Who’s at risk and what can be done about it? A review of the literature on the social isolation of different groups of seniors

Definitions

The concept of social isolation is multi-faceted and overlaps with other related concepts such as loneliness, social vulnerability, social inclusion and exclusion, social connectedness, and social capital. Social isolation is commonly defined as a low quantity and quality of contact with others, and includes "number of contacts, feeling of belonging, fulfilling relationships, engagement with others, and quality of network members" (Nicholson, 2009). Socially isolated persons lack social contacts, social roles and mutually rewarding relationships (Keefe et al., 2006; North Sky Consulting Group Ltd, 2013). The severity of isolation can vary from person to person depending upon whether it is voluntary or involuntary, permanent or chronic, episodic or temporary.^{Footnote 6} While researchers define social isolation differently, most agree that it has a more objective dimension to it than loneliness, which depends upon an individual's subjective reactions to their environment and relationships rather than the more general condition of having few social contacts. Still, in the literature on social isolation, the terms are often used interchangeably. While this review focuses on social isolation, findings on loneliness and other related concepts are sometimes referred to where they help provide a more complete picture of the situation of specific groups of seniors.

Research scope and methods

To guide the literature review, two key research questions were identified:

1. What is known about the different groups of seniors who may be vulnerable to social isolation? (Note that the groups themselves were already identified through NSC consultations).
2. What does the literature say about the risk^{Footnote 7}correlates of social isolation – and how these factors interact – among these groups of seniors?

A thorough review of the scholarly and program/policy literature was undertaken, with approximately 150 articles, reports, books and book-chapters identified and reviewed. An exhaustive list of search terms was developed and used to identify the literature. While a majority of articles and reports were in English and came from Canada or the United States, a significant sample of French-language articles was also included in the report (with a focus on the Canadian literature and particularly influential reports from Europe). One hundred and twenty pieces of literature were reviewed in English and 30 in French.

Measuring the overall prevalence and impact of social isolation brings with it a series of methodological challenges. First, a major stumbling block for both researchers and practitioners is how to find the isolated, who – by definition – are not connected and are difficult to find (Newall, 2015). Frontline workers also report that linguistic and cultural barriers – along with stigma surrounding isolation – can make it even more difficult to reach certain groups of seniors (North Sky Consulting Group Ltd, 2013; see also section below on Immigrant Seniors). The overall incidence of isolation may therefore be under-reported. A distinct but related problem is obtaining accurate data on the size of population groups affected by social isolation. In this review, for instance, data on certain population groups, such as LGBT seniors, seniors who are caregivers, seniors with mental health issues and seniors with health challenges – are primarily based upon self-reporting in surveys. Self-reporting can lead to under-estimations, particularly for LGBT seniors, who are reluctant to be open about their sexual orientation or gender identity (Tang, 2015).

Second, social isolation is a complex phenomenon associated with multiple factors. While some researchers deploy multivariate analyses to isolate causal factors, this is not always possible (or desirable, depending upon the purpose of the research).^{Footnote 8} Thus, some studies may identify factors associated with social isolation without being able to determine whether they are causal factors per se, or the extent to which they are generalizable, or how exactly they interact. Third, even when large-scale studies are conducted, the relationship between social isolation and other factors can be reciprocal or bidirectional in nature, further compounding the challenge of determining causality. For example, the literature on social isolation clearly demonstrates that social isolation can have an impact on various measures of health, but it is equally well established that health challenges can further exacerbate social isolation. Fourth, there may be cumulative or synergetic effects across risk factors that combine to influence social isolation and related experiences within and between vulnerable groups of seniors.

This review takes into consideration these various challenges and seeks to contextualize research findings as much as possible. It takes a general approach to understanding the dynamics of social isolation by looking at the interaction of multiple risk factors that can create a cascading of risk^{Footnote 9} that render an individual – or a group of seniors – particularly vulnerable. Taken as a whole, the accumulation of risk factors can place seniors at multiple jeopardy of experiencing loneliness and social isolation.

Aboriginal seniors

Based on the National Household Survey (NHS), the Aboriginal population of Canada was estimated at 1,400,685 people in 2011, representing 4.3% of the total Canadian population (Statistics Canada, 2013a).^{Footnote 15} Although only 6% of the Aboriginal population is aged 65 and over, (compared to 16% in the Canadian population), the proportion of Aboriginal seniors is increasing over time due to declines in mortality (more survivors to older ages) and lower fertility (smaller younger cohorts).^{Footnote 16} Aboriginal seniors are a highly diverse group with about half living in cities, and the other half in rural and remote areas, mostly on reserve (Wister and McPherson, 2014).^{Footnote 17} Recommendations to tackle social isolation need to recognize the unique characteristics of these groups.

Aboriginal seniors have faced numerous challenges over the course of their lives that may follow them into older age. These include higher levels of unemployment and poverty, substandard housing, higher rates of crime and abuse, separation from families during the residential school period, higher mortality and morbidity rates, and fewer health care, residential and community services (Wilson et al., 2011; Wister and McPherson, 2014). In addition, as a result of the colonial experience, they have experienced marginalization, especially those living on reserve in more rural and remote areas (Richmond, 2007). This makes them a highly vulnerable group and one prone to risks associated with social isolation. Yet, research into social isolation among Aboriginal seniors is scarce, with most literature focusing on access to health and community care or social inclusion among Aboriginal people in general (Health Canada, 2009).

Approaches to addressing aging-related needs among Aboriginal communities from the vantage point of social inclusion have identified complex barriers and challenges (Habjan et al., 2012; Haskell and Randall, 2009; Wilson et al., 2011). Haskell and Randall (2009), for instance, note how social connectedness at the interpersonal level has often been fragmented due to traumatic experiences (e.g., family violence, suicide, substance abuse, poverty, dislocation, etc.). Treloar et al. (2014) identify three situational barriers that can undermine the experience of social inclusion of Aboriginal people in cancer treatments: low socio-economic security, lack of trust in health care provision, and lack of knowledge about the system of cancer treatment (health literacy). To overcome these challenges, health-care providers must acknowledge them and create practical and symbolic responses in partnership with Aboriginal people, communities and health organisations.

Similarly, Habjan et al. (2012) identified multiple barriers regarding health and social support, including lack of a family caregiver, limited access to professional and health care services and providers, educational and training needs, barriers linked to community and government relationships, and relevant and safe care. Beatty and Berdahl (2011) have also identified multiple factors that lead to health challenges among Aboriginal seniors, including poverty, lack of housing, lack of nursing homes, and education and literacy barriers, as well as limited policymaker and caregiver knowledge of the needs of Aboriginal seniors. To address these complex barriers, multi-level changes are required.

