms of elders. Grieving any recent loss was significantly associated with the total depression score, as well as with the worry subscale and the Dysphoric Mood subscale. Our findings support previous studies that found that older adults grieving a loss reported more depressive symptoms (Adams et al., 2004a; Glass et al., 1997), including the oldest-old group (Meller et al., 1997). As prior research on grief and bereavement has shown (Bonanno and Kaltman, 2001), the oldest-old commonly experience feelings of Dysphoric Mood and worry about the present and future due to recent losses, as do the young-old population.

Second, the oldest-old group in congregate housing experienced grief from various stressful life events, including loss of spouse, friends, family, and health and relocation. Grieving about relocating and grieving about health-related losses were most significantly associated with depressive symptoms. Grieving about having moved household (presumably into the congregate housing) was predictive of a higher total GDS score, suggesting that feeling grief from relocation has a greater impact on depressive symptoms in the oldest-old group. Relocation involves changes in physical and social environments and the oldest-old may face greater challenges in acclimatizing themselves to new environments regardless of their desire or willingness (Glass et al., 1997). By moving to a new place, they may find it difficult to maintain their old relationships and lose interest in developing new relationships. They may feel sad about leaving their previous homes and giving up some of the freedoms and benefits of living in the community.

Our findings also revealed that different types of grief are significantly associated with different types of depressive symptoms. Grieving about loss of health was predictive of the Hopelessness and Cognitive Impairment subscales on the GDS. Perceived health implies time to death rather than chronological age in the oldest-old group in the last stage of life (Bravell and Berg, 2008). It is possible that losing health implies a shortened amount of time remaining, resulting in feelings of hopelessness about the present and the future. In addition, feeling a loss of heath may affect perceived cognitive abilities, which is a significant risk factor for depressive symptoms in older adults (Djernes, 2006; Margrett et al., 2010).

Unlike previous studies (De Beurs et al., 2001; Glass et al., 1997; Meller et al., 1997), we found that grieving over the loss of spouse, family, and friends was not significantly related to depressive symptoms. A previous study on the oldest-old group showed no association between negative life events, including death of family members, and depression (Jeon and Dunkle, 2009). One interpretation of these findings is that as people get older and have already experienced many deaths of family or friends, such losses may no longer be surprising or as affectively charged. Instead, it seems that day-to-day activities, such as eating, sleeping, and other activities for survival, may be more critical issues to the oldest-old group.

Following the stress process model, we found that other risk factors (i.e. physical and cognitive functioning and loneliness) were associated with depression, consistent with previous studies (Forsell et al., 1998; Meller et al., 1997; Skoog, 1993). Many in our oldest-old sample population needed help to maintain their daily activities due to impairments in their physical abilities and cognitive functioning, which may impact their perceived quality of life. The loss of social and personal resources resulting in emotional and social loneliness also plays a significant role in depression. Although the oldest-old group may have experienced and coped with many more losses throughout their lives than the young-old group and may have greater resilience, they are also affected by the absence of a social network, everyday feelings of boredom, social marginalization, and emotional loneliness (Adams et al., 2004a; Kwag et al., 2011; Pinquart and Sorensen, 2001). Perhaps because of a smaller overall close social network, in an earlier study we found that the oldest-old participants were more invested in spending time with old friends and family members than the younger-old participants were. This study contributes to the understanding of the stress process model, in particular the effects of grief due to various losses and the risk factors for depressive symptoms in the oldest-old living in congregate housing. Nevertheless, there are four main limitations to the current study. First, we utilized cross-sectional data. Despite the higher level of mortality in the oldest-old group, longitudinal studies of grief and depression are needed to fully understand the process of grief and the effects of grief on depression. Second, the results might not generalize to all the oldest-old because oldest-old who were not as physically and psychological capable may not voluntarily participate in this study. Third, although we used a list of specific types of losses, we only included a small number of items. It would be beneficial to use a more detailed measure such as the Geriatric Scale of Recent Life Events (Kahana et al., 1982) to allow for a greater understanding of the effects of grief due to various types of loss. Fourth, we used a convenience sample of oldest-old living in congregate housing and most participants in this study were Caucasian and female. Therefore, future research should consist of a longitudinal study with more detailed measures on a more diverse group, including racial/ethnic minorities. In spite of the previously mentioned limitations, our findings highlight the need to provide tailored interventions to cope with losses and deal with depressive symptoms in order to meet their individual needs. In order to alleviate stress from relocation in the oldest-old group, they should be encouraged to maintain old relationships with their friends, family, or close neighbors through phone calls or invitations to the facilities. In particular, the newly moved oldest-old would benefit greatly from visits by old acquaintances and opportunities to meet other residents of the facilities, such as support groups and meaningful social activities. Also, service providers at congregate housing facilities should understand the propensity for feelings of loneliness and depression and the individual needs (i.e. coping skills, emotional and social support, and physical help) of this group. Such efforts would reduce stress levels and improve emotional distress in the oldest-old living in congregate housing.

