Social Capital and Loneliness Among the Very Old Living at Home and in Institutional Settings

Abstract

Objective: The aim of this study was to investigate the association between aspects of social capital and loneliness among the very old living at home and in institutional settings. Method: Half of those aged 85 years, and all 90- and 95-year-olds and older in urban and rural municipalities in northern Sweden and western Finland were invited to participate in a cross-sectional population-based study in 2005-2007. A sample of 483 participants who completed the study was included in the analyses. Results: Loneliness was experienced by 55% percent of those living in institutional settings often or sometimes and 45% of those living in their own homes. Loneliness was closely related to living alone, to depression, and to region (northern Sweden). Discussion: Social capital and loneliness are context dependent (i.e., geographical or living environment). Among the very old, the link between social capital resources and loneliness is also highly influenced by health status.

Keywords

social capital loneliness very old Finland Sweden

Introduction

Loneliness is experienced at all life stages of the life course and is not unique to the world of older people (Dykstra, 2009). However, loneliness is often considered to be a problem of growing older. Older people are at higher risk of increasing health problems as well as social isolation through loss of spouse, close family members and friends. In addition, residential relocation is common, moving from a family home to an institution, which implies a change in informal relations. All these factors are associated with an increased risk of experiencing loneliness (Dykstra, 2009; Hawkley & Cacioppo, 2010; Jylhä & Saarenheimo, 2010). Although researchers have provided evidence that social resources, including social capital, are linked to various aspects of well-being such as absence of loneliness (Islam, Merlo, Kawachi, Lindström, & Gerdtham, 2006; Kim, Subramanian, & Kawachi, 2008; Nyqvist, Forsman, Giuntoli, & Cattan, 2012), few studies have focused on the very old, or people aged 85 and above.

There are challenges in defining loneliness due to the differing theoretical taxonomies used to explain the concept. Frequently loneliness is described as a subjective, unpleasant, and distressing phenomenon stemming from a discrepancy between an individual’s desired and achieved levels of social relations (de Jong Gierveld & Van Tilburg, 2010; Perlman & Peplau, 1982; Victor, Grenade, & Boldy, 2005; Victor, Scambler, & Bond, 2009). Loneliness could therefore be said to arise from the perception of a mismatch between the actual and expected quality and quantity of social interaction and resources such as social capital.

The theoretical concept of social capital was made popular in particular by Putnam (1993), who defines it as “features of social organisation, such as trust, norms and networks, that can improve the efficiency of society by facilitating coordinated actions” (p. 167). According to Putnam (1993, 2000), civic engagement and social participation are key sources of cooperation that advance the collective welfare of society. However, the emphasis on formal social networks within society, as suggested by Putnam, is problematic particularly when focusing on older people (Nyqvist et al., 2012). Social capital in terms of participation may be quite different in later life when health deteriorates and the possibilities to engage in community life are limited. Furthermore, informal social interactions and activities between close friends and within family have been established as the key social capital factors for health and well-being in older people (Forsman, Herberts, Nyqvist, Wahlbeck, & Shierenbeck, 2013). Therefore, when it comes to the very old, the interaction between individuals within various living contexts at the micro or individual level (family and friends) may be as important as community or society level social capital.

Social capital incorporates structural (quantity) and cognitive aspects (quality) of resources. Whereas structural social capital describes the networks, relationships, and institutions that bring groups of people together, cognitive social capital represents the subjective side of social capital and encompasses values and perceptions. Although the causal relationship between structural and cognitive aspects is unclear (Hooghe & Stolle, 2003; Lindström, 2004; Putnam, 2000), both aspects are important for health and well-being, and most likely also for the experience of loneliness.

Social Capital and Loneliness Among the Very Old in Various Contexts

To our knowledge social capital in relation to loneliness among the very old has not been studied previously. Among older people in general, aspects of structural social capital have been associated with loneliness (Litwin & Shiovitz-Ezra, 2011). Nevertheless, well-known indicators of social capital, such as social networks and social support, have coincided with well-known health resources. Research in this area suggests that living alone (Victor, Scambler, Bond, & Bowling, 2000; Wenger & Davies, 1996), widowhood (Dykstra & de Jong Gierveld, 1999; Pinquart, 2003; Tijhuis, de Jong Gierveld, Feskens, & Kromhout, 1999) as well as lack of contacts with friends and neighbors (Bondevik & Skogstad, 1998; Holmén, Ericsson, Andersson, & Winblad, 1992) are risk factors for loneliness. Several studies have reported an association between social support, the quality and quantity of social networks and loneliness (de Jong Gierveld, 1998; Penninx et al., 1999; Stephens, Alpass, Towers, & Stevenson, 2011). Loneliness has also been found to be more common among those living in institutions (Drageset, Kirkevold, & Espehaug, 2011; Savikko, Routasalo, Tilvis, Strandberg, & Pitkälä, 2005; Tijhuis et al., 1999; Tsai & Tsai, 2011), although the significance of social resources among those living in their own homes compared with those living in institutions have not been addressed.

An exception is a study by Prieto-Flores, Forjaz, Fernandez-Mayoralas, Rojo-Perez, and Martinez-Martin (2011) who examined factors associated with loneliness in noninstitutionalized and institutional older people aged 60 and above in Spain. They noticed similarities as well as disparities in the explanation of loneliness within the two study groups. Depression and functional dependence were significant factors for loneliness in both groups whereas other factors were specific to one or the other group. To live without a partner was linked to loneliness among the noninstitutionalized older adults while contacts with family, friends, or neighbors were linked to loneliness among residents in care settings. This contradicts the results from Bondevik and Skogstad’s (1996) study, who found that few contacts with children, grandchildren and other family members were associated with loneliness among community residents, but not among residents in care settings. A possible explanation for these contradictory findings could be the contextual differences, in terms of geographical locations and the residential settings. In the current study, the connection between social capital and experienced loneliness is examined in a Nordic context among the very old living either at home or in institutional settings.