Numerous studies have shown that Aboriginal communities can harness community-based resources especially through participatory and action-oriented approaches that empower seniors, their families, and their communities (Wister and McPherson, 2014). At the same time, a paradox exists with respect to social support and health among Aboriginal people – while they experience higher rates of social support within Aboriginal communities, this important social determinant of health has not resulted in better health status for Aboriginal seniors (Richmond, 2009). Fewer personal (especially economic) and interpersonal resources likely contribute to this paradox. Additionally, Richmond (2009) has shown that there can be stress associated with the supportive roles community members are expected to fill. Due to the small size of many Aboriginal communities and their relative isolation, community members who are also nurses or social workers may have difficulties escaping their professional roles. This can have negative consequences for those providing support as well as those receiving it.

The level of social isolation experienced by older Aboriginal people is not known, and it is likely that there is a tremendous degree of diversity depending on the community, and type of senior (e.g., middle-class urban dweller, senior living in an isolated reserve community, etc.). However, cultural and structural factors, coupled with high rates of significant traumatic events over the life course, and trust and understanding of the health and community care systems, may increase the risk of loneliness and social isolation. Low education and health literacy, low income or poverty, geographic location and isolation, poor health status, cultural separation, a higher likelihood of experiencing traumatic events, and fragmented informal and formal support systems create a cascading of risk for social isolation among Aboriginal seniors. Approaches to reducing the risk of social isolation need to address a number of risk factors that appear to be integrally connected to historic, cultural and social marginalization and rooted in economic disadvantage.

Seniors who are caregivers

Caregivers are the backbone of the continuing care system in Canada. In 2012, 8 million Canadians, or 28% of the population aged 15 and over, provided care to family members or friends with a long-term health condition, a disability or problems associated with aging (Turcotte, 2013). Among these caregivers, the majority provided care to a senior.

Although literature shows positive effects of caregiving, there is a substantial body of research that has established negative effects on the caregiver, especially when the intensity of care and related stressors increases, such as when the care recipient experiences serious physical and/or cognitive impairment (Pinquart and Sörensen, 2003; Robinson et al., 2009). Negative effects on the caregiver include burden, depression, stress, financial problems, poor health, loneliness and social isolation (Chappell and Funk, 2011).

While there is a vast literature on caregivers, not much of it has explored the issue of social isolation. Those studies that have been conducted suggest that caregiving in general does not lead to feelings of loneliness and social isolation per se. However, caregivers who experience lower levels of caregiving satisfaction experience more loneliness and social isolation. When caregiving is particularly demanding – when it entails heavy or live-in 24-hour care, for instance, or caring for someone with moderate or severe dementia or a serious physical disability, or when the care competence of the caregiver is low or challenged – it can result in feeling alone or being disconnected from the wider community of family, friends and neighbours (Lecovich, 2015; Robinson et al., 2009). Spousal caregivers may also be at greater risk of experiencing decreased social support and loneliness (Centre for Addiction and Mental Health, 2010). A qualitative study on caregiving in Elliot Lake, a retirement community in northern Ontario, found that caregivers' opportunity to participate in social activities is significantly reduced by the inability to leave the care recipient alone, and that caregivers can feel a great sense of isolation in the post-caregiving stage as they need to find new identities, social roles and activities (Sherman and Lacarte, 2012). Caregivers who are looking after spouses with dementia face the additional challenge of, in the words of one expert, "waking up every day with someone who has nothing to say."^{Footnote 18}

Caregivers in remote areas may face particular challenges to isolation and loneliness. Qualitative research undertaken by Chappell et al. (2008) found that caregivers in remote areas faced a number of challenges, such as small isolated populations, severe winters, non-existent community programs, lack of community facilities in which to meet with others, and costly travel for services.

There are also a number of ameliorating factors that can protect against the negative impact of caregiving on health and well-being, including loneliness and social isolation. For instance, Robinson et al. (2009) found that lack of formal care services and financial difficulty were by far the strongest predictors of negative consequences of caregiving. Similarly, accessibility barriers can reduce social participation opportunities for both care recipients and caregivers, and lack of information on available services can increase caregiver burden and affect social participation (Sherman and Lacarte, 2012).

Thus, the provision of helping and financial support through informal assistance, respite services, home care or related services, and income and tax relief programs can have an enormous impact on the reduction of loneliness and social isolation of caregivers in these extreme situations (Sims-Gould and Martin-Matthews, 2007; Sussman and Regehr, 2009).^{Footnote 19}

Immigrant seniors

Canada possesses a significant population of immigrants from a diversity of countries and regions. In 2011, approximately 22% of the Canadian population was born outside of Canada, including 4.5 million seniors.^{Footnote 20} Generally, immigrant seniors in Canada have been here for a long time since people tend to migrate when they are relatively young. Indeed, seniors aged 65 and older represented a relatively smaller proportion of new arrivals between 2006 and 2011 (only 3.3%). According to the Canada Mortgage and Housing Corporation (2012), the more recent the period of immigration, the lower the probability that a senior immigrant lives alone. This finding may be partly explained by the fact that many recent immigrant seniors live with members of their family who sponsored them when they came to Canada. Furthermore, among certain immigrant groups, it is traditional for elderly people to live with their children or other relatives.

De Jong Gierveld et al.’s (2015b) study found that, on average, older immigrants are significantly lonelier than older adults who are Canadian-born. Among the immigrant population of seniors, however, there is considerable diversity. Those who originate from countries with different languages and cultures than Canada are significantly lonelier than their Canadian-born counterparts while those who share English or French as a first language are not.^{Footnote 21} Interventions targeting the senior immigrant population must therefore take into account language and ethnicity.

This finding is reinforced by the experience of frontline workers. Among older socially-isolated immigrants in Vancouver, for instance, program administrators of the SUCCESS Seniors Quality of Life Outreach Project identified lack of proficiency in English as a major risk factor. Consequently, the program organizes activities and events in the languages of its clientele (primarily Cantonese and Mandarin) (Li, 2010).^{Footnote 22}

Other factors were also shown to be significant in de Jong Gierveld et al. (2015a), particularly sense of belonging to the local community and frequency of social contacts (though not frequency of contact with relatives). Perhaps surprisingly, loneliness was also significantly associated with having a network of people who speak one’s native language. The authors speculate that this could be because such networks reinforce a feeling of cultural loss, or may be too small (and insular) to help the person connect to a broader network of people across a broader geographic space. In France, immigrant seniors seem to be at greater risk of experiencing several difficulties, including poverty and isolation (Alisaid-Guérain, 2015).

