In 1954, Richard Doll and Austin Bradford Hill published a paper in the British Medical Journal tracking the smoking habits and the death records of British doctors. Their finding, that smoking causes lung cancer, is one of the foundations of modern public health. It took decades to become policy. The tobacco industry funded counter-research, argued that correlation was not causation, and framed regulation as an attack on personal freedom.
Eventually the evidence won. Warning labels went on packages. Taxes went up. Advertising was banned. Smoking rates fell. The campaign against tobacco is one of the great public health victories of the twentieth century.
Loneliness is the harder version of that problem. There is no product to regulate, no company to sue, no package to put a warning on. The harm is done by an absence.
In 2023 the U.S. Surgeon General, Vivek Murthy, issued an advisory on loneliness and isolation that put the mortality risk of being socially disconnected on a par with smoking up to 15 cigarettes a day. The advisory ties poor social connection to a 29 percent higher risk of heart disease, a 32 percent higher risk of stroke, and a 50 percent higher risk of dementia in older adults, along with weakened immune function and higher all-cause mortality. Later that year the World Health Organization launched a Commission on Social Connection, saying that a lack of social connection carries a risk of early death equivalent to or greater than smoking, excessive drinking, physical inactivity, obesity and air pollution.
The 15 cigarettes figure rests on work by Julianne Holt-Lunstad, a psychologist at Brigham Young University, whose meta-analyses on social connection and mortality are among the most cited in the field. The number that came out of her 2015 review was a 26 percent increase in the likelihood of death associated with loneliness. The advisory's own wording is narrower than the version that circulates: comparable to smoking up to 15 cigarettes per day. Dropping the qualifier turns a range into a slogan.
In Canada, Statistics Canada found that 49 percent of those who said they always or often felt lonely rated their mental health as fair or poor. Among those who said they rarely or never feel lonely, the figure was 7 percent. The 2024 Benefits Canada Healthcare Survey found that 38 percent of benefits plan members are experiencing a general sense of loneliness, isolation and social disconnection: more than one in three working Canadians with an employer benefits plan.
Remote work, adopted out of necessity during the pandemic, is now a permanent feature of the labour market, and the loneliness data on it is uncomfortable for its champions. Gallup's 2024 State of the Global Workplace report found that fully remote employees report higher levels of loneliness (25 percent) than those who work exclusively on-site (16 percent).
The Canadian Coalition for Seniors' Mental Health published the first Canadian clinical guidelines on social isolation and loneliness in older adults in 2024. Cholesterol, blood pressure and diabetes have had Canadian guidelines for decades. Loneliness got its first set two years ago.
Part of the reason for the lag is that loneliness does not fit the diagnostic model. There is no blood test for it and no imaging scan, so it gets filed as a social problem rather than a medical one. The biology is not in dispute. Chronic loneliness keeps the body's stress response switched on, which raises cortisol, drives inflammation and suppresses immune function.
Social prescribing is one model of a response. It started in the United Kingdom, where it has been part of the National Health Service since 2019. Instead of prescribing only medication, a clinician refers the patient to a link worker, who connects them with something in the community: a walking group, a choir, a community garden, a cooking class. The prescription is literal: it goes on paper.
Kate Mulligan, at the University of Toronto's Dalla Lana School of Public Health, has written about how this is being built in Canada. In a 2023 Healthcare Quarterly paper, she and her co-authors describe social prescribing as a practical tool for addressing the social determinants of health through supported referrals to community services, and introduce the Canadian Institute for Social Prescribing.
The referral is the easy half. Someone still has to run the walking group next month and the month after, and that work sits with community organizations and volunteers rather than with the health system. We know loneliness kills, we know it affects a large share of the population, and we know roughly what the solutions look like. The response has been fragmented, underfunded and treated as a nice-to-have.
The campaign against tobacco ran for decades and moved on several fronts at once: legislation, taxation, advertising bans, public education, reform of how the health system handled smokers. It worked because governments treated smoking as a public health emergency.
Loneliness is a public health emergency of the same order. The Surgeon General has said so, the WHO has said so, and the clinical evidence backs them. Canada's response has been to study it further.
There is a particular cruelty to loneliness as a health crisis. A person who smokes can be identified and offered a cessation program. A person who is lonely looks like everyone else: they go to work, they buy groceries, and from the outside they may appear to have a full social life. Loneliness accumulates silently, the way plaque builds in arteries, until the damage shows up as something else.
An annual physical checks blood pressure, cholesterol and blood sugar. It does not ask whether the patient has spoken to another person this week, or whether anyone would notice if they did not turn up somewhere tomorrow. The CCSMH guidelines recommend asking. Whether the medical system adopts that is another matter.
A cigarette-scale problem calls for a cigarette-scale response. Warning labels, in this case, would be public awareness that treats loneliness as a health risk rather than a personal failing. Taxes would be sustained funding for community infrastructure: third places, gathering spaces, the link workers and local organizations that social prescribing runs on. Advertising bans would be an honest accounting of the social costs of remote work, suburban sprawl and the defunding of community institutions.
We have the research, and the clinical guidelines are newly written and ready to be used. What we do not have is the political will to treat loneliness the way we treated tobacco: as something that kills people.