Exercise for depression: type, intensity and setting

A 41-year-old teacher returns to the consultation six weeks after starting an antidepressant for a moderate depressive episode. Her sleep has improved a little, but she still has no energy and barely leaves the house. She asks whether she should "start going to the gym", because she read that exercise works as well as medication, and wants to know if she could swap one for the other.

The question is common, and the answer has more nuance than the headlines suggest. There are hundreds of clinical trials on exercise in depression, and the most recent reviews show consistent effects. At the same time, the quality of these trials is limited, and guidelines place exercise alongside psychotherapy and antidepressants, not in place of them.

This text shows what the trials say, which modalities and intensities have the most support, how exercise compares with established treatments, where it fits in the sequence of care and what precautions to take when prescribing it.

Key points

• In a network meta-analysis of 218 studies, walking or running, yoga and strength training had the largest effects on depression (Noetel, 2024).
• The effect was greater at higher intensity, but light activity also brought clinically relevant benefit (Noetel, 2024).
• In comparative trials, exercise had an effect similar to psychotherapy and to antidepressants, with evidence of limited certainty (Clegg, 2026).
• The Canadian guideline recommends supervised exercise as first-line treatment in mild depression and as an adjunct in moderate depression (Lam, 2024).
• More physical activity is also associated with a lower risk of developing depression (Pearce, 2022).
• Suicide risk assessment remains mandatory and has its own management (Lam, 2024).

Summary

• Why talk about exercise in depression
• What the trials show about the effect
• Which type of exercise and which intensity
• Where exercise fits in treatment
• The limits of the evidence
• Exercise to prevent depression
• How to prescribe in practice
• How NEXO helps train this reasoning
• Frequently asked questions
• Take-home messages

Why talk about exercise in depression

Depression is one of the leading causes of disability worldwide, and access to treatment is unequal. The authors of the largest recent meta-analysis note that treatment coverage reaches about half of people with depression in high-income countries and one in five in low- and lower-middle-income countries (Noetel, 2024). Many patients also respond only partially to established treatments.

Exercise is appealing for several reasons. It is available almost anywhere, brings physical benefits independent of mood and can be combined with the other treatments (Lam, 2024). Some patients also prefer an approach without medication, and the 2026 Cochrane review cites this preference as one of the reasons for studying exercise (Clegg, 2026).

But appealing is different from proven. The clinical question has several parts: does exercise reduce symptoms? By how much? Which type and intensity? How does it compare with what already works? And in whom should it be the first choice, a complement or a secondary option? The following sections answer one at a time.

What the trials show about the effect

The network meta-analysis

The meta-analysis published in the BMJ in 2024 gathered 218 studies with 14,170 participants who met criteria for major depression (Noetel, 2024). Compared with active controls, such as usual care or placebo pill, the effects were moderate for walking or running (g of -0.62), yoga (-0.55), strength training (-0.49), mixed aerobic exercise (-0.43) and tai chi or qigong (-0.42) (Noetel, 2024).

Exercise appeared equally effective in people with and without other diseases and at different levels of baseline severity (Noetel, 2024). The authors conclude that these modalities could be considered alongside psychotherapy and antidepressants as core treatments.

Other reviews

An umbrella review of 97 systematic reviews, with 1039 trials and 128,119 participants, found a median effect of -0.43 of physical activity on depression in adults from varied populations (Singh, 2023). The greatest benefit appeared precisely in people with depression.

A meta-analysis of 41 studies with 2264 participants estimated a large effect, with a standardized mean difference of -0.946, which corresponds to a number needed to treat of 2 (Heissel, 2023). When the analysis was restricted to the 12 studies at low risk of bias, the effect fell to -0.666, still moderate, with a number needed to treat of 2.8 (Heissel, 2023).

The Cochrane review updated in 2026 found, across 57 trials, a standardized mean difference of -0.67 in favor of exercise compared with no treatment or control (Clegg, 2026). In the seven trials with allocation concealment, intention-to-treat analysis and blinded assessment, the effect was smaller, at -0.46 (Clegg, 2026).

Which type of exercise and which intensity

The network meta-analysis allowed the modalities to be compared with each other. Walking or running, yoga and strength training stood out, and yoga and strength training were the best tolerated modalities, with fewer dropouts than active controls (Noetel, 2024).

Strength training deserves special attention. A meta-analysis of 33 trials with 1877 participants found a moderate reduction in depressive symptoms, with an effect of 0.66 (Gordon, 2018). The prescribed volume, the participants' health status and the gain in strength were not associated with the effect size. This suggests that the benefit for mood appears even without large gains in strength.

There are signs that the ideal type varies from person to person. The effects of strength training and cycling seemed greater in women, and those of yoga, tai chi and aerobic exercise combined with psychotherapy seemed greater in men (Noetel, 2024). Yoga and aerobic exercise with psychotherapy seemed more effective in older participants, and strength training in younger ones. These differences carry great uncertainty and serve more to guide the conversation than to decide.