As a segment of the immigrant population, qualitative research undertaken by Vatz Laaroussi (2013) on older refugee women in Montreal suggests that migrational trajectories may have an impact on the experience of social isolation. Vatz Laaroussi found that first-generation exiles who came to Quebec in the 1960s and 70s with higher levels of education and autonomy were unlikely to experience isolation compared to women who migrated from refugee camps in the 1980s and 90s who tended to be more culturally distant from their children and grandchildren.^{Footnote 23}

Some immigrant senior groups may also possess traits that provide a measure of protection against social isolation. Ng and Northcott (2015) found that familial living arrangements of South Asian immigrants, who typically live in extended families, were an important factor protecting against isolation in Edmonton. What mattered most, however, was not the living arrangement per se, but the amount of waking time spent alone at home and the quality of family relationships.^{Footnote 24}

Limited research has also been undertaken on factors related to successful aging among immigrant groups with implications for the issue of social isolation. Noubicier and Charpentier (2013), for instance, studied the perceptions of senior immigrant women from Sub-Saharan Africa in Montreal towards aging, and found that social engagement, intergenerational relationships, financial autonomy, and faith were all key elements of successful aging.^{Footnote 25}

In some cases, identifying seniors from immigrant groups who experience social isolation may be challenging. As part of its Age Friendly Initiative, the City of Nanaimo (British Columbia) sought to identify and remove barriers to social isolation for seniors in the community (North Sky Consulting Group Ltd, 2013). While only a small number of socially-isolated senior immigrants were identified, the authors speculate that referral agencies and others may have been hesitant to identify isolated immigrant seniors out of the fear that this could somehow jeopardize their immigration status.

Other factors may also interact with social isolation among senior immigrant groups. In the broader social isolation literature, for instance, there is a convincing body of knowledge suggesting that younger immigrant women are at high risk of social isolation. From the limited published work, however, gender differences among immigrant seniors could not be identified as a compounding factor.^{Footnote 26} Little is also known about the comparative prevalence of isolation among specific groups of immigrant seniors. Although more research is required on immigrant seniors and social isolation, it is clear that strategies to address this issue should consider culture, language, and the broader familial environment.

LGBT seniors

The LGBT community is a heterogeneous group comprised of lesbian, gay, bisexual and transgender individuals. According to census estimates, there are approximately 335,000 LGBT seniors in Canada, though this figure is understood to be an underestimate. Self-reporting is limited because many LGBT seniors are reluctant to be open about their sexual orientation or gender identity (Tang, 2015).

While many social and legislative changes in Canada have led to a considerable increase in the recognition of the rights of LGBT Canadians (Brotman et al., 2010), the history of the LGBT community is tainted in Canada and around the world by homophobic discrimination, stigmatization and exclusion by society (Brotman et al., 2010; Cassidy, 2007; Groupe SOS, SOS homophobie and AIDES, 2013). Yet LGBT seniors, many of whom were once required to keep their identity or sexual orientation secret, are still afflicted by the consequences of their traumatic experiences; and many forms of discrimination are still ongoing.

Discrimination towards older LGBT adults can be related to numerous factors; for example, as a result of having a sexual orientation that is different from the majority, aging, being a woman, having HIV/AIDS, or identifying with a sub-culture. Such discrimination can have a multiplier rather than a cumulative effect (Hébert et al., 2012; Orel, 2014). For certain LGBT seniors, living with multiple forms of discrimination poses additional challenges when they need support from their friends and families, or when they are looking for a suitable place to live (Orel, 2014). This complicates their access to services, such as health care, and has a direct impact on their overall health (Porter and Krinsky, 2014).

Indeed, there is considerable literature demonstrating that LGBT people experience discrimination in accessing health services. Several studies carried out in the 1990s and early 2000s revealed that older LGBT adults stated that they experienced discrimination in their interactions with health care professionals (Brotman et al., 2010), a situation that appears to be unchanged (Orel, 2014; Stein et al., 2010).

A study in the United States showed that discrimination can prevent LGBT seniors from seeking out resources (Orel, 2014), which may have serious consequences on their well-being (Brotman et al., 2010). Trans seniors who have not completed their physical transformation have even been forced to revert to behaviours associated with their original gender (clothing, etc.) (Latham, 2016). These experiences can lead to social isolation.

Additionally, LGBT seniors may have weak primary networks on which they can count on in situations of lost autonomy (Groupe SOS, SOS homophobie and AIDES, 2013; Stein et al., 2010). This may be due to the fact that they did not have children or that their network does not accept their LGBT identity (Orel, 2014). LGBT individuals who reveal their identity once they are older risk living separated from their friends and family (Groupe SOS, SOS homophobie and AIDES, 2013), which leads to a fear of coming out or a tendency to keep a certain distance in order to avoid rejection (Orel, 2014). Moreover, LGBT seniors who are out and maintain good relations with their friends and family indicate that these ties seem to break down as they get older for no apparent reason (Orel, 2014). The social network of LGBT seniors is therefore most often comprised of other LGBT persons (Orel, 2014). For some of them, in particular lesbians, friends come to be considered as family members (Gabrielson et al., 2014).

Furthermore, the spread of HIV in the 1980s directly affected the primary network of several LGBT people as a result of the death of their significant others, which contributed to the breakdown of their social network and their increased isolation (Groupe SOS, SOS homophobie and AIDES, 2013; Larkin, 2008; Orel, 2014). The weak social network, combined with difficult social recognition leads some LGBT people to dread the moment when they will lose their autonomy (Orel, 2014). Some consider suicide in order to avoid finding themselves in this type of situation (Witten, 2014, quoted by Zelle and Arms, 2015).

Other factors are also important. Isolation is apparently easier to overcome in an urban setting, where the LGBT community is larger and more visible (Erdley et al., 2014). This poses major challenges in a country like Canada, where urban settings are limited. Social isolation seems to have a more direct effect on homosexuals and bisexuals who hide their identity or don’t fully accept it than on those who assume their identity more openly (Camenzind, 2012). Additionally, the more that LGBT seniors live anonymously, the harder it is for services to reach them (Camenzind, 2012). This difficulty is also experienced by trans seniors depending on when their identity change took place (Hébert et al., 2012). Those who begin it later in life have much more difficulty with psychological and physiological adaptation, as well as with social acceptance (Cook-Daniels, 2001, quoted by Hébert et al., 2012).

As they age, older homosexual adults, and in particular men, also report feeling excluded from their community, which has a strong sub-culture that values youth and sexuality (Cassidy, 2007). Some seniors therefore have their need to socialize met in open religious communities, where they can become involved and have social relationships that are not focused on sexuality (Cassidy, 2007).