Footnotes

Funding

This study was funded by a Hartford Geriatric Faculty Scholars Award to Dr. Kathryn Betts Adams.

Author biographies

Heehyul Moon, MSW, is a PhD candidate at Mandel School of Applied Social Sciences, Case Western Reserve University, Cleveland, Ohio, USA.

Kathryn Betts Adams, PhD, LISW-S, is Associate Professor of Social Work at Mandel School of Applied Social Sciences, Case Western Reserve University, Cleveland, Ohio, USA.

Amy Restorick Roberts is a PhD candidate at Mandel School of Applied Social Sciences, Case Western Reserve University, Cleveland, Ohio, USA.

References

Adams

K.B.

Roberts

(2010) ‘Reported Coping Strategies and Depressive Symptoms among African American and White Residents of Congregate Housing’, Journal of Gerontological Social Work 53(6): 473–94.

Adams

K.B.

Matto

H.C.

Sanders

(2004b) ‘Confirmatory Factor Analysis of the Geriatric Depression Scale’, The Gerontologist 44(6): 818–26.

Adams

K.B.

Sanders

Auth

E.A.

(2004a) ‘Loneliness and Depression in Independent Living Retirement Communities: Risk and Resilience Factors’, Aging and Mental Health 8: 475–85.

Beekman

Deeg

van Tilburg

Smit

J.H.

Hooijer

van Tilburg

(1995) ‘Major and Minor Depression in Later Life: A Study of Prevalence and Risk Factors’, Journal of Affective Disorders 36: 65–75.

Blazer

Burchett

Service

George

(1991) ‘The Association of Age and Depression among the Elderly: An Epidemiologic Study’, Journal of Gerontology: Medical Sciences 46: 210–15.

Bonanno

Kaltman

(2001) ‘The Varieties of Grief Experience’, Clinical Psychology Review 21: 705–73.

Bravell

M.E.

Berg

(2008) ‘Health, Functional Capacity, Formal Care, and Survival in the Oldest Old: A Longitudinal Study’, Archives of Gerontology and Geriatrics 46(1): 1–14.

Brink

T.L.

Yesavage

J.A.

Lum

Heersema

Adey

Rose

T.L.

(1982) ‘Screening Tests for Geriatric Depression’, Clinical Gerontologist 1: 37–44.

Bulut

(2009) ‘Late-Life Depression: A Literature Review of Late-Life Depression and Contributing Factors’, Anales de Psicología 25(1): 21–6.

10.

De Beurs

Beekman

Geerlings

Deeg

van Dyck

van Tilburg

(2001) ‘On Becoming Depressed or Anxious in Late Life: Similar Vulnerability Factors but Different Effects of Stressful Life Events’, British Journal of Psychiatry 179: 426–31.

11.

De Jong-Gierveld

van Tilburg

(1999) Manual of the Loneliness Scale. Amsterdam, The Netherlands: Vrije Universiteit.