In practice, the best modality is usually the one the person can keep up. Patient preference weighs heavily in a treatment that depends on adherence, as the experience with listening to the patient's schedule before proposing any change shows.

Which intensity

In the network meta-analysis, the effects were proportional to the prescribed intensity (Noetel, 2024). Light activity, such as walking and hatha yoga, still brought a clinically relevant effect, with g of -0.58, but the expected effect was greater for vigorous activity, such as running and interval training, with g of -0.74 (Noetel, 2024).

The umbrella review reached a similar conclusion: physical activity of higher intensity was associated with greater improvement in symptoms (Singh, 2023). It also observed that longer interventions had smaller effects, perhaps because of a drop in adherence over time.

This does not mean that every depressed patient should start with vigorous training. Lack of energy, anhedonia and slowing make starting difficult, and an overly ambitious goal can become one more reason for frustration. The Canadian guideline recommends low to moderate intensity as a starting point (Lam, 2024), and progression comes later, as the person tolerates it.

Exercise compared with psychotherapy and antidepressant

The question from the professor in the opening scene, whether exercise can replace medication, depends on this comparison. The 2026 Cochrane review found ten trials with 414 participants that compared exercise with psychotherapy, with little or no difference in effect at the end of treatment (Clegg, 2026). Five trials with 330 participants compared exercise with pharmacological treatment and also found little or no difference.

The SMILE trial is a classic example. Two hundred and two adults with major depression were randomly assigned to supervised group exercise, home exercise, sertraline or placebo for 16 weeks (Blumenthal, 2007). Remission rates were 45 per 100 with supervised exercise, 40 with home exercise, 47 with sertraline and 31 with placebo.

The same study shows the main caution. The difference between the active treatments and placebo was not statistically significant, and the authors point out that part of the response comes from expectation, attention and monitoring (Blumenthal, 2007). In depression, placebo and nonspecific factors carry great weight in any treatment.

In addition, direct comparisons are few and small. The certainty of the Cochrane evidence for these comparisons is limited (Clegg, 2026). The most honest reading is that exercise has an effect of the same order of magnitude as the treatments established in short-term trials, not that it is equivalent in every situation.

Where exercise fits into treatment

The 2023 Canadian guideline for major depression is one of the most detailed on the topic (Lam, 2024). It recommends supervised exercise of low to moderate intensity, for 30 to 40 minutes, 3 to 4 times a week, for at least 9 weeks. In this form, exercise is a first-line treatment as monotherapy for mild depression and a second-line treatment, as an add-on, for moderate depression.

The same guideline points out that exercise can reduce suicidal ideation and has good effects when combined with antidepressants (Lam, 2024). It also cites benefits with supervised aerobic activity of moderate to high intensity in adolescents and young adults, low to moderate intensity in middle-aged and older women, and high-intensity interval training in adults.

The guideline's exercise recommendations are limited to mild and moderate depression (Lam, 2024). In severe cases, with psychotic symptoms or high suicide risk, the established treatment comes first, and exercise can come later as a complement.

In the opening scene, the teacher has a moderate episode and has been in treatment for six weeks. Exercise comes in as an add-on: she keeps the antidepressant, and exercise adds its effect on mood, energy and physical health. The decision to reduce or stop the medication follows the usual criteria of response and treatment duration.

Situation	Place of exercise	Source
Mild depression	First line as monotherapy, supervised	Lam, 2024
Moderate depression	Second line, as an add-on to the established treatment	Lam, 2024
Comparison with psychotherapy	Similar effect in small trials	Clegg, 2026
Comparison with antidepressant	Similar effect in few trials, with limited certainty	Clegg, 2026
Modalities with more support	Walking or running, yoga and strength training	Noetel, 2024
Format	Supervised and in groups, at moderate intensity	Heissel, 2023

The limits of the evidence

The main problem is the quality of the trials. In the network meta-analysis, only one study met the Cochrane criteria for low risk of bias, and confidence in the evidence was low for walking or running and very low for the other modalities (Noetel, 2024). In the umbrella review, most reviews had critically low methodological quality (Singh, 2023).

Blinding is nearly impossible. People who exercise know they are exercising, and the expectation of improvement can inflate the effect. Trials with blinded allocation or assessment showed smaller effects, both in strength training (Gordon, 2018) and across the set of trials (Clegg, 2026).

The long-term effect is uncertain. In the nine trials with long follow-up, the evidence on maintenance of benefit was very uncertain (Clegg, 2026). Many patients stop exercising when the study ends, and adherence outside the research setting tends to be lower.

None of this cancels the central finding. The effects appear in different reviews, with different methods, and persist, even if smaller, in the higher-quality trials (Heissel, 2023). The balanced reading is the same as for any imperfect evidence: acknowledge the probable benefit without promising more than the data support, a caution discussed in the text on why we get the estimate wrong.

Exercise to prevent depression

Beyond treating it, physical activity is associated with a lower risk of developing depression. A meta-analysis of 49 prospective cohorts with 266,939 people found a lower chance of depression in those with a high level of activity, with an adjusted odds ratio of 0.83 (Schuch, 2018). The protective effect appeared in young people, adults and older adults and in different regions of the world.

Another meta-analysis, with 15 cohorts and 191,130 people, described a curve with a larger gain in the first levels of activity (Pearce, 2022). Those who accumulated half the recommended volume had a risk 18 per 100 lower than those who did no activity at all, and those who met the recommended volume had a risk 25 per 100 lower. Above that, the additional gain was small and uncertain.

These data come from observational studies and are subject to reverse causality: people with early depressive symptoms tend to move less. Even so, the pattern matches the WHO general recommendation that some activity is better than none (Bull, 2020).

How to prescribe in practice

Before prescribing

Suicide risk assessment comes before any recommendation. The Canadian guideline includes this assessment in the first steps of care and notes that the risk of suicide attempt in people with major depression is five times higher than in the general population (Lam, 2024). This management is its own thing and does not depend on exercise; the text on what conversation captures that the scale does not capture details this reasoning.

It is also worth reviewing the clinical conditions that limit exertion and the medications in use. Some psychotropic drugs interfere with temperature regulation, and exercise on hot days calls for attention, as discussed in the section on medications that worsen the response to heat. Heat is also associated with worsening of psychiatric conditions, the topic of the text on who is most vulnerable.

The prescription

The format with the most support is supervised and in a group (Heissel, 2023). Supervision helps with adherence, and the group adds social contact, which is often lacking in depression. When the supervised format is not feasible, walking combined with small goals and record keeping is a reasonable starting point.

The initial goal should be modest and concrete: days, times and duration defined with the patient. The Canadian guideline suggests 30 to 40 minutes, 3 to 4 times a week (Lam, 2024), but starting with less and progressing is better than not starting. The choice among walking, yoga and strength training can follow the patient's preference and physical condition (Noetel, 2024).

Follow-up

The follow-up visit should assess symptoms and adherence together. If symptoms improve and the person keeps up the exercise, the strategy is working. If adherence has dropped, it is worth understanding the reason before insisting. If symptoms do not improve within a few weeks, the answer is to review the entire treatment, and time used with judgment helps in this decision, as in the text on how to communicate the wait.

It is also necessary to avoid turning exercise into guilt. For someone with depression, an unmet goal can reinforce the feeling of failure. The useful message is that any increase counts and that exercise is one part of treatment, not a test of willpower. Avoiding exaggerated promises is part of quaternary prevention.

Finally, the diagnosis of depression should be reviewed when the picture does not follow the expected course. Hypothyroidism, anemia, sleep apnea and medication effects can produce fatigue and low mood, and keeping other hypotheses open is part of reviewing the hypothesis with each new piece of data.

A script for the teacher in the scene

Returning to the 41-year-old teacher, the conversation can follow a simple order. First, acknowledge the partial improvement and explain that exercise comes as reinforcement of the current treatment, not as a replacement (Lam, 2024). Second, confirm that there is no suicidal ideation or recent worsening that changes the priority of the visit.

Third, ask what she has enjoyed doing and what fits her routine. If she likes walking, walking is a start with good support in the trials; if she prefers something indoors, yoga and strength training were well tolerated (Noetel, 2024). A supervised group near work or home helps maintain frequency (Heissel, 2023).

Fourth, agree on a small goal for the next two weeks, such as three short walks on fixed days, and record what was done. Fifth, schedule the follow-up to review symptoms and adherence, without treating the goal as a test. If she cannot get started, the next step is to understand the barrier, not to increase the pressure.

Over the weeks, progression can follow tolerance: more minutes, more days and, if she wants, more intensity, since higher intensity was associated with a greater effect in the trials (Singh, 2023). The antidepressant follows the usual criteria for duration and response.

It is also worth agreeing with her what to do in bad weeks. Missing a walk does not cancel the agreement, and resuming the next day counts for more than making up for it with a long session. When depression worsens, the activity can become lighter and shorter, keeping the habit alive until energy returns. This prior agreement reduces guilt and protects adherence, which is the weak point of any exercise prescription.

Common mistakes

The first mistake is promising that exercise replaces treatment. Direct comparisons are few and of limited certainty (Clegg, 2026), and swapping an antidepressant that is working for exercise exposes the patient to relapse without need.