Aims

Given that the very old is the fastest growing age group in most European countries (Eurostat, 2011), and that serious or moderate loneliness is common among the very old compared with younger old people (50% and 20%-30%, respectively; Dykstra, 2009), knowledge of the factors that reduce loneliness is an increasingly important public health issue. The experience of loneliness and social capital is likely to be value and context related (Jylhä & Jokela, 1990; Putnam, 2000), implying that the meaning of social capital for loneliness may be different in different geographical regions across Europe. In this paper the focus is on two Nordic regions. Although the influence of social capital on health and well-being has been assessed in previous research, this relationship has not yet been examined in-depth among the very old. Using the theory of social capital, the interaction between the older person and his or her social environment on a micro level will be examined, while also considering the residential and geographical context. The aim of this study is to investigate the links between social capital and loneliness among the very old living either at home or in institutional settings.

Method

Sample

The data used in this study were collected in the period 2005-2007 as part of the Gerontological Regional Database and Resource Centre project (GERDA project), which is a continuation of the Umeå 85+ study. The overall aim of the multidisciplinary project was to map living and health conditions of older people in the Bothnia region, that is, on both sides of the Gulf of Bothnia, in Västerbotten in Sweden and in Pohjanmaa in Finland. As part of the project a random sample comprising half of the 85-year-olds, and the total population of 90-year-olds and ≥ 95-year-olds was selected for participation from the urban municipality of Umeå and five rural municipalities in the county of Västerbotten in Sweden and in the municipalities of Vaasa and Mustasaari in Finland. Participants were selected from the population records, acquired from the National Tax Board in Sweden and the Population Register Centre in Finland. Of the total sample of 962, Finnish participants numbered 350 (36.4%) and Sweden participants 612 (63.6%). Seventy-two died before request, and so 890 were asked to participate. Participation was declined by 181 persons, with similar proportions for men and women. Nonrespondents were on average younger than respondents. Answers to questions on loneliness were not completed by 226 persons, who mainly lived in institutional care and were not able to answer mainly due to severe cognitive impairment. The final sample consisted of 483 participants of whom 334 (69.1%) lived in their own homes (with or without home care services) and 149 (30.9%) in institutional settings with access to staff and nurses at all hours. A flowchart of the intended participants and the final study sample is presented in Figure 1.

Figure 1.

Flowchart of the study population.

Procedure

One data collector contacted every selected individual by phone or in person to confirm consent with participation and to schedule a home visit. The data were collected by physicians, medical students, physiotherapists, and nurses. Full participation comprised a structured interview and assessments with the participant and, where applicable, an interview with a relative and/or with care staff. Interviews and assessments for each participant were performed by a single data collector. Medical records from hospitals, general practitioners, and from care institutions were also used.

Variables

Feeling of loneliness

Loneliness was assessed by the question “Do you ever feel lonely?” with four response alternatives: often, sometimes, seldom, never. A similar single question has been used in previous research of older adults (see Victor et al., 2005). In line with other studies, we dichotomized the variable (lonely, not lonely) by combining often and sometimes into one category and seldom and never into the other (e.g., Holmén et al., 1992; Routasalo, Savikko, Tilvis, Strandberg, & Pitkala, 2006).

Social Capital Resources

Structural

Having children (yes, no) and the respondents’ living situation (living together with someone [spouse, children/grandchildren, other], alone) were included as measures of structural social capital. The frequency of social contacts was measured by two questions: the frequency of being visited during a normal week and the frequency of visiting during the previous week. For the analyses of the frequency of being visited, two times or more were combined into one category, and 0 to 1 time into another, whereas the frequency of visiting was categorized as 0 time and 1 time or more. These contact frequency variables were dichotomized based on a median split of the distribution of the data. In addition, the frequency of phone contacts with relatives and friends was assessed (every day, every week, every month, and rarely/never). This item was dichotomized with the first two response alternatives as “every week” and the latter two as “monthly or less often.”

Cognitive

In this study, having a good friend to talk to when needed (yes, no) and a perception of frequency of social contacts were used as indicators of cognitive social capital, which capture the more subjective or perceptual aspects of the concept. The frequency and perception of social contacts were measured by two items: “How often do you feel that someone visits you?” and “How often do you feel that you have visited someone?” For each item the respondents had four response alternatives: often, sometimes, seldom, never. For the analyses, often and sometimes were combined into one category and seldom and never into another.

Health resources

Health factors (comprising depression, functional limitations, and cognitive impairments) are considered important constraints for being socially engaged and were assessed in this study as control variables in examining the relationship between social capital and loneliness.

The Mini Mental State Examination (MMSE; Folstein, Folstein, & McHugh, 1975) was used to assess the cognitive capacity of the participants. The scale has a maximum score of 30 with a score of 23 or less considered to indicate impaired cognition (Grut, Fratiglioni, Viitanen, & Winblad, 1993). The total MMSE score was divided into two categories (cutoff 23/24). Dependence in activities of daily living (ADL) was assessed by using two items from the ADL Staircase (KATZ Index of ADL; Katz, Ford, Moskowitz, Jackson, & Jaffe, 1963); dependence in cleaning (yes, no) and bathing (yes, no), which measured Instrumental ADL (I-ADL) and Personal ADL (P-ADL), respectively. Depressive symptoms were assessed by the 15-item Geriatric Depression Scale (GDS-15; Sheikh & Yesavage, 1986). Scores on the GDS ranged from 0 (no depressive symptoms) to 15 (severe depressive symptoms). A GDS score of 5 or higher indicated depressive symptoms (Lesher & Berryhill, 1994).

Sociodemographic variables

The sociodemographic characteristics in this study included type of residence (own home [own house, apartment], institutional setting [nursing home, group dwellings for people with dementia]), gender, age (85-year-olds, 90-year-olds, ≥95-year-olds), region (Västerbotten, northern Sweden and Pohjanmaa, western Finland), and educational level (low [0-6 years] and high [7+ years]).

