Movement

I’ve often said to patients that if I could recommend only one thing for their brain, it would be exercise. Nothing else on the list does as much, in as many ways, for as little cost.

What movement does to the brain

Exercise acts on the brain by several routes at once, which is much of why it works so well.

Two of those routes are the subjects of the last two articles. Regular aerobic activity lowers blood pressure, improves vascular function and increases cerebral blood flow, protecting the small vessels that supply the deep white matter. It also improves insulin sensitivity more reliably than anything else available without a prescription.

The third route is more direct. Exercise increases production of brain-derived neurotrophic factor, a growth factor that supports the survival of existing neurons, the birth of new ones in the hippocampus, and the remodelling of synaptic connections. Higher fitness is consistently associated with larger hippocampal volume, and randomised aerobic training trials in older adults have shown increases in hippocampal and grey matter volume over months rather than years.

The important point here is that as we age, the hippocampus shrinks, and shrinks faster in Alzheimer’s disease. In these trials the hippocampus grew.

The evidence

At population level the association is about as consistent as anything in this field. People who are more physically active develop dementia at lower rates, across large cohorts, in many countries, over decades of follow-up. Physical inactivity has held its place on the Lancet Commission’s list through every revision.

The trial evidence supports it. Meta-analyses of exercise trials in cognitively healthy older adults show real cognitive benefits, with the effect driven mainly by aerobic training. And the strongest prevention trial we have — FINGER, which combined exercise with diet, cognitive training and vascular risk management over two years in at-risk older adults — showed measurably better cognitive trajectories, sustained at long-term follow-up.

Start early enough for it to matter

Exercise trials in people who already have Alzheimer’s disease have been largely disappointing. A sixteen-week aerobic programme in mild-to-moderate disease produced no change in hippocampal volume or cortical thickness. A six-month cycling trial found effects that were very small and that reversed once the intervention stopped.

That pattern should feel familiar by now. It is the same shape as the anti-amyloid drugs and the same shape as semaglutide, and it says the same thing: intervening after the disease is established achieves very little. It is not an argument against exercise. It is an argument for prevention — for starting in the decades before the damage is already done.

Why this is where I tell people to start

Beyond the direct evidence, exercise has a property that nothing else on this list has. It improves almost every other risk factor at the same time.

It lowers blood pressure. It improves insulin sensitivity and reduces visceral fat. It improves sleep quality and depressive symptoms. Each of those is an independently established dementia risk factor with its own evidence base. A patient who starts exercising is not pulling one lever. They are pulling five.

And the risk-benefit calculation is unlike anything in pharmacology. It costs nothing, the side effects are largely musculoskeletal and manageable, and the benefits outside the brain — cardiovascular, metabolic, skeletal, functional — are not in dispute by anyone.

What kind, and how much

The evidence is strongest for aerobic exercise of moderate intensity, sustained over months. Most of the trials that produced structural brain change used something in the range of three sessions a week, thirty to fifty minutes, at an intensity where conversation is possible but a little effortful. I have written separately about zone 2 training, which describes the same intensity more precisely.

There is a case for pushing harder occasionally. The release of brain-derived neurotrophic factor after a session is intensity-dependent, and interval work produces a larger response than steady moderate effort, with lactate appearing to be part of the trigger. Whether that translates into better long-term cognitive outcomes is not established — the trials that grew hippocampal volume used moderate continuous training, and in older adults the two approaches have produced similar results. But it is a reasonable argument for including one or two harder efforts a week alongside the steady work, rather than doing everything at an easy pace.

Resistance training has been less studied for cognitive outcomes. That is not an argument against it — muscle mass, bone density and functional independence matter enormously for how the last decades of life go, and the case for it stands on its own.

The dose-response relationship suggests most of the benefit arrives early. Going from no exercise to some exercise, even if modest, has a much larger effect than going from modest exercise to intense exercise.

Practical steps for this week

1. Establish the baseline honestly. Not what you intend to do, but what you actually did last week. Most phones already track this. If the number is small, that is useful information rather than a reason for self-reproach.
2. Aim for three aerobic sessions a week, thirty minutes each. Brisk walking counts if it is brisk enough that talking is a little effortful. This is roughly the dose used in the trials that showed structural brain change. If you can, make one of those sessions harder — some intervals, or a hill.
3. Add resistance work twice a week if you can. Body weight is enough to begin with. The cognitive evidence is thinner, but the case for preserving muscle into later life does not depend on it.
4. Choose something you will still be doing in a year. Consistency matters more than intensity. An activity you tolerate beats an optimal programme you abandon in March.

References

1. Erickson KI et al. Exercise training increases size of hippocampus and improves memory. PNAS. 2011.
2. Northey JM et al. Exercise interventions for cognitive function in adults older than 50: a systematic review with meta-analysis. British Journal of Sports Medicine. 2018.
3. Ngandu T et al. FINGER: a multidomain intervention to prevent cognitive decline. The Lancet. 2015.