Some earlier reports had suggested that antidepressants generally only improve mood in people with severe depression.

But that might be because those studies weren't precise enough to pick up on smaller changes in symptoms that can still make a difference for people with milder forms of the disease, researchers said.

"I think there's a valid concern... that if someone has not-that-severe depression that hasn't lasted that long, maybe it will get better itself or with therapy," said Dr. David Hellerstein, from the New York State Psychiatric Institute and Columbia University, who worked on the study.

Still, he said the question of whether or not to prescribe medication shouldn't necessarily come down to how severe the depression is, but how long symptoms have lasted.

People with "transient depression" that will improve with diet or exercise or after a few weeks of therapy "shouldn't be taking the risk of being on meds," he told Reuters Health.

"But people who have more persistent depression should be evaluated for treatment and medicine should be one of the options," even when the depression is more modest.

Hellerstein and his colleagues collected data from six studies done at the state's psychiatric institute between 1985 and 2000. Those included 825 people with non-severe, long-lasting depression enrolled in trials that compared symptoms with antidepressant treatment versus a placebo.

In three of the six studies, patients taking an antidepressant improved more on a widely-used scale of depression symptoms and severity than those taking a placebo, and in four studies, a higher percentage of patients taking antidepressants went into remission, meaning they were no longer considered to have clinically-significant depression.

Depending on the particular drug and study, the researchers calculated that between three and eight people with non-severe depression would have to be treated with an antidepressant for one to benefit substantially from it.

That, they wrote in the Journal of Clinical Psychiatry, is "a range considered by researchers as sufficiently robust to recommend treatment."

The drugs tested in those studies included Prozac, as well as older and now less-popular medications known as monoamine oxidase inhibitors and tricyclic and tetracyclic antidepressants. It's hard to know how well the findings would apply for newer antidepressants, the researchers said.

The results don't mean that everyone with mild depression should be on an antidepressant, a psychiatrist not involved in the study pointed out.

"People with these milder depressions also respond well to counseling and psychotherapy and can respond well to exercise," said Dr. Michael Thase, from the University of Pennsylvania School of Medicine in Philadelphia.

"This is basically saying, these antidepressants aren't that good, and you should also consider other treatment options and don't just focus on the thing that's the easiest," he told Reuters Health.

The researchers said that some combination of antidepressants and talk therapy is considered most effective in depression treatment -- but getting therapy is often more expensive and time-consuming than medication.

Talk therapy can run $100 or more per session, while generic brands of antidepressants usually cost about $20 per month. Drugs may come with side effects, including insomnia and stomach aches, but they're usually minor, according to Hellerstein.

Still, people on antidepressants should be followed closely by a doctor to see how they're responding to treatment, he said.

Several of the authors of the current study reported having received funding for other research projects from drug companies that make antidepressants.

One recent study found that some depressed people on the antidepressant Cymbalta did worse than the comparison placebo group -- but the majority got some benefit (see Reuters Health story of December 9, 2011).

"I believe the basic finding that drugs are more effective than placebo," Thase said.

But, "The benefits of antidepressants may not be that dramatic in patients with milder depressions for whom many other (non-drug) strategies can also be considered."

SOURCE: bit.ly/yVBEdk Journal of Clinical Psychiatry, online December 27, 2011.

• A Permanent Economic Emergency