Exercise and Physical Activity for Mental Health: Dose-Response Relationships, Neurobiological Mechanisms, and Condition-Specific Efficacy Data

Physical exercise has emerged as one of the most robust transdiagnostic interventions in mental health care. Once relegated to lifestyle advice offered as an afterthought, exercise now occupies a position of genuine clinical significance backed by hundreds of randomized controlled trials, large-scale meta-analyses, and mechanistic neuroscience research. The evidence base has matured to the point where major clinical guidelines — including those from the National Institute for Health and Care Excellence (NICE), the American Psychiatric Association (APA), and the Royal Australian and New Zealand College of Psychiatrists (RANZCP) — explicitly recommend structured physical activity as part of treatment for depression, anxiety disorders, and other psychiatric conditions.

What distinguishes the contemporary evidence from earlier enthusiasm is specificity. We now have data on dose-response relationships (how much exercise is needed, and whether more is always better), head-to-head comparisons with pharmacotherapy and psychotherapy, neurobiological mechanisms that go well beyond the outdated "endorphin hypothesis," and condition-specific effect sizes that allow clinicians to make informed treatment decisions. This article synthesizes this evidence with the rigor and specificity required for clinical application.

Critically, exercise is not a panacea. Its efficacy varies substantially across diagnoses, and effect sizes are moderated by exercise type, intensity, supervision, and patient characteristics. Understanding where exercise works well, where it serves as a useful adjunct, and where the evidence is weak or absent is essential for responsible clinical integration.

Monoamine Neurotransmitter Systems

Acute exercise robustly increases synaptic availability of serotonin (5-HT), norepinephrine (NE), and dopamine (DA) through multiple pathways. Tryptophan hydroxylase activity increases with exercise, enhancing serotonin synthesis. Chronic exercise upregulates serotonin transporter expression in the dorsal raphe nuclei and increases 5-HT1A autoreceptor desensitization, effects that parallel the mechanism of selective serotonin reuptake inhibitors (SSRIs). The noradrenergic effects — mediated through locus coeruleus activation — contribute to improvements in arousal regulation, attention, and stress responsivity. Dopaminergic upregulation, particularly in the mesolimbic pathway, may underlie improvements in anhedonia, motivation, and reward processing observed in exercising patients with depression.

Neurotrophic Factors and Neuroplasticity

One of the most well-established mechanisms involves brain-derived neurotrophic factor (BDNF). Aerobic exercise produces reliable increases in peripheral BDNF concentrations, with a meta-analysis by Szuhany, Bugatti, and Otto (2015) reporting a moderate effect size (Hedges' g = 0.46) for the acute BDNF response to single exercise bouts. BDNF supports hippocampal neurogenesis, long-term potentiation, and synaptic plasticity — processes that are impaired in major depressive disorder (MDD). Erickson et al. (2011) demonstrated that a 12-month aerobic exercise program increased hippocampal volume by approximately 2% in older adults, effectively reversing 1-2 years of age-related atrophy. This hippocampal volumetric increase correlated with improvements in spatial memory and with increased serum BDNF.

Other neurotrophic factors modulated by exercise include vascular endothelial growth factor (VEGF), insulin-like growth factor 1 (IGF-1), and fibroblast growth factor 2 (FGF-2), all of which contribute to angiogenesis, neurogenesis, and neuronal survival in the hippocampus and prefrontal cortex.

HPA Axis Regulation and Inflammatory Pathways

Exercise normalizes hypothalamic-pituitary-adrenal (HPA) axis dysregulation, a hallmark of depression and PTSD. Regular exercise reduces basal cortisol levels and blunts cortisol reactivity to psychological stressors. In parallel, exercise exerts potent anti-inflammatory effects: it reduces circulating levels of pro-inflammatory cytokines (IL-6, TNF-α, CRP) while increasing anti-inflammatory myokines released from contracting skeletal muscle. Given that approximately 25-30% of patients with MDD show elevated inflammatory biomarkers, this mechanism may be particularly relevant for the inflammatory subtype of depression.

