News · Heart & Metabolic
Loneliness tracked heart and metabolic disease across 33 countries
Pooling six aging cohorts covering 81,978 people, researchers found loneliness preceded cardiometabolic disease, and unhealthy habits explained only 5% of the link.
Based on a peer-reviewed multicohort study in the European Journal of Preventive Cardiology
- Researchers pooled six aging cohorts spanning 33 countries, including studies in China, England, the US, Europe, Korea and Mexico, publishing in the European Journal of Preventive Cardiology.
- Among 81,978 participants aged 45 and over, 15,750 developed cardiometabolic disease over a median five years.
- Loneliness was associated with higher risk of any cardiometabolic disease (aHR, 1.23; 95% CI, 1.20-1.26).
- The association was stronger for multiple conditions (aHR, 1.48; 95% CI, 1.37-1.60) than for a single one (aHR, 1.18; 95% CI, 1.15-1.22).
- Unhealthy lifestyle explained only 5.19% of the association, so behavior is not the main pathway.
- Low socioeconomic status, unhealthy habits and loneliness together carried the highest risk (adjusted HR, 1.89).
- Observational. Early undiagnosed illness can cause social withdrawal, which would produce this pattern in reverse.
- Loneliness was self-reported and its meaning varies across 33 countries and cultures.
That loneliness is bad for the heart has been reported often enough to become background noise, mostly from single cohorts in single countries. A study in the European Journal of Preventive Cardiology pooled six of them across 33 countries and then asked the question that usually goes unasked: through what?
The aim was to identify the association between loneliness and cardiometabolic disease (CMD) across diverse settings, particularly cardiometabolic multimorbidity, and investigate whether unhealthy lifestyles and socioeconomic status (SES) help explain or amplify their relationships.
Six cohorts, one analysis
The breadth is the point. Data were obtained from six cohort studies across 33 countries - the major national aging studies of China, England, the United States, continental Europe, Korea and Mexico.
Those six were deliberately built to be comparable with one another, which is what makes pooling them defensible. Eligible participants were aged 45 years or older.
The scale is substantial: among 81,978 participants, with a median follow-up of five years, 15,750 participants developed CMD. Nearly one in five.
The association
Loneliness came first, disease came later, and the relationship held across very different societies.
Loneliness was associated with higher risk of any CMD, at 1.23 - roughly a quarter higher risk.
The more interesting number is what happened when the researchers looked at people accumulating several conditions rather than one. There was a stronger association for multimorbidity than for single CMD: 1.48 against 1.18.
A dose-like pattern of that kind, where the association strengthens as the outcome gets more severe, is one of the things that makes an observational finding harder to dismiss.
The pathway that was not there
This is the part worth sitting with, because it contradicts the intuitive story.
The obvious explanation for loneliness harming the heart is behavior: lonely people smoke more, drink more, move less, eat worse. Test that and you would expect lifestyle to carry most of the effect.
It carried almost none. Unhealthy lifestyles mediated 5.19% of the association with any CMD.
Five percent. Whatever connects loneliness to cardiometabolic disease in this data, it is overwhelmingly not the fact that lonely people have worse habits.
Where disadvantage compounds
The third question was whether social circumstances make things worse, and they did.
Individuals experiencing a combination of low SES, multiple unhealthy lifestyle factors, and loneliness demonstrated the greatest cardiometabolic disease risk, at 1.89 - the risks stacking rather than simply adding.
What this cannot establish
It is observational, and the most serious alternative explanation is reverse causation. Heart disease and diabetes develop over years before anyone records a diagnosis, and the early phase brings fatigue, breathlessness and reduced activity. People withdraw. Five years of median follow-up is not long enough to rule out that much of the loneliness here was an early symptom rather than a cause.
Loneliness was self-reported, and its meaning does not transfer cleanly across 33 countries. Admitting loneliness carries different weight in Shanghai, Seoul and Sheffield, and questionnaire translation only partly bridges that.
The mediation figure also deserves care. It quantifies how much of the association runs through lifestyle as measured, which in these cohorts means broad self-reported categories rather than precise records. Crude measurement of a mediator will understate its role.
Why it still matters
The authors position it as building on prior evidence that has largely come from single-cohort studies, and their findings highlight loneliness as a global risk factor rather than a local one.
The mediation result is the piece with practical bite. If loneliness harmed health mainly through smoking and inactivity, the response would be to target those. It does not appear to, which means telling lonely people to exercise more addresses about a twentieth of the problem.
That is a harder conclusion, because the remaining 95% has no obvious prescription attached to it.
People also ask
What is cardiometabolic disease?
An umbrella covering heart attack, stroke, heart disease and type 2 diabetes - conditions that share risk factors and often arrive together. Multimorbidity means having more than one. That distinction matters here, because the association with loneliness was considerably stronger for accumulating several conditions than for developing just one.
How can loneliness affect the heart if not through habits?
The mediation analysis found lifestyle explained about 5% of the association, which leaves most of it unexplained. Proposed biological routes include chronic activation of the stress response, raised inflammatory markers, elevated blood pressure and disrupted sleep. None of these is established as the mechanism here. This study measured how much behavior explains, which is a different and more answerable question than what does.
Is loneliness the same as being alone?
No, and the distinction is central to this field. Loneliness is the subjective experience of a gap between the connection you want and the connection you have. Social isolation is the objective count of contacts. People can be isolated without feeling lonely, and lonely in a crowded household. This study measured the felt experience, which is generally the better predictor of health outcomes.
Could the causation run the other way?
Very plausibly, in part. Early undiagnosed heart disease or diabetes causes fatigue, breathlessness and reduced activity, all of which shrink someone's social world before any diagnosis is recorded. With a median follow-up of five years, some of the disease counted at the end was probably developing at the start. That would produce exactly this association without loneliness causing anything.
What can someone actually do about it?
This is general information rather than medical advice, and the honest position is that evidence for loneliness interventions is thinner than evidence that loneliness matters. Approaches with some support include structured group activities built around a shared purpose rather than around socializing itself, and psychological therapy addressing the thought patterns that sustain loneliness. Persistent loneliness is worth raising with a clinician, particularly alongside low mood.
References
- Wang DY, Xie L, Wu Q, et al. Loneliness, unhealthy lifestyle, and incident cardiometabolic disease among middle-aged and older adults across 33 countries: a multicohort study. European Journal of Preventive Cardiology (2026).
- Centers for Disease Control and Prevention. Health Effects of Social Isolation and Loneliness.
- National Institute on Aging. Loneliness and Social Isolation: Tips for Staying Connected.