Researchers who analysed data from previous studies found people who exercised experienced a "moderate" reduction in their depressive symptoms compared to those who did other activities, such as using relaxation techniques, or received no treatment.
"This review provides some additional evidence that there may be some benefit (to exercise)," Dr Gillian Mead, the study's senior author from the University of Edinburgh in Scotland, told Reuters Health.
A 2009 review from the Cochrane Collaboration, an international organisation that evaluates medical research, found similar results, but more studies looking at the link between exercise and depression have since been published.
"We'd become aware of some new trials in the area and – in general – the Cochrane review should be kept updated if there is new evidence that may lead to changes," Mead said. About 1 in 10 Americans reports being depressed, according to the US Centres for Disease Control and Prevention.
The most popular treatments for depression include anti-depressant medications and psychotherapy. Mead and her fellow researchers write in The Cochrane Library, however, that many people prefer alternative treatments, and some doctors recommend exercise as a potential option.
Single measurement for comparison
For the new review, they searched databases for all medical trials conducted through March 2013 that compared exercise among adults with depression to other activities or no treatment. Overall, the researchers were able to combine data from 35 trials that included 711 people who were randomly assigned to an exercise program and 642 who were randomized to comparison groups. Because the studies used various scales to assess depression, they converted the results into a single measurement to compare people in exercise and non-exercise groups.
Using that measurement, a difference between groups of 0.2 represents a small effect, 0.5 a moderate effect and 0.8 a large effect. Mead's team found a 0.62-point difference in depressive symptoms favouring people who exercised.
In one of the included trials from 2007, for instance, researchers found 45% of people who took part in supervised exercise no longer met the criteria for depression after four months, compared to 31% taking an inert placebo pill.
In another trial from 2002, 55% of older people experienced a significant decline in depression symptoms after 10 weeks of exercise, compared to 33% who attended informational talks during that time.
The difference between groups, however, was greatly diminished when the review authors only analysed data from the six trials that were considered high quality. Still, exercise appeared to reduce depressive symptoms as much as psychotherapy or antidepressant medications. But Mead cautioned that those findings are only based on data from a small number of trials. "One has to be careful saying it was as effective as other therapies," she said.
More information needed
She added that it's still unknown how exercise affects depression."There are lots of ideas about potential mechanisms, but I don't think there is enough evidence in the literature that one mechanism applies more than another," Mead said.
The researchers were also unable to say which type of exercise is best, but Mead said previous reviews have recommended people choose an activity that they'll stick with over the long run.
"Once people are prescribed exercise or they choose exercise, the big challenge is to make the exercise real," Dr Madhukar Trivedi, who has studied the effect of exercise on depression but wasn't involved with the new research, told Reuters Health.
"If the recommendation from the treating clinician is that you should be exercising with some frequency and intensity... it's important that the patient follow that regimen week after week," Trivedi, a professor of psychiatry at the UT South-western Medical Centre in Dallas, said.
Michel Lucas, who was also not involved with the new review but has studied the topic before, said studies tend to show a dose-response relationship between exercise and depression.
"The dose is very important. If you're walking at a very slow pace, this has no effect," Lucas, a visiting scientist at the Harvard School of Public Health in Boston, told Reuters Health.