Social Isolation, Loneliness, and Living Alone: Identifying the Risks for Public Health

Follow-up on: Pantell M, Rehkopf D, Jutte D, Syme SL, Balmes J, Adler N. Social isolation: a predictor of mortality comparable to traditional clinical risk factors. Am J Public Health. 2013;103(11):2056–2062.

The extraordinary rise of living alone is among the most significant social changes of the modern world. Consider that, until the middle of the 20th century, not a single society in the history of our species sustained large numbers of people living alone for long periods of time. Today, however, living alone is ubiquitous in developed, open societies, with one-person households accounting for more than 40% of all households in Scandinavian nations such as Sweden and Finland; more than one third of all households in France, Germany, and England; and more than one quarter of all households in the United States, Russia, Canada, Spain, and Japan.¹ There’s good reason to believe that this spike in living alone, and particularly aging alone, affects health and health care. But we don’t yet have enough research to understand exactly how.

One possibility is that there is a causal link between living alone, being socially isolated, and feeling lonely. But these are three distinct conditions, and experiencing one (living alone) does not necessarily mean experiencing one or both of the others (being isolated or feeling lonely). For example, when I interviewed more than 300 people for my book, Going Solo, many told me that nothing had made them lonelier than being in a bad marriage. Moreover, survey data show that, on average, Americans who live alone spend more time with friends and neighbors and volunteer in civic organizations more often than married people.¹ Unfortunately, journalists, scholars, and health care providers often conflate living alone, feeling lonely, and being isolated, and the result is widespread confusion about each condition.

The article by Pantell et al. published three years ago in AJPH made a significant contribution to our knowledge of the risks of social isolation, and added clarity to an important public health debate.² The authors carefully specified their measure of isolation: the Social Network Index from the landmark Berkman and Syme study of social networks and health. Unlike previous studies, which drew on local samples, this research used data from nationally representative samples, the Third National Health and Nutrition Examination Survey and the National Death Index. The key findings—that social isolation was a predictor of mortality on par with smoking, obesity, elevated blood pressure, and high cholesterol—were therefore particularly robust. And the policy recommendation, that health care providers assess isolation in their regular patient examinations, has clear potential to save, or extend, lives.

Of course, identifying the risks of social isolation is just the first step of addressing it as a public health problem. After that, we need to know how common isolation is in different places and subpopulations, and what can be done to reduce it. Here, it is especially important that we not conflate living alone, being isolated, and feeling lonely, because each condition requires a particular form of diagnosis and a specific intervention.

Sociologists have a long history of measuring social networks and social isolation, and we know a lot about its prevalence. For decades, research based on the most reliable social surveys has indicated that social isolation is a rare condition. As Fischer and Phillips report, from the 1980s through today, several studies, using a variety of measures for isolation (including number of confidants and close friends, frequency of social interaction, and access to social support), suggest that fewer than one in 20 Americans are socially isolated.³ Recently, however, McPherson et al. reported a dramatic spike in social isolation, such that one in four Americans had no confidant by 2004.⁴ This finding has been widely cited, and the article has been massively influential in the public sphere. But careful scrutiny of the article by Fischer and others has established that the finding about the rise of isolation is an artifact based on flawed data, from which we should draw no inferences.⁵

Social isolation may be less widespread in the general population than many believe, but it is more common among some people and places than others. Older people are at higher risk for isolation because of physical frailty as well as deaths in their family and friendship networks.⁶ American men are more likely to be socially isolated than women.⁷ They have smaller social networks and less stable contact with children and relatives over the life course, and they face heightened risks of isolation if they divorce or become widowers.⁸ Men’s isolation is not universal, however; in other countries, such as Mexico and Spain, women are more likely to lack social contact. Some reports show that lesbian, gay, bisexual, and transgender people face a heightened risk of social isolation, in part because they are less likely to have children and more likely to be estranged from family.

The risks of social isolation depend not only on who you are, but also on where you live. Certain social environments foster social isolation, while others promote local contact and mutual support. When I did ethnographic fieldwork to explore the factors that increased or decreased the risk of death during the Chicago, Illinois, heat wave, I discovered that poor neighborhoods with a robust social infrastructure—residential density, busy sidewalks, commercial activity, well-maintained public spaces, and community organizations—experienced few fatalities, whereas comparably poor areas with a depleted social infrastructure suffered the city’s highest mortality rates.⁸

Since individual, group, and neighborhood conditions determine who is living alone, policies for reducing social isolation should attend to each of these levels. I endorse the recommendation by Pantell et al. that clinicians assess whether their patients are isolated.² But what kinds of assistance can health care providers offer to those who are truly on their own? Here, again, it depends on the condition. For relatively healthy people at risk for isolation, such as widows and widowers, older single men, and older single lesbian, gay, bisexual, and transgender people who live alone, a warning about the danger of isolation and simple encouragement to be socially active may help promote social interaction. Those who suffer from isolation and loneliness are vulnerable to a vicious cycle that leads to social withdrawal, and they would likely benefit from psychological care as well as social activity.

Certain people at risk for isolation need more support. People who are aging alone in impoverished areas with degraded social infrastructure would benefit from neighborhood revitalization, but that would require considerable investment from the public and private sectors, and there is little reason to think either will come through soon. Old, frail, and reclusive people who live alone may require home care and specialized services such as meal delivery or social visits. In these cases, care workers should understand that they likely serve as a vital source of interaction. They should be trained to recognize when an isolated person is in danger and no longer able to live alone, and how to connect that person to appropriate sources of support. Unfortunately, home services are expensive, and poor people living in poor communities often suffer from inadequate attention and support. There are a handful of volunteer-based organizations attempting to fill the care gap (for instance, Little Brothers, Friends of the Elderly), but not enough to substantially reduce isolation at the national or international levels. As the population ages and the number of people aging alone grows, societies throughout the world will need to develop new supportive housing programs and new forms of elderly care. Few nations have recognized this challenge, and fewer still have made real investments in devising solutions.

At this point, most policy recommendations for reducing isolation are speculative only. We lack sound research on the effectiveness of proposed interventions for social isolation, in different contexts and with different populations. It’s time for public health scholars to take on this important challenge.