12.

Djernes

J.K.

(2006) ‘Prevalence and Predictors of Depression in Populations of Elderly: A Review’, Acta Psychiatrica Scandinavica 113(5): 372–87.

13.

Duke University Center for the Study of Aging and Human Development (1978) Multidimensional Functional Assessment: The OARS Methodology. Durham, NC: Duke University Press.

14.

Folstein

McHugh

(1975) ‘Mini-mental State’, Journal of Psychiatric Research 12(3): 189–98.

15.

Forsell

Jorm

A.F.

Winblad

(1998) ‘The Outcome of Depression and Dysthymia in a Very Elderly Population: Results from a Three-year Follow-up Study’, Aging & Mental Health 2: 100–4.

16.

Glass

Kasl

Berkman

(1997) ‘Stressful Life Events and Depressive Symptoms among the Elderly’, Journal of Aging & Health 9: 70–89.

17.

Gostynski

Ajdacic-Gross

Gutzwiller

Michel

Herrman

(2002) ‘Depression bei Betagten in der Schweitz’, Nervenartz 73: 851–60.

18.

Hybels

C.F.

Blazer

D.G.

Pieper

C. F.

(2001) ‘Toward a Threshold for Subthreshold Depression: An Analysis of Correlates of Depression by Severity of Symptoms Using Data from an Elderly Community Sample’, The Gerontologist 41(3): 357–65.

19.

Jeon

H.S.

Dunkle

R. E.

(2009) ‘Stress and Depression among the oldest-old: A Longitudinal Analysis’, Research on Aging 31(6): 661–87.

20.

Kahana

Fairchild

Kahana

(1982) ‘Adaptation’, in Mangen

D. J.

Peterson

W. A.

(eds) Research Instruments in Social Gerontology, Vol. I: Clinical and Social Psychology, pp. 145–93. Minneapolis: University of Minnesota Press.

21.

Kaplan

G.A.

Reynolds

(1988) ‘Depression and Cancer Mortality and Morbidity: Prospective Evidence from the Alameda County Study’, Journal of Behavioral Medicine 11(1): 1–13.

22.

Katz

Ford

A. B.

Moskowitz

R. W.

Jackson

Jaffe

(1963) ‘Studies of Illness in the Aged: The Index of ADL: A Standardized Measure of Biological and Psychosocial Function’, Journal of the American Medical Association 12: 914–19.

23.

Kwag

H.K.

Martin

Russel

Franke

Kohut

(2011) ‘The Impact of Perceived Stress, Social Support, and Home Based Physical Activity on Mental Health among Older Adults’, The International Journal of Aging and Human Development 72(2): 137–54.

24.

Lavretsky

Kumar

(2002) ‘Clinically Significant Non-major Depression: Old Concepts, New Insights’, American Journal of Geriatric Psychiatry 10: 239–55.

25.

Lyons

J.S.

Strain

J.J.

Hammer

J. S.

Ackerman

A. D.

Funlop

(1989) ‘Reliability, Validity, and Temporal Stability of the Geriatric Depression Scale in Hospitalized Elderly’, International Journal of Psychiatry in Medicine 19: 203–9.

26.

McCall

N.T.

Parks

Smith

Pope

Griggs

(2002) ‘The Prevalence of Major Depression or Dysthymia among Aged Medicare Fee-for-Service Beneﬁciaries’, International Journal of Geriatric Psychiatry 17: 557–65.

27.

Margrett

Martin

Woodard

J. L.

Miller

L. S.

MacDonald

Baenziger

Siegler

I. C.

Davey

Poon

(2010) ‘Depression among Centenarians and the Oldest Old: Contributions of Cognition and Personality’, Gerontology 56(1): 93–9.

28.

Meller

Fichter

M. M.

Schröppel

(1997) ‘Risk Factors and Psychosocial Consequences in Depression of Octo- and Nonogenarians: Results of an Epidemiological Study’, European Archives of Psychiatry and Clinical Neuroscience 247: 278–87.

