select search filters
briefings
roundups & rapid reactions
Fiona fox's blog
A meta analysis published in the British Journal of Sports Medicine looks at the effect of exercise on depression and anxiety symptoms.
Dr Brendon Stubbs, NIHR Clinical Lecturer, King’s College London (KCL), said:
“This umbrella review confirms what we know from a very large and consistent body of evidence showing that almost all forms of physical activity and structured exercise are associated with meaningful improvements in depression and anxiety symptoms, across the lifespan and in both clinical and subclinical populations.
“However, an important caveat is needed when interpreting claim made in the press release that exercise is better than medication or talking therapies. This review does not include head-to-head trials against pharmacotherapy or psychotherapy, and 87% of the included meta-analyses are rated as low or critically low quality (using AMSTAR-2). These factors undermine this statement, especially when making theoretical comparisons with typically much higher quality medication/psychotherapy trials/ papers.
“Moreover, what we do know from head-to-head trials or network meta-analyses is that exercise, medication and psychotherapy generally show similar effect sizes in mild cases, rather than one clearly outperforming the others.
“The key message to the public or people with anxiety or depression is that exercise is a credible, evidence-based option that can sit alongside medication and talking therapies, and in some cases be an effective first step, particularly when tailored to individual preferences and circumstances. The good news we keep hearing from all of these reviews is that the type of exercise does not make much difference, they all have a potential benefit.”
Prof David Curtis, Honorary Professor, UCL Genetics Institute, University College London (UCL), said:
“The main limitation with this study is that the only patients included were those who were willing to be entered into a randomised trial of exercise treatment. Patients with more severe illness would presumably have preferred other treatments which are readily available, such as antidepressants, and would simply not have volunteered to participate in the research.
“The authors cannot claim that exercise is an effective treatment for depression or anxiety in general. The study shows that exercise is helpful for patients who are well enough to attempt it as a form of treatment and it certainly seems reasonable that it should be offered as one possible approach. Other patients, though, will benefit from other interventions which have been shown to be safe and effective.”
Prof Jonathan Roiser, Professor of Neuroscience & Mental Health, University College London (UCL), said:
“This is an overview paper on the beneficial effects of physical activity on anxiety and depression. It examines previous reviews focused on clinical trials, and includes studies looking at both diagnosed and non-clinical groups, excluding studies of people with physical health conditions. Perhaps unsurprisingly, the headline results do not provide much new information over the many dozens of reviews on this topic published over the past decade: 1) physical activity is effective for anxiety and depression; 2) the effect for anxiety (medium) is smaller than that for depression (large); 3) low-intensity activity is more effective for anxiety; 4) aerobic exercise delivered in a supervised group format is more effective for depression. Some of the results relating to anxiety seemed internally inconsistent, although the authors do not comment on this.
“The authors state that these findings should encourage more widespread prescription of physical activity for depression and anxiety, which is in line with NHS guidelines. However, unfortunately they do not discuss in any detail the flaws of most clinical trials of physical activity for anxiety and depression, which, to quote the authors, result in most work “being categorised as critically low or low quality.” Some of the main issues include inadequate/inappropriate control groups which fail to account for expectancy or placebo effects (including so-called “waitlist” control – i.e., no treatment at all – which is known to be actively harmful), and small sample sizes.
“Unfortunately, the authors make a superficial, non-statistical, comparison with trials of other mental health treatments – medication and talking therapies – even claiming in the press release: “all forms of exercise are as good as, or better than, medication and talking therapies”. This is grossly misleading, because: 1) it is not even supported by the authors’ own analyses (e.g. neither moderate nor vigorous activity was found to be effective for anxiety); 2) even if it was, the poor quality of control conditions in many trials of physical activity, compared to the more rigorous controls in trials of other treatments, makes such a comparison meaningless; 3) there are very few head-to-head trials of medication/psychotherapy versus physical activity, and none of these were included in the current review. Certainly, no one who is currently receiving medication or psychotherapy for anxiety or depression should be stopping their treatment on the basis of this paper.”
Professor Anna Whittaker, Professor of Behavioural Medicine, Centre for Research & Innovation in Sport, University of Stirling, said:
“This new comprehensive pooling of existing meta-analyses reinforces the importance of people considering exercise as one potential treatment option for depression and anxiety. Traditional medication and talking therapies can still be very helpful for some individuals and should not be discontinued abruptly without taking medical advice, but patients should also talk to their GPs about all their treatment options, to see if exercise prescription could be a potential treatment. This paper in the BMJ shows there is a substantial weight of evidence to encourage health professionals to prescribe exercise as a treatment for some people with depression and anxiety. The research considers data from people with diagnosed and undiagnosed symptoms of depression and anxiety, and suggests exercise is effective for both severities but somewhat more effective for non-diagnosed what might be considered less severe symptoms – this evidence provides an effective direct route to treat symptoms in some people without having to take medication or join a waiting list for talking therapies. The data underline the importance of group supervised aerobic activity for people with depression as well as indicating some key groups where this might be the most effective – younger adults and post-natally. The evidence for anxiety is strong but less is known due to there being less data but from what is available, this is encouraging given that short duration lower intensity exercise can be beneficial. This study cannot firmly comment on the ideal exercise duration and intensity for these conditions given the range of definitions of both across the literature, but there are some indications of effects of different types. The usual message of anything being better than nothing still applies and this paper also points to the gap in supporting clinicians to be confident in prescribing exercise. This is where a collaborative approach with exercise psychologists and physical activity experts might be needed.”
‘Effect of exercise on depression and anxiety symptoms: systematic umbrella review with meta-meta-analysis’ by Neil Richard Munro et al. was published in British Journal of Sports Medicine at 23:30 UK time Tuesday 10 February 2026.
DOI: 10.1136/bjsports-2025-110301
Declared interests
Dr Brendon Stubbs: “Editorial Board of the NPJ Exercise Medicine and Health, Journal of Physical Activity and Health, Ageing Research Reviews, Mental Health and Physical Activity, The Journal of Evidence Based Medicine, and The Brazilian Journal of Psychiatry. Brendon has received honorarium from a co-edited book on exercise and mental illness (Elsevier), an education course and unrelated advisory work from ASICS and FitXR LTD.”
Prof David Curtis: “I have no conflict of interest.”
Prof Jonathan Roiser: “No relevant conflicts of interest to declare.”
Professor Anna Whittaker: “No conflicts of interest.”
For all other experts, no reply to our request for DOIs was received.