The literature therefore suggests that LGBT seniors are particularly vulnerable when it comes to social isolation, though precise numbers are often lacking. Given the multilayered challenges they encounter related to complex and ongoing histories of discrimination, LGBT seniors face many challenges that contribute to isolation and loneliness.

Seniors living alone

Among the Canadian population, living alone becomes increasingly more common as people age, especially for older women. According to the 2011 census, 31.5% of women aged 65 and over lived alone compared to 16% of senior men. Among those aged 85 and over, 36.6% of women lived alone compared to 21.7% of men (Statistics Canada, 2012). This is due to the higher life expectancy of women, men marrying younger women, lower rates of remarriage among women, and increasing preferences for independent living over time (Wister and McPherson, 2014). Older women are also more likely to live in institutions: among those aged 85 and over, 35.2% of women compared to 22.6% of men lived in institutions (Statistics Canada, 2012).

The association between living alone, social isolation and loneliness is complex. Living or not living with others does not necessarily result in a lack of social contact (Victor et al., 2000), yet it may be a sufficient cause. Living alone has been associated with fewer social contacts and lower social support, partly because people who live alone are not married or partnered, thereby lowering the amount of proximal support available (Sinha, 2014). Widowed, divorced and never married seniors have been shown to experience higher rates of loneliness and social isolation than married seniors (de Jong Gierveld et al., 2015a; Keefe et al., 2006; National Seniors Council, 2014a and 2014b). Recent findings from the English Longitudinal Study on Aging (ESLA)^{Footnote 27} revealed that household size is inversely related with prevalence of loneliness: seniors who live in larger households are less likely to be lonely (Iparraguirre, 2015).

Seniors living alone are also much more likely to be living in low income, which may increase the risk of being isolated (see below). Similarly, while living in a deprived neighbourhood increases the risk of all seniors of living in isolation, the effect is even greater among seniors who live alone. Transportation issues may also arise for single or widowed seniors who have never learned to drive (British Columbia Ministry of Health, 2004).

Data from the Canadian Community Health Survey – Healthy Aging, revealed that people living alone are particularly likely to need help with transportation, and that inadequate access to transportation or difficulty getting around may be a barrier to social participation (Turcotte, 2012).

While living alone and social isolation and loneliness share many of the same correlates (e.g., age, gender, widowhood, childlessness, entry into care, and mortality), they differ in terms of various resource-based risk factors, especially those linked to social class, physical and mental health status (Victor et al., 2000; Holt-Lunstad et al., 2015). Thus, it may not be living alone that contributes to social isolation and loneliness, but rather, other risk factors associated with both (Perissinotto and Covinsky, 2014). For instance, Ng and Northcott (2015) found that, among a sample of South Asian seniors living in Edmonton (who tend to live in extended families – see above), loneliness was associated with the quality of relationships and walking time, rather than living arrangement per se. Also, in a recent systematic review, Holt-Lunstad et al. (2015) found that mortality risk was increased by 29%, 26%, and 32% respectively for persons experiencing social isolation, loneliness and living alone, controlling for other correlates. This underscores the importance of examining the cumulative effects of risk and protective factors underlying social isolation and loneliness.

There is a need for research that can disentangle the common and distinct causes and consequences of living alone, loneliness and social isolation, before living alone can be used as a proxy or correlate of the latter.

Seniors living in remote or rural areas

Approximately 23% of all seniors in Canada live in rural areas, a pattern that is similar to the overall population since four out of five Canadians lived in urban areas in 2011 (Canada Mortgage and Housing Corporation, 2012). Large-scale studies on the overall prevalence of social isolation in rural areas in Canada are virtually non-existent, and few studies have compared differences between urban and rural areas. Kobayashi et al. (2009) found that as many as 17% of older adults, living primarily in small-town and small-city environments in British Columbia, were socially isolated.^{Footnote 28}

Gender differences were also identified, with rural older men having smaller support networks and lower utilization rates of health and social services compared to rural women, who used more services and received more support from family and friends. The authors conclude that geographic and gender differences require targeted programs to help individuals sustain or enhance social integration.

For seniors living in rural areas with advanced cancer, lack of community support and access to services can lead to isolation and loneliness (Duggleby et al., 2011). Inadequate resources, loss in community and devaluing of rurality have also been linked to loneliness and a negative view of one’s self among rural senior women in Ontario. Ethnicity and gender were additional negative factors, with Aboriginal women pointing to a long history of discrimination and exclusion (Panazzola and Leipert 2013). In focus groups conducted by Hall et al. (2003), the inadequacy or non-existence of transportation services in rural areas was brought up as a source of isolation and loneliness. Keating et al. (2013) also found that loneliness among long-time residents in rural areas of newly affluent communities may have been linked to the fact that they were no longer able to take part in activities and organizations due to financial challenges (see also section on seniors with low income). The link between lack of transportation and isolation in rural areas has been noted elsewhere in the literature (Keefe et al., 2006). (See above for findings on isolation and living in remote areas among caregivers).

The way in which social isolation interacts with mental health may also vary by living environment. While seniors in rural, urban, and small-town environments exhibit similar levels of depressive symptoms (about 11.5% of the senior population), the factors associated with depression vary significantly across different environmental settings.

For seniors in rural areas, living alone, perceiving one's income as inadequate, and having a functional impairment were associated with depressive symptoms (St. John et al., 2006) in contrast to a different set of factors for urban and small-town seniors.^{Footnote 29}

Despite the limited research on rurality and social isolation among seniors, the literature suggests that social isolation has an impact on the wellbeing of rural seniors that cannot be assimilated to the more general experience. More comparative research is required in this area.

Low-income seniors and those living in poverty

Approximately 4.7 million people – or 13.8% of the population – lived with low income in 2012 (Statistics Canada, 2014).^{Footnote 30} After 20 years of reductions, Canada’s elderly poverty rate rose between the mid-1990s and late 2000s, particularly among older women (The Conference Board of Canada, 2013). Drawing on data from Statistics Canada, Shillington (2016) reports that, low income among Canadian seniors has been steadily rising, from a low of 4% in 1995 to 11% – or one in nine seniors – in 2013. For seniors living alone, the poverty rate is particularly high.

Over the same period, the rate rose from 9% to 28% —almost one in three persons—for single women seniors, and from 6% to 24% for single male seniors. Many other groups of seniors are also more likely to experience poverty, including seniors with less education, seniors with intermittent work histories and low wages, older immigrants, Aboriginal seniors, seniors with chronic health problems, and seniors with disabilities (National Seniors Council, 2009).