Endocannabinoid System

The "runner's high" — once attributed to endorphins, which do not readily cross the blood-brain barrier — is now better explained by exercise-induced increases in endocannabinoid signaling, particularly anandamide (AEA). Circulating AEA levels increase significantly after moderate-intensity aerobic exercise (typically at 70-80% maximum heart rate), producing anxiolytic and mildly euphoric effects through CB1 receptor activation in the prefrontal cortex and amygdala.

Psychological Mechanisms

Beyond neurobiology, exercise operates through several well-supported psychological mechanisms: self-efficacy enhancement (consistent with Bandura's theory, as mastery experiences in physical tasks generalize to broader competence beliefs), behavioral activation (exercise disrupts avoidance patterns and increases engagement with positively reinforcing activities, a core mechanism of behavioral therapy for depression), distraction and rumination interruption (exercise demands attentional resources that compete with depressive rumination), and social interaction (group-based exercise provides social contact that counteracts isolation). The relative contribution of each mechanism varies by condition and exercise modality.

The dose-response relationship between exercise and mental health outcomes is nonlinear and condition-specific. A landmark analysis by Choi et al. (2019), using Mendelian randomization with data from over 600,000 individuals, established that objectively measured physical activity has a causal protective effect against depression, with the equivalent of 15 minutes of vigorous activity or 1 hour of moderate activity per day associated with a 26% reduction in depression risk.

Intensity

For depression, moderate-to-vigorous intensity exercise (defined as 50-85% of maximal heart rate or a rating of perceived exertion of 12-16 on the Borg scale) produces the largest effect sizes. The DEMO-II trial (Helgadóttir et al., 2016) compared light, moderate, and vigorous exercise and found that moderate and vigorous groups showed significantly greater reductions in depression severity than the light exercise group, though both active conditions outperformed the control. Importantly, high-intensity interval training (HIIT) has shown comparable or superior effects to moderate continuous training in some studies, potentially offering a time-efficient alternative.

For anxiety disorders, moderate-intensity exercise appears optimal. High-intensity exercise can transiently increase arousal and anxiety in some patients — particularly those with panic disorder — making graduated exposure to physical arousal sensations an important clinical consideration.

Frequency and Duration

The most common effective protocol in clinical trials involves 3-5 sessions per week, 30-60 minutes per session, for 8-16 weeks. A dose-response meta-analysis by Pearce et al. (2022) in the British Journal of Sports Medicine, analyzing 1,039 RCTs with over 128,000 participants, found that the antidepressant benefits of exercise plateau at approximately 1,000 MET-minutes per week (roughly equivalent to 150 minutes of brisk walking or 75 minutes of running), aligning with WHO physical activity guidelines. However, benefits began accruing at doses well below this threshold — even 500 MET-minutes per week produced clinically meaningful improvements.

Exercise Type

Both aerobic exercise (running, cycling, swimming) and resistance training (weightlifting, resistance bands) demonstrate antidepressant effects. A meta-analysis by Gordon et al. (2018) specifically examined resistance exercise training (RET) and found a significant antidepressant effect with an overall effect size of ES = 0.66 (95% CI: 0.48-0.83), with effects present regardless of health status or whether participants had clinically significant depressive symptoms. Notably, RET effects did not depend on improvements in strength, suggesting mechanisms beyond physical conditioning.

Yoga — combining physical postures, breathwork, and meditative components — has a separate evidence base. Cramer et al. (2017) reported a moderate effect for yoga versus control conditions in depression (SMD = -0.56). Yoga appears particularly effective for anxiety-related conditions, possibly due to its emphasis on interoceptive awareness and parasympathetic activation through regulated breathing.

Mind-body exercises such as tai chi and qigong have shown smaller but significant effects, primarily in older adult populations and for generalized anxiety symptoms.

Major Depressive Disorder (MDD)

The evidence for exercise in depression is the strongest of any psychiatric condition. The Pearce et al. (2022) umbrella review reported an overall effect size of SMD = -0.43 (95% CI: -0.58 to -0.29) for exercise versus control conditions, with effects increasing for supervised interventions (SMD = -0.62). The number needed to treat (NNT) for clinically significant improvement is approximately 4-5, comparing favorably with both antidepressants (NNT = 5-8 in the STAR*D trial) and psychotherapy.

