Antidepressants might benefit only people with more severe depression, according to a new study published January 6 in the Journal of the American Medical Association (JAMA) and reported by The New York Times.

According to the Times, there have been lingering questions about the effectiveness of antidepressants for people with less severe depression. Studies conducted by the pharmaceutical industry and published in peer-reviewed scientific journals tend to show a benefit in mildly depressed people. The Times also reports that companies “buried”—or did not publish—studies that did not show a benefit.

To shed light on this debate, Jay Fournier, MA, from the University of Pennsylvania in Philadelphia, and his colleagues conducted a review of six studies of antidepressants involving 728 depressed people. About half had milder symptoms and half had more severe symptoms of depression. Three of the studies compared the selective serotonin reuptake inhibitor (SSRI) Paxil (paroxetine) with a placebo, and three studies compared the older tricyclic antidepressant Tofranil (imipramine) with a placebo.

Fournier’s team found that a placebo was about as affective as either of the two drugs in people with mild to moderate depression. The drugs did work, however, in people with more severe depression. In fact, they worked best in people with very severe depression.

“I think the study could dampen enthusiasm for antidepressant medications a bit, and that may be a good thing,” Erick H. Turner, MD, a psychiatrist at Oregon Health and Science University, said in the Times article. “People’s expectations for the drugs won’t be so high, and doctors won’t be surprised if they’re not curing every patient they see with medications.” But he cautioned: “The findings shouldn’t dampen expectations so much that people refuse to even try medication.”