Volunteering Actually Helped Lonely Older Adults. Here's the Trial That Tested It

Most advice about loneliness rests on something that sounds obvious but is rarely tested properly: do things with other people, and you’ll feel less alone. A trial published in The Lancet Healthy Longevity in November actually tested it, on a specific activity, with a control group. That matters more than it sounds like it should.

Why this trial is unusual

Almost everything written about loneliness interventions — join a club, volunteer, get out more — comes from studies that watched what happened to people who did the thing, without comparing them to people who didn’t. If lonely people who start volunteering feel better a year later, it’s tempting to credit the volunteering. But maybe they were also getting less depressed, or their health improved, or they moved somewhere with more community life, and the volunteering had nothing to do with it.

A randomised controlled trial removes that ambiguity, at least partly, by comparing people who were assigned to do something with people who were assigned to a control condition, so the only reliable difference between the groups is the intervention. The HEAL-HOA trial did exactly that. It recruited lonely older adults in Hong Kong, and tested whether structured volunteering and prosocial engagement reduced loneliness compared with a control group that didn’t get the intervention.

This kind of study is rare in the loneliness literature. The National Academies noted in 2020 that most existing interventions for isolated older adults have been evaluated with small, uncontrolled designs, which makes it hard to say what actually works versus what merely correlates with feeling better. HEAL-HOA is one of the exceptions.

What it found

The trial’s headline result: prosocial engagement and volunteering reduced loneliness among the older adults assigned to it, compared with the control group. That’s a real effect, measured against a comparison group, not just a before-and-after story.

It’s worth being precise about what this doesn’t tell us. This was one trial, in one place, with older adults specifically. It doesn’t prove volunteering works for a 32-year-old who just moved cities, or that it works everywhere, or that the effect lasts for years. What it does establish is that a specific kind of structured, purposeful social activity can move the needle on loneliness in a controlled comparison — not just a plausible story people tell about themselves after the fact.

That’s a meaningfully different claim than “volunteering is good for you,” which everyone already believed and which turns out to be much harder to demonstrate than it sounds.

Why volunteering, specifically, might work

There’s a reasonable theory for why this particular activity would outperform something looser, like “meet new people” or “get out of the house more.” A qualitative review of social prescribing programmes — where doctors refer patients to community activities rather than medication — found that participants described benefits going beyond simple contact with other humans. What actually helped was restored meaningful participation and purpose. People didn’t just want company. They wanted to matter to something.

Volunteering has that built in. You show up, you’re needed, there’s a task, and the task doesn’t depend on you being charming or making small talk land. That review’s broader finding was that structured, purposeful group activity seems to work better than contact alone — being in a room with people isn’t the same as being useful alongside them.

This lines up with something people who study loneliness keep circling back to: loneliness isn’t just about the number of people around you. It’s about feeling seen, needed, and connected to something outside yourself. A volunteering shift gives you all three without requiring you to generate them socially from scratch, which is exactly the part that feels hardest when you’re lonely.

The gap this fills

The American Heart Association’s 2022 scientific statement on social isolation made an unusually blunt admission for a scientific body: the biggest problem in this field isn’t understanding that isolation and loneliness harm health. Julianne Holt-Lunstad’s 2015 meta-analysis had already put numbers on that — loneliness and isolation both raise the odds of early death, with effects that hold up even after adjusting for existing health conditions. The problem is that almost nobody has rigorously tested what actually reverses it.

That’s the gap HEAL-HOA is chipping away at. It’s a small trial, in a specific population, testing one intervention. But it’s evidence, not intuition, and the field badly needs more of exactly this kind.

What this means for an actual Tuesday

If you’re lonely and looking for something concrete to do, this is a reasonable steer: look for volunteering that has a defined role, not a vague “meet people” event. The trial’s design — structured, purposeful, recurring — is the part worth copying, more than the specific activity.

That could be:

A regular shift at a food bank, where you’re assigned a task and a time. A library programme that needs someone to run a weekly reading session. A local mutual aid group that needs someone to make deliveries on a fixed schedule. The common thread isn’t the cause you care about. It’s that someone is expecting you, at a set time, to do a specific thing, and that thing is useful whether or not you feel like talking that day.

This matters for a reason that has nothing to do with virtue. Unstructured socialising requires you to generate the energy and the topic and the invitation yourself, which is exactly the thing that’s hardest to do when you’re already isolated. A volunteering shift removes that decision. You don’t have to be interesting. You have to show up and do the thing, and the social contact happens as a side effect of being useful, which is a much lower bar than most people assume “getting out there” requires.

Where the theory could be wrong

It would be a mistake to read this trial as proof that volunteering is a cure for loneliness in general. It tested older adults in Hong Kong who were recruited specifically because they were lonely, using a structured programme designed for the trial. Whether the same effect shows up for a lonely 28-year-old working from home, or a newly divorced 50-year-old in a city where they know almost no one, hasn’t been tested the same way.

There’s also a question the trial doesn’t answer: how much of the effect comes from volunteering specifically, versus from any structured, recurring, purposeful activity with the same shape. It’s plausible that a weekly choir rehearsal or a community garden shift would do something similar, because the mechanism — purpose, structure, recurrence — isn’t unique to charitable work. The trial tested volunteering because that’s what it tested, not because volunteering has some special property nothing else shares.

That uncertainty is honest, and it’s also freeing. If the mechanism is structure and purpose rather than the specific act of volunteering, then the useful move isn’t to feel obligated to find a cause you care about. It’s to find any recurring commitment where you’re expected, where there’s a task, and where showing up doesn’t depend on your mood that day.

The smallest version of this

Don’t look for the perfect volunteering opportunity that matches your values and your schedule and feels meaningful on day one. Look for the nearest one that repeats weekly and has a clear task attached to it. A shelter that needs someone Thursday mornings. A school that needs a reading helper on Tuesdays. The specificity of the task is doing more work than the nobility of the cause. Show up four times before deciding whether it’s working — one trial’s worth of evidence says that’s a better bet than most of the alternatives you’ve already tried.