Do antidepressants work?

In January 2008, the New England Journal of Medicine (NEJM) published a study that seemed to question the effectiveness of antidepressants. Since then, I've been fielding occasional questions from patients who are taking these medications. Understandably, they want to know whether the study was correct, and whether they should reconsider taking their antidepressant.

Although the study provides an interesting analysis of how antidepressant results are reported, it does not say anything new about antidepressant effectiveness. The best way to make decisions about antidepressant therapy remains weighing your doctor's advice and your personal assessment of the risks and benefits of treatment.

The article demonstrates a phenomenon sometimes called "publication bias." In the NEJM paper, the researchers compared the results of antidepressant studies registered with the FDA against those published in peer-reviewed journals. They confirmed what previous investigators have found: positive studies are most likely to be published, while negative findings are less likely to make their way into print. The authors reached these conclusions after reviewing 74 antidepressant studies that were reviewed by the FDA as part of the medication approval process. The studies evaluated 12 antidepressants approved between 1987 and 2004.

The FDA considered 38 of the studies to be positive, meaning that the drug being evaluated offered significant benefit when compared to placebo. Of these, 37 were published in journals. But only three of the 36 studies the FDA classified as producing negative or questionable results were published in a way that acknowledged these findings. Of the remaining 33, 22 never got published, and 11 were published in a way that might lead readers to believe the outcomes were positive. The source of the publishing bias is not known, but could be attributed to researchers' discouragement about negative results, industry sponsors' economic incentives to publish positive news, journal editors' and reviewers' eagerness to showcase what is new or upbeat - or some combination.

Taken together, the published studies gave the impression that - for patients in the studies - antidepressants were about a third more effective than they were when all FDA data were considered. In contrast, when unpublished data were taken into account, the drugs did not seem to provide much more benefit than a placebo. The authors (and, in press interviews, the NEJM's editor-in-chief) emphasized that this skewed publication pattern deprives clinicians and patients of information needed to make good treatment decisions.

A larger issue is that although it is interesting to lump antidepressant studies together, that approach can give a misleading impression about effectiveness, because depression is not a single illness; rather, it is several illnesses. Some forms of depression (for example, more severe types or the depression that occurs in bipolar disorder) tend to respond very well to antidepressant medication. But the drugs are not as beneficial for milder forms of depression or depression that is integral to a person's personality. Experts point out that it is this second group of patients who are often subjects in antidepressant studies.

And because every depression is different, it is frequently necessary to persist and try several reasonable options or combinations of drug therapy and psychotherapy to get relief. Imagine, for example, a collection of 10 locks and 10 keys. If they are randomly matched in one trial, only a few locks will open - even if each lock has a key that works. After several tries, however, the right matches may be found.

We are still far from finding all the right keys to depression treatment. Clinicians and patients must have all study data and be given the tools to understand the results - without being oversold. But the NEJM study should not be cause for alarm or spur a major shift in treatment. Rather, it serves as a reminder to appreciate the limits of our knowledge, to tailor treatment to individual needs and - if you find it - to stick with what works.

Have you found antidepressants to be helpful for you or your friends? What other strategies have helped you or someone you know to live with depression?

Dr. Michael Miller has been on staff of the Beth Israel Deaconess Medical Center, a large teaching hospital in Boston, for more than 25 years. He is also an Assistant Professor of Psychiatry at Harvard Medical School.

Depression
Nearly 1 in 10 adults will suffer from some form of depression in a given year, affecting not only them, but also their friends and family. Thankfully, years of research and recent breakthroughs have made this serious illness easier to treat. With Understanding Depression, a special health report from Harvard Medical School, you can stay up-to-date on the latest information on depression symptoms and treatments to improve your life-or the life of someone close to you.





Are you Living With Depression? Connect with others with similar health concerns and issues. Click here to join the group.

This content is not intended to substitute for personalized medical advice, diagnosis, or treatment from your healthcare provider. Read our full disclaimer.