Create Account | Sign In: Author or Forum

Search Symptoms

Category: Pharmacy | Psychiatry

Back to Articles

Antidepressants do they work? A Review of the Review

Dr. K. Eisele

Dr. K. Eisele's avatar

Published online: July 27, 2008

Disclosures of potential conflicts of interest and author contributions are found at the end of this article.

BACKGROUND: This article is in response to an announcement by researchers quoted in many UK publications this week, stating that the class of anti-depressants known as SSRIs (selective serotonin reuptake inhibitors) and SNRIs (serotonin-norepinephrine reuptake inhibitors) are no more effective than placebos. I looked at the NHS (National Health Services) website to get more information about this new research, and found that this was a systematic review of studies previously done on these medicines. It was not a randomized controlled trial (RCT) in and of itself, which is the "gold standard" for scientific research; it was, basically, a book report.

The specific medications included in this review of the literature are: fluoxetine, venlafaxine, nefa-zodone, paroxetine, sertraline, and citalopram. However, of the 47 RCTs supplied by the FDA, the results of only 35 were included in the review. FDA data on fluoxetine, venlafaxine, nefazodone, and paroxetine was used. Data on the other anti-depressants came from post-FDA-approval studies. The review found that "antidepressants improved symptoms more than placebo, and this difference was statistically significant". The rest of the story resulting in the announcement that these medicines don't work comes down to the researchers' own interpretations. For example, anti-depressants were more effective for more severely depressed patients, but the researchers report that the more seriously depressed patients did not respond as well to placebo as those with milder depression. This is one way to look at it, but one could also conclude that those test subjects with severe depression were possibly those who would show neurochemical depression if such a test existed.

On the NHS website, click here, the strengths and limitations of the study are pointed out. They range from only drug-company data was reviewed, to the researchers had to use estimates because some of the figures were missing. Perhaps the biggest limitation is that symptoms of depression are highly individualized and the review only takes into account symptoms that are measured by one scale. This implies that only one type of depression is really being measured, and from the results of the report, it seems that those with the most severe of symptoms are the true test subjects since these individuals had the best response to anti-depressant therapy.

All this begs the question, I think, about the effectiveness of anti-depressants on those with mild depression. In modern psychiatry residency programs, it is widely preached that without a serious impairment to a prospective patient's ability to function, there can be no diagnosis. And, if there is no diagnosis, there is hardly a reason for medication. In addition, according to evidence-based medicine, which is the standard of care, the best treatment for depression is medication in combination with psychotherapy.

The NHS also emphasizes that "the assessment of severity [of one's depression] is a skilled task and an individual's response to treatment may vary". The NHS is not in favor of anyone stopping their anti-depressant medication without first speaking to their physician.

Conflict of interest statement

None to report.

Dr. K. Eisele. Antidepressants do they work? A Review of the Review. Doctors Lounge. Available at: Accessed September 24 2022.


Kirsch I, Deacon BJ, Huedo-Medina TB, et al. Initial Severity and Antidepressant Benefits: A Meta-Analysis of Data Submitted to the Food and Drug Administration. PLoS Med 2008; 5(2)

Previous: Pregnancy Planning in the Weight Loss Surgery Patient Next: No health insurance? Where to go during the financial crisis

Reader comments on this article are listed below. Review our comments policy.

Submit your opinion: