Introduction/Rationale

Increased life expectancy and population aging are altering the composition of societies on a global basis. Consequently, countries such as Canada face new questions about how to best meet the needs of a wide range of older people. One of the issues at the forefront of these conversations is that of social isolation. Older people who are socially isolated are often separated from their communities, and known to experience adverse outcomes to health and well-being (Bowling & Grundy, Reference Bowling and Grundy1998; Keefe, Andrew, Fancey, & Hall, Reference Keefe, Andrew, Fancey and Hall2006; Victor, Scambler, Bond, & Bowling, Reference Victor, Scambler, Bond and Bowling2000; Wister, Reference Wister2014). Although research over the past two decades has uncovered clear links between social isolation and risk factors for older people (Nicholson, 2010; Valtorta & Hanratty, Reference Valtorta and Hanratty2016), the conversation and response to social isolation among older people remains incomplete.

This growing interest in social isolation has resulted in a number of international and domestic programs to reduce social isolation among older people. At the international level, institutions such as the World Health Organization (WHO) have flagged social isolation as a key social and policy issue for aging (International Federation on Ageing, 2012; World Health Organization, 2015). In Canada, social isolation has been identified as a national level priority. In 2013, The National Seniors Council completed an investigation of social isolation among seniors to determine how to reduce the issue of isolation in later life ^{Footnote 1} – a task directed by the-then Minister of State (Seniors), Employment and Social Development Canada (ESDC), and the Minister of Health (The National Seniors Council, 2014). In 2015, ESDC launched a special funding opportunity through the New Horizons for Seniors program to fund pan-Canadian projects aimed at reducing social isolation among older people. ^{Footnote 2} At the provincial level in Canada, provinces such as British Columbia and Ontario have also identified social isolation as a priority issue facing older people (British Columbia Ministry of Health, 2004; Health Quality Ontario, 2008).

Despite research that outlines the risks and adverse outcomes of social isolation in later life, and government initiatives that seek to address isolation, there are several key obstacles that may impede policy and program efforts. First, researchers working from a variety of disciplinary backgrounds continue to struggle with how to define and measure social isolation. There is little consensus on the meaning of social isolation (Coyle & Dugan, Reference Coyle and Dugan2012; Valtorta & Hanratty, Reference Valtorta and Hanratty2016; Victor et al., Reference Victor, Scambler, Bond and Bowling2000), and some confusion exists between, for example, isolation and loneliness. Second, there is uncertainty about how to prevent or reduce social isolation in later life. The evidence demonstrating how social isolation may be effectively reduced is thin, making it difficult for researchers and decision makers to determine a plan of action (Findlay, Reference Findlay2003). Third, the complex nature of social isolation makes it difficult to approximate prevalence. It is estimated that somewhere between 19 per cent and 24 per cent of older people in Canada experience some level of isolation, and that over 30 per cent of older Canadians are at risk of isolation (Keefe et al., Reference Keefe, Andrew, Fancey and Hall2006; The National Seniors Council, 2014). This phenomenon, however, is not unique to Canada but is considered to affect older people, families, and communities on an international level (World Health Organization, 2015). Changes to global migration patterns, fertility rates, and the shrinking of family sizes are factors considered to increase social isolation on a global scale (World Health Organization, 2015). Although recognition of social isolation represents a first step in improving the lives of older people, we argue that the conversation needs to be extended to view social isolation among older people as a social and cultural phenomenon.

Risks and Outcomes

Social isolation has been linked to a number of individual-level risk factors, and has long been recognized as dangerous to the health and well-being of older people (e.g., see Brown [Reference Brown1960] and Munnichs [Reference Munnichs1964]). Individual-level risks for older people include health and wellness-related factors such as complex health conditions (Kobayashi, Cloutier-Fisher, & Roth, Reference Kobayashi, Cloutier-Fisher and Roth2009; The National Seniors Council, 2014) and mental health concerns (Buffel, Rémillard-Boilard, & Phillipson, Reference Buffel, Rémillard-Boilard and Phillipson2015; Elder & Retrum, Reference Elder and Retrum2012). Personal risk factors also include advanced age (e.g. 75 and older), losing the ability to drive, being widowed or divorced, and having no children (Buffel et al., Reference Buffel, Rémillard-Boilard and Phillipson2015; The National Seniors Council, 2014). Belonging to a group with minority status is also considered a significant risk factor, with older people who belong to LGBTQ+, language minority, or a racial minority group, identified as having an increased risk of social isolation (Elder & Retrum, Reference Elder and Retrum2012).

The literature is clear that social isolation can have negative effects on older people’s physical/mental health and well-being, and that a bi-directional relationship exists whereby health challenges can impact isolation. Older people who are socially isolated are at an increased risk of physical health conditions that include higher rates of circulatory conditions including hypertension (Tomaka, Thompson, & Palacios, Reference Tomaka, Thompson and Palacios2006) and coronary heart disease (Heffner, Waring, Roberts, Eaton, & Gramling, Reference Heffner, Waring, Roberts, Eaton and Gramling2011), increased rates of falls and re-hospitalization (Mistry et al., Reference Mistry, Rosansky, McGuire, McDermott, Jarvik and Grp2001; Nicholson, Reference Nicholson2012), and pre-mature mortality (Bowling & Grundy, Reference Bowling and Grundy1998; Steptoe, Shankar, Demakakos, & Wardle, Reference Steptoe, Shankar, Demakakos and Wardle2013). Research has also identified that social isolation can be detrimental to older people’s mental health and well-being, and that physical and mental health concerns are both a risk factor for and a negative outcome of social isolation (Wister, Reference Wister2014). Most notably, social isolation has been linked to the development of depression and/or depressive symptoms (Alspach, Reference Alspach2013; Beach & Bamford, Reference Beach and Bamford2014; Nicholson, Reference Nicholson2012), and in some cases, may increase the risk of suicide (Conwell, Van Orden, & Caine, Reference Conwell, Van Orden and Caine2011). Social isolation, and specifically a lack of social engagement, has also been linked to dementia and cognitive decline (Barnes, De Leon, Wilson, Bienias, & Evans, Reference Barnes, De Leon, Wilson, Bienias and Evans2004; Fratiglioni, Paillard-Borg, & Winblad, Reference Fratiglioni, Paillard-Borg and Winblad2004).

Family and community-level risks for social isolation also exist, and tend to be discussed as having a compound effect with individual risks. At the family level, identified factors for risk of social isolation include particular household and family characteristics, such as low household income and living alone (Elder & Retrum, Reference Elder and Retrum2012; Kobayashi et al., Reference Kobayashi, Cloutier-Fisher and Roth2009). Becoming socially isolated may be associated with changing household structures, such as transitioning into residential care, or having household members die or move out (Buffel et al., Reference Buffel, Rémillard-Boilard and Phillipson2015; Elder & Retrum, Reference Elder and Retrum2012). At the social and community level, risks of social isolation for older people include physical barriers and/or poor urban design (Buffel et al., Reference Buffel, Rémillard-Boilard and Phillipson2015; Elder & Retrum, Reference Elder and Retrum2012), lack of accessible transportation, and too few opportunities for meaningful social participation (The National Seniors Council, 2014).

These family and community-level risks are beginning to be considered as creating negative community-level outcomes that extend beyond individual health and well-being. As more older people become isolated and closed off from other people, their communities miss out on the important contributions that older people make, and social cohesion within society can be negatively impacted (Hortulanus, Machielse, & Meeuwesen, Reference Hortulanus, Machielse and Meeuwesen2006). Through this process, older people may become invisible, and/or reinforce negative age-related stereotypes (Falletta & Dannefer, Reference Falletta, Dannefer, McLeod, Lawler and Schwalbe2014). As Buffel et al. (Reference Buffel, Rémillard-Boilard and Phillipson2015) have stated, communities with socially isolated older people experience a “weakening of social bonds” across generations (p. 13). Such findings suggest the need for a careful reconsideration of the assumptions and understandings that guide current approaches, accompanied by an analysis of social isolation as a social and cultural phenomenon.

Defining Social Isolation – The Individual and the Collective

Despite a broad understanding that social isolation is a complex phenomenon comprising a range of contributing factors, and a variety of uses in policy and research contexts, ^{Footnote 3} research and responses to social isolation among older people have tended to focus on the individual level (Bachrach, Reference Bachrach1980; Lowenthal, Reference Lowenthal1964). This has taken place through an objective/subjective distinction, and in many cases, a privileging of the individual, objective-level criterion. Most notably, early definitions of social isolation – such as those used in public health – focused on counting the objective number of social connections and network attachments of an individual (Berkman & Syme, Reference Berkman and Syme1979). These types of definitions led to social isolation typically being framed as an exclusively objective measure of social embeddedness, and distinguished from the experience of loneliness, which is usually defined as being entirely subjective (Victor et al., Reference Victor, Scambler, Bond and Bowling2000; Wenger, Davies, Shahtahmasebi, & Scott, Reference Wenger, Davies, Shahtahmasebi and Scott1996). Such distinctions are not surprising considering the-then dominance of psychological perspectives, focused on objective factors such as cognition, motivation to engage with other people, and other potential psychological barriers such as poor mental health (Bassuk, Glass, & Berkman, Reference Bassuk, Glass and Berkman1999; Elder & Retrum, Reference Elder and Retrum2012; Nicholson, Reference Nicholson2009). Authors have since challenged some of this work, suggesting that subjective feelings accompanying a lack of interaction are also a key component of social isolation among older people (Ackley & Ladwig, Reference Ackley and Ladwig2004; Lien-Gieschen, Reference Lien-Gieschen1993; Nicholson, Reference Nicholson2009).

Over time, the definition and uses of social isolation shifted to be more inclusive of the subjective dimensions of older people’s experiences. Approaches from the early 2000s onward began to move away from defining social isolation as entirely objective or subjective, and towards the development of multi-pronged definitions that included objective and subjective dimensions. This meant that definitions began to incorporate feelings and subjective measures of “aloneness” as a means to provide balance to objective indicators and outcomes, and to recognize that an objective measure of social contact may not always be indicative of isolation (Pettigrew, Donovan, Boldy, & Newton, Reference Pettigrew, Donovan, Boldy and Newton2014). Nicholson (Reference Nicholson2009), for example, outlined a useful definition that draws together objective and subjective components into five key attributes: (1) number of contacts; (2) belonging; (3) inadequate relationships (non-fulfilling); (4) engagement; (5) quality of network members. Although extremely useful in building a model that blends subjective and objective dimensions, this conceptualization misses the experience of social isolation as a social and cultural phenomenon that occurs across time, is structured in particular ways, and is shared and/or collective.

Considering the relationship between policy and practice, definitions that are individually oriented are likely to be matched with individual-level interventions. What this means for current ideas and practices is that the tendency to focus on individual and objective measures or factors of social isolation tends to overlook the social and cultural nature of social isolation, and in doing so, misses both the processes that produce isolation, and the connections with poverty, inequality, and exclusion. Although overlooking marginal groups is unintended, especially when considered within population health models characterized by discourses of prevention, participation, and inclusion, the way of operating in relation to social isolation configures isolation as an individual problem with largely individual causes. Part of the issue here is that the understandings and responses to social isolation are located in a context characterized by increasing individualism, fragmentation of community life, and individualized risks (see Bauman, Reference Bauman2000; Beck, Reference Beck1992; Giddens, Reference Giddens1990). Such interpretations overlook the social and cultural context within which isolation operates and is understood, and the social processes that give rise to social isolation and the experiences thereof. As such, current approaches to social isolation risk concealing trajectories of inequality – especially those that occur in relation to age and disadvantage – and the problems of participation and access that may underpin social isolation and exclusion. The problem of taking an individualized approach to social isolation – in particular, as it applies to marginalized or disadvantaged groups – is that older people may be blamed for their situations, and for failing to adequately integrate into their communities. The individualization of social isolation thereby reinforces exclusion and obscures the shared negative community-level outcomes. From this point of view, it becomes clear that a more social reading on social isolation that includes a macro or community-oriented perspective is needed if the root causes of isolation among older people are to be addressed. To do this, research must reach into more detailed understandings of social isolation as socio-cultural, political, economic, and spatial. This includes taking account of trajectories into social isolation, and how experiences may differ across contexts, settings, and social locations in late life.

Suggestions for Change – Expanding the Conversation

We suggest that existing understandings and policy/practice approaches to social isolation would benefit from a sociological and cultural reading of social isolation, and an expansion to include social and structural dimensions that are apparent in the field of aging and gerontology. Although a larger analysis of social isolation among older people within the context of rising individualism (see Bauman, Reference Bauman2000; Beck, Reference Beck1992; Giddens, Reference Giddens1990) and declining community opportunities and supports (see, e.g., Bellah, Madsen, Sullivan, Swidler, & Tipton, Reference Bellah, Madsen, Sullivan, Swidler and Tipton2007; Putnam, Reference Putnam2001) is not possible in this research note, a number of directions can be taken in the short term. Specifically, the current individualized definitions and responses to social isolation among older people would benefit from the inclusion of three dimensions, including time/duration, place/space, and inequality/exclusion.

The first dimension is duration and time. Social isolation is an experience with many temporal components, and variations across individuals and groups. Timing may be especially significant with respect to social isolation in later life. In some cases, the onset of social isolation and loneliness in late life may occur in alignment with a major transitional event, such as following the loss of a spouse or friend (Beach & Bamford, Reference Beach and Bamford2014). For others, the experience of social isolation may be more reflective of a lifelong pattern of few meaningful social supports, illness, or mental health issues (Machielse, Reference Machielse, Hortulanus, Machielse and Meeuwesen2006, Part 1, Chapter 2). Although such circumstances are occasionally listed as being risk factors for social isolation, researchers could benefit from taking a closer look at age relations (see Calasanti, Reference Calasanti, Biggs, Lowenstein and Hendricks2003), the trajectories of inequality across the life course, and the temporal variations in onset. Both the timing of onset and the duration of the experience are important temporal factors. Buffel and co-authors (Buffel et al., Reference Buffel, Rémillard-Boilard and Phillipson2015) distinguished between situational – a temporary reduction in the size of a social network – and chronic social isolation, a longer term of social isolation. Likewise, Grenier, Sussman, Barken, Bourgeois-Guérin, and Rothwell’s (2016) research on homelessness has demonstrated patterns and trajectories of inequality that occur across time. These types of temporal distinctions and life course trajectories could have potentially noteworthy implications for prevention and intervention efforts, especially where the intersection of age and time are concerned.

Second, place and space feature prominently in older people’s experiences of social isolation. Within gerontology, spatial perspectives have explored the intersection of aging and place at micro (e.g., how an individual navigates a space as they age), meso (e.g., the age-friendliness of a community), and macro levels (e.g., the impact of globalization on population aging) (Wahl & Oswald, Reference Wahl, Oswald, Bengston and Settersten2016). Recently, spatial analysis has gained considerable momentum as a result of the global age-friendly cities movement (see World Health Organization, 2007) and is useful in rethinking experiences of isolation. For example, spatial perspectives can highlight differences between older people who live in the community and in institutional settings such as long-term care. Community-dwelling older people may experience an overall greater risk as a result of living alone (Kobayashi et al., Reference Kobayashi, Cloutier-Fisher and Roth2009), whereas living in a residential care facility may further isolate older people, geographically and socially, from their families (Cannuscio, Block, & Kawachi, Reference Cannuscio, Block and Kawachi2003).

A further issue with regards to place and space is the notion of rural versus urban isolation (Keating, Swindle, & Fletcher, Reference Keating, Swindle and Fletcher2011; Scharf & Bartlam, Reference Scharf, Bartlam and Keating2008). Rurality, or living in a rural region, is sometimes considered a risk factor for social isolation, whereas living in an urban region is sometimes framed as being mutually exclusive with social isolation (Locher et al., Reference Locher, Ritchie, Roth, Baker, Bodner and Allman2005; Tomaka et al., Reference Tomaka, Thompson and Palacios2006). In other words, it is sometimes falsely believed that a person cannot experience social isolation if they are living within close proximity to other people. Although rurality or living in institutional care brings unique challenges that may not be applicable across contexts (e.g., living at home in an urban setting), many older people residing in urban regions do in fact experience social isolation. Indeed, living in a disadvantaged urban neighbourhood may be a significant risk factor for isolation and loneliness among older people despite the fact that they may be surrounded by many people (Scharf, Phillipson, Smith, & Kingston, Reference Scharf, Phillipson, Smith and Kingston2002). Likewise, living in an urban environment that is inaccessible and/or excludes older people (i.e., age unfriendly) is an important risk factor for isolation among older people.

Third, considerations of inequality and exclusion have much to offer understandings and approaches to social isolation in Canada and abroad. As mentioned, research and programming have taken a highly individualized response to social isolation, often failing to address the social processes that lead to the isolation of older people. A wider lens, however, would begin to account for broader social and cultural shifts towards individualization, as well as for how these produce and sustain exclusion and inequality across the life course and into late life. Generally speaking, social exclusion occurs when individuals are disconnected or detached from “mainstream society” (Walsh, Scharf, & Keating, Reference Walsh, Scharf and Keating2016). Accordingly, social isolation in later life can be seen as a by-product of structural developments and/or accumulated disadvantage or inequalities (Machielse, Reference Machielse, Hortulanus, Machielse and Meeuwesen2006, Part 1, chapter 2). A closer analysis of age, age relations, and time – especially time spent in disadvantage – deepens insight on social isolation and exclusion in later life. Indeed, the fact that disadvantaged older people are more likely to become socially isolated suggests a connection with marginalization, inequality, and social exclusion. This can be seen in examples whereby particular groups are overrepresented as socially isolated or excluded. For example, research has highlighted how older people who identify as LGBTQ+ are overrepresented among those who are isolated in later life (Addis, Davies, Greene, & Macbride-Stewart, Reference Addis, Davies, Greene and Macbride-Stewart2009; Guasp, Reference Guasp2011), and how older people from minority ethnic and language groups who are excluded from services and resources experience a heightened risk of social isolation (Jopling, Reference Jopling2015). Building on this connection between isolation and exclusion provides valuable insights into social relations that produce and sustain isolation and exclusion among older people. This lens of inequality and exclusion shifts the focus from addressing risk factors at the individual level (e.g., living alone), to understanding the social processes (e.g., social and structural causes, age relations) and risk factors (e.g., disadvantaged neighbourhoods) that exist at the macro, population, and group level.

Conclusion – An Extended Approach to Social Isolation

Many important aspects should be considered in moving forward with an extended conversation that recognizes societal transitions, structured and shared aspects of social isolation, and disadvantage over time. In the short term, we suggest broadening existing approaches to include three aspects that could be integrated into policy and practice: First, social isolation is a temporal experience whereby experiences of isolation or risk of isolation vary greatly in onset and duration over a lifetime. Overlooking the temporal and situated life course aspects of the experience fails to recognize the heterogeneity that exists across the lives of diverse groups of older people, and how responses may need to be better suited – or timed – according to key transitional moment, onset, and/or duration. Second, geographical, spatial, and place-based factors can greatly affect social isolation. Living conditions, neighbourhoods, institutions, and other spatial factors, such as accessibility and perceived safety, should also be accounted for when designing research and policy interventions. Third, conversations must begin to take into account the relationship between social isolation, inequality, and exclusion. By failing to recognize that social isolation is closely linked with processes of exclusion, and may be connected with patterns of inequality across the life course, we conceal the social conditions and root causes that underpin the problem.

We suggest that conversations be broadened to include larger social trends, and to link current understandings with our proposed dimensions of social isolation in order to respond more fully to the needs of a diverse range of older people, including vulnerable or marginalized groups. Such a broadening would contribute to a more nuanced definition of social isolation, provide context to better assess the prevalence and differences within experiences of social isolation among older people, and broaden the scope of research and planning to ensure that risk factors and outcomes at both individual and collective level are addressed. At the same time, we note the importance that any program or response must move forward in a way that does not stigmatize or blame those who have been isolated or excluded.