We already know that keeping active is good for our heart, muscles and general health.

But could the type of physical activity we do also make a difference to our risk of dementia?

New research published in The Lancet Public Health has taken a closer look at this question — with some surprising results.

Researchers brought together evidence from 74 studies involving more than 4.2 million people and found that people with high overall levels of physical activity had a 20% lower risk of developing dementia compared with those with low levels of activity.

But when they looked at where that physical activity came from, they found something particularly interesting.

Exercise and activity performed during our leisure time were associated with a lower risk of dementia, while physical activity at work did not show the same relationship.

So could going for a walk, cycling, gardening or dancing really be different from spending all day on your feet at work?

Let’s look at what the researchers found.

What did the new study find?

The study was a systematic review and meta-analysis published in The Lancet Public Health in 2026.

Rather than carrying out a single experiment, researchers brought together evidence from 74 previous studies involving 4,227,297 people.

The participants had a median age of just over 67, and the studies followed people for a median of around 10 years.

When people with high levels of overall physical activity were compared with those with low levels, higher activity was associated with:

• 20% lower risk of all-cause dementia

• 21% lower risk of Alzheimer’s disease

• 29% lower risk of vascular dementia

Those are substantial differences.

But they don’t mean exercise has been proven to reduce an individual’s dementia risk by these exact percentages.

Most of the evidence was observational. Researchers observed people’s activity levels and what happened to their health over time rather than randomly assigning people to exercise or not exercise.

So the results show an association between greater physical activity and lower dementia risk, rather than proving that physical activity alone caused the difference.

But does the type of activity matter?

This is where the study becomes particularly interesting.

Physical activity doesn’t just mean deliberate exercise. We can be active during our leisure time, at work, while travelling or commuting, and around the home.

When the researchers examined these different types of activity separately, they found very different associations with dementia risk:

Leisure-time activity: around 24% lower risk

Household activity: around 15% lower risk

Occupational activity: around 20% higher observed risk

Commuting activity: around 10% higher observed risk

However, these figures aren’t all supported by equally strong evidence. In particular, the household finding came from only one study, so considerably more research is needed before firm conclusions can be drawn from it.

Leisure-time activity stood out

Of the different types, leisure-time physical activity produced one of the clearest favourable associations.

People with high levels of leisure-time activity had around a 24% lower risk of all-cause dementia compared with those with low levels. Leisure-time activity can take many forms, including:

• brisk walking

• cycling

• swimming

• dancing

• gardening

• exercise classes

• sport

• jogging

You certainly don’t have to be running marathons.

Current NHS advice recommends that adults aged 65 and over aim for at least 150 minutes of moderate-intensity activity each week, or 75 minutes of vigorous activity if already active.

The NHS also recommends activities that improve strength, balance and flexibility on at least two days a week, and being physically active every day — even if some of that activity is light.

What about being active at work?

Here’s where the findings become surprising.

People with high levels of occupational physical activity had around a 20% higher observed risk of dementia compared with those reporting low levels.

Before concluding that physical work is somehow bad for the brain, however, there are some important reasons to be cautious.

People with physically demanding jobs may differ from those with less physical jobs in many other ways, including education, income, working hours, stress, diet, environmental exposures and opportunities for recreational exercise.

So the study does not show that having a physically demanding job causes dementia.

Instead, it raises an intriguing question: why doesn’t physical activity at work appear to have the same relationship with dementia risk as activity we choose to do in our leisure time?

Researchers sometimes refer to a similar phenomenon as the physical activity paradox.

What about walking or cycling to get somewhere?

The researchers also looked at physical activity used for transport or commuting.

Higher levels of commuting-related physical activity were associated with around a 10% higher observed risk of dementia.

Again, this does not mean that walking or cycling somewhere causes dementia.

Commuting habits are closely connected with where people live, occupation, income, education, access to transport and many other factors. The finding therefore adds to the puzzle rather than providing evidence that active travel is harmful.

Why might leisure-time exercise be different?

We don’t yet know for certain.

One possibility is that leisure-time exercise often happens in relatively short periods followed by recovery.

A brisk walk, bike ride or swim can raise the heart rate and challenge the cardiovascular system before allowing the body time to recover.

Physical work can be very different. Someone may spend many hours standing, lifting, carrying or performing repetitive tasks without necessarily raising their heart rate in the same way as purposeful aerobic exercise.

Occupational activity can also involve prolonged physical work without the same opportunities for recovery.

There may also be differences in the intensity and type of movement involved.

These are possible explanations rather than conclusions from the study, however, and more research is needed to understand why the associations appear to differ.

Why might exercise be good for the brain?

Our brains don’t exist separately from the rest of our bodies.

Many of the things that help protect our cardiovascular health are also associated with brain health.

Regular physical activity can help maintain cardiovascular fitness and is associated with healthier blood pressure, blood sugar regulation and body weight.

That matters because conditions including high blood pressure, diabetes and cardiovascular disease are themselves associated with dementia risk.

Research also suggests that regular exercise can influence blood flow to the brain and biological processes involved in maintaining healthy brain tissue, although exactly how these mechanisms translate into dementia prevention is still being investigated.

The connection is sufficiently well established that both the NHS and World Health Organization include physical activity within their advice on reducing the risk of dementia and cognitive decline.

New WHO guidance puts brain health in the spotlight

The World Health Organization published the second edition of its guidelines on reducing the risk of cognitive decline and dementia in July 2026.

The updated guidance takes a life-course approach to brain health and considers a wide range of potentially modifiable risk factors.

These include physical inactivity, smoking, harmful alcohol consumption, diet, obesity, diabetes, high blood pressure, high cholesterol, hearing loss, depression, stroke, traumatic brain injury, vision impairment, sleep and air pollution.

The important point is that dementia risk isn’t determined by one single behaviour. Age and genetics play an important role, alongside our health, lifestyle and environment.

There is also no guaranteed way to prevent dementia. But there are things we can do throughout life that are associated with better general and brain health — and staying physically active is one of them.

You don’t have to become an athlete

Perhaps the most useful message from all this research is that looking after our health doesn’t require becoming a marathon runner.

The NHS recommends regular moderate activity such as brisk walking, cycling, dancing or swimming, alongside activities that help maintain strength as we get older.

And if you currently do very little exercise, simply becoming a little more active can be a good place to start.

So, does the type of exercise matter? We can’t say for certain yet. This research shows an association rather than proof, and the evidence for the individual types of activity remains limited.

But while nobody can promise that a daily walk, swim or bike ride will prevent dementia, there are already plenty of good reasons to keep moving.

And this latest research raises an intriguing possibility: when it comes to our long-term brain health, how we get our physical activity may matter as well as how much we do.