Across more than 200 randomized trials, people with depression who were assigned to exercise finished with symptom scores that were moderately lower than those of people assigned to a control condition. That is the headline of the largest synthesis to date, and it is also where the certainty of the evidence starts to thin out. Here is what the primary papers and the federal guideline say about how much exercise helps, which kinds, and what the studies cannot tell you.

The size of the effect, in plain words

The BMJ published a systematic review and network meta-analysis on February 14, 2024 that pooled 218 randomized trials with 14,170 adults who met a clinical threshold for depression. Against active controls, walking or jogging reduced depressive symptoms with a Hedges g of about 0.62, yoga 0.55, strength training 0.49, mixed aerobic exercise 0.43, and tai chi or qigong 0.42. By the convention researchers use, 0.2 is small, 0.5 is moderate and 0.8 is large, so these are moderate effects, in the same range the review found for exercise added to an SSRI (0.55) or to psychotherapy (0.54). Dance scored highest (0.96) but rested on five small trials with 107 mostly female participants, which the authors flag as too thin to lean on. The review received no funding.

A Cochrane review published January 8, 2026 reached a similar place by a different route. Pooling 57 trials with 2,189 participants that compared exercise with no treatment or a control, it found a standardized mean difference of 0.67 at the end of treatment. In ten trials with 414 people, exercise and psychotherapy came out about even (0.03, moderate certainty), and in five trials with 330 people, exercise and antidepressant medication came out about even (0.11, low certainty). The authors note that long-term follow-up data are scarce. The update itself had no grant funding; some authors were partly supported by a UK National Institute for Health and Care Research collaboration.

For milder depression, a meta-analysis in Frontiers in Psychology published April 14, 2026 pooled 37 trials with 3,110 participants with mild-to-moderate or subthreshold symptoms and found a larger pooled effect (g of 0.86), with aerobic exercise (0.91), mind-body exercise (0.88) and resistance training (0.84) close together. Heterogeneity was high and a test for publication bias was positive, and the authors rated certainty as low.

Which kinds and how hard

The BMJ analysis found that effect grew with prescribed intensity: light activity such as walking or hatha yoga carried a g of 0.58, vigorous exercise 0.74. A dose-response by weekly energy expenditure could not be pinned down because the credible intervals were wide. Strength training and cycling showed larger effects in women; yoga, tai chi and aerobic exercise combined with psychotherapy showed larger effects in men. Yoga worked better in older participants and strength training in younger ones. These are study-level patterns, not head-to-head tests, and the authors warn they can be confounded.

How sure anyone can be

Here the picture is weaker than the effect sizes suggest. In the BMJ review, only one of 218 trials met Cochrane criteria for low risk of bias, mainly because participants and staff cannot be blinded to whether someone is exercising. Confidence in the evidence was rated low for walking or jogging and very low for every other form. A test for publication bias was significant. Cochrane rated its main comparison low certainty for similar reasons. None of this means the effect is not real; it means the true size could be smaller than the pooled number, and expectancy may account for part of it.

What the federal guideline says

The second edition of the Physical Activity Guidelines for Americans, issued by the Department of Health and Human Services, lists reduced risk of depression as a long-term benefit of physical activity for adults, reduced anxiety and better sleep quality among the benefits that follow a single session, and reduced symptoms of anxiety and depression among the benefits for people who already have those conditions. The guideline's general adult target is 150 to 300 minutes a week of moderate-intensity aerobic activity plus muscle-strengthening activity on at least two days. The guideline does not set a separate dose for depression.

What the evidence does not show

The trials measured symptom scores, mostly over weeks to a few months, in people who agreed to be randomized to exercise. They do not show how long the benefit lasts, whether exercise prevents a relapse, or how it performs in severe or treatment-resistant depression, which most trials excluded. The comparisons with medication and therapy rest on a handful of small trials. And because a trial of walking cannot be blinded, some of the measured effect may be expectation. The question to bring to a clinician is not whether to swap treatment for exercise but which form and intensity fits alongside the treatment already in place, and how to track whether it is helping.

The 988 Suicide and Crisis Lifeline is reachable in the United States by calling or texting 988, or by chat at chat.988lifeline.org, 24 hours a day, free and confidential.

This article reports research and expert guidance. It is not medical advice. Talk to your clinician before changing medication, treatment or a medical diet.

Sources