Reduces symptoms of depression, with moderate effects in trials that mostly tested aerobic exercise.

Reduces symptoms of depression, with moderate effects in trials that mostly tested aerobic exercise.

Mindfulness programs modestly reduce depression symptoms, with an effect a little smaller than for anxiety.

EPA-rich fish oil (about 1 g a day) eases depressive symptoms a little; DHA-only products do not.

Bright light lowers depression symptoms in trials, though most tested light boxes rather than morning sunlight.

Lifting weights reduces symptoms of depression in trials, though the studies are small and the best-designed ones show smaller effects.

Small improvement in low mood in healthy adults across 20 trials; effects vary by strain.

Breathing practices reduced depressive symptoms by a small-to-moderate amount across 18 randomised trials, mostly in people without a diagnosis.

Eases depressive symptoms, mainly as an add-on to standard care, based on small trials.

May ease mild to moderate depression, mostly from small trials.

Eases self-rated depression symptoms in small trials, mostly from Iran, but clinician-rated scores did not improve and the newest independent trial found no effect.

In a 12-week trial of 67 adults with depression, a Mediterranean-style diet programme with dietitian support helped more people reach remission than social support alone (32% vs 8%).

Adults taking 7,000 or more steps a day had about a 31% lower risk of developing depression than those taking fewer, in studies that followed people over time.

May ease depressive symptoms a little, mostly in people who are deficient; large trials in healthy adults were null.

May ease low mood in stressed adults, but trials are small and mostly not in clinical depression.

The pooled effect of expressive writing on depression, anxiety and stress was very small, and trials in people with cancer found no effect on depression.

Irregular bed and wake times go with more depression and anxiety symptoms in studies, but this shows a link, not cause.