The landmark SMILE trial (Blumenthal et al., 1999) randomized 156 older adults with MDD to supervised aerobic exercise (3 sessions/week at 70-85% HRR for 16 weeks), sertraline (50-200 mg), or combination treatment. At 16 weeks, remission rates were comparable: 60.4% for exercise, 65.5% for sertraline, and 68.8% for combination. At 10-month follow-up, the exercise group had significantly lower relapse rates (8% vs. 38% for medication alone, p = 0.01), suggesting exercise may confer durable protective effects. The TREAD trial (Trivedi et al., 2011) further demonstrated that higher doses of exercise (16 kcal/kg/week) produced greater remission rates (28.3%) than lower doses (15.5%) in SSRI partial responders, supporting exercise as an augmentation strategy.

Anxiety Disorders

Meta-analytic data support a moderate effect of exercise on anxiety symptoms. Stubbs et al. (2017) reported an effect size of SMD = -0.58 (95% CI: -0.80 to -0.35) for exercise versus control in people with anxiety disorders. Effects were larger for diagnosed anxiety disorders than for subclinical anxiety. For generalized anxiety disorder (GAD), aerobic exercise shows promise but head-to-head data against first-line treatments (CBT, SSRIs/SNRIs) are limited. For panic disorder, the evidence is nuanced: exercise can serve as an interoceptive exposure to panic-like sensations (elevated heart rate, sweating, breathlessness), which may be therapeutically beneficial when framed appropriately but may initially provoke anxiety.

Post-Traumatic Stress Disorder (PTSD)

Exercise for PTSD is an emerging evidence area. Rosenbaum et al. (2015) conducted a meta-analysis finding significant reductions in PTSD symptoms with exercise interventions (SMD = -0.35). The mechanisms may involve extinction learning facilitation (exercise increases BDNF, which enhances fear extinction consolidation), reductions in hyperarousal through parasympathetic retraining, and improved sleep quality. Exercise appears most effective as an adjunct to trauma-focused therapies (CPT, PE) rather than a standalone treatment.

Schizophrenia and Psychotic Disorders

Firth et al. (2015) conducted a meta-analysis of exercise interventions in schizophrenia, finding significant improvements in total psychiatric symptoms (SMD = -0.39), positive symptoms, negative symptoms (especially when supervised), and cognitive function. Aerobic exercise at moderate-to-vigorous intensity for ≥90 minutes per week significantly improved cardiorespiratory fitness (VO2max) and global cognition, including working memory and social cognition. Given the profound cardiometabolic burden of antipsychotic medications, exercise serves dual psychiatric and physical health objectives in this population.

ADHD

Acute bouts of moderate-intensity aerobic exercise improve executive function, attention, and inhibitory control in both children and adults with ADHD, likely through catecholaminergic upregulation in prefrontal circuits. Effect sizes for cognitive outcomes are moderate (d = 0.50-0.70). However, exercise has not been shown to be a sufficient standalone treatment for ADHD and is best conceptualized as an adjunctive strategy that may reduce medication requirements.

Conditions with Weaker or Insufficient Evidence

Bipolar disorder: Limited RCT data exist. Exercise appears safe during euthymic and depressive phases but requires caution during manic episodes, where excessive activity may be symptom-driven rather than therapeutic. Eating disorders: Exercise prescription is clinically complex, as compulsive exercise is a core feature of anorexia nervosa and bulimia nervosa. Exercise should not be prescribed in underweight patients or those with active compulsive exercise patterns. Substance use disorders: Emerging evidence suggests moderate benefits for reducing cravings and improving mood, but effect sizes are small and dropout rates in exercise trials are high in this population. OCD: Very limited data; aerobic exercise may reduce general distress but has not been shown to specifically reduce obsessional or compulsive symptoms.

The clinical utility of exercise is best understood in direct comparison with established treatments. Several studies and meta-analyses provide this data, though methodological challenges (blinding is impossible for exercise) temper interpretation.

Exercise vs. Antidepressant Medication

The SMILE trial, as noted, found comparable remission rates between exercise and sertraline at 16 weeks for MDD in older adults. A more recent network meta-analysis by Noetel et al. (2024), published in the BMJ and encompassing 218 RCTs with 14,170 participants, found that walking/jogging, yoga, strength training, and mixed aerobic exercise were all more effective than SSRIs for depression, though confidence intervals overlapped for several comparisons. Walking/jogging had the highest effect size among exercise modalities (SMD = -0.62), followed by yoga (SMD = -0.55) and strength training (SMD = -0.49). SSRIs showed an effect of approximately SMD = -0.26 in this network.

These findings should be interpreted carefully. Most exercise trials have high risk of performance bias (participants know they are exercising), which likely inflates effect sizes compared to double-blind pharmacotherapy trials. Additionally, exercise trials often include participants with mild-to-moderate depression, whereas medication trials more frequently enroll severe cases.

Exercise vs. Cognitive Behavioral Therapy (CBT)

Direct comparisons are scarce. Indirect evidence from network meta-analyses suggests that exercise and CBT have broadly similar effect sizes for mild-to-moderate depression (both typically in the SMD range of -0.40 to -0.70). For anxiety disorders, CBT retains a clear advantage as the first-line treatment, with more robust evidence across specific diagnoses (GAD, social anxiety disorder, panic disorder, OCD). Exercise may serve as an effective adjunct to CBT, particularly for patients who are physically deconditioned or who have prominent somatic anxiety symptoms.

Exercise as Augmentation

Perhaps the strongest clinical case for exercise is as an augmentation strategy for partial responders to first-line treatments. The TREAD trial demonstrated that adding exercise to SSRI treatment in partial responders produced additional remission. Mota-Pereira et al. (2011) found that adding a walking program to treatment-resistant depression (defined as non-response to at least two adequate antidepressant trials) produced response in 21% and remission in 26% of previously treatment-resistant patients — a notable finding for a population with characteristically poor outcomes.

Not all patients respond equally to exercise interventions, and identifying moderators of treatment response is a clinical priority.

Baseline Severity

Paradoxically, several meta-analyses report larger effect sizes in populations with higher baseline symptom severity. This may reflect greater room for improvement (regression to the mean) or a genuine signal that exercise's neurobiological effects are more impactful when baseline neurotransmitter and neurotrophic function is more impaired. The Pearce et al. (2022) analysis found that effect sizes were largest for individuals with depression, anxiety, and psychological distress compared to the general population.

Supervision and Accountability

Supervised exercise consistently outperforms unsupervised exercise in clinical trials. Supervised interventions in the Noetel et al. (2024) BMJ review showed significantly larger effects, likely reflecting adherence support, proper intensity calibration, and the therapeutic relationship with the exercise professional. This has direct implications for clinical practice: simply advising a patient to "get more exercise" is substantially less effective than prescribing a structured, supervised program.

Exercise Modality Preferences

Patient preference for exercise type moderates adherence and, consequently, outcomes. Allowing patients to choose their preferred modality does not appear to diminish efficacy and may improve long-term adherence. The Noetel et al. (2024) analysis found comparable effects across modalities (walking, running, yoga, strength training, dance), suggesting that the best exercise is the one the patient will actually perform.

Age

Effect sizes for exercise in depression are generally consistent across age groups, though exercise may be particularly impactful in older adults, where it simultaneously addresses physical deconditioning, social isolation, cognitive decline, and cardiometabolic risk — all of which contribute to late-life depression. In adolescents, effect sizes tend to be slightly smaller, potentially due to adherence challenges and the complexity of adolescent depression.

Inflammatory Biomarkers

Emerging evidence suggests that patients with elevated baseline inflammatory markers (CRP, IL-6) may preferentially respond to exercise, consistent with exercise's anti-inflammatory mechanism. This aligns with the broader precision psychiatry effort to match treatments to biological subtypes.

Adherence and Dropout

Dropout rates in exercise trials for depression average 18-20%, comparable to psychotherapy trials and lower than some pharmacotherapy trials. However, real-world adherence outside of trial contexts is substantially lower. Major barriers include fatigue (a core symptom of depression), low motivation and anhedonia, physical health limitations, and lack of access to facilities or supervision.

