Dangerous Placebos: The Depressing Thing About Antidepressants Is That They’re Placebos With Side Effects

Views 10548

Disappointingly, suicide rates among people who take antidepressants are not statistically any different than suicide rates for people who take a placebo.

The decades long experiment in antidepressant drugs has been a failure. Disappointingly, suicide rates among people who take antidepressants are not statistically any different than suicide rates for people who take a placebo. And while the drugs reduced the symptoms of depression by 40.7%, placebos reduced the symptoms by a shockingly similar 30.9% (Arch Gen Psychiatry 2000;57:311-7).

And when an analysis of antidepressant drugs for adults was undertaken, it found that the selective serotonin reuptake inhibitors (SSRIs) have no clinically meaningful advantage over placebo. The study concluded that they “have not been convincingly shown to affect the long term outcome of depression or suicide rates” (BMJ 2005;331:155-7).

When a meta-analysis looked at all the clinical trials of antidepressant drugs that were submitted to the FDA, it found that there was no difference between antidepressant drugs and placebo for moderate depression. For very severe depression, there was only a relatively small benefit to the drugs. But the researchers were careful to point out that the small benefit the drug had compared to the placebo for very severe depression was not because of the drug working better but because of the placebo working worse (PLoS 2008;doi.org/10.1371/journal.pmed.0050045).

Another fascinating study looked at seventy-four studies on antidepressant drugs that had been registered with the FDA. It found that 37 of 38 studies with positive results were published. But, of the 36 studies with negative results, 22 of them were never published, and eleven of them were published with the interpretation spun to make the negative results appear to be positive. Only three negative studies got published as negative studies. The researchers say that the published literature—the stuff people actually get to see—made it look like 94% of the studies conducted on antidepressant drugs yielded positive results. However, according to the FDA’s analysis of the actual studies, only an unimpressive 51% were positive (NEJM 2008;358:252-260).

This problem of not publishing negative studies or spinning them to sound positive is, alarmingly, not uncommon. A recent systematic review found that 56% of medical studies are guilty of putting a positive spin on negative studies (PLoS Biol 2017;15(9):e2002173).

An earlier study of all the published and unpublished studies submitted to the FDA between 1987 and 1989 found that of the 47 studies submitted on the six most prescribed antidepressants (Prozac, Zoloft, Paxil, Effexor, Serzone and Celexa), only twenty found superiority over placebo (Prevention and Treatment 2002;5;dx.doi.org/10.1037/1522-3736.5.1.523a).

And what about antidepressant drugs for children with depression? In 2001, a study was published on the SSRI Paxil (paroxtine). The study was funded by the pharmaceutical company that made the drug, GlaxoSmithKline. It concluded that Paxil was safe and effective for the treatment of depression in adolescents (J Am Acad Child Adolesc Psychiatry2001;40:762-72). Now the data from that study has been reanalyzed by an unbiased group of researchers from Toronto. They disagree.

This study was important because, since its publication, it has played a significant role in the case for using SSRIs to treat teenage depression. The original paper compared the SSRI Paxil to the older antidepressant imipramine and placebo. Though it admitted that neither parents nor the teenagers in the study rated either drug as significantly superior to the placebo, it still concluded that Paxil, but not imipramine, was significantly better than placebo and that it is safe and effective for treating major depression in adolescents.

However, when the unbiased researchers re-examined the data, they found that Paxil was neither statistically nor clinically superior to the placebo for any of the studied outcomes. What's worse is that it found not only that the original conclusion about efficacy was wrong, so was the original conclusion about safety. The increased risk of harm from taking Paxil was clinically significant, including thoughts of suicide and other serious side effects (BMJ 2015;351:h4320).

Prozac, the most famous antidepressant drug of all, was approved by the FDA based on the four positive studies. All Elli Lily, the pharmaceutical company that makes Prozac, had to do was not tell anybody about the six negative studies. In those six unreported studies, Prozac failed to beat a placebo. The combined result of the original ten studies on Prozac show that for about 90% of people, Prozac is no better than a placebo.

All of these studies suggest that antidepressant drugs are not superior to placebo. Now an innovative study has proven that Prozac’s performance is only as good as the placebo effect.

The placebo effect is the well-known benefit conferred by the belief that an intervention is going to work; the nocebo effect is the less know negative effect caused by the belief that the intervention is not going to work. Prozac, this study found, is as good as the placebo effect and as bad as the nocebo effect.

This intriguing study looked at two previous studies of Prozac. There were two parts to the analysis. The first included 673 people with major depressive disorder who responded to 12 weeks of Prozac. The important detail, though, is that the study was not blinded: they knew they were on Prozac.

At the end of the successful 12 weeks, all 673 people were told that they would now either continue on their Prozac or be switched to a placebo for another year. This time, the study was blinded. No one knew if they were getting the Prozac or the placebo, so doubt—the nocebo effect—was introduced: confidence in their treatment was shaken.

After doubt was introduced by the possibility that they were only on a placebo, depression—as measured by the Hamilton Depression Scale (HAMD)—significantly worsened in both groups: those on the placebo and those on the Prozac. And the worsening depression was essentially equal in both groups (.46 for Prozac and .48 for placebo for a significance of P<.001 in each).

