Critical Psychiatry Textbook, Chapter 8: Depression and Mania (Affective Disorders) (Part Three)


Editor’s Note: Over the next several months, Mad in America is publishing a serialized version of Peter Gøtzsche’s book, Critical Psychiatry Textbook. In this blog, he discusses the latest fad, ketamine, and how depression pills cause dependence and withdrawal. Each Monday, a new section of the book is published, and all chapters are archived here.

A pill bottle on its side spills white tablets out (photo)
“Pills” by The Javorac on Flickr.

The latest fad in psychiatry is esketamine. It is the S-enantiomer or mirror image of ketamine, a dissociative hallucinogen used as a general anaesthetic for over 50 years.

Esketamine induces dissociative anaesthesia, a trance-like state providing pain relief, sedation, and amnesia. Ketamine is commonly used as a street drug,294 often called a recreational drug even though it is not particularly recreational to be a drug addict.

In 2019, two psychiatrists praised esketamine for treatment resistant depression in BMJ.295 Together with colleagues, I responded that common sense tells us that a drug cannot possibly have a dramatic effect on depression within the first day of treatment unless something is terribly wrong.296

Esketamine was approved by FDA in March and by EMA in December 2019. It was not mentioned in any of the textbooks, not even in the one from 2021.20 But the 2018 book mentioned ketamine under hallucinogens in a section about drug abuse.18:77

Ketamine seems to work mainly through stimulation of opioid receptors. In a cross-over trial of 12 very severely depressed patients (Hamilton score of 27.4), a colossal “effect” was observed the first day post-infusion of ketamine, a reduction of 22.3, which corresponds to an effect size of 7.0.294 An effect size of this magnitude is totally unheard of in psychopharmacology, and in the rest of medicine as well. For placebo-controlled trials of depression pills, the effect size is only 0.2 to 0.3, and it takes weeks before this can be measured.268,271,273

When ketamine was supplemented with naltrexone (an opioid antagonist), the reduction was smaller, 5.6. As prolonged opioid use can cause depression,298 long-term use of esketamine might increase the risk of chronic depression.

The two unduly enthusiastic psychiatrists wrote that the effects of ketamine and esketamine fit with “modern theory that depression emerges from an impoverished neural network rather than serotonin deficiency.”295 It is not clear what they meant by this, and newness does not make a theory any more reliable than the discarded hypothesis about a chemical imbalance causing depression (see Chapter 4).

We wrote that we are convinced it could be demonstrated that alcohol, morphine, cocaine, and ecstasy also exert an “effect” on depression within the first day, but that does not make these substances acceptable. They may have acute euphoric effects, but frequent and long-term use often results in dysphoric mood states.

The dream of a quick fix for depression never stops. It has become popular to discuss another hallucinogen for depression, psilocybin, produced by fungi, and even LSD (lysergic acid diethylamide) is being dusted off. In 2020, the authors of a systematic review reported positive results and concluded that LSD is “a potential therapeutic agent in psychiatry.”299

This is unbelievable. Will psychiatrists ever learn from their mistakes? Psychedelic drugs are not the answer to psychiatric disorders. They make things worse.

Depression pills lead to dependence

After the authorities, at long last, in the 1980s, more than 20 years after it had been documented that benzodiazepines cause dependence, admitted that the huge consumption of benzodiazepines was a public health disaster and had started to warn against them, usage went down.264 At the same time, the American Psychiatric Association tightened the criteria for substance dependence, very conveniently just before the SSRIs appeared on the market.304

I have often wondered how much corruption was involved, as this change in the criteria must have been worth many billions of dollars for the companies.

The changes were major. Before 1987, dependence meant development of tolerance to a substance or withdrawal symptoms, which is how most people would define it. But from 1987, at least three criteria out of nine were needed and a time criterion was also added.304 From being very simple, it became highly complicated, arbitrary and judgmental, and no one can remember all these criteria or apply them consistently from case to case.7:239

For example: “A great deal of time” (how much?); “substance often taken” (how often?); “Important social, occupational, or recreational activities given up” (what is important and who decides on this?); “Frequent intoxication or withdrawal symptoms” (how frequent?); “Substance often taken to relieve or avoid withdrawal symptoms” (this criterion is meaningless; if a patient misses just one dose of paroxetine, it can elicit withdrawal symptoms305—does “often” for paroxetine mean taking three paroxetine pills a day? Hardly).