Aerobic Exercise Protocol for Depression (Based on Clinical Trial Evidence)

• Frequency: 3-5 sessions per week
• Intensity: Moderate to vigorous (50-85% heart rate reserve, or RPE 12-16 on Borg 6-20 scale)
• Duration: 30-45 minutes per session (including 5-minute warm-up and 5-minute cool-down)
• Type: Walking, jogging, cycling, swimming, or elliptical training — patient preference guides selection
• Duration of program: Minimum 8 weeks; 12-16 weeks for optimal effects; maintenance recommended indefinitely
• Supervision: Preferably supervised by an exercise physiologist, qualified personal trainer, or physical therapist, especially during initial weeks

Resistance Training Protocol for Depression

• Frequency: 2-3 sessions per week
• Intensity: Moderate (60-80% of one-repetition maximum)
• Volume: 8-12 repetitions per set, 2-3 sets per exercise, 6-10 exercises targeting major muscle groups
• Progression: Progressive overload (gradual increase in weight/resistance) is incorporated over weeks
• Duration of program: Minimum 8 weeks

Yoga Protocol for Anxiety and Depression

• Frequency: 2-3 sessions per week
• Duration: 60-90 minutes per session
• Style: Hatha yoga or Iyengar yoga are most studied; Bikram (hot) yoga has limited psychiatric evidence
• Components: Asanas (postures), pranayama (breathwork), and brief meditation/relaxation
• Duration of program: 8-12 weeks

Graduated Approach for Severely Deconditioned Patients

Patients with severe depression, those on sedating medications, or those with significant physical deconditioning may not tolerate standard protocols initially. A graduated approach is recommended:

• Weeks 1-2: 10-15 minutes of light walking, 3 times per week
• Weeks 3-4: Increase to 20 minutes, introduce moderate intensity
• Weeks 5-8: Gradually progress to 30-45 minutes at moderate-to-vigorous intensity
• Motivational support: Motivational interviewing techniques, behavioral activation scheduling, and self-monitoring (step counting, workout logs) significantly improve adherence

Exercise is generally safe and has a favorable side-effect profile compared to pharmacotherapy. However, it is not without risks and limitations.

Physical Risks

• Musculoskeletal injury: The most common adverse event, particularly in previously sedentary individuals. Risk is mitigated by proper warm-up, gradual progression, and supervision.
• Cardiac events: Rare but serious. Patients over 50 or those with cardiovascular risk factors should undergo medical clearance (e.g., PAR-Q+ screening or physician assessment) before initiating vigorous exercise.
• Rhabdomyolysis: Very rare, primarily associated with extreme or unaccustomed exertion. Relevant for patients on statins or certain psychotropic medications that may increase risk.
• Overtraining syndrome: Excessive exercise without adequate recovery can paradoxically worsen mood, impair sleep, and elevate cortisol — counteracting the therapeutic intent.

Condition-Specific Contraindications

• Anorexia nervosa and active eating disorders: Exercise should generally be restricted or carefully controlled during acute treatment and weight restoration phases. Compulsive exercise is a symptom, not a treatment, in this context.
• Active mania/hypomania: Increased physical activity may be symptom-driven and can perpetuate sleep deprivation. Exercise is not contraindicated per se but requires clinical judgment and monitoring.
• Acute suicidality: Exercise should not delay implementation of safety planning, medication, or intensive psychotherapy. It is not appropriate as a primary intervention for acute suicidal crisis.
• Severe psychomotor retardation: Patients in the throes of melancholic depression may be functionally unable to initiate exercise without substantial support and graduated programming.

Methodological Limitations in the Evidence Base

The exercise-mental health literature has well-recognized limitations: impossible blinding (participants know they are exercising), expectancy effects (people may expect exercise to improve mood), attention control confounding (exercise groups often receive more therapeutic contact than control groups), and publication bias (positive results are more likely to be published). When exercise is compared against active controls (e.g., stretching, relaxation) rather than waitlist controls, effect sizes are smaller, typically in the range of SMD = -0.20 to -0.35 for depression.

