The Cardio Prescription
If you tell a doctor, therapist, or personal trainer that you feel anxious, the advice follows a familiar script: go for a run, try yoga, sign up for a cycling class, maybe consider meditation. The clinical logic is well-rehearsed, because aerobic exercise boosts endorphins, lowers cortisol, and has decades of evidence behind it. The American College of Sports Medicine centers its mental health guidelines on aerobic activity, and most research reviews on exercise and anxiety follow the same playbook. Nobody in that chain of advice says go deadlift.
But in 2017, a team at the University of Limerick collected every randomized controlled trial that had tested whether lifting weights reduces anxiety, pooling sixteen studies and 922 participants into a single quantitative synthesis. The pooled result was unambiguous: resistance exercise training significantly reduced anxiety symptoms, producing a Hedges' d effect size of 0.31 that sits squarely in the range where SSRIs and cognitive behavioral therapy typically land for anxiety treatment. The finding held across ages, sexes, and workout protocols, yet it remains one of the least-discussed results in exercise science.
What They Actually Did
Brett Gordon and his colleagues searched five databases for published RCTs that assigned participants to either a resistance training program or a non-active control condition and measured anxiety using validated instruments before and after the intervention. They identified 16 qualifying studies involving 922 people with an average age of 43, about two-thirds of whom were women, drawn from populations as diverse as healthy college students, older adults, people with fibromyalgia, and cardiac rehabilitation patients. The studies varied widely: some ran for six weeks while others lasted six months, some used machines while others relied on free weights, and intensities ranged from light to near-maximal.
Despite all this variation, the pooled effect was statistically consistent, with resistance training reducing anxiety at p < 0.001 and showing low heterogeneity across studies (I² = 28.3%). Sampling error alone explained 77.7% of the variance between results, which means the sixteen studies mostly agreed with each other about the direction and magnitude of the anxiety-reducing effect. Whatever mechanism links barbell work to lower anxiety, it appears to operate reliably across populations and workout configurations.
The Effect That Rivals Medication
The overall effect size was d = 0.31 across all participants, but among healthy people without a diagnosed physical or mental illness, the effect jumped to d = 0.50 (95% CI: 0.22–0.78), a magnitude considered clinically meaningful by standard psychiatric benchmarks. Meta-analyses of SSRIs for generalized anxiety disorder typically report effect sizes between d = 0.2 and d = 0.4 when compared against placebo, and cognitive behavioral therapy falls in a similar range, meaning resistance training produces anxiety reduction of comparable magnitude to the interventions clinicians already prescribe.
Gordon made the comparison explicit to Reuters: "The effect size of these reductions is comparable to that of frontline treatments such as medication and psychotherapy," adding that resistance training "is a low-cost behavior with minimal risk." The National Institute of Mental Health estimates that 19.1% of American adults experience an anxiety disorder in any given year, roughly 50 million people. If even 10% added resistance training twice a week and experienced the average observed effect, approximately 5 million people would see a clinically meaningful reduction in anxiety symptoms without a single prescription.
The 2024 Twist
The Gordon meta-analysis did not directly compare resistance training to aerobic exercise, but a 2024 systematic review by Azevedo and colleagues did, separating studies by exercise mode and analyzing effects on anxiety independently. The results inverted the conventional hierarchy: resistance and mixed-mode exercise showed a statistically significant reduction in anxiety symptoms (SMD = -0.83, p = 0.005), while aerobic exercise alone fell short of significance (SMD = -0.56, p = 0.090). The between-group difference was not statistically significant, so we cannot claim resistance training is definitively superior to cardio for anxiety, but the direction runs exactly opposite to what most clinicians and clinical guidelines would predict.
The Strongest Case Against
The most serious limitation is the control condition: all sixteen trials compared resistance training to non-active controls such as waitlists, usual care, or no exercise at all, which means the studies demonstrate that lifting weights beats doing nothing rather than proving it beats running, swimming, or yoga. The d = 0.31 effect could shrink or vanish entirely in a head-to-head comparison with aerobic training.
The "comparable to medication" framing also relies on a cross-study comparison rather than a direct trial, because SSRIs are benchmarked against pill placebos rather than against inactivity, and comparing an exercise effect measured against a waitlist to a drug effect measured against a sugar pill is not the same as randomizing people to barbells versus sertraline. No such trial exists. The sample size also warrants caution: while 922 participants across 16 RCTs is reasonable for exercise science, it is modest compared to pharmaceutical efficacy trials with tens of thousands of enrollees.
What We Didn't Prove
This evidence does not establish that resistance training is superior to aerobic exercise for anxiety, because the subgroup comparison from Azevedo et al. is suggestive but not conclusive, with the between-group difference falling short of statistical significance. We also do not know the optimal dose: the Gordon meta-analysis found no significant moderation by workout frequency, duration, or intensity, which could mean any amount helps or could reflect insufficient statistical power in a pooled sample of this size.
The mechanism linking resistance training to reduced anxiety remains unclear, with proposed pathways including increased self-efficacy from measurable strength gains, neurobiological changes in brain-derived neurotrophic factor, improved sleep quality, and the social reinforcement of gym environments, none of which has been confirmed as the primary driver. People who volunteer for resistance training research may also differ from the general population in motivation or baseline fitness, and this selection bias could inflate the observed effect.
The Bottom Line
The evidence base for resistance training as an anxiolytic is not yet as deep as the evidence for aerobic exercise, but it is substantial, it is growing, and the effect sizes sit in the same neighborhood as the treatments clinicians already prescribe. If you already lift weights, this is a reason you might not have known you were doing something measurably good for your mental health. If you hate running but can tolerate a barbell, the research says that counts.
What You Can Do
- Start with two sessions per week. The WHO recommends at least two days of muscle-strengthening activity, and the Gordon meta-analysis found that training frequency did not moderate the anxiolytic effect, so you do not need to live in the gym to benefit.
- Keep intensity moderate. Studies using 60–70% of one-rep max or 8–12 repetitions showed robust anxiety-reducing effects, meaning you do not need to train to failure or chase personal records to see a mental health benefit.
- Track how you feel, not just what you lift. A simple mood log before and after sessions can make the anxiolytic effect visible, because it tends to appear within the first weeks of consistent training.
- Don't abandon other treatments. Resistance training is a complement rather than a replacement for therapy or medication; talk to your provider before adjusting any existing treatment plan.
- Consider it a gateway. If cardio feels punishing, resistance training offers an alternative entry point into exercise-based mental health benefits with a different skill curve and reward structure.