Poverty and low income have been associated with experiences of loneliness and social isolation among the general population (Stewart et al., 2007). However, there have been few studies that focus specifically on older adults. Yet, older adults are more likely to live with low income than most other age groups, and low income has been associated with poorer health outcomes and lower quality of life indicators among seniors (Wister and McPherson, 2014). Thus, seniors living with fewer financial resources are faced with a number of potential negative outcomes which have been linked to loneliness and social isolation (Keefe et al., 2006; National Seniors Council, 2014a and 2014b).

After controlling for a large number of other risk factors, de Jong Gierveld et al. (2015a) found that income exhibited a weak but statistically significant inverse association with loneliness among seniors. A review of the literature on the social participation of seniors carried out by Raymond et al. (2008) also revealed that income is negatively associated with isolation, and a representative study in Belgium revealed that social isolation among seniors with low income was 29 percentage points higher than those with higher income (Vandenbroucke et al., 2012). Low income or poverty may often work in combination with other related risk factors to place people at ‘multiple jeopardy.’ One recent study has established a connection between low income, social isolation and nutritional risk among older adults, and these associations were especially strong among senior men who were members of a racial minority group (Locher et al., 2005). Research has also shown that low income is often associated with low self-esteem, which can inhibit the search for new social contacts (Pinquart and Sörensen, 2000).

In addition, MacCourt et al., (2011) found that poverty limits the ability of seniors to have adequate housing, while Gilmour (2012) found that retired seniors living in low income housing were more likely to be lonely. A report on housing for marginalized people in France found that aging and poverty put seniors at risk of isolation. While inadequate housing not adapted to the needs of seniors can be a major contributing factor to social isolation, adequate housing can be a major preventative factor (Haut comité pour le logement des personnes défavorisées, 2012).^{Footnote 31}

One researcher also found that seniors in public long-term care (LTC) facilities were more likely than those in private facilities, where residents have greater economic and familial resources, to experience isolation (Trépied, 2012). Living in a deprived neighbourhood can also promote loneliness, especially among seniors who are living alone (British Columbia Ministry of Health, 2004). Seniors living on a low income in deprived neighbourhoods may have safety concerns related to leaving their dwelling (Abbott and Sapsford, 2005). This can be particularly challenging since many seniors look to their neighbourhoods to meet both practical and emotional needs (Freiler, 2004). Gentrification may also have an impact on increasing loneliness. Research in rural Canada, for example, found evidence of loneliness among long-time residents of newly affluent communities who were no longer able to take part in activities and organizations and for whom financial challenges limited social connections (Keating et al., 2013).^{Footnote 32}

Older people who live alone are more likely to be near or below the poverty line than other older people. These seniors, in turn, face greater barriers to social participation, particularly if they live in rural or small-town environments, where there are often fewer social and leisure activities available (Kobayashi et al., 2009). One study of low-income senior women living alone in British Columbia found that, while their support networks were comparable to other groups, they relied more heavily on family support; greater efforts were required to diversify both their informal and formal sources of support since small family networks can be quickly overwhelmed (Ryser and Halseth, 2011).

Conversely, high income has been shown to be associated with less loneliness, possibly because those with higher income have commercial networks that can be used for social opportunities and because they are also in a better position to return support from others (Pinquart and Sörensen, 2000). Early evidence from the English Longitudinal Study on Aging has also shown that isolation decreases as wealth increases (Shankar, 2015). In the General Social Survey, Statistics Canada found that, although household income does not have a large influence on the number of close relatives and friends, Canadians in the highest income bracket ($150,000 or more) have larger networks of "other" friends (double the size of those with a household income of $20,000 or less) (Sinha, 2014).

In recognition of the complex interaction between poverty, isolation and other risk factors, some policymakers have adopted a comprehensive approach toward addressing social isolation among seniors with a strong economic dimension. France, for example, recently adopted a law related to societal adaptation to aging that focuses on reducing social inequality and isolation (Assemblée nationale, 2014). The law builds on previous efforts to address social isolation, including the national mobilization against the isolation of seniors,^{Footnote 33} a program which was embedded in the national plan to fight against poverty and for social inclusion (Serres, 2013). (Prior to the adoption of the law, a study on the social benefit for seniors found that the majority of recipients were socially isolated, and that the benefit was not sufficient to lift them out of poverty (Chaput et al., 2007)). In a related vein, Greenfield et al. (2015) argue that age-friendly community initiatives have the potential to address multiple issues related to older adults’ health and wellbeing, including economic insecurity and social isolation (the concept of age-friendly communities is discussed in Section 7.1.).

Seniors with mental health issues (including Alzheimer’s and other dementias)

Mental health has been defined as the capacity of each of us to feel, think and act in ways that enhance an individual’s ability to enjoy life and deal with life challenges. Mental illnesses, in contrast, are seen as alterations in thinking, mood or behaviour associated with some significant distress and impaired functioning (Standing Committee on Social Affairs, Science and Technology, 2004). The literature also includes cognitive and age-related mental health issues as part of the mental health spectrum. For example, the Mental Health Commission of Canada, as outlined in Guidelines for Comprehensive Mental Health Services for Older Adults in Canada, promotes a holistic approach considering mental illnesses as including Alzheimer’s disease and age-related dementias, serious and persistent mental illnesses complicated by aging issues, and mental illnesses that occur for the first time in old age.

Approximately 20% of seniors living in the community and up to 90% of those living in institutions are estimated to have a mental health problem or illness (Public Health Agency of Canada, 2010). The most common conditions include: depression, dementia, anxiety, delirium, and delusional disorders. Persons diagnosed with mental health issues tend to have fewer social resources, poor social competencies and symptoms that can pose barriers to maintaining connections to family, friends or community (Smith and Hirdes, 2009).

The stigma attached to an individual experiencing a mental illness can be a powerful agent in separating them from social contacts and social support systems (MacCourt et al., 2011; National Seniors Council, 2014a and 2014b). Several studies have shown that there are higher rates of reported loneliness amongst seniors with a mental illness, particularly dementia (e.g., British Columbia Ministry of Health, 2004; Victor et al., 2015). In one recent survey of people with dementia in the UK – the majority of whom were aged 65 and over – nearly two-thirds of those living alone reported feeling lonely, and a third said they had lost friends (Kane and Cook, 2013).

Mental health and mental illness among seniors can be connected to aspects of the aging process as well as situational factors that are associated with end of life experiences. Often, mental illness can be incurred or magnified due to stressors associated with advanced age, such as declines in resilience, physical health, low perceived control over health, death of a partner, moving to a new environment, financial difficulties, social isolation or loneliness (Newall et al., 2013 and 2014; Wister and McPherson, 2014). In this sense, the relationships between social isolation and mental health problems are often related to other aging-related challenges for many individuals.