The new criteria took the power away from the patients to decide for themselves if they had become dependent on depression pills. The time criterion is also foolish. Symptoms should have persisted for at least one month or should have occurred repeatedly over a longer time. Very many patients are dependent on psychiatric drugs without fulfilling the time criterion. They might have tried to stop a few times but quickly resumed treatment and decided never to try again because of the abstinence symptoms they experienced. According to the time criterion, such patients are not dependent, although they are the ones who are the most dependent.

The new criteria are a smokescreen that serve to deflect attention away from the fact that SSRIs and SNRIs cause dependence. We found in our research that withdrawal symptoms were described with similar terms for benzodiazepines and SSRIs and were very similar for 37 of 42 identified symptoms.304 When Lundbeck, which sells several depression pills, was interviewed about our findings, the company called it “nonsense” that people could become dependent on SSRIs.306

The worst argument I have heard—also from professors of psychiatry—is that patients are not dependent because they don’t crave higher doses. If that were true, smokers are not dependent on nicotine because they don’t increase their consumption of cigarettes, and every smoker could stop smoking overnight, with no ill effects.

To describe similar problems as dependence for benzodiazepines but as “withdrawal reactions” or the even milder term “discontinuation symptoms,” invented by Eli Lilly,307:65 for SSRIs is irrational, and for the patients it’s just the same. It can be very hard for them to stop either type of drug. A survey showed that 57% of 493 Danish patients agreed to the sentence: “When you have taken antidepressants over a long period of time it is difficult to stop taking them,”89 and in another survey, 55% of 1,829 patients in New Zealand taking depression pills mentioned withdrawal effects, which 25% described as severe.308

A systematic review showed that half of the patients experience withdrawal symptoms; half of those with symptoms experience the most extreme severity rating on offer; and some people experience withdrawal for months or even years.136 A survey of 580 people reported that in 16% of the patients, the withdrawal symptoms lasted for over three years.136

According to Lundbeck, patients who say they have difficulty coming off the drugs talk nonsense,306 and according to psychiatry professor Lars Kessing, such patients are ignorant.89 Not much respect for patients in psychiatry.

The American Psychiatric Association’s Textbook of Psychiatry from 1999 stated that, not long ago, most patients would recover from a major depressive episode, whereas now “depression is a highly recurrent and pernicious disorder.”5:161 But the fact is that the disease has not changed. The psychiatrists and other doctors have failed to understand that they themselves have created an iatrogenic disaster because of their use of depression pills.7:256 The apparent “chronicity” in mental disorders is an artefact of the medications used.

This was shown in a study of 172 patients with recurrent depression who had been in remission for at least 10 weeks since their last episode.309 Of those who continued to take drugs, which they were supposed to do according to the guidelines, 60% relapsed in two years. The relapse rate was similar for intermittent users (64%) whereas it was 46% for those who did not take drugs and only 8% in those who did not take drugs and received psychotherapy. Differences in disease severity could not explain these results, so they were not due to confounding by indication.

Another paper showed that people with uncomplicated episodes of depression (lasting no longer than two months and not including suicidal ideation, psychotic ideation, psychomotor retardation, or feelings of worthlessness) were hardly more likely to have a further episode within 12 months than people with no history of depression, and the relapse rates are very low (3.7% versus 3.0%).103 Other data show the same.7:256 In the article “Medicalising Unhappiness,” Allen Frances wrote: “Watchful waiting over multiple visits can enable doctors to see if the problems will resolve without intervention.”103 This is true for all major psychiatric disorders.

We have known for over 50 years that depression pills cause dependence, and the patients have known it, too, but even 50 years after we knew it, the dependence problem was still being trivialised by the UK Royal College of Psychiatrists and the National Institute for Health and Care Excellence (NICE).7:76 The Royal College of Psychiatrists prioritised the interests of the College and the profession it represents over the wellbeing of patients when they took down an incriminating survey that totally contradicted what they postulated as soon as we had sent a complaint to them. What they falsely claimed was that, “We know that in the vast majority of patients, any unpleasant symptoms experienced on discontinuing antidepressants have resolved within two weeks of stopping treatment.”

When the College refused to correct the error, we made our complaint public, and the BBC’s Radio 4 program, Today, covered it on 3 October 2018. The College refused to participate in the programme.

