Exercise

What the evidence actually shows

In adults with diagnosed anxiety disorders, several forms of structured exercise reduced anxiety severity compared with a waiting list, at moderate certainty. Resistance training (strength work) and mind-body exercise (such as yoga and tai chi) had the strongest signals, with aerobic exercise also showing a positive, moderate-certainty effect.

Those effects are big enough to reject the idea that exercise is only good for general wellbeing. It is a real treatment signal. But the studies mostly compared exercise against a waiting list rather than against an active treatment, and only a few were at low risk of bias, so the numbers describe an average, not a guaranteed personal result.

Why "resistance training is best" overstates it

One ranking put resistance training at the top. It is tempting to turn that into "so do strength training first." The evidence does not support that leap. A ranking like this is a probability of where a modality tends to land, not proof that it clearly beats the others. The resistance-training result came from a small set of studies, and the lead did not hold up when weaker trials were removed; mind-body exercise moved ahead under some analyses.

The stable message is that several feasible types of exercise can reduce anxiety. Which one to choose is better decided by safety, preference, access, and support than by a ranking that shifts when the weakest studies are set aside.

Diagnosed anxiety and everyday anxiety symptoms are not the same

The clearest evidence is for people with a diagnosed anxiety disorder. For people with elevated anxiety symptoms but no formal diagnosis, the picture was more mixed: only mind-body exercise reached significance, at low certainty. So the strong reading applies to diagnosed disorders, and the symptom-only case should be described more cautiously.

Where mind-body exercise ends and mindfulness begins

Yoga and tai chi combine movement with breathing and mental focus, so they overlap with mindfulness practices. Here they count as exercise, because the intervention is structured physical activity. Their exercise evidence does not transfer to meditation-only practice or to the named mindfulness and acceptance courses, which rest on different studies. If movement is the treatment, it belongs on this page; if a structured meditation course is the treatment, see mindfulness and acceptance.

What taking part involves

"Just exercise" undersells it. The trials tested structured, repeated programs, commonly around three sessions a week for about eight weeks, not a one-off instruction to move more. Program design, progression, and sticking with it are part of the treatment. Practical needs can include:

• equipment, coaching, or a safe space, especially for resistance training;
• time, transport, cost, privacy, and comfort in group settings, which can be a real barrier in social anxiety;
• screening or adaptation for heart, joint, pregnancy, or other health considerations;
• attention to panic-related fear of a fast heartbeat or breathlessness, which can itself make exercise feel threatening at first.

How it fits with other care

Guidelines place structured exercise as a complement or add-on rather than a stand-alone replacement. For panic disorder in particular, guidance frames it as something added to standard treatment rather than used alone. It is not merely a footnote, because the diagnosed-disorder signal is real. It is also not a proven substitute for CBT, exposure therapy, or medication, because the studies were not built to show that.

What the evidence does not settle

• The single best type of exercise for an individual.
• How long the benefit lasts beyond short study endpoints.
• How much benefit survives when exercise is compared against an active treatment rather than a waiting list.
• How well the structured trial programs translate to real-world access and adherence.