At the same time, mental health issues of any kind may follow individuals throughout their life in a continual or episodic manner. An earlier onset of mental illness may lead to greater barriers in developing strong social contacts and maintaining social engagement with time, which can be carried forward into old age (Smith and Hirdes, 2009). The association between social isolation and mental health, therefore, needs to be examined from a life-course perspective as well as a dynamic one.

Seniors who become socially isolated, in turn, are at greater risk of mental illness given that social support is a protective factor for mental health problems (Chappell and Funk, 2011). For instance, Iliffe et al. (2007) found that social isolation was a predictor of depressed mood and perceived health among seniors who were not living in institutional settings. Similarly, in a review of the literature on suicide prevention among seniors, Saïas et al. (2013) found that isolation is a major risk factor associated with suicide among seniors. Social contact also attenuates two factors that are closely associated with suicide – depression and sleep disorders (and thus acts as a protective factor). There is also a connection between social isolation and addiction. Graziani (2010), for example, has shown that alcohol abuse among seniors can be a response to solitude, isolation or the loss of social support. The association between social isolation and mental health is therefore complex and bidirectional.

While mental health outcomes appear to be more associated with feelings of loneliness than social isolation per se (Coyle and Dugan, 2012), there is need for more research into the potentially cumulative and reciprocal effects of social isolation and mental health over the later life courses of individuals. Kobayashi et al. (2009) have found, for instance, that socially isolated adults are significantly different from their non-socially isolated counterparts in terms of several factors related to psychological well-being, including having higher levels of distress, lower coping abilities, and more negative feelings.

Seniors with health challenges or disabilities

As people age, they may experience physical symptoms associated with illness, frailty or the aging process. For others, disabilities may be a life-long or long-term experience. Although physical health issues and disabilities are distinct factors, both may be associated with physical vulnerability to social isolation. In Canada, most persons aged 65 and over have at least one chronic illness, and a majority have more than one chronic illness (Public Health Agency of Canada, 2010). Additionally, approximately 45% of seniors report a mild, moderate or severe disability, with this rising to 75% among those aged 85 and over (Statistics Canada, 2007).

While a majority of older adults rate their perceived health, well-being and daily functional ability as high even in the face of chronic illness or aging-related disability, many seniors nonetheless face challenges to daily functioning and maintaining connections with their family, friends and community. These impairments include pain, limited mobility due to sight, strength, and balance problems, loss of breath, general weakness, or fatigue, and other indicators of frailty and mobility impairment (Wister and McPherson, 2014). In the Canadian Community Health Survey – Healthy Aging (CCHS), health reasons were the most common reason given by seniors 75 years and over who were unable to participate in activities as much as they would have liked (Turcotte, 2012).

In a study on social isolation among seniors in British Columbia, Kobayashi et al. (2009) found that individuals who report being in fair or poor health have considerably higher odds of being socially isolated than those who report being in excellent, very good, or good health. Data from the CCHS revealed that the odds of being lonely were significantly higher for seniors who reported urinary incontinence than for those who did not, even when age, sex, education and living arrangements were taken into account (Ramage-Morin and Gilmour, 2013).^{Footnote 34} Keefe et al. (2006) also found that higher number of chronic illness was associated with higher levels of vulnerability to isolation even when controlling for other factors. Earlier research from the 1990s showed that chronic illness was one of the main factors linked to social isolation (see Hall et al., 2003).

Although physical health has been associated with social isolation and loneliness, the directionality of the relationship remains ambiguous. Most research have examined the effects of social isolation on a variety of health outcomes, including perceived health, functional health (disability), chronic illness, and accessing health care (Cornwell and Waite, 2009; de Jong Gierveld et al., 2015a; Kobayashi et al., 2009; Newall et al., 2015).^{Footnote 35}

By definition, social isolation can restrict connectedness with social networks, which can result in fewer resources, lower social support, poorer health behaviours and lower access to health and community care system, all of which result in poorer health outcomes and lower levels of healthy aging (Keefe et al., 2006; National Seniors Council, 2014a and 2014b). Social isolation can also act as an aggravating factor in the diagnosis of disabilities among seniors – with fewer family members and friends, socially isolated individuals who develop sensorial problems as they age are likely to go untreated (Covelet, 2007).

Physical vulnerabilities can also limit mobility, which can compound transportation difficulties, particularly in areas where there is inadequate transportation provision. In the CCHS, when controlling for all other factors, seniors who travelled mostly by driving their car were more likely to participate than those who used any other form of transportation. Seniors who were mainly passengers and did not have a licence and those who used accessible transit or taxis had the lowest participation rates (Turcotte, 2012). In qualitative research undertaken by Hall et al. (2003), seniors indicated that the cessation of driving due to health problems was a problem of particular significance, especially in rural areas (see also North Sky Consulting Group Ltd, 2013). Some also mentioned the stigma of being dependent upon others as a barrier to asking for assistance. According to the World Health Organization, transportation is in turn a determinant of health because of the role it plays in seniors’ independence and ability to access resources (Keefe et al., 2006).

* The extent to which each factor has been linked to social isolation among particular groups is discussed in the previous sections. As noted throughout, some factors are more generalizable depending upon the methodologies of specific studies.

Age-friendly communities

In the Council’s previous report on social isolation (National Seniors Council, 2014a), the notion of age-friendly communities was raised as providing an important conceptual and practical approach to tackling many seniors’ issues, including social isolation. This approach is again noted here as it has the potential to mitigate many of the general risk factors that seniors from different walks of life experience. Developed by the World Health Organization (WHO), the concept of age-friendly communities (AFCs) addresses eight key domains of community living that enable seniors with varying needs and capacities to live in security, good health and to participate fully in society.^{Footnote 37}

These include:

• Transportation
• Housing
• Social participation
• Respect and social inclusion
• Civic participation and employment
• Communication and information
• Community support and health services, and
• Outdoor spaces and buildings

AFCs provide opportunities to improve the quality of life not just for seniors, but for the whole community. As the Chief Public Health Officer of Canada (CPHO) (Public Health Agency of Canada, 2010) has noted, however, AFCs require not just increased service delivery, but also community awareness of services for seniors. "Developing outreach strategies for programs and services for seniors," the CPHO notes, "will require identifying which populations are underutilizing services and targeting attention to those seniors (and their networks) in program marketing plans." Changing the lived environment so that there are more meeting spaces for seniors and places to rest (including outdoor benches) are modest albeit important steps (Caradec, 2006). Underlying the success of AFCs is their ability to secure funding and resources over time (Greenfield et al., 2015). While age-friendly policies have the potential to combat ageism, a barrier to their implementation can also be discriminatory attitudes towards older people (Edwards, 2015).^{Footnote 38}

Many provinces in Canada have undertaken age-friendly community initiatives. Garon et al. (2014) examined two case studies in Quebec and identified several factors that can determine success or failure. Among other things, they found that a pluralistic, participatory, and collaborative approach among stakeholders was essential (as opposed to a traditional top-down approach). Key to the success, according to one of the cases they looked at, was the identification of common goals and the development of a simple model on which there was consensus among stakeholders.

