MELISSA BLOCK, host:
Many of the millions of Americans who take antidepressants struggle with debilitating depression, but many of them are not severely depressed. It turns out that few studies have looked directly at the effects of antidepressants on people with those milder symptoms.
NPR's Alix Spiegel reports on new research that does just that.
ALIX SPIEGEL: Since Prozac was introduced to the American public in 1987, there's been a huge amount of research on antidepressants. Small armies of patients have offered themselves up for testing and gone through a complex screening process.
Robert DeRubeis is a psychological researcher at the University of Pennsylvania. And he says when part of that process is a standardized test, which measures your level of depression:
Dr. ROBERT DERUBEIS (Psychological Researcher, University of Pennsylvania): The interviewer asks a series of questions and the patient responds to those questions. And on each of those items, the interviewer assigns a rating of zero, if it's not present, up to the maximum number four.
SPIEGEL: Now, any score over 19 indicates severe depression, and DeRubeis says that in most cases, antidepressant studies won't accept you unless you're actually higher than 20.
Dr. DERUBEIS: It's just been the practice of pharmaceutical trials that studies of antidepressant medications include patients with severe or very severe depression.
SPIEGEL: This week in the Journal of the American Medical Association, DeRubeis publishes a study which looks not at the very severely depressed but at people with mild and moderate depression.
Gregory Simon is a mental health researcher at the Group Health Research Institute in Seattle, who says this issue is important because many patients treated with antidepressants fall into exactly that category.
Dr. GREGORY SIMON (Psychiatrist, Group Health Research Institute): My best guess would be that about half the people treated by doctors out in the community would probably fall into that severe or very severe range, and about half of them would fall into the moderate or possibly even mild range.
SPIEGEL: In fact, a recent survey showed that almost 70 percent of depressed patients on medication had moderate to mild symptoms. So why only study those with very serious problems?
People who are mildly ill often recover without medication if they get some attention, even a placebo. And so, testing a drug in that group will make it look like the drug is not as effective.
Professor Brett Deacon of the University of Wyoming is a psychologist who does research on depression treatments.
Dr. BRETT DEACON (Psychologist, University of Wyoming): Because drug companies fund the vast majority of clinical trials, and obviously drug companies are interested in having their clinical trials show the efficacy of their antidepressant medications, and they're more likely to have that favorable outcome if they can recruit very severely depressed patients.
SPIEGEL: But it's not just about drug companies getting striking results so that they can get a drug approved by the FDA. Gregory Simon says it's also about doing right by the patients involved in research.
Dr. SIMON: Ethically, there's some obligation to say how can we do this research the most efficiently so that we don't enroll a lot more people than we need to in a study where they're randomly assigned to get a placebo?
SPIEGEL: So what did the JAMA study conclude? DeRubeis compared people who took real pills to people who got placebo pills. And he says that it appears that antidepressants do genuinely improve mood in severe cases. But DeRubeis says people with mild depression didn't really do so well.
Dr. DERUBEIS: The difference of the benefit of the medication over placebo was rather small, nearly zero.
SPIEGEL: But Dr. Philip Wang, deputy director of the National Institute of Mental Health, says consumers with mild symptoms shouldn't necessarily be scared off of drugs by these results.
Dr. PHILIP WANG (Deputy Director, National Institute of Mental Health): They don't not work for everyone, and they don't work for everyone. I think buried within the group are people who do respond if they have mild or moderate depression.
SPIEGEL: And so, Dr. Wang says clinicians just need to closely monitor patients who don't seem to respond to medication and give them an alternative treatment if necessary.
Alix Spiegel, NPR News, Washington.