Exercise and anxiety: what the review of 79,551 participants found

How much evidence supports exercise as an intervention for anxiety? To answer this question, Neil Richard Munro and colleagues collected systematic reviews that had already combined numerous trials. Their study was published in the British Journal of Sports Medicine in 2026.

What kind of study was it?

This was an umbrella review with a meta-meta-analysis. It included 63 reviews contributing 81 meta-analyses, 1,079 studies and 79,551 participants. The dataset contained 800 studies of depression with 57,930 participants and 258 studies of anxiety with 19,368 participants.

A large dataset does not remove problems in the original studies. An umbrella review combines published results and inherits differences in diagnoses, scales, exercise programmes, comparison groups and risk of bias.

What did it find?

In the main analysis, exercise was associated with a reduction of 0.61 standard deviations in depressive symptoms and 0.47 standard deviations in anxiety symptoms. These figures express an average difference between groups measured on different scales; they do not mean that every participant improved by 61% or 47%.

The authors found effects across different ages and populations. Subgroup analyses suggested a larger effect for aerobic exercise and, for anxiety, for lower-intensity interventions lasting up to eight weeks. These results do not establish one optimal prescription for everybody. Subgroup comparisons combine different programmes and samples and are less certain than the overall result.

What does it not show?

The review did not conduct trials comparing exercise directly with psychotherapy or medication. Its findings therefore do not establish that exercise is better than those treatments. Researcher Brendon Stubbs also noted that many included reviews were rated as low or critically low in quality.

Nor is physical activity necessarily free, accessible or almost free of adverse effects for everyone. Disability, physical illness, pain, available time, environment and cost may limit a person’s options. Activity also needs adaptation where there is an injury risk or medical condition.

How can the result be applied?

The study supports physical activity as one option within care. It does not require a particular discipline or high intensity. Walking, swimming, cycling, dancing, strength work or adapted activity may be suitable, depending on health and preference.

When anxiety is severe, causes panic attacks, prevents necessary activities or occurs with depression, exercise is not a substitute for assessment. It may be used alongside psychotherapy, medication or other measures recommended for that person.