The Age-Friendly Communities in Canada: Community Implementation Guide – Toolbox (Nova Scotia Centre on Aging, 2012) presents useful and practical tips on how such communities can become a reality. The Public Health Agency of Canada (PHAC) promotes the uptake of AFC in Canada and provides a variety of tools and resources related to age-friendly communities. In 2015, PHAC has also prepared the Age-Friendly Communities Evaluation Guide to provide communities with practical information on how to use indicators to measure progress and evaluate their age-friendly initiatives.

Promising interventions for senior in general

With these high-level policy considerations in mind, the review ends with a brief overview of some of the promising interventions that have been identified in the literature that was consulted to tackle social isolation and reconnect seniors to their communities. While most of this literature focuses on seniors in general, findings on specific groups of seniors are referenced, where such findings are available. For information on various federal and provincial programs that address social isolation, the NSC’s Report on the Social Isolation of Seniors (2014) cites several relevant examples.^{Footnote 39}

A brief note on the limitations of this literature is in order. Although interventions targeting social isolation have increased in number over the last decade, the number of high quality evaluation studies of programs remains relatively low, limiting the evidence to fully assess their impact. Writing in 2003, Findlay noted that “research evidence to support the belief that intervention can counteract social isolation and its adverse effects on older people is almost non-existent.” Responding to the dearth in evidence, Sabir et al. (2009) offered practice-based suggestions for a social isolation research agenda, including efficacy studies of multicomponent interventions.^{Footnote 40} While the state of knowledge has gradually increased, there are still important limitations to be noted. As Jopling (2015) has recently suggested, evidence on the impact of interventions to address social isolation exists in a spectrum, though “even where the evidence is of a lower quality it can be an important step in the development of a firmer understanding of what works." International networks of researchers and practitioners have also begun to coalesce around the topic of social isolation, including the UK Campaign to End Loneliness.^{Footnote 41} The Campaign has launched a Research Hub to increase and develop the evidence base on the issue of loneliness in older age.

Available evaluation studies indicate that there is a high degree of diversity in terms of the types of interventions that are assessed, the methodology used, and the quality of the evaluation (Dickens et al., 2011). Research designs range from randomized control and quasi-experimental studies to a variety of qualitative-evaluation approaches.^{Footnote 42}

Further compounding the challenges, socially isolated individuals can be hard to reach to take part in evaluations (with the most isolated being the most invisible), diversity across seniors means that program impacts can be variable and funds for large-scale evaluations are often lacking. The result is that evaluations often have small and sometimes biased samples. Finally, researchers use different conceptual frameworks for classifying different types of interventions, which makes the task of comparing programs more difficult. Qualitative research has helped fill the void, but many studies are small-scale and it is not always clear whether findings can be generalized. Lurking in the background of many of these studies is the question as to whether or not program success is scalable – that is, can positive results be replicated at a larger level.

Still, important lessons can be gleaned from the literature. Systematic reviews of randomized control and quasi-experimental studies suggest that there is moderate evidence demonstrating the success of programs targeting social isolation in seniors.^{Footnote 43} Common characteristics of successful programs and types of programs are listed in the table below.

Image description of the Figure 3 : Successful Ways of Combatting Social Isolation

Characteristics of programs:

• Based on coherent theory
• Involve seniors in all steps
• Use participatory approaches
• Target groups of individuals who share common characteristics
• Target loneliness and social isolation directly
• Use multiple interventions
• Train and support co-ordinators and frontline providers
• Mobilize community resources
• Involve nurses and health professionals as gatekeepers and advocates

Types of programs:

• Group activities
• One-on-one interventions
• Support-provision
• Arts and culture based activities
• Leisure activities
• Inter-generational activities
• Educational activities
• Friendship programs
• Telephone support
• Gatekeeper programs
• Internet groups
• Support groups
• Religious activities

Outcome:

• Reduced social isolation

In terms of characteristics, programs that are based on a coherent theoretical basis of how to address social isolation have a greater chance of being successful, as are those that involve seniors in all programming steps, from planning to implementation, and evaluation. Participatory approaches that involve seniors in decision making and group activities are also important. Programs that target groups of individuals who share common characteristics – such as the various groups of seniors discussed in this review – can be fruitful, as well as those that target loneliness and social isolation directly (as opposed to targeting these indirectly while focusing on another issue). Programs that offer more than one method of intervention – such as group activities or individual support – have greater potential. Providing co-ordinators and frontline providers with good training and support is also important, as are mobilizing community assets to promote innovative approaches that pool resources, and involving nurses or allied health professionals as gatekeepers and advocates.

Jopling’s (2015) study on social isolation programs in the United Kingdom, conducted under the auspices of the UK Campaign to End Loneliness, further identifies the “structural enablers” that help facilitate success. These include designing programs at the neighbourhood level, using asset-based community development approaches – which focuses on identifying and mobilizing individual and community assets – creating opportunities for seniors to volunteer, and promoting age-positive approaches among key organizations and institutions within a local area. Joplin provides several useful descriptions of programs that have adopted these approaches.

In terms of the types of programs that are successful, the literature suggests that many different kinds of interventions can have a positive impact, including group activities, one on-one interventions and support provision (though group activities may have a greater impact according to the World Health Organization, 2015), arts and culture based activities, leisure activities, inter-generational activities, educational activities, friendship programs, telephone support, gatekeeper programs, internet groups, support groups, and religious activities. Thus, a range of different kinds of interventions can yield success – including those that use technology to create new communities – though in-person group activities are frequently mentioned as being particularly impactful.

In one systematic review, the Ministry of Health and Long-Term Care Ontario (2008) examined two different kinds of interventions – in-person group support activities and technology assisted interventions – and found that both showed positive results. Similarly, McDaid et al. (2015) looked at empirical studies testing the effectiveness of multiple programs and found that there was a small evidence base demonstrating that all of them had positive effects. These included arts and culture-based activities, friendship programs, intergenerational activities, and what they referred to as “Third Age” education. Toepoel (2013) examined data from a representative survey in the Netherlands and found that, among adults 65 years of age and older, cultural activities, reading books, and hobbies contribute to decreasing loneliness.