Later, the Royal Society of Medicine launched a podcast series where the opening topic was about depression pills and withdrawal. One of the two interviewees was psychiatrist Sir Simon Wessely, president of the Royal Society of Medicine (and recent president of the College). Wessely rejected any link between depression pills and suicide and stated, categorically, that depression pills are “not addictive.”

Despite the psychiatrists’ steadfast denial of the facts, things changed. In September 2019, Public Health England published a 152-page evidence review making important recommendations, including for services to assist people coming off depression pills and other psychiatric drugs, and about better research and more accurate national guidelines.310 NICE updated its guidelines in line with the evidence the following month.136

Drug companies don’t care about patient safety if it could harm sales.6,7 Psychiatric leaders don’t care about patient safety if it could threaten their own reputation, the guild they represent, or the flow of money they receive from drug companies. This corruption of a medical specialty also permeates our authorities, which rely heavily on specialists when issuing guidelines and only make changes if critics make a lot of public noise about the wrongdoing.

When the profession cannot avoid addressing public criticism, the replies are often revealing. I have described7:16,311 how I was met with ad hominem attacks and false and highly misleading scientific arguments302 from the upper echelons of the UK psychiatric establishment after I gave a keynote lecture in 2014 at the opening meeting of the Council for Evidence-based Psychiatry in the House of Lords, chaired by the Earl of Sandwich: Why the Use of Psychiatric Drugs May Be Doing More Harm Than Good. The other speakers, psychiatrist Joanna Moncrieff and anthropologist James Davies, gave similar talks and have written books critical of mainstream psychiatry.3,4,312,313

Number needed to treat is highly misleading

When psychiatrists want to praise their drugs, they often refer to the number needed to treat (NNT) to benefit one patient, but it is so misleading for psychiatric drugs that any such information should be ignored.8:85

Technically, NNT is calculated as the inverse of the benefit difference. If 60% have improved on drug and 50% on placebo, NNT = 1/(0.6-0.5) = 10. Here are the main problems:

  1. NNT is derived from flawed trials, with cold turkey withdrawal in the placebo group, insufficient blinding, and industry sponsorship with data torture and selective publication.6-8
  2. NNT only takes those patients into account that have improved by a certain amount. If a similar number of patients have deteriorated, there is no NNT, as it would be infinite (1 divided by zero is infinite). If a drug is useless and only makes the condition after treatment more variable, so that more patients improve and more patients deteriorate than in the placebo group, the drug would seem effective based on NNT.
  3. NNT opens the door to additional bias. If the chosen cut-off for improvement does not yield the desired result, other cut-offs can be tried till the data confess. Such manipulations with the data during the statistical analysis, where the prespecified outcomes are changed after company employees have seen the data, are very common, also in psychiatry.6,7,137,279,300
  4. NNT is only about a benefit and completely ignores that drugs have harms, which are much more certain to be experienced than their possible benefits. Thus, in a mathematic sense, the NNT should be negative, but I have never seen a negative NNT in the literature. The NNT is -2 for sexual harms of depression pills (as I will discuss further), which means that for every two people we do not treat, we will spare one from getting sexually harmed.
  5. If benefits and harms are combined in a preference measure, it is not likely that an NNT can be calculated because psychiatric drugs do more harm than good. We can only calculate the number needed to harm (NNH). Dropouts during trials of depression pills illustrate this. Since 12% more patients drop out on drug than on placebo,301 the NNH is 8.

When the top UK psychiatrists in 2014 tried to convince the world that depression pills are highly effective, they did not take any of these flaws into account.302 They claimed that depression pills have an impressive effect on recurrence, with an NNT of around 3 to prevent one recurrence.302 It was not recurrence but withdrawal symptoms in the placebo group. As only two patients are needed to get one with withdrawal symptoms,136 there cannot exist an NNT to prevent recurrence, only an NNH, which is 2.

There cannot exist an NNT in other depression trials either, as the difference between drug and placebo in flawed trials is about 10%,303 or an NNT of 10, which is far less than the NNH.

These issues apply to all psychiatric drugs. Thus, NNTs in psychiatry are bogus. They don’t exist.


To see the list of all references cited, click here.


Mad in America hosts blogs by a diverse group of writers. These posts are designed to serve as a public forum for a discussion—broadly speaking—of psychiatry and its treatments. The opinions expressed are the writers’ own.


Mad in America has made some changes to the commenting process. You no longer need to login or create an account on our site to comment. The only information needed is your name, email and comment text. Comments made with an account prior to this change will remain visible on the site.