Exercise is, by the available evidence, the most consistently underprescribed psychiatric intervention. The data span depression, anxiety, cognitive preservation, sleep architecture, executive function, and dementia risk reduction — and the effect sizes for moderate-intensity aerobic exercise in mild-to-moderate depression are comparable to those of first-line antidepressants in meta-analyses, with substantially better side-effect profiles. The longevity-psychiatry frame takes this seriously: exercise is not an adjunct or a wellness suggestion, it is a primary intervention with mechanistic depth and quantifiable dose-response. The clinical task is to prescribe it with the same specificity used for medication — modality, intensity, frequency, duration, progression — and to follow up on adherence the way a clinician follows up on a titration.
The mechanism runs primarily through brain-derived neurotrophic factor (BDNF) and a small family of related signals. Aerobic exercise reliably elevates peripheral and central BDNF, particularly in the hippocampus, where BDNF supports synaptic plasticity, long-term potentiation, and adult neurogenesis in the dentate gyrus. Erickson's randomized trial in older adults showed that one year of aerobic exercise produced a 2% increase in hippocampal volume — reversing the roughly 1–2% annual atrophy expected at that age — while a stretching control group continued to lose volume. The accompanying BDNF elevation and memory improvement made the mechanism plausible rather than merely correlational. IGF-1, VEGF, and lactate signaling contribute additional pathways; the picture is multi-mechanism but BDNF is the central node.
The cognitive evidence is stronger than most psychiatrists appreciate. The FINGER trial (multidomain lifestyle intervention with exercise as a major component) showed cognitive benefit in at-risk older adults. The SPRINT-MIND trial showed cognitive benefit from intensive blood pressure control, and exercise contributes to that pathway. Observational data link cardiorespiratory fitness to dementia risk reduction with hazard ratios in the 0.5–0.7 range for the fittest tertile versus the lowest. The mechanistic plausibility (BDNF, vascular, metabolic, inflammatory pathways converge on cognition) and the dose-response (more fitness, more benefit, with diminishing returns at the high end) place exercise as one of the most robust modifiable factors for cognitive trajectory across decades.
The mood evidence supports exercise as monotherapy or augmentation in mild-to-moderate depression. Meta-analyses consistently show effect sizes in the 0.4–0.7 range for aerobic exercise versus inactive controls, with comparable results for resistance training. The SMILE trials by Blumenthal showed exercise comparable to sertraline at four months and superior at long-term follow-up because the exercise group sustained the behavior change. Anxiety responds similarly, particularly generalized anxiety and panic, with both aerobic and resistance modalities producing measurable benefit. The clinical move is to offer exercise as a real treatment option, not as a should-also-do-this addendum after the prescription is written.
The prescription is specific. Aerobic component: 150 minutes per week of moderate-intensity activity (brisk walking, cycling, swimming) or 75 minutes of vigorous activity (running, intervals), or a mix. Resistance component: two sessions weekly covering major muscle groups. Intensity matters — the cognitive and mood benefits scale with intensity within tolerable ranges, and zone-2 work plus occasional higher-intensity intervals is the pattern most consistently associated with cardiorespiratory fitness improvement. Frequency matters more than session duration for adherence; daily short sessions outperform weekly long ones in most patients. Progression matters — the prescription that does not advance produces plateaued benefit.
The barriers are real and addressable. Time, energy, depression itself as a barrier to behavior change, joint pain, fear of injury, lack of access. The clinical conversation explores what is realistically available — does the patient have a walking-safe neighborhood, access to a pool, a gym membership, a friend or family member who would join? — and writes a starting prescription matched to current capacity rather than to an idealized program. The most common error is prescribing a level of activity the patient cannot start; the second is failing to follow up and adjust. The most effective version of this work treats exercise like a medication that needs titration and ongoing management. The clinician who does not engage with exercise is leaving one of the most powerful tools in the psychiatric toolkit unused.