Analyses

The distribution (%) of all variables included in the study was calculated by the type of residence (own home and institutional settings; Table 1). Pearson’s chi-square tests were conducted to analyze variations in loneliness according to social capital and separately for those living in their own homes and in institutional settings (Table 2). Logistic regressions were used to analyze the factors associated with experienced loneliness (Tables 3 and 4) and the results were presented as odds ratios (ORs) and their 95% confidence intervals (CIs). While a conventional level of significance such as 0.05 can fail in identifying variables known to be important (Mickey & Greenland, 1989), we included social capital variables in Table 2 with a p value < .150 in the multivariate logistic regression analysis (Tables 3 and 4). In addition, the sociodemographic variables as well as health variables associated with loneliness (p < .150) were included in the regression analyses. Participants with missing values were excluded (n = 44 and n = 37, respectively, for those living at home and in institutional settings) in the multivariate analyses. Finally, an analysis was also run using the total sample of the very old (N = 493) for comparisons. The first model included only the type-of-residence variable to assess the impact of institutionalization on loneliness. In the second model, the confounding effects of social capital (p < .150) and sociodemographic variables (p < .150) were examined. The statistical program IBM SPSS statistics, version 19.0, was used for the analyses.

Table 1.

Sociodemographic Characteristics, Social Capital, Health Characteristics, and Feelings of Loneliness Among the Very Old Living at Home and in Institutional Settings.

	Total (n = 483)		At home (n = 334)		Institution (n = 149)		Significance level
	n	%	n	%	n	%	Significance level
Gender
Men	149	30.8	119	35.6	30	20.1	p = .001
Women	334	69.2	215	64.4	119	79.9	p = .001
Age group
85	185	38.3	151	45.2	34	22.8	p < .001
90	177	36.6	130	38.9	47	31.5
95+	121	25.1	53	15.9	68	45.6
Region
Västerbotten, northern Sweden	329	67.1	224	67.1	105	70.5	p = .458
Pohjanmaa, western Finland	154	31.9	110	32.9	44	29.5	p = .458
Educational level
Low (0-6)	322	71.1	217	68.7	105	76.6	p = .086
High (7+)	131	28.9	99	31.3	32	23.4	p = .086
Living situation
With someone	101	21	101	30.3	149	100	p < .001
Alone	381	79	232	69.7	0	0	p < .001
Having children
Yes	413	86.2	285	85.8	128	87.1	p = .718
No	66	13.8	47	14.2	19	12.9	p = .718
Have visited at least once during the last week
Yes	182	43	158	43	24	20.2	p < .001
No	241	57	146	57	95	79.8	p < .001
Frequency of visits during a week
2+ times	256	60.4	189	62.2	67	55.8	p = .229
0-1 time	168	39.9	115	37.80	53	44.2	p = .229
Telephone contact
Every week	374	83.7	278	87.4	96	74.4	p = .001
Monthly or less often	73	16.3	40	12.6	33	25.6	p = .001
Perception of having visited
Often/sometimes	198	43.2	164	50.6	34	25.4	p < .001
Seldom/never	260	56.8	160	49.4	100	76.4	p < .001
Perception of being visited
Often/sometimes	372	80.2	259	79.2	113	82.5	p = .419
Seldom/never	92	19.8	68	20.8	24	17.5	p = .419
Having a good friend to talk to
Yes	329	72.1	240	74.1	89	67.4	p = .151
No	127	27.9	84	25.9	43	32.6	p = .151
Depression (GDS-15)
No (0-4)	333	71.8	247	76.5	86	61	p = .001
Yes (5-15)	131	28.2	76	23.5	55	39	p = .001
Cognitive function (MMSE)
Not impaired (24-30)	232	48.3	195	58.7	37	25	p < .001
Impaired (0-23)	248	51.7	137	41.3	111	75	p < .001
Functional ability
Independent in cleaning (KATZ)
Yes	185	38.4	174	52.1	11	7.4	p < .001
No	297	61.6	160	47.9	137	92.6	p < .001
Independent in bathing (KATZ)
Yes	308	63.9	273	81.7	35	23.6	p < .001
No	174	36.1	61	18.3	113	76.4	p < .001
Feeling of loneliness
Often	55	11.4	27	8.1	28	18.8	p = .005
Sometimes	178	36.8	124	37.1	54	36.2
Seldom	109	22.7	78	23.4	31	20.8
Never	141	29.1	105	31.4	36	24.2

Note. GDS = Geriatric Depression Scale; MMSE = Mini Mental State Examination.

Table 2.

The Distribution (%) of Feelings of Loneliness by Social Capital Among the Very Old Living at Home and in Institutional Settings.

	At home (n = 334)		χ²	Institution (n = 149)		χ²
	Lonely (n = 150)	Not lonely (n = 184)	χ²	Lonely (n = 83)	Not lonely (n = 66)	χ²
Living situation
With someone	16.8	83.2	45.7, df = 1, p < .001	NA	NA	NA
Alone	56.9	43.1	45.7, df = 1, p < .001	55.7	44.3	NA
Having children
Yes	44.6	55.4	0.08, df = 1, p = .77	57.0	43.0	0.62, df = 1, p = .43
No	46.8	53.2	0.08, df = 1, p = .77	47.4	52.6	0.62, df = 1, p = .43
Have visited at least once during the last week
Yes	46.2	53.8	0.003, df = 1, p = .96	41.7	58.3	4.46, df = 1, p = .035
No	45.9	54.1	0.003, df = 1, p = .96	65.3	34.7	4.46, df = 1, p = .035
Frequency of visits during a week
2+ times	44.4	55.6	1.71, df = 1, p = .191	69.7	40.3	0.12, df = 1, p = .733
0-1 time	52.2	47.8	1.71, df = 1, p = .191	56.6	43.4	0.12, df = 1, p = .733
Telephone contact
Every week	47.5	52.5	4.31, df = 1, p = .038	55.2	44.8	0.056, df = 1, p = .813
Monthly or less often	30.0	70.0	4.31, df = 1, p = .038	57.6	42.4	0.056, df = 1, p = .813
Perception of having visited
Often/sometimes	41.7	58.3	5.33, df = 1, p = .021	57.5	42.5	0.091, df = 1, p = .763
Seldom/never	57.4	42.6	5.33, df = 1, p = .021	54.2	45.8	0.091, df = 1, p = .763
Perception of being visited
Often/sometimes	37.2	62.8	7.67, df = 1, p = .006	44.1	55.9	3.32, df = 1, p = .068
Seldom/never	52.5	47.5	7.67, df = 1, p = .006	62.0	38.0	3.32, df = 1, p = .068
Having a good friend to talk to
Yes	46.4	53.6	0.229, df = 1, p = .632	52.6	43.8	0.218, df = 1, p = .691
No	43.4	56.6	0.229, df = 1, p = .632	60.5	39.5	0.218, df = 1, p = .691

Table 3.