Children and Adolescents

A meta-analysis by Bailey et al. (2018) found a small-to-moderate effect of exercise on depression in youth (SMD = -0.37). School-based programs that integrate physical activity into daily routines show better adherence than clinic-based programs. For adolescents with mild-to-moderate depression, NICE guidelines recommend structured group physical activity as a first-line intervention before medication. Team sports and group-based activities may provide additional social benefits. The recommended dose is at least 60 minutes of moderate-to-vigorous physical activity daily — consistent with general pediatric guidelines — with structured exercise programs 3 times per week for therapeutic purposes.

Older Adults (≥65 years)

Exercise is especially valuable in this population for several reasons: (1) late-life depression is common (prevalence 10-15% in community-dwelling older adults) and often undertreated; (2) older adults are more vulnerable to antidepressant side effects (falls from orthostatic hypotension, hyponatremia from SSRIs, anticholinergic burden); (3) exercise simultaneously addresses cognitive decline, falls risk, cardiovascular disease, and social isolation. The Erickson et al. (2011) study's hippocampal volume findings are particularly relevant here. Combined aerobic and resistance training programs appear optimal. Tai chi and yoga are effective and well-tolerated alternatives for frail older adults or those with mobility limitations. Balance training should be incorporated to reduce falls risk.

Perinatal Mental Health

Exercise during pregnancy is both safe and beneficial for perinatal depression and anxiety. ACOG (American College of Obstetricians and Gynecologists) recommends at least 150 minutes per week of moderate-intensity aerobic activity during pregnancy in the absence of obstetric contraindications. A meta-analysis by Davenport et al. (2018) found that prenatal exercise reduced the odds of prenatal depression by 67% (OR = 0.33, 95% CI: 0.21-0.53) and postpartum depression by approximately 40%. This is clinically significant given the reluctance of many pregnant patients to take psychotropic medications. Exercise type should avoid supine positions after the first trimester, contact sports, and activities with high fall risk. Walking, swimming, stationary cycling, and prenatal yoga are safe and effective options.

Patients with Serious Mental Illness (SMI)

Individuals with schizophrenia, schizoaffective disorder, and severe bipolar disorder have a life expectancy 15-20 years shorter than the general population, primarily due to cardiovascular disease exacerbated by antipsychotic-induced metabolic syndrome. Exercise programs in this population require higher levels of supervision, motivational support, and accommodation for medication-related sedation and weight gain. Despite these challenges, structured exercise programs show meaningful improvements in both psychiatric symptoms and cardiometabolic risk factors. Dropout can be reduced through integrated care models where exercise programming is embedded within mental health service delivery.

One of the most attractive features of exercise as a mental health intervention is its accessibility and low cost relative to psychotherapy and pharmacotherapy. Walking requires no equipment, no referral, and no insurance authorization. However, this framing can obscure important access barriers.

Cost

Walking and bodyweight exercises are free. Gym memberships range from $10-100+/month in the United States. Supervised exercise programs by qualified professionals (exercise physiologists, personal trainers) cost $40-150/session, comparable to psychotherapy. Insurance coverage for exercise interventions remains very limited in most countries, though some health systems (e.g., the UK's NHS Social Prescribing model, Australia's Exercise & Sports Science Australia referral pathway) are developing reimbursement mechanisms. In the United States, Medicare began covering Diabetes Prevention Programs that include exercise components, and some insurers cover gym memberships through wellness programs.

Qualified Professionals

Accredited Exercise Physiologists (AEPs, ACSM-certified clinical exercise physiologists, or equivalent) are the most qualified professionals to deliver exercise interventions for psychiatric populations. They hold at minimum a bachelor's degree in exercise science with clinical training. In the UK, the British Association of Sport and Exercise Sciences (BASES) provides relevant accreditation. Mental health professionals (psychologists, psychiatrists, social workers) should collaborate with exercise professionals rather than prescribing exercise without knowledge of training principles, contraindications, and progression.