Individual program evaluations demonstrate positive results though they are usually based on small samples. An evaluation of Contact the Elderly (2008), for instance, indicated that a majority of participants felt less lonely as a result of the program. The program, which boasts 320 groups across the United Kingdom, organizes monthly Sunday afternoon teas between volunteers and seniors. Likewise, an evaluation of the Seniors Centre without Walls program, which provides free social and educational programming for older adults via telephone in Manitoba, revealed that the vast majority of participants felt more connected and less isolated (Newall and Menec, 2015). TeleVisit, a California-based non-profit, adopts a similar approach by facilitating group activities among seniors via teleconference and easy to use tablets.

In addition to these studies and evaluations, there is an emerging descriptive literature providing information on programs that tackle social isolation. For example, the World Health Organization (2015) describes several programs in Europe that have adopted innovative approaches, such as the Telephone Rings at 5 in Portugal, Multigenerational Centres in Germany, Cité Seniors in Switzerland, and Men’s Sheds in Australia and Ireland. Beach and Bamford (2015) also provide a useful summary of multiple projects in England including Walking Football, the Culture Club, and the Seafarers Project. The Belgium-based, Fondation Roi Baudouin, which provides funding to community initiatives to address social isolation, describes 20 separate projects that have been undertaken, their successes and challenges. Projects range from collective suppers, groups for cultural outings, and cross-cultural discussion groups to collective pampering sessions, inter-generational groups to promote knowledge exchange, and pen-pal groups between seniors and children (Teller and Dorsselaer, 2007). In France, social cafés have been established in several places where seniors can get together, converse, and participate in different activities, including intergenerational ones (Saci, 2012).

The organization, Pour Que l’Esprit Vive, hosts monthly musical concerts for seniors in care facilities in rural France as a way to encourage socialization and combat isolation. Both contemporary and older forms of music are played (Desjobert, 2012). Social actions centres (clubs du Centre d’Action Sociale), where physical activities are organized for seniors, and open houses (Maison Ouverte), where seniors can drop in to participate in activities, also exist in France (Hallier-Nader, 2011).

The annual reports of organizations tackling social isolation can also be extremely instructive. Les Petits Frères (2016), an organization that is active in 15 regions of Quebec, provides annual updates on its many activities, including inter-generational pairing of older people with young volunteers for social activities (jumelage), holiday celebrations, regional trips, and other ongoing outings and activities.

Finally, there are multiple toolkits that provide useful suggestions on how to start programs to address social isolation, including the Nelson Allan Project Toolkit, which provides practical information for grassroots groups and community organizations (Nelson Allan Project, 2015). Innoweave (2015) also provides practical information on how projects addressing social isolation among seniors can be scaled up and make use of social innovation techniques. Within the context of the UK Campaign to End Loneliness, the Hidden Citizens project provides practical steps to service providers on identifying those who experience loneliness (Goodman et al., 2015). The Campaign has also developed a guide to assist practitioners with evaluation.

Promising interventions for specific groups of seniors

In terms of the evaluation literature on programs addressing social isolation among specific groups of seniors, the literature is mostly suggestive. The following illustrates some potential promising interventions, however as previously stated, the evidence-base for some of these is still very much in its early stages, and this is not meant to be an exhaustive list:

• For Aboriginal seniors, activities that incorporate traditional activities and foods, including going out on the land, may be an effective way of addressing isolation and improving health outcomes. Some organizations that provide services to Aboriginal seniors, including Saanich First Nations Adult Care Society and Peter Ballantyne Cree Nation, connect seniors with youth, emphasizing the important role of First Nations Elders and storytelling (Health Council of Canada, 2013).
• For seniors who are caregivers, telephone-support groups may be an effective way of reducing social isolation (Stewart et al. 2006).^{Footnote 44} A study on caregiving in Elliot Lake recommended that, to enhance social participation among caregivers, the needs of care recipients must be met. Other recommendations included organizing social activities with both the caregiver and recipient involved, as well as activities that involve only the caregiver.
• For spousal caregivers caring for partners with dementia, adult day programs are an effective way of providing respite (Sussman and Regehr, 2009).
• For immigrant seniors, group activities provided in their main language can help break down linguistic barriers. As previously noted, the SUCCESS Seniors Quality of Life Outreach Project in Vancouver offers both recreational and educational activities in Mandarin and Cantonese to socially isolated seniors. In Belgium, immigrant and non-immigrant seniors are grouped together through the Mémoire Vivante program for cross-cultural learning as a way to overcome isolation (Teller and Dorsselaer, 2007). Similarly, in France, cafés for seniors adopt a mixed cultural and social approach to promote activities between ethnic groups and across generations and social classes (Saci, 2012).
• For older women living alone, educational activities via telephone have demonstrated some success based on participant self-reporting in the context of the Seniors Centre without Walls program in Manitoba (Newall and Menec, 2015). Widowhood bereavement support groups are also effective (Stewart et al., 2001).
• For seniors with dementia and their caregivers, memory cafés that provide naturally forming support groups can be effective (McFadden and Koll, 2014). Intergenerational programs such as the intergenerational choir for college students and persons with dementia and their family caregiver also show signs of success (Harris and Caporella, 2014).
• For LGBT seniors, countering discrimination is imperative. Training of health care staff involving LGBT cultural competency has demonstrated good results and has improved reception of and openness towards older LGBT adults, as well as consideration for their reality and specific needs (Porter and Krinsky, 2014). Training centres that address the experiences of LGBT individuals and their needs should be encouraged (Porter and Krinsky, 2014), including for social workers, who are particularly well placed to provide services to LGBT seniors and advocate on their behalf (Erdley et al., 2014).
• Inter-generational programs for LGBT seniors can also help build solidary and counter isolation. To counter ageism and social isolation in the LGBT community, a New York organization, Services and Advocacy for Gay, Lesbian, Bisexual and Transgender Seniors (SAGE), pairs LGBT seniors with LGBT volunteers with similar life experiences (age, orientation / sexual identity) (Larkin, 2008.)

Although programs and initiatives addressing social isolation seem to be multiplying with the growing awareness of this important issue, programs targeting specific groups of seniors, with some exceptions, are rare. While general programs are necessary, targeted ones could provide further impetus to efforts by stakeholders to combat social isolation. In both cases, evaluation efforts are essential to further develop the collective knowledge of the effectiveness of different types of interventions.