Multivariate Logistic Regression Model on the Probability of Feeling Lonely Among the Very Old Living in Their Own Home (n = 290).

	OR	95% CI
Gender
Men	1.00
Women	1.04	[0.57, 1.90]
Age group
85	1.00
90	1.09	[0.59, 2.0]
95+	1.12	[0.47, 2.66]
Region
Pohjanmaa, western Finland	1.00
Västerbotten, northern Sweden	2.98**	[1.58, 5.64]
Living situation
With someone	1.00
Alone	7.80***	[3.77, 16.15]
Perception of having visited
Often/sometimes	1.00
Seldom/never	1.57	[0.86, 2.87]
Perception of being visited
Often/sometimes	1.00
Seldom/never	1.78	[0.85, 3.73]
Telephone contact
every week	1.00
Monthly or less often	0.50	[0.18, 1.35]
Depressive symptoms (GDS-15)
No (0-4)	1.00
Yes (5-15)	4.90***	[2.40, 10.02]
Cognitive function (MMSE)
Not impaired (24-30)	1.00
Impaired (0-23)	1.12	[0.61, 2.05]
Functional ability
Dependent in cleaning (KATZ)
No	1.00
Yes	1.49	[0.81, 2.72]

Note. Model χ² = 90.513, p < .000, Nagelkerke R² = .358; percentage of correct prediction 71.4%. OR = odds ratio; CI = confidence interval; GDS = Geriatric Depression Scale; MMSE = Mini Mental State Examination.

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

Table 4.

Multivariate Logistic Regression Model on the Probability of Feeling Lonely Among the Very Old Living in Institutional Settings (n = 112).

	OR	95% CI
Frequency of visiting during the last week
1+ time	1.00
0 time	1.09	[0.31, 3.78]
Perception of being visited
Often/sometimes	1.00
Seldom/never	0.72	[0.23, 2.21]
Depressive symptoms (GDS-15)
No (0-4)	1.00
Yes (5-15)	9.43***	[3.44, 25.85]
Cognitive function (MMSE)
Not impaired (24-30)	1.00
Impaired (0-23)	2.30	[0.74, 7.22]
Functional ability
Dependent in cleaning (KATZ)
No	1.00
Yes	4.61	[0.60, 35.56]
Dependent in bathing (KATZ)
No	1.00
Yes	0.65	[0.16, 2.69]

Note. Model χ² = 33.973, p < .000, Nagelkerke R² = .352; percentage of correct prediction 76.8%. OR = odds ratio; CI = confidence interval; GDS = Geriatric Depression Scale; MMSE = Mini Mental State Examination.

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

Ethics

The GERDA data collection was approved by the Ethics Committee of the Faculty of Medicine at Umeå University (Registration Number 05-063M) and the Ethics Committee of Vaasa Central Hospital (Registration Number 05-87).

Results

The distribution (%) of all the included variables was reported by type of residence (Table 1). A higher proportion of those living in institutional settings were women (79.9% vs. 64.4%) and aged 95+ (45.6% vs. 15.9%). No statistical difference was found between the very old living at home and in institutional settings with respect to region and educational level. All of the institutionalized very old were rated as not living with someone, such as spouse and partner, compared with 69.7% of those living at home. A higher proportion of the very old in institutions did not visit anyone during the previous week (79.8% vs. 57%) nor did they have a perception of having visited anyone (76.4% vs. 49.4%). Furthermore, fewer older adults living in institutions had weekly telephone contacts with friends and relatives (74.4% vs. 87.4%). No statistical differences between the two groups were found with regard to having children, frequency of visits and quality of social contacts, such as perception of being visited and having a good friend to talk to. More people living in institutional settings reported depressive symptoms (39% vs. 23.5%), cognitive impairment (75% vs. 41.3%) and functional instrumental and personal limitations (92.6% vs. 47.9%, and 76.4% vs. 18.3%, respectively). Finally, people in institutions more often reported feelings of loneliness (often and sometimes; 55% vs. 45.2%). In the next analyses, feelings of loneliness are dichotomized into lonely (often, sometimes) and not lonely (seldom, never).

The results in Table 2 show that the distribution of loneliness varied among the very old with respect to type of residence and different aspects of social capital. Among the very old living at home, significant differences in loneliness could be found between living situation (p < .001), perceptions of having visited (p = .021) and having being visited (p = .006), and telephone contact (p = .038). Among those reporting loneliness, 16.8% lived together with someone compared with 56.9% of those living alone, and loneliness was more common among those with few perceived social visits. However, those who experienced loneliness were more likely to have frequent telephone contacts with relatives and friends. A significant difference in loneliness was found among the very old living in institutional settings in terms of the frequency of visits during the previous week (p = .035). Of those who did not visit anyone during the previous week 65.3% reported feelings of loneliness, compared with 41.7% of those who did visit someone.

