CATEGORY: STRESS & MIND-BODY
Put exercise head to head against talking therapy, and it keeps up. Put it against antidepressant pills, and it keeps up there too. That’s the verdict of a newly updated Cochrane review pooling 73 randomized trials with nearly 5,000 adults who had depression: exercise brought a moderate drop in symptoms compared with no treatment, and roughly the same relief as psychotherapy or medication.1 One honest caution comes with it. In the handful of most rigorously run trials, the benefit looked smaller, though it didn’t disappear.
Seventy-three trials, one question
Cochrane reviews are the slow, skeptical audits of medicine, and this one has been asking the same question since 2008. The 2026 update adds 35 new trials. Every study in the pile used a random draw, not personal preference, to decide who got the exercise program. That matters. People who choose to exercise differ from people who don’t in dozens of ways, and the draw wipes those differences out. Ten trials, with 414 people, compared exercise directly against psychological therapy and found essentially no difference between them, a result the reviewers rate with moderate confidence. Five trials, with 330 people, compared it against antidepressants and again found no clear gap, though the evidence there is thinner.
Lighter workouts did at least as well
You don’t have to punish yourself. Programs of light-to-moderate intensity were linked to as much improvement as vigorous ones, with the largest gains seen in people who completed somewhere between 13 and 36 sessions. Mixed programs, and ones built around resistance work, appeared to edge out plain aerobic routines. Resistance work means moving against a load, like lifting a weight or pushing against a band. Side effects were rare and minor, mostly the occasional strained muscle or sore joint. People stuck with exercise about as well as they stuck with any other treatment.
Where the evidence stays soft
Most of the 73 trials were small, many under 100 people, and plenty had design weaknesses. When the reviewers kept only the seven strictest trials, covering 447 people, the benefit shrank to something more modest while still favoring exercise. Few studies followed anyone after the program ended, so nobody knows how long the effect lasts. The comparisons with therapy and pills rest on a few hundred people each, enough to say the treatments look similar, not enough to settle fine distinctions. A separate 2024 analysis of 218 trials reached the same broad conclusion, with walking, jogging and strength training among the forms that worked.2
Key Takeaways
- Across 73 randomized trials, exercise reduced depression symptoms moderately compared with no treatment, and about as much as talking therapy (moderate-certainty evidence) or antidepressants (less certain, fewer trials).
- Light-to-moderate intensity worked at least as well as vigorous exercise, and finishing roughly 13 to 36 sessions went with the biggest improvements.
- The strictest trials showed a smaller benefit, and almost none followed people long term, so the durability of the effect is unknown.
References
- Clegg A, Hill JE, Mullin DS, et al. Exercise for depression. Cochrane Database of Systematic Reviews. 2026;(1):CD004366. link
- Noetel M, Sanders T, Gallardo-Gómez D, et al. Effect of exercise for depression: systematic review and network meta-analysis of randomised controlled trials. BMJ. 2024;384:e075847. link