Exercise for Depression: What Treatment Comparisons Can Tell Us

Comparing exercise with antidepressants sounds like a contest with one winner, but the trials differ in participants, controls and support. A pooled symptom effect cannot decide an individual's treatment on its own. The 2023 umbrella review and the 2024 BMJ network analysis support an important role for physical activity while raising questions about evidence quality and indirect comparisons. This article reads those comparisons as information for a treatment discussion, rather than a reason to replace prescribed care with a generic exercise plan.

How the review was built

An umbrella review combines results from existing systematic reviews. Its breadth is useful, but it also inherits their weaknesses and may include overlapping primary trials. Singh and colleagues searched twelve databases and included reviews of randomized physical-activity interventions. The populations ranged from healthy adults to people with mental health or other medical conditions; one pooled result should not be treated as a treatment effect for every diagnosis.

That breadth matters. Rather than asking whether jogging helps healthy students, the review aggregated what hundreds of trials reported across ages, conditions, and exercise types, from walking programs to resistance training and yoga.

What the numbers showed

Across all populations, physical activity produced a medium-sized reduction in depression symptoms, with a median standardized effect size of minus 0.43, and a similar effect for anxiety at minus 0.42. Psychological distress fell even more, with an effect size of minus 0.60 compared with usual care. In plain terms, the average person who exercised finished a trial noticeably better off than the average person who did not.

The researchers reported that these reductions were comparable to, or slightly larger than, those typically observed for psychotherapy and leading medications. Benefits were largest in people with depression, HIV, and kidney disease, in pregnant and postpartum women, and in healthy adults. Higher-intensity activity was associated with greater improvements, while effects appeared to shrink in interventions running longer than 12 weeks.

Putting the effect size in context

Standardized effect sizes are easy to misread. A value between minus 0.4 and minus 0.6 is considered moderate and roughly means the exerciser scored better than about two-thirds of the non-exercising comparison group. The University of South Australia's press materials summarized the aggregate result as making physical activity around 1.5 times more effective than counselling or leading medications, a framing drawn from comparing pooled effects across reviews rather than from a single head-to-head trial.

A 2024 network meta-analysis in The BMJ sharpened the comparison. Noetel and colleagues combined 218 randomized trials with roughly 14,000 participants diagnosed with major depressive disorder. Walking or jogging produced an effect of about minus 0.62 versus usual care, yoga about minus 0.55, and strength training about minus 0.49, all in the same range as cognitive behavioral therapy and selective serotonin reuptake inhibitors.

The BMJ network analysis focused on depression and combined direct and indirect comparisons across a network of trials. That design can compare options that were not all tested against one another in the same study, but it depends on assumptions about trial comparability. It does not establish that every exercise program equals a particular drug for a particular patient.

Why movement affects mood

Several mechanisms plausibly overlap. Exercise raises brain-derived neurotrophic factor, a protein that supports neuron growth and plasticity and is often low in depression. It also triggers endorphins and endocannabinoids, dampens systemic inflammation, and helps normalize stress-hormone rhythms governed by the hypothalamic-pituitary-adrenal axis. Sleep quality improves too, and sleep loss is tightly linked to low mood.

Psychological pathways matter as well. Structured programs provide routine, mastery experiences, and often social contact, all of which counter the withdrawal and helplessness that depression reinforces.

The limitations deserve weight

The review's own quality assessment was sobering: 77 of the 97 included meta-analyses were rated as having critically low confidence using the AMSTAR 2 tool. Most trials compared exercise with usual care rather than against antidepressants directly, participants cannot be blinded to exercise, and small trials with positive results are more likely to be published. A commentary accompanying the paper noted that many included trials enrolled non-clinical populations, which limits how far the findings extend to psychiatric patients.

Smaller effects in longer interventions do not reveal the reason for the difference. Adherence, populations and study design may contribute. The review does not show that exercise stops helping after a fixed number of weeks.

Bring a feasibility question to the treatment discussion

A useful question is what activity remains possible when symptoms are difficult, and what support would make participation more likely. That may involve accompaniment, transport, a quieter environment or a smaller starting task. Record whether the proposed program is an addition to care and who will review symptoms. An average standardized effect does not predict how quickly one person will improve or whether a particular medicine should change. The practical contribution here is to connect evidence with access and follow-up: a theoretically effective activity cannot help through a schedule or setting the person cannot use.

What it means in practice

The authors argued for a central role for physical activity in care. Turning that conclusion into a treatment plan requires attention to symptom severity, preference, access and other treatment. The appropriate starting activity may be very modest when depression makes ordinary tasks difficult.

Exercise can be considered alongside established care, without using a pooled comparison to justify stopping medication. Discuss treatment changes with the clinician overseeing that care. Improvement in symptoms is the outcome to follow, rather than whether a headline says one treatment won.

A workable plan needs an alternative for difficult days. A short walk with someone trusted may be more accessible than a demanding class; someone else may prefer a structured appointment. These are planning options, not trial-proven prescriptions for a particular individual. Reassess symptoms and feasibility with the treating professional.