Reporting on loneliness and how people find their way out of itTools & apps
Be Social.

The Science

What the Heart Association Actually Found About Loneliness

The American Heart Association's 2022 scientific statement put a number on what loneliness does to your heart. Here's what the number means and what it doesn't.


Photograph · Pexels

In August 2022, the American Heart Association published a scientific statement with a specific claim in it: being socially isolated or lonely raises your risk of a heart attack, a stroke, or dying from either by about 30%. It’s worth sitting with that number for a second, because it’s easy to file loneliness under “sad” and much harder to file it under “cardiovascular risk factor.” The statement, led by Crystal W. Cene on behalf of several AHA councils, is doing exactly that second thing.

The specifics: a 29% increased risk of heart attack and death from heart disease, and a 32% increased risk of stroke. If you already have coronary heart disease or you’ve had a stroke, isolation and loneliness predict a worse outcome — including a higher chance of a second stroke.

That’s not a wellness statistic. That’s the kind of number that gets a risk factor added to a doctor’s checklist alongside blood pressure and cholesterol.

Isolation and loneliness are not the same thing, and the difference matters

The statement is careful to separate two things people usually lump together. Social isolation is a fact about your life: how many people you see, how often, how thin your network is. Loneliness is a feeling: the gap between the connection you have and the connection you want. You can be isolated without feeling lonely — plenty of people live alone by choice and are fine. And you can be surrounded by people and still feel lonely, which is its own kind of common and its own kind of miserable.

The AHA statement looks at both, separately, and finds each one carries its own cardiovascular risk. They’re not interchangeable, but they’re not independent either — a person who is objectively isolated is more likely to end up lonely too, and the two seem to compound.

This lines up with a broader pattern in the research. A 2015 meta-analysis in Perspectives on Psychological Science found isolation, loneliness, and living alone all independently predicted earlier death, with the effects holding up even after adjusting for existing health conditions. The heart association’s contribution is to narrow that broad mortality signal down to a specific mechanism: the heart and the brain, and how they seem to respond to disconnection over time.

The honest gap: nobody has proven that fixing it fixes the risk

Here’s the part that a lot of coverage of this statement skipped, and it’s the most important part. The AHA is explicit that the evidence connecting isolation to cardiovascular harm is much stronger than the evidence for what to do about it. They name the absence of intervention evidence as the central gap in the field. In plain terms: researchers are fairly confident isolation is bad for your heart. They do not yet have solid trials showing that a specific programme, app, or habit lowers that risk by getting people less isolated.

That doesn’t mean nothing works. It means the causal chain from “join a walking group” to “lower stroke risk” hasn’t been tested the way a cholesterol drug has been tested. The statement is a description of a problem, backed by strong evidence, attached to a solution that is still mostly a hypothesis.

That’s a useful thing to know, because it should change how you read any confident claim — including ones on this site — that a particular fix will protect your heart. The honest version is: strong connection is associated with better outcomes, isolation is associated with worse ones, and the mechanism plausibly runs through stress hormones, sleep, blood pressure, and inflammation. Nobody has run the trial that nails down exactly how much of the risk reverses if you get less isolated.

Who the statement flags as most exposed

The AHA newsroom summary of the statement calls out older adults and socially vulnerable groups specifically — people who’ve lost a spouse, people who’ve retired out of a work social network, people whose mobility or income limits how easily they get out. This tracks with what the U.S. Surgeon General’s 2023 advisory said more broadly: mortality risk from disconnection is now being compared to the risk from smoking up to 15 cigarettes a day. That’s not a metaphor being used for shock value. It’s the register the AHA and the Surgeon General are both now using, because the size of the effect earns it.

What this actually changes about your evening

You don’t need a cardiology trial to justify calling someone. But this research does argue for treating your social contact like you’d treat a blood pressure reading — something to notice, not just something to feel bad about in the abstract.

Concretely: if you’ve noticed your week doesnn’t reliably include a conversation that isn’t transactional — not “thanks, bye” at the coffee shop, but an actual back-and-forth with someone who knows you — that’s the pattern the AHA is talking about. The fix isn’t a grand life overhaul. It’s usually one recurring thing: a Tuesday call with the same person, a Thursday group that meets whether or not you feel like going.

The reason repetition matters here isn’t just habit-formation cliché. Isolation, as the AHA defines it, is about the structure of your contact over time — not one good conversation, but a pattern. A single great dinner with an old friend doesn’t undo a year of otherwise empty weeks, any more than one salad undoes a year of fast food. What seems to matter is the baseline you return to.

So the useful move isn’t “go be more social.” It’s smaller than that: pick one person, pick one fixed slot in the week, and don’t leave it up to how you feel on the day. The evidence for the harm of isolation is solid. The evidence for what fixes it is thinner, but everything researchers do have points toward the boring, repeatable thing rather than the dramatic one.

Where this comes from