The second mistake is prescribing vaguely. "Do exercise" rarely changes behavior; type, days, duration and format defined with the patient are more likely to work, and the Canadian guideline gives concrete parameters for this (Lam, 2024).

The third mistake is giving up early. The guideline speaks of at least 9 weeks of supervised exercise (Lam, 2024), and improvement can take time, as with other depression treatments.

How NEXO helps train this reasoning

In NEXO, the case of the day is fictional, with a real clinical presentation and up to six progressive clues, and you propose the diagnosis by searching for the ICD-10 code.

The limits of use of the game are described on the NEXO Economy page. If you research the teaching of clinical reasoning, the page for Researchers explains how to get in touch. Professionals who write or review cases find information on Curator and on Creators, and institutions can learn about the proposal on the partnerships page. For the most frequent reasons for consultation, see the most common codes in primary care.

Frequently asked questions

Can exercise replace the antidepressant?

In small trials, exercise had an effect similar to that of the antidepressant, with limited certainty (Clegg, 2026). The Canadian guideline places it as first line only in mild depression (Lam, 2024).

Which exercise works best?

Walking or running, yoga and strength training had the largest effects, and yoga and strength were the best tolerated (Noetel, 2024).

Does it need to be intense?

Higher intensity brought a greater effect, but light activity also had a clinically relevant benefit (Noetel, 2024).

Does strength training help even without gaining strength?

It helps. The effect of strength training on depression was independent of the gain in strength (Gordon, 2018).

How long does it take to have an effect?

The Canadian guideline recommends at least 9 weeks of supervised exercise (Lam, 2024).

Does exercise prevent depression?

More active people have a lower risk of depression in cohort studies, with a greater gain at the first levels of activity (Pearce, 2022).

Take-home messages

• Exercise reduces depressive symptoms in clinical trials, with a moderate effect in most reviews.
• Walking or running, yoga and strength training have the most support, and the patient's preference weighs in the choice.
• Higher intensity brings a greater effect, but light activity also helps and is a good starting point.
• In mild depression, supervised exercise is a first-line option; in moderate depression, it enters as an adjunct.
• The quality of the trials is limited, blinding is difficult and the long-term effect is uncertain.
• Suicide risk assessment comes before any prescription and has its own management.
• Supervision, group, small goals and follow-up increase adherence.

Play the case of the day on NEXO.

References

1. 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. PMID 38355154. DOI 10.1136/bmj-2023-075847.
2. Singh B, Olds T, Curtis R, et al. Effectiveness of physical activity interventions for improving depression, anxiety and distress: an overview of systematic reviews. Br J Sports Med. 2023. PMID 36796860. DOI 10.1136/bjsports-2022-106195.
3. Heissel A, Heinen D, Brokmeier LL, et al. Exercise as medicine for depressive symptoms? A systematic review and meta-analysis with meta-regression. Br J Sports Med. 2023. PMID 36731907. DOI 10.1136/bjsports-2022-106282.
4. Clegg AJ, Hill JE, Mullin DS, et al. Exercise for depression. Cochrane Database Syst Rev. 2026. PMID 41500513. DOI 10.1002/14651858.CD004366.pub7.
5. Gordon BR, McDowell CP, Hallgren M, et al. Association of Efficacy of Resistance Exercise Training With Depressive Symptoms: Meta-analysis and Meta-regression Analysis of Randomized Clinical Trials. JAMA Psychiatry. 2018. PMID 29800984. DOI 10.1001/jamapsychiatry.2018.0572.
6. Blumenthal JA, Babyak MA, Doraiswamy PM, et al. Exercise and pharmacotherapy in the treatment of major depressive disorder. Psychosom Med. 2007. PMID 17846259. DOI 10.1097/PSY.0b013e318148c19a.
7. Pearce M, Garcia L, Abbas A, et al. Association Between Physical Activity and Risk of Depression: A Systematic Review and Meta-analysis. JAMA Psychiatry. 2022. PMID 35416941. DOI 10.1001/jamapsychiatry.2022.0609.
8. Schuch FB, Vancampfort D, Firth J, et al. Physical Activity and Incident Depression: A Meta-Analysis of Prospective Cohort Studies. Am J Psychiatry. 2018. PMID 29690792. DOI 10.1176/appi.ajp.2018.17111194.
9. Lam RW, Kennedy SH, Adams C, et al. Canadian Network for Mood and Anxiety Treatments (CANMAT) 2023 Update on Clinical Guidelines for Management of Major Depressive Disorder in Adults. Can J Psychiatry. 2024. PMID 38711351. DOI 10.1177/07067437241245384.
10. Bull FC, Al-Ansari SS, Biddle S, et al. World Health Organization 2020 guidelines on physical activity and sedentary behaviour. Br J Sports Med. 2020. PMID 33239350. DOI 10.1136/bjsports-2020-102955.