Significant differences in loneliness were found between region and perceived loneliness among those living in their own home (data not shown). Feelings of loneliness among those living at home were more common in Sweden compared with Finland (51.3% and 31.8%, respectively, p = .001). A strong association between depression and loneliness was found among the very old living at home and in institutional settings. A higher proportion with depressive symptoms reported experiencing loneliness (75.0% and 83.6%, in noninstitutions and institutions, respectively, p < .001) than those with no depressive symptoms. Among the very old living in institutional settings, the experience of loneliness was also more common among those with impaired cognitive cognition (60.4% vs. 40.5%, p = .036) as well as among those with functional dependence in cleaning, I-ADL (58.4% vs. 18.2%, p = .01) and bathing, P-ADL (60.2% vs. 40.0%, p = .036).

Finally, we created a logistic regression model where the factors associated with loneliness (p < .150) in the bivariate analyses were included separately for those living in their own homes and in institutions. As shown in Table 3 the very old in northern Sweden, and those with depressive symptoms were more likely to report feelings of loneliness. The most powerful independent variable associated with experienced loneliness was living alone (OR = 7.80, CI = [3.77, 16.15]). Perceptions of contacts and telephone contacts were no longer significantly associated with feelings of loneliness in the logistic regression model. In Table 4, depressive symptoms (OR = 9.43, CI = [3.44, 25.85]) was the most powerful independent variable associated with loneliness among those living in institutional settings. None of the other variables were significant. Finally, when running the analysis with the total sample (data not shown) and controlling only for type-of-residence, the results showed that loneliness was more common among those in institutional settings than among those living at home (OR = 1.53, CI = [1.05, 2.28]). When including the social capital and sociodemographic variables in the model the OR in loneliness between the two groups decreased close to one (OR = 1.04, CI = [0.55, 1.97]).

Discussion

The purpose of this study was to assess the relationship between indicators of social capital and feelings of loneliness among the very old. We examined this relationship in a residential and a geographical context, that is, in regular housing and in institutions as well as in two geographical settings (northern Sweden and western Finland). An important finding was that a large proportion of the very old experienced loneliness, which is consistent with the findings of other studies. Loneliness was reported by as many as 55% of the very old in institutions often or sometimes and 45% of those living at home. Nineteen percent and 8% reported that they often felt lonely in institutional and home settings, respectively. In our multivariate analyses we found that living alone was closely related to loneliness among those living in their own home. In addition, the very old in Sweden experienced loneliness more frequently than their peers in Finland. Depression was related to loneliness in both groups. None of the selected social capital indicators were independently associated with loneliness among those living in institutional settings.

Social capital in relation to health and well-being has rarely been studied among the very old (Nyqvist, Gustavsson, & Gustafson, 2006). This is noteworthy given that social capital is of particular relevance also to older age groups (Forsman et al., 2013; Forsman, Nyqvist, & Wahlbeck, 2011; Gray, 2009; Nyqvist et al., 2012; Poulsen et al., 2012). Drawing on Putnam’s work, social capital benefits an individual because of his or her membership in a certain community. To be socially included in one’s community is important for the quality of life and well-being in later life and may be of even more importance as people age and their mobility declines. In our study, the community environment was conceptualized as where people live; in their own homes or in institutional settings. Our results clearly showed that social capital resources were significantly different between the two groups (Table 1). Older people at home had more access to social contacts, whereas the quality aspects of social contacts did not differ significantly between the two groups. Interestingly, of the social capital indicators included in our study, only living alone was related to loneliness among those living in their own home, whereas none of the social capital indicators were related to loneliness among those living in institutional settings.

Pinquart and Sörensen’s (2001) systematic review found that older people in institutions reported more loneliness than people living in their own homes, which is in line with the results of our study. They argued that factors such as reduced social support and poor health could be an explanation to higher experienced loneliness within institutions rather than institutionalization per se (Pinquart & Sörensen, 2001). It seems reasonable to assume that poor health and frailty limit the possibility to interact with peer residents and to be engaged in social activities and outings. As expected, our results showed that the very old living in institutions were more frail in terms of depressive symptoms and functional and cognitive impairments. Although research has shown that institutional care settings lack high levels of social interaction and activities (Mattiasson & Andersson, 1997; Ruuskanen & Parkatti, 1994), Hubbard, Tester, and Downs (2003) suggests that residents do interact and communicate. However, the type of interaction may take different forms when one is frail and in need of help and it is likely that our study did not succeed in capturing the essential interactions.

Besides the importance of peer contacts within the care facility, it has also been shown that social contacts with staff is just as important as a social resource as contacts with family and friends outside the care facility (Carpenter, 2002). The contacts within the institutions were, however, not surveyed in the present study. A recent study suggests that long-lasting friendship is very important for well-being among older people living in nursing homes and the loss of close friends is irreplaceable (Forsman et al., 2013). It may also be difficult to develop new friendships if many residents suffer from cognitive limitations or if the older persons living in the institution have little in common. Nevertheless, in our study, 67% of the very old living in institutional settings indicated that they had a good friend to talk to, although this was not linked to reduced loneliness.

In later life, older people can expect a decline in social capital that is likely to increase feelings of loneliness. The very old in particular are at high risk of losing their partners and friends through death and to cognitive and functional limitations, which reduces the possibility to maintain social capital. Our study shows that age as such or cognitive and functional limitations did not increase the risk for experienced loneliness, although depressive symptoms did, which is consistent with previous research (Drageset et al., 2011; Prieto-Flores et al., 2011; Savikko et al., 2005). Our findings suggest that depression increases the risk for experienced loneliness, although a reverse relationship is also likely; loneliness may cause depression, which has been demonstrated in other research (Heikkinen & Kauppinen, 2004). Depression may act as a barrier for social engagement (Fessman & Lester, 2000) and has a causal link to various social, physical, and psychological problems. In a population-based study, high levels of social capital have been related to reduced depression among older people (Forsman et al., 2013), suggesting that social capital may modify the relationship between depressive symptoms and loneliness. Strategies to reduce depression and to increase social capital may thus have implications for experienced loneliness among the very old living in their own homes and in institutional settings.