Implementation Models

• Exercise referral schemes: Primary care clinicians refer patients to community exercise programs (well-established in the UK and Australia)
• Integrated care: Exercise programming embedded within mental health services (e.g., gym facilities in psychiatric hospitals or community mental health centers)
• Digital delivery: App-based and telehealth exercise programs have shown preliminary efficacy, particularly for increasing physical activity levels, though effect sizes for mental health outcomes are smaller than in-person supervised programs
• Prescription pads: Standardized exercise prescription tools (e.g., the Exercise is Medicine initiative by the American College of Sports Medicine) enable clinicians to write structured exercise prescriptions specifying type, intensity, duration, and frequency

Barriers to Uptake

Despite strong evidence, exercise remains underutilized in psychiatric practice. Surveys of psychiatrists reveal that while the majority believe exercise is beneficial, fewer than 25% routinely prescribe it with specific parameters. Key barriers include: lack of training in exercise prescription among mental health professionals, lack of reimbursement infrastructure, patient symptom severity (fatigue, anhedonia, psychomotor retardation), and the cultural separation between mental health services and exercise/sports medicine.

The following landmark studies and reviews form the backbone of the exercise-mental health evidence base:

• SMILE Trial (Blumenthal et al., 1999): The first major RCT comparing exercise to antidepressant medication (sertraline) in older adults with MDD. Demonstrated comparable acute efficacy and superior long-term outcomes for exercise.
• TREAD Trial (Trivedi et al., 2011): Demonstrated dose-response effects of exercise as augmentation to SSRI treatment in MDD, supporting higher-dose exercise for greater benefit.
• DEMO-II Trial (Helgadóttir et al., 2016): Compared light, moderate, and vigorous exercise in depression, showing intensity-dependent effects.
• Choi et al. (2019) — Mendelian Randomization Study: Using genetic instruments in 611,583 individuals, established a causal protective effect of physical activity against depression, strengthening causal inference beyond observational data.
• Pearce et al. (2022) — BJSM Umbrella Review: Analyzed 97 systematic reviews covering 1,039 RCTs and 128,000+ participants, confirming exercise as effective for depression, anxiety, and distress with moderate-to-large effects.
• Noetel et al. (2024) — BMJ Network Meta-Analysis: The most comprehensive network meta-analysis to date (218 RCTs, 14,170 participants), placing exercise modalities in a comparative effectiveness framework alongside SSRIs and psychotherapy.
• Gordon et al. (2018) — Resistance Exercise Training Meta-Analysis: Established antidepressant effects of resistance training specifically, with an effect size of 0.66.
• Firth et al. (2015) — Exercise in Schizophrenia Meta-Analysis: Demonstrated benefits for psychiatric symptoms and cognition in psychotic disorders.
• Erickson et al. (2011) — Hippocampal Volume Study: Showed that aerobic exercise increased hippocampal volume and improved memory in older adults, providing critical neuroanatomical evidence.

The evidence supports the following clinical conclusions:

• Exercise is an effective treatment for depression with effect sizes comparable to first-line pharmacotherapy and psychotherapy, particularly for mild-to-moderate severity. NNT is approximately 4-5.
• Exercise is effective for anxiety disorders with moderate effect sizes, though it should generally be positioned as adjunctive to CBT and/or pharmacotherapy for diagnosed anxiety disorders.
• Exercise is a promising adjunct for PTSD, schizophrenia, and ADHD, with emerging evidence supporting neurobiological mechanisms specific to each condition.
• The optimal dose for depression appears to be 150+ minutes per week of moderate-to-vigorous activity, though benefits begin at lower thresholds. Both aerobic exercise and resistance training are effective.
• Supervision matters: Supervised, structured programs outperform advice to exercise independently.
• Exercise should be prescribed with specificity — type, intensity, frequency, duration — analogous to pharmacological prescribing. Vague recommendations to "be more active" are insufficient.
• Exercise is not appropriate as a standalone treatment for severe depression with psychotic features, active suicidality, bipolar mania, or eating disorders with active compulsive exercise.
• The risk-benefit profile is exceptionally favorable: exercise provides cardiometabolic, cognitive, and psychosocial benefits beyond its psychiatric effects, with minimal adverse effects when properly prescribed.

As the field moves toward precision psychiatry, future research should identify biomarkers (inflammatory profile, BDNF genotype, HPA axis function) that predict exercise response, enabling truly individualized exercise prescriptions. In the interim, the current evidence base is more than sufficient to justify routine integration of structured physical activity into mental health treatment planning across most diagnostic categories.