29.

Minicuci

Maggi

Pavan

Enzi

Crepaldi

(2002) ‘Prevalence Rate and Correlates of Depressive Symptoms in Older Individuals: The Veneto Study’, Journal of Gerontology: Medical Sciences 57: 155–61.

30.

Mojtabai

Olfson

(2004) ‘Major Depression in Community-dwelling Middle-aged and Older Adults: Prevalence and 2- and 4-years Follow-up Symptoms’, Psychological Medicine 34: 623–34.

31.

National Institute on Aging (2007) ‘Why Population Aging Matters: AGlobal Perspective’, available online at: http://www.nia.nih.gov/NR/rdonlyres/9E91407E-CFE8-4903-9875-D5AA75BD1D50/0/WPAM.pdf (accessed June 2011).

32.

Païvarintä

Verkkoniemi

Niinistö

Kivelä

S.-L.

Sulkava

(1999) ‘The Prevalence and Associates of Depressive Disorders in the Oldest-old Finns’, Social Psychiatry and Psychiatric Epidemiology 34: 352–9.

33.

Pàlsson

S.P.

Östling

Skoog

(2001) ‘The Incidence of First-onset Depression in a Population Followed from the Age of 70 to 85’, Psychological Medicine 31: 1159–68.

34.

Pearlin

Lieberman

Menaghan

Mullan

(1981) ‘The Stress Process’, Journal of Health and Social Behavior 22: 337–56.

35.

Pinquart

Sorensen

(2001) ‘Influences on Loneliness in Older Adults: A Meta-analysis’, Basic and Applied Social Psychology 23(4): 245–68.

36.

Reeker

(1997) ‘Personal Meaning, Optimism, and Choice: Existential Predictors of Depression in Community and Institutional Elderly’, The Gerontologist 37: 709–16.

37.

Roberts

Kaplan

G. A.

Shema

S. J.

Strawbridge

W. J.

(1997) ‘Prevalence and Correlates of Depression in an Aging Cohort: The Alameda County Study’, Journal of Gerontology: Social Sciences 52: 252–8.

38.

Ryan

Shea

M. E.

(1996) ‘Recognizing Depression in Older Adults: The Role of the Dietitian’, Journal of the American Dietetic Association 96(10): 1042–53.

39.

Skoog

(1993) ‘The Prevalence of Psychotic, Depressive and Anxiety Syndromes in Demented and Non-demented 85-year–olds’, International Journal of Geriatric Psychiatry 8: 247–53.

40.

Thoits

P.A.

(2006) ‘Personal Agency in the Stress Process’, Journal of Health and Social Behavior 47(4): 309–23.

41.

Valvanne

Juva

Erkinjuntti

Tilvis

(1996) ‘Major Depression in the Elderly: A Population Study in Helsinki’, International Pyschogeriatrics 8: 437–43.

42.

Watson

L.C.

Lewis

C.L.

Kistler

C. E.

Amick

H. R.

Boustani

(2004) ‘Can We Trust Depression Screening Instruments in Healthy “Old-old” Adults?’, International Journal of Geriatric Psychiatry 19: 278–85.

43.

Welsh

K.A.

Breitner

Magruderhabib

(1993) ‘Detection of Dementia in the Elderly Using Telephone Screening of Cognitive Status’, Neuropsychiatry, Neuropsychology and Behavioral Neurology 6: 103–10.

44.

Yesavage

Brink

T. L.

Rose

Lum

Huang

Adley

Leirer

(1983) ‘Development and Validation of a Geriatric Depression Screening Scale: A Preliminary Report’, Journal of Psychiatric Research 17: 37–49.

45.

Zimmerman

S.I.

Munn

Koenig

(2006) ‘Social Work Practice in Assisted Living Settings’, in Berkman

D’Ambruoso

(eds) Handbook of Social Work in Health and Aging, pp. 677–84. New York: Oxford University Press.