Older people who lived alone reported experiences of loneliness more often than those who lived together with someone, which has also been found in other research (Victor et al., 2000; Wenger & Davies, 1996). Of those not living alone, the majority lived with their spouse and a few lived with their adult children. Those living with an adult child were more likely to experience loneliness than those living with a spouse (data not shown). This suggests that marital status represents an important context within which to understand the connection between social capital and loneliness among those still living in their own homes. Marriage may change the types of networks to which individuals belong and in which they participate. It may be that single older people have to be more engaged and proactive in maintaining social resources compared with married and cohabiting couples. Having children did not explain the experience of loneliness. While this contradicts findings from some previous research (Felton & Berry, 1992; Routasalo et al., 2006), it corroborates other research suggesting that older people rely more on friends than adult children when they feel lonely (Bondevik & Skogstad, 1998; Holmén et al., 1992).

This study allowed us to examine loneliness in a Swedish and Finnish context. Cross-national studies on loneliness in Europe reveal that there are wide variations between countries (Fokkema, de Jong Gierveld, & Dykstra, 2012; Jylhä & Jokela, 1990; Sundström, Fransson, Malmberg, & Davey, 2009; Walker, 1993; Yang & Victor, 2011). In southern European countries, a higher prevalence of loneliness has been observed than in northern Europe. There is no solid basis for explaining the observed cross-country differences. However, it has been suggested that value systems and the ideology of individualism of the Nordic countries (Jylhä & Jokela, 1990), which all belong to the same Social Democratic welfare regime type (Esping-Andersen, 1990), may lower the levels of experienced loneliness. While living alone is more typical in northern Europe, feelings of loneliness are also lower compared with the more collectivistic societies of southern Europe. A comparison in loneliness between urban areas in Sweden, Finland and Denmark showed no significant differences (Heikkinen, Berg, & Avlund, 1995), supporting the assumption that the Social Democratic regime type may have an impact on levels of loneliness. Interestingly, our results showed differences in loneliness in northern Sweden and western Finland, so that noninstitutionalized older people in northern Sweden felt lonelier than their peers in Finland.

One reason for these differences may be that our study region is relatively rural, except for two cities in northern Sweden and western Finland, respectively. In addition, the northern region of Sweden is a less populated area than the western parts of Finland. Given that the experience of loneliness tends to be more common in rural areas than in cities (Savikko et al., 2005), this could explain our results. However, when comparing urban and rural areas in Sweden, loneliness was more common in urban areas than in rural (data not shown). It is likely that a high migration from rural areas to the city of Umeå in Sweden, a fast growing city with a young population structure, increases social isolation, and the level of loneliness in later life when old friends and family are not close geographically. This issue should be addressed in future research.

Our results can be said to have a number of important implications for policy, service provision, and practice in the field of elderly care, as well as for future research. First, loneliness and depressive symptoms are common among the very old and targeted preventive strategies to reduce loneliness could be expected to reduce depression as well. Second, living alone is identified as a high risk factor for experienced loneliness and intervention strategies aimed at reducing social isolation and loneliness for older single people living in their own homes should be implemented. Third, the very old living in institutional settings are not alone, that is, they do not lack social contacts, but they feel lonely, suggesting that more focus need to be put on social inclusion and social relationships within the care facilitates. Finally, experienced loneliness differs between regions of Finland and Sweden and that is an issue that warrants further investigation.

Strengths and Limitations

One of the strengths of this study is that the very old living in their own home and in institutions were included in the home interview survey, enabling comparisons between different living contexts. We underline the special social prerequisites of the two living contexts when comparing the two groups. As shown in the flowchart (Figure 1), the drop-out rate from our final study sample was relatively high as could be expected when studying the very old. However, the distribution by type of residence and age of those who agreed to participate (n = 709) corresponded to the distribution in the final study sample (n = 483) indicating that the results can be reliable for the very old living in their own homes as well as within institutional settings.

The cross-sectional nature of the data used does not enable evidence-based conclusions on how social capital and loneliness are causally related. A longitudinal data set would allow analyses of the relationships over time. Furthermore, we used a single-item self-report loneliness rating question instead of scales as for example the Jong Gierveld Scale (de Jong Gierveld & Kamphuis, 1985) or the UCLA Loneliness Scale (Russell, 1996), though both approaches have their special advantages and disadvantages, as discussed by Victor et al. (2005). It should further be noted that the difference in the prevalence of experienced loneliness between the very old living within institutional settings and in their own homes may be an issue of selection bias. Older people experiencing loneliness tend to be at higher risk of moving to an institution (de Jong Gierveld, 1998).

The social capital indicators used here focused on social isolation, although the definition by Putnam (2000) also includes aspects of trust and norms. However, these features were not surveyed in the present study. Furthermore, the indicators that were used are not fully validated, even though similar items have been used elsewhere (Islam et al., 2006; Kim et al., 2008; Nyqvist et al., 2012; Stone, 2001). The issues related to the validation of social capital measurements are however a reoccurring problem in this research field (van Deth, 2008).

Although dichotomization of variables implies a loss of information, we decided to reduce the number of response categories to gain statistical power. We analyzed the frequency of social contacts also as continuous variables; however, this did not change the results significantly. To include items on contacts within an institutional setting, such as peers and staff members, should be one direction for future research (Carpenter, 2002), while another study should look at including items on losses of social and personal resources, which both have been proven significant in other studies (Forsman et al., 2013; Victor et al., 2005).

Conclusion

Studying social capital among the very old requires examination in context, recognizing that the meaning of social capital for experienced loneliness may vary between living communities. Our study shows that loneliness in the very old entails specific challenges, which need to be carefully addressed, related to individual characteristics and to community-level features, such as attitudes and values shared in various geographical contexts. Strategies to alleviate loneliness should be specifically designed to enhance social interactions among those living alone as well as among those in institutional settings.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by grants from King Gustav V and Queen Victoria’s Freemasons Foundation; Swedish Research Foundation for Aging-Related diseases; The Research Foundation for the Faculty of Medicine at Umeå University; The Borgerskapet of Umeå Research Foundation; The Detlof Research Foundation; Swedish Dementia Association; The Alzheimer Foundation, Interreg IIIA Mitt-Skandia; the Strategic Research Programme in Care Sciences, Sweden, The Botnia-Atlantica Regional Development Fund; the Swedish Research Council (2005); and The Västerbotten County Council. The work by Fredrica Nyqvist was financially supported by the Academy of Finland (Project 250054) as part of the FLARE-2 programme.