This study shows two things. The people on Prozac only got better as long as they knew they were on Prozac. That is, the benefit is attributable to a placebo effect. As soon as they thought they even might be on a placebo, the benefit disappeared: the nocebo effect. The researchers say that the worsening of the depression after the introduction of a possible placebo was significantly associated with the degree of improvement when they knew they were getting Prozac. That means they got as much worse when they thought they might not be getting Prozac as they got better when they knew they were.

The authors conclude that changes in expectations affect changes in depression. Which is a delicate way of saying that Prozac is just a placebo with dangerous side effects (J Clin Psychiatry 2014;75(10):1040-6).

And it is those side effects that make the charge against antidepressant drugs even more severe. It is not just that they are really nothing more than expensive placebos, they are also dangerous placebos. Prescribing antidepressant drugs over placebos introduces, not benefit, but significant risk.

A recently published article documents that, in the case of antidepressant drugs, the drugs may be as dangerous as the disease they are intended to treat (Psychother Psychosom 2016;85:270-288). The article comprehensively reviewed the side effects of latest generation of antidepressant drugs: selective serotonin reuptake inhibitors (SSRI), serotonin noradrenaline reuptake inhibitors (SNRI), bupropion, mirtazapine, trazodone, agomelatine, vilazodone, levomilnacipran and vortioxetine.

Despite the popularity of prescribing these drugs for depression—7% of visits to the doctor end in a prescription for an antidepressant—a significant number of people with depression do not get adequately better on their medication. And one reason for that failure is because so many of them have to discontinue their drugs due to side effects. Amazingly, up to 43% of people with depression discontinue their medication because of side effects.

Some of the most common side effects, according to the article, include gastrointestinal side effects, liver toxicity, weight gain, diabetes, sexual dysfunction, osteoporosis, eye diseases and suicide.


Approximately half of all people who take newer generation antidepressants experience gastrointestinal side effects. Nausea, diarrhoea, indigestion, GI bleeding and abdominal pain are amongst the most common side effects of antidepressants. So common is GI bleeding on short term use of 7-28 days of antidepressants that the authors of the review recommend that, based on this statistic, the same well-known cautions for NSAID’s should be given for antidepressants.

Liver Toxicity

For people taking SSRIs and SNRIs, the incidence of drug-induced liver toxicity is 0.5-1%. Higher doses are more likely to cause toxicity. That means that antidepressants cause liver toxicity in 1 out of every 100 or 200 people who use them. When you consider how many people use them, that’s alarming.

Weight Gain

Compelling evidence indicates that the use of most antidepressants may increase weight in a significant proportion of patients who use them and that they increase the risk of obesity.


There is some evidence that antidepressants increase the risk of diabetes. Studies have had mixed results, but a 2013 systematic review and meta-analysis of 8 studies found a 50% increase in the risk of diabetes (Diabetes Metab Res Rev2013;29(4):273-84).

Sexual Dysfunction

All SSRIs and SNRIs have been shown to have significant sexual side effects. Sexual dysfunction affects as many as 50-70% of people taking SSRIs.


The use of SSRIs is associated with reduced bone mineral density and a higher risk of fracture. A recent meta-analysis of 13 studies found a 72% increased risk of fracture with SSRIs (Osteoporos Int 2012;23(1):365-75). In some studies, there is a full 98% increased risk of fracture on Prozac, Zoloft, Paxil and Celexa.

Eye Diseases

SSRIs increase the risk of angle-closure glaucoma by as much as nearly 6 times. The newer generation antidepressants also increase the risk of cataracts by about a third.


A 2016 meta-analysis (BMJ 2016 27;352:i65) of 70 studies found a significant 139% increased risk of suicide for children and adolescents: that’s more than double the risk of suicide. The same study found a 179% increase in aggression for children and adolescents and more than double the risk of agitation. Adults also experience twice the rate of agitation.

When all of this data is put together, the evidence suggests that antidepressant drugs are no better than placebos but alarmingly more dangerous. The scope of side effects is much wider than usually discussed. So, the most depressing thing about antidepressants is that they may just be very dangerous placebos.

If you found this article interesting, join Linda and Ted’s newsletter and get The Natural Path delivered to your in-box every month. The Natural Path is Your Guide to Good Health & Vitality: Cutting Edge Research Made Easy. Subscribe today, and get the latest research to keep you and your family healthy.

Disclaimer: This article is not intended to provide medical advice, diagnosis or treatment. Views expressed here do not necessarily reflect those of GreenMedInfo or its staff.
Sayer Ji
Founder of GreenMedInfo.com

Subscribe to our informative Newsletter & receive The Dark Side of Wheat Ebook

Our newsletter serves 500,000 with essential news, research & healthy tips, daily.

Download Now

The Dark Side of Wheat

This website is for information purposes only. By providing the information contained herein we are not diagnosing, treating, curing, mitigating, or preventing any type of disease or medical condition. Before beginning any type of natural, integrative or conventional treatment regimen, it is advisable to seek the advice of a licensed healthcare professional.

© Copyright 2008-2023 GreenMedInfo.com, Journal Articles copyright of original owners, MeSH copyright NLM.