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 lack of efficacy and poor evidence base for drugs used for psychosis. Each Monday, a new section of the book is published, and all chapters are archived here.
Psychosis drugs are the poster child of psychiatry and were highly praised in the textbooks. We are told that, before the advent of them, many patients needed to live the rest of their lives in hospitals and other institutions;16:222 the discovery of the pills in the 1950s meant that many psychotic patients clearly improved their quality of life, enabling their dismissal from the institutions and reintegration into society;20:416 patients who were previously tortured by their disease and were aggressive could now live alone or in protected housing;18:307 psychosis pills led to a decrease in hospital beds.16:616
We are told that chlorpromazine was a revolution in the treatment of psychotic disorders16:560 and it contributed in particular to emptying psychiatric hospitals;18:307 and—before chlorpromazine, lithium, depression pills, and benzodiazepines, the seriously ill patients spent most of their lives in isolated institutions, behind locked doors, with barred windows, and physical force was used—but the development of psychiatric drugs in the 1950s revolutionised the treatment.17:644
Psychiatrists propagate this narrative all over the world to gain support for their specialty but all the above is wrong.1,4-8 There were no references for the extravagant claims, but it has been documented that the pills had nothing to do with the emptying of the asylums.1:155,3:53,147,148 Furthermore, it is impossible for drugs that—according to the standard scale for evaluating the effect on the psychosis—do not have clinically relevant effects (see just below) to produce such dramatic outcomes.
Since the “emptying of the asylums” is the core argument for the claimed revolution in psychiatric drug treatment that started with chlorpromazine in 1954, I shall explain why it is wrong. The misconception stems in particular from flawed studies in New York.148 The authors noted that the populations in asylums fell after 1954 and ascribed this to drug treatment. Better studies were conducted in Michigan and California by other authors who compared treated and untreated patients. They found that the drugs did not increase discharge rates.
In 1985, a study debunked the myth totally.148 It covered all US states and compared two nine-year trends in discharge rates, 1946 to 1954 with 1955 to 1963. The mean percentage change in discharge rates was 172 before chlorpromazine, a little higher than with chlorpromazine, 164.
There are no supportive studies of the myth from other countries either. In England, inpatient populations began to decline before the drugs were introduced; in France, inpatient populations increased for 20 years after the drugs were introduced;148 and in Norway, inpatient numbers did not change with the introduction of the drugs.3:54
The Joint Commission on Mental Illness and Health, commissioned by the US Congress, wrote in 1961 that “Drugs have revolutionized the management of psychotic patients in American mental hospitals,” quoting the misleading New York studies and avoiding mentioning the better designed Michigan study even though it was available.148 It was politically expedient to dupe the population this way, painting a false picture of huge progress in psychiatry.
Psychosis pills don’t have clinically relevant effects on psychosis
One textbook noted that the strongest evidence in psychopharmacology is for the effect of psychosis pills in the acute phase of schizophrenia and for relapse prevention, as they markedly reduce the risk of relapse.16:560 It claimed that the pills improve prognosis and survival in most patients,16:222 and that it is essential to know which biological processes in the brain the pills influence in order to offer the most optimal medical treatment.16:216
All of this is wrong. Robert Whitaker once wrote to me that it requires extraordinary mental gymnastics by the psychiatrists to conclude that these drugs, which cause obesity, metabolic dysfunction, diabetes, tardive dyskinesia, lethal cardiac arrhythmias, and so on, protect against death. They don’t; they kill many people,7:307 which I shall explain below.
It is impossible to offer a better treatment by knowing more about biological brain processes when the drugs do not have clinically relevant effects on the psychosis apart from tranquillising the patients, which is an unspecific effect.
Virtually all placebo-controlled trials of psychosis drugs are seriously biased by cold turkey withdrawal effects in the placebo group, which occur when the psychosis drug the patient is already on gets withdrawn before randomisation. These iatrogenic harms are usually avoided in the actively treated group. The reason that Janssen could claim that its bestseller risperidone didn’t cause more extrapyramidal (muscular) harms than placebo was the abrupt withdrawal of the previous psychosis drug, which inflicted these effects on the placebo group to such an extent that one in six patients got them.1:276 The companies needed to show that their drugs reduced psychotic symptoms and they made some of the placebo patients psychotic by withdrawing their drug cold turkey.4:45,31,149
I have only found two trials where none of the patients had received a psychosis drug before. One was from China and appeared to be fraudulent.150 It compared olanzapine with placebo in patients with first-episode schizophrenia.151 The patients needed to have a score on the Positive and Negative Syndrome Scale (PANSS) of at least 60 to be included. However, the score before treatment was only about 9, even though by definition it must be at least 30 (the lowest score is 1 and there are 30 items). The score increased to 71.3 in the olanzapine group and to 29.4 in the placebo group. The authors reported that olanzapine was effective although patients on placebo fared much better. Furthermore, a difference of 42 in PANSS is implausibly large. In the placebo-controlled trials in submissions to the US Food and Drug Administration (FDA) of newer psychosis drugs, including olanzapine, the difference was only 6.152
The only trial that doesn’t appear to be fraudulent and wasn’t flawed by withdrawal effects was published in 2020, 70 years after the discovery of the first psychosis drug, chlorpromazine.153 It randomised 90 patients with a first-episode psychosis (FEP) with a duration of untreated psychosis of less than six months to risperidone, paliperidone, or placebo.
However, even after 70 years, the psychiatrists weren’t yet ready to draw the obvious conclusions of their results. They wrote that the differences were “small and clinically trivial, indicating that treatment with placebo medication was no less effective than conventional antipsychotic treatment” (P = 0.95). They noted that “the immediate introduction of antipsychotic medication may not be required for all cases of first episode psychosis” with the reservation that “this finding can only be generalised to a very small proportion of FEP cases at this stage, and a larger trial is required to clarify whether antipsychotic-free treatment can be recommended for specific subgroups of those with FEP.”
What the authors should have written is something like this: “Our study was small, but it is unique because it only included patients who had not been treated with a psychosis drug before. We found that psychosis drugs are ineffective in patients with untreated psychosis. This is great progress for patients, as these drugs are highly toxic and make it difficult for them to come back to a normal life. Based on the totality of the evidence we have, the use of psychosis drugs in psychosis cannot be justified.”
The authors of a 2011 Cochrane review of psychosis pills for early episode schizophrenia pointed out that the available evidence doesn’t show that the drugs are effective.154 This is one of the few Cochrane reviews of psychiatric drugs that can be trusted. Apart from the cold turkey problem, Cochrane reviews in schizophrenia include trials in a meta-analysis where half of the data are missing.
This Cochrane review noted that twice as many patients on chlorpromazine than on placebo were rehospitalised within three years, risk ratio 2.3 (1.3 to 4.0). There were also fewer rehospitalisations in the placebo group at the one-year follow-up in the famous trial funded by the US National Institute of Mental Health (NIMH), which was published in 1964, but the difference wasn’t quantified, and the original data appear to have been lost.154 These data totally contradict the psychiatric narrative that psychosis pills emptied the asylums.
In trials supposed to be double-blind, but which are not blind in practice, investigators may report positive effects that only exist in their imagination. This occurred in the NIMH 1964 study, which is still highly cited as evidence that psychosis drugs are effective.