Across 218 trials and 14,170 people, exercise helps depression by a real but moderate amount, and harder sessions helped more than easy ones (Meta-analysis). A 2026 Cochrane review found that in the few direct tests, it did about as well as talk therapy or antidepressants (Review). Add it to your current care with your clinician’s help. Do not use it to replace that care.
What Counts as Exercise Therapy
In these studies, exercise therapy means a set program, often led by a coach. It is given to people diagnosed with depression, or with high scores on a symptom checklist.
NICE writes treatment guidance for the NHS in England. Its version is a group class run by a trained practitioner. The class usually meets more than once a week for 10 weeks, with about 8 people in the group (Guideline).
Trials judge success in two ways. One is a symptom scale, a questionnaire scored in points. The other is remission, when symptoms drop low enough to count as well.
Two kinds of study come up below. A treatment trial takes people who are already depressed and decides by coin flip who exercises. A prevention study follows thousands of people for years and tracks who later becomes depressed. It can show a link, but it cannot prove cause. Some of the link may run the other way, since people who already feel low may move less (Meta-analysis).
One more problem affects every number below. In a drug trial, some people get a dummy pill, so nobody knows who got the real thing. You cannot hide a workout like that. People know they are exercising, and hope alone can lift their scores. The Cochrane team rated every exercise trial as exposed to this problem, and many relied on people rating their own mood (Review).
How Much Exercise Helps Depression
On average, exercise helps by a moderate amount, and by less in the best-run studies.
The Cochrane review combined 57 trials and 2,189 people that compared exercise with no treatment or with a comparison activity. Exercise may give a moderate drop in symptoms, but the authors had low confidence in that (Review).
The benefit shrinks when you keep only the strongest trials. Just 7 trials, with 447 people, passed three strict quality tests. One test was that the people scoring symptoms did not know who had exercised. In those 7, the benefit was still there, but about a third smaller and less certain (Review).
A 2023 sports medicine review found bigger numbers. It combined 41 trials and 2,264 people and found a large benefit, just as large in people with diagnosed major depression (Meta-analysis). On a standard depression scale scored by clinicians, exercisers ended nearly 5 points lower than the comparison groups. The authors put it this way: for every two people treated with exercise, about one can expect a big drop in symptoms.
Supervision seemed to matter. Programs led by a coach or trainer showed a large benefit (Meta-analysis). The 6 trials of people exercising on their own did not show a clear benefit. With so few trials, that is hard to judge either way.
In the 12 trials with the fewest flaws, the benefit dropped from large to moderate. For about every three people who exercised, one more had a big improvement than in the comparison groups (Meta-analysis). The authors also checked for disappointing studies that never got published. Allowing for those shrank the benefit again, to small.
In practice, a moderate benefit means many people feel clearly better. It does not mean the depression is gone.
Exercise vs Antidepressants
Only a few small trials have compared the two directly. In the Cochrane review, 10 trials with 414 people compared exercise with talk therapy. There was probably little or no difference at the end of treatment (Review). Five trials with 330 people compared exercise with antidepressant drugs. There may be little or no difference there too, but the authors were less sure.
Side effects were uncommon. Exercisers had some muscle and joint injuries, plus some reports of depression. People on the drug sertraline reported diarrhea, sexual problems and tiredness.
The biggest recent direct test is the Dutch MOTAR study. It followed 141 adults with depression, anxiety or both for 16 weeks (Study). One group took an antidepressant, escitalopram or sertraline. The other did group running outdoors, 45 minutes a session, two to three times a week. About the same share reached remission: 44.8% on pills and 43.3% with running.
Running also helped the body. By the end, the groups differed on weight, waist size, blood pressure, heart rate and heart rate variability, a marker of heart health (Study). The gap came from both sides. Runners improved, while the pill group got a bit worse, gaining about 3 kg on average over 16 weeks.
But the study has a big weakness. Only 22 of the 141 people, about 15%, let a coin flip pick their treatment (Study). The rest chose for themselves. In the end, 45 people took pills and 96 ran. The pill group was also more depressed to begin with. Psychiatrists from outside the study, quoted in medical news, flagged these limits, and one called them “very important” (Report).
Sticking with it was harder for the runners. About 82% of the pill group were still taking their medicine at 16 weeks. Only 52% of runners hit their attendance target, and 15% never started (Study). The two goals were measured differently, but the gap is large.
An older US trial gave 156 adults with major depression 4 months of aerobic exercise, the antidepressant sertraline, or both (Trial). At 4 months, about the same share of each group had recovered.
Six months after treatment ended, the exercise group was doing best. Among those who had recovered, 8 in every 100 in the exercise group slipped back into depression. That compares with 38 in every 100 on the drug alone and 31 in every 100 on both (Trial). But most people in the drug groups had stopped their pills by then. So it mostly compares exercise with stopping medication, not with staying on it.
Exercise can also go alongside pills. The 218-trial analysis looked at exercise added to an SSRI, the most common type of antidepressant. That pairing came out about as helpful as talk therapy (Meta-analysis). It also came out ahead of an SSRI alone. That comparison is indirect, though, and the authors had low confidence in most of these results. NICE lists adding group exercise as an option when antidepressants alone have not worked well enough (Guideline).
Do not stop or swap your medication to exercise instead. Stopping an antidepressant suddenly can cause its own problems, such as withdrawal symptoms or a return of depression. Plan any change to a prescription with the clinician who wrote it.
Which Kinds Work Best
The 218-trial analysis also ranked exercise types, each compared against a real alternative (Meta-analysis). Three came out on top: walking or jogging, yoga, and strength training. Mixed aerobic workouts and tai chi or qigong helped too, just a little less. Dance showed the biggest benefit of all. But that came from only 5 small studies with 107 people, so treat it as a hint.
Some types were easier to stick with. Strength training and yoga were the only two where fewer people dropped out than in the comparison groups (Meta-analysis). The best fit may also vary by person. Strength training seemed to work better in women and younger people, and yoga in men and older people. The authors call these early hints that still need testing (Meta-analysis).
A 2018 strength training review combined 33 trials and 1,877 people and found a moderate drop in symptoms (Meta-analysis). For about every 4 people who lifted, one more improved than in the comparison groups.
You may not need to get much stronger to benefit. The benefit was not clearly tied to how much strength people gained, or to how much training they were given (Meta-analysis). Trials where the people scoring symptoms did not know who lifted showed a smaller benefit, but still a real one. If you also lift for long-term health, our post on strength training and longevity covers the dose for that goal.
How Hard and How Often
The 218-trial analysis found a clear pattern for effort: the harder the prescribed sessions, the bigger the benefit (Meta-analysis). Vigorous work, such as running or interval training, did best. Light activity, such as walking or gentle yoga, still gave a meaningful benefit. The total amount of exercise per week showed no clear pattern.
The DOSE trial tested amounts directly (Trial). It took 80 adults aged 20 to 45 with mild to moderate depression through 12 weeks of supervised aerobic exercise. A coin flip set each person’s weekly calorie burn at a high or a low level, spread over 3 or 5 days. A separate group did stretching instead, as a stand-in for a dummy pill.
The higher dose, sized to public health advice, cut symptom scores by 47%. The low dose cut them by 30%, about the same as stretching at 29%. Exercising on 3 days or on 5 days made no difference (Trial).
A 2008 review turned that into advice: aim for about 150 minutes a week of moderate exercise, like brisk walking (Review). It adds that DOSE suggests fewer, longer sessions could work too. That is the authors’ advice, and it rests largely on one small trial.
Prevention studies point the same way, with one big difference: they show links, not proof. A 2022 review of 15 long-term studies and 191,130 adults found the biggest gains at the low end (Meta-analysis). Half the recommended weekly activity was linked to an 18% lower risk of depression. The full amount, about 2.5 hours of brisk walking a week, was linked to a 25% lower risk. Going past that seemed to add little. The studies also disagreed with each other a fair amount.
A Norwegian study followed 33,908 adults with no mental health symptoms for 11 years (Study). Even 1 hour of exercise a week was linked to fewer new cases. Most of the protection came at low levels, however hard people worked. If exercise really is the cause, the authors estimate 12% of cases might have been avoided with at least that hour.
So for lowering the risk of depression, a little activity at any effort is linked to a benefit. For treating it, the trials favor harder sessions.
A Weekly Prescription to Try
This sample week is built from those trials. Bring it to your doctor or therapist as a starting point, on top of your current care.
- Three sessions a week, 45 to 50 minutes each. Three days worked as well as five in DOSE (Trial). A later trial aimed for DOSE’s higher dose. Its hospital patients on exercise bikes averaged about 146 minutes a week in 3 sessions (Trial).
- Two of them as brisk walking, jogging or running. Go at a pace that makes you breathe hard, since harder sessions helped more (Meta-analysis).
- One or two as strength training or yoga. Both ranked near the top, and only these two had fewer dropouts than the comparison groups (Meta-analysis).
- Do it with a group or a coach if you can. Supervised programs showed the clearest benefit (Meta-analysis). NICE offers group exercise as one of the first options for milder depression (Guideline).
- Give it at least 10 to 12 weeks. NICE’s programs run about 10 weeks, and DOSE ran 12.
On a low-energy day, do less rather than nothing, since light activity still helped (Meta-analysis). NICE also says any regular activity, like walking, swimming, dancing or gardening, could lift your sense of wellbeing. It adds that the benefit can be greater outdoors (Guideline).
After the program, keep going. In the older US trial, people who kept exercising on their own were less than half as likely to be depressed at 10 months (Review). They chose for themselves whether to keep going, though, so this part shows a link, not proof (Trial).
If you have thoughts of harming yourself, get help now: call your local emergency number or a crisis line.
What the Trials Can’t Tell Us
You cannot hide a workout from the person doing it, and many trials relied on people rating their own mood (Review).
Most trials are small and have flaws that make a treatment look better than it is. In the 218-trial analysis, only one study met Cochrane’s test for a trial with few flaws (Meta-analysis). Confidence was rated low for walking or jogging, and very low for the other types.
The best trials show smaller benefits. That held in the Cochrane review (Review), the sports medicine review (Meta-analysis) and the strength training review (Meta-analysis).
Few trials checked back later. Only 9 trials, with 405 people, measured long-term effects against a comparison group, and the results were very uncertain (Review).
Starting is hard when you are depressed. In MOTAR, 15% of the runners never began (Study). For more severe depression, NICE says options with more therapist contact should be weighed first (Guideline).
Overall, the benefit is real, but it is likely smaller than the biggest numbers suggest.
Frequently Asked Questions
Is Exercise Better for Depression Than Antidepressants?
Not clearly. In the few direct tests, it did about as well. The Cochrane review found there may be little or no difference, based on 5 small trials (Review). Exercise is best used alongside treatment. Never stop an antidepressant without talking to your prescriber.
How Much Exercise per Week for Depression?
A common target is about 150 minutes a week of moderate aerobic exercise. You can fit that into fewer, longer sessions (Review). In the DOSE trial, that dose cut symptoms more than a low dose did, and three days worked as well as five (Trial). Harder sessions seemed to help more (Meta-analysis).
Does Exercise Prevent Depression?
It may lower the risk. Long-term studies link even 1 hour a week to fewer new cases (Study). About 2.5 hours a week is linked to about a quarter lower risk (Meta-analysis). These studies show a link, not proof that exercise caused it.
How Much Does Exercise Help Depression?
A moderate amount, on average. Across 57 trials, exercise may give a moderate drop in symptoms compared with no treatment, though the authors had low confidence in that (Review). In the best-run trials, the benefit was about a third smaller.
Key Takeaways
- Real but moderate benefit: Across 218 trials and 14,170 people, exercise eased depression by a moderate amount (Meta-analysis).
- Similar in direct tests: In the few direct tests, exercise did about as well as talk therapy or antidepressants (Review).
- Intensity matters: Harder sessions helped more, while extra weekly minutes showed no clear pattern (Meta-analysis).
- Walking, yoga and strength lead: These three ranked best, and only yoga and strength had fewer dropouts than the comparison groups (Meta-analysis).
- A weekly starting dose: About 150 minutes a week is a common target, and 3 days worked as well as 5 (Review).
- Add, don’t swap: When antidepressants alone fall short, NICE lists adding group exercise as an option to choose with your clinician (Guideline).
Add Exercise to Your Treatment Plan
Exercise is cheap and has few side effects. NICE’s committee found some evidence that group exercise works and is good value for milder depression (Guideline). NICE adds that it can help your physical health and avoids the possible side effects of medication (Guideline).
Bring this plan to your next appointment, and ask whether a group class or a coach is available near you. Pick the kind of exercise you are most likely to enjoy. Start smaller than you think you need to, and give it a few months.
Exercise will not fix depression on its own, but the evidence says it deserves a place in your plan.
This article is for educational purposes and is not medical advice. Talk to a qualified clinician before changing your health regimen.
Leave a Reply