The one study that actually put loneliness and smoking side by side, in the same people, found smoking the stronger predictor of death — not an equal one. But that doesn't mean the "loneliness is as bad as smoking" idea came from nowhere: two earlier, much larger analyses found that weak social relationships raise mortality risk by a margin researchers themselves describe as comparable to other major, well-established risk factors for death.
In brief:
- A 2021 study that measured loneliness, social isolation, and smoking directly in the same two large cohorts found smoking predicted death more strongly than either social measure.
- Two earlier meta-analyses found loneliness and social isolation each raise mortality risk by roughly a quarter to a third — a magnitude the researchers call comparable to other well-established risk factors, without naming smoking specifically.
- The "as deadly as smoking" framing is a popularization of that comparable-magnitude finding, not a literal result either underlying paper reported.
Where does the "loneliness is as bad as smoking" idea come from?
It traces back to a pair of meta-analyses by psychologist Julianne Holt-Lunstad and colleagues. The first, in 2010, pooled 148 prospective studies covering 308,849 people and found that having stronger social relationships was associated with a 50% higher likelihood of survival over the follow-up period — an effect that held up across age, sex, health status, cause of death, and how long each study followed people.¹ A second analysis, in 2015, focused specifically on loneliness and social isolation and found that each raised mortality risk by a comparable amount: about 26% for loneliness, 29% for objective social isolation, and 32% for living alone, after controlling for other factors.² Both papers reached the same conclusion in almost the same words: the effect of poor social connection on mortality risk is "comparable with well-established risk factors for mortality." Neither paper named smoking specifically — that comparison became widely repeated shorthand in health reporting and public health messaging afterward, not a finding either paper stated directly.
Did anyone actually test loneliness against smoking directly?
Yes, in 2021. Researchers wanted to check whether the popularized comparison held up when loneliness, social isolation, and smoking were measured in the same people and compared head-to-head, rather than pulled from separate reviews of different study populations.³ They used two large, independent datasets: UK Biobank, with 466,876 participants followed for a mean of 6.6 years, and the English Longitudinal Study of Ageing, with 7,505 older adults followed for a mean of 7.7 years. In both datasets, for death from any cause and from cancer specifically, the order of association strength was consistent: loneliness came first (weakest), then social isolation, then cigarette smoking (strongest). For cardiovascular death specifically, smoking and social isolation showed similar associations, though loneliness did not track as closely. The researchers' own conclusion was direct: contrary to the cross-review comparison that had circulated, poor social integration turned out to be less strongly linked to total mortality than cigarette smoking — not equally linked.
What the research says
Holt-Lunstad and colleagues, pooling 148 studies and 308,849 participants, found stronger social relationships associated with a 50% higher likelihood of survival (OR 1.50, 95% CI 1.42–1.59), an association that was strongest for complex measures of social integration and weakest for simply living alone versus with others.¹ A follow-up analysis from the same group, isolating loneliness and social isolation specifically, found each raised mortality risk on its own: about 26% for loneliness and 29% for social isolation, with living alone at 32%.² Separately, Batty and colleagues directly compared loneliness, social isolation, and smoking in two large cohorts (UK Biobank, n=466,876; the English Longitudinal Study of Ageing, n=7,505) and found smoking the strongest predictor of all-cause and cancer mortality in both, with loneliness the weakest of the three.³
The limits matter here. All three studies are observational: they can show that people with weaker social connection tend to die sooner, not that loneliness itself causes that outcome rather than, for example, pre-existing illness making both connection and survival harder. The 2010 and 2015 meta-analyses also pool many different studies that measured "loneliness" or "social isolation" in different ways, which is part of why the exact size of the effect varies from study to study. And the 2021 direct comparison, while more rigorous on the head-to-head question, still relies on a single self-reported measure of loneliness and smoking status taken at one point in time, not a lifetime history of either.
What you can do this week
- Audit actual contact, not just relationship status. The studies above measured how connected people were — frequency of contact, breadth of social ties — not simply whether someone was married, single, or living alone; that distinction is where living-alone status showed the weakest association of the three measures.
- Bring it up at a checkup, not just in conversation. Loneliness and social isolation are measurable with short screening questions, the same way blood pressure or weight are — worth raising with a clinician directly rather than treating as something to privately push through.
- Weigh depth over breadth if time is limited. The strongest associations in the research above came from complex measures of social integration, not simply whether you live with someone — a few relationships with real contact outperformed a checkbox about living arrangement.
The headline comparison to smoking doesn't hold up under a direct test. What does hold up, across every study above, is that social connection measurably predicts how long people live — by a margin researchers themselves put in the same range as other health risks worth taking seriously.