References

Bondevik

Skogstad

(1996). Loneliness among the oldest old: A comparison between residents living in nursing homes and residents living in the community. International Journal of Aging & Human Development, 43, 181-197.

Bondevik

Skogstad

(1998). The oldest old, ADL, social network, and loneliness. Western Journal of Nursing Research, 20, 325-343.

Carpenter

B. D.

(2002). Family, peer, and staff social support in nursing home patients: Contributions to psychological well-being. Journal of Applied Gerontology, 21, 275-293.

de Jong Gierveld

(1998). A review of loneliness: Concept and definitions, determinants and consequences. Reviews in Clinical Gerontology, 8, 73-80.

de Jong Gierveld

Kamphuis

F. H.

(1985). The development of a Rasch-type loneliness scale. Applied Psychological Measurement, 9, 289-299.

de Jong Gierveld

Van Tilburg

(2010). The De Jong Gierveld short scales for emotional and social loneliness: Tested on data from 7 countries in the UN generations and gender surveys. European Journal of Ageing, 7, 121-130.

Drageset

Kirkevold

Espehaug

(2011). Loneliness and social support among nursing home residents without cognitive impairment: A questionnaire survey. International Journal of Nursing Studies, 48, 611-619.

Dykstra

(2009). Older adult loneliness: Myths and realities. European Journal of Ageing, 6, 91-100.

Dykstra

P. A.

de Jong Gierveld

(1999). Differential indicators of loneliness among elderly: The importance of type of partner relationship, partner history, health, socioeconomic status and social relations. Tijdschrift Voor Gerontologie En Geriatrie, 30, 212-225.

10.

Esping-Andersen

(1990). The three worlds of welfare capitalism. Cambridge, UK: Polity.

11.

Eurostat. (2011). Europe in figures. Eurostat Yearbook 2011. Luxembourg: Publications Office of the European Union. Retrieved from http://epp.eurostat.ec.europa.eu/cache/ITY_OFFPUB/KS-CD-11-001/EN/KS-CD-11-001-EN.PDF

12.

Felton

B. J.

Berry

C. A.

(1992). Do the sources of the urban elderly’s social support determine its psychological consequences? Psychology and Aging, 7, 89-97.

13.

Fessman

Lester

(2000). Loneliness and depression among elderly nursing home patients. International Journal of Aging & Human Development, 51, 137-141.

14.

Fokkema

de Jong Gierveld

Dykstra

P. A.

(2012). Cross-national differences in older adult loneliness. The Journal of Psychology, 146, 201-228.

15.

Folstein

M. F.

Folstein

S. E.

McHugh

P. R.

(1975). Mini-mental state: A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12, 189-198.

16.

Forsman

A. K.

Herberts

Nyqvist

Wahlbeck

Shierenbeck

(2013). Understanding the role of social capital for mental well-being among older adults. Ageing & Society. 33, 804-825.

17.

Forsman

A. K.

Nyqvist

Wahlbeck

(2011). Cognitive components of social capital and mental health status among older adults: A population-based cross-sectional study. Scandinavian Journal of Public Health, 39, 757-765.

18.

Gray

(2009). The social capital of older people. Ageing & Society, 29, 5-31.

19.

Grut

Fratiglioni

Viitanen

Winblad

(1993). Accuracy of the mini-mental status examination as a screening test for dementia in a Swedish elderly population. Acta Neurologica Scandinavica, 87, 312-317.

20.

Hawkley

L. C.

Cacioppo

J. Y.

(2010). Loneliness matters: A theoretical and empirical review of consequences and mechanisms. Annals of Behavioral Medicine, 40, 218-227.

21.

Heikkinen

R.-L.

Berg

Avlund

(1995). Depressive symptoms in late life. Journal of Cross-Cultural Gerontology, 10, 315-330.

22.

Heikkinen

R.-L.

Kauppinen

(2004). Depressive symptoms in late life: A 10-year follow-up. Archives of Gerontology and Geriatrics, 38, 239-250.

23.

Holmén

K. F.

Ericsson

K. F.

Andersson

L. F.

Winblad

(1992). Loneliness among elderly people living in Stockholm: A population study. Journal of Applied Nursing, 17, 43-51.

24.

Hooghe

Stolle

(2003). Generating social capital: Civil society and institutions in comparative perspective. New York, NY: Palgrave Macmillan.

25.

Hubbard

Tester

Downs

M. G.

(2003). Meaningful social interactions between older people in institutional care settings. Ageing & Society, 23, 99-114.

26.

Islam

M. K.

Merlo

Kawachi

Lindström

Gerdtham

(2006). Social capital and health: Does egalitarianism matter? A literature review. International Journal for Equity in Health, 5(1), 3. doi:10.1186/1475-9276-5-3

27.

Jylhä

Jokela

(1990). Individual experiences as cultural: A cross-cultural study on loneliness among the elderly. Ageing & Society, 10, 295-315.

28.

Jylhä

Saarenheimo

(2010). Loneliness and ageing: Comparative perspectives. In Dannefer

Phillipson

(Eds.), The SAGE handbook of social gerontology (pp. 317-328). London, England: SAGE.

29.

Katz

Ford

A. B.

Moskowitz

R. W.

Jackson

B. A.

Jaffe

M. W.

(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, 185, 914-919.

30.

Kim

Subramanian

S. V.

Kawachi

(2008). Social capital and physical health. A systematic review of the literature. In Kawachi

Subramanian

S. V.

Kim

(Eds.), Social capital and health (pp. 139-190). New York, NY: Springer.

31.

Lesher

E. L.

Berryhill

J. S.

(1994). Validation of the Geriatric Depression Scale–Short Form among inpatients. Journal of Clinical Psychology, 50, 256-260.

32.

Lindström

(2004). Social capital, the miniaturisation of community and self-reported global and psychological health. Social Science & Medicine, 59, 595-607.

33.

Litwin

Shiovitz-Ezra

(2011). Social network type and subjective well-being in a national sample of older Americans. The Gerontologist, 51, 379-388.

34.

Mattiasson

Andersson

(1997). Quality of nursing home care assessed by competent nursing home patients. Journal of Advanced Nursing, 26, 1117-1124.

35.

Mickey

R. M.

Greenland

(1989). The impact of confounder selection criteria on effect estimation. American Journal of Epidemiology, 129, 125-137.

36.

Nyqvist

Forsman

A. K.

Giuntoli

Cattan

(2013). Social capital as a resource for mental well-being in older people: A systematic review. Aging & Mental Health, 17, 394-410. doi:10.1080/13607863.2012.742490

37.

Nyqvist

Gustavsson

Gustafson

(2006). Social capital and health in the oldest old: The Umeå 85+ study. International Journal of Aging and Later Life, 1, 91-114.

38.

Penninx

B. W. J. H.

van Tilburg

Kriegsman

D. M. W.

Boeke

A. J. P.

Deeg

D. J. H.

van Eijk

J. T. M.

(1999). Social network, social support, and loneliness in older persons with different chronic diseases. Journal of Aging and Health, 11, 151-168.

39.

Perlman

Peplau

(1982). Theoretical approaches to loneliness. In Peplau

Perlman

(Eds.), Loneliness: A sourcebook of current theory, research and therapy (pp. 123-134). New York, NY: John Wiley.

40.

Pinquart

(2003). Loneliness in married, widowed, divorced, and never-married older adults. Journal of Social and Personal Relationships, 20, 31-53.

41.

Pinquart

Sörensen

(2001). Influences on loneliness in older adults: A meta-analysis. Basic and Applied Social Psychology, 23, 245-266.

42.

Poulsen

Siersma

V. D.

Lund

Christensen

Vass

Avlund

(2012). Impact of social capital on 8-year mortality among older people in 34 Danish municipalities. Journal of Aging and Health, 24, 1203-1222.

43.

Prieto-Flores

Forjaz

M. J.

Fernandez-Mayoralas

Rojo-Perez

Martinez-Martin

(2011). Factors associated with loneliness of noninstitutionalized and institutionalized older adults. Journal of Aging and Health, 23, 177-194.

44.

Putnam

R. D.

(1993). Making democracy work: Civic traditions in modern Italy. Princeton, NJ: Princeton University Press.

45.

Putnam

R. D.

(2000). Bowling alone: The collapse and revival of American community. New York, NY: Simon & Schuster.

46.

Routasalo

P. E.

Savikko

Tilvis

R. S.

Strandberg

T. E.

Pitkala

K. H.

(2006). Social contacts and their relationship to loneliness among aged people—A population-based study. Gerontology, 52, 181-187.

47.

Russell

(1996). The UCLA Loneliness Scale (Version 3): Reliability, validity, and factor structure. Journal of Personality Assessment, 66, 20-40.

48.

Ruuskanen

J. M.

Parkatti

(1994). Physical activity and related factors among nursing home residents. Journal of the American Geriatrics Society, 42, 987-991.

49.

Savikko

Routasalo

Tilvis

R. S.

Strandberg

T. E.

Pitkälä

K. H.

(2005). Predictors and subjective causes of loneliness in an aged population. Archives of Gerontology and Geriatrics, 41, 223-233.

50.

Sheikh

J. I.

Yesavage

J. A.

(1986). Geriatric Depression Scale (GDS): Recent evidence and development of a shorter version. Clinical Gerontologist, 5, 165-173.

51.

Stephens

Alpass

Towers

Stevenson

(2011). The effects of types of social networks, perceived social support, and loneliness on the health of older people: Accounting for the social context. Journal of Aging and Health, 23, 887-911.

52.

Stone

(2001). Measuring social capital: Towards a theoretically informed measurement framework for researching social capital in family and community life (No. 24). Melbourne: Australian Institute of Family Studies.

53.

Sundström

Fransson

Malmberg

Davey

(2009). Loneliness among older Europeans. European Journal of Ageing, 6, 267-275.

54.

Tijhuis

de Jong Gierveld

Feskens

Kromhout

(1999). Changes in and factors related to loneliness in older men: The Zutphen elderly study. Age and Ageing, 28, 491-495.

55.

Tsai

(2011). Changes in depressive symptoms, social support, and loneliness over 1 year after a minimum 3-month videoconference program for older nursing home residents. Journal of Medical Internet Research, 13, e93. doi:10.2196/jmir.1678

56.

van Deth

J. W.

(2008). Measuring social capital. In Castiglione

van Deth

J. W.

Wolleb

(Eds.), The handbook of social capital (pp. 150-176). Oxford, UK: Oxford University Press.

57.

Victor

Grenade

Boldy

(2005). Measuring loneliness in later life: A comparison of differing measures. Reviews in Clinical Gerontology, 15, 63-70.

58.

Victor

Scambler

Bond

(2009). The social world of older people: Understanding loneliness and social isolation in later life. Maidenhead, UK: Open University Press.

59.

Victor

Scambler

Bond

Bowling

(2000). Being alone in later life: Loneliness, social isolation and living alone. Reviews in Clinical Gerontology, 10, 407-417.

60.

Walker

(1993). Age and attitudes: Main results from a Eurobarometer survey. Bruxelles, Belgium: Commission of the European Communities.

61.

Wenger

G. C.

Davies

(1996). Social isolation and loneliness in old age: Review and model refinement. Ageing & Society, 16, 333-358.

62.

Yang

Victor

(2011). Age and loneliness in 25 European nations. Ageing & Society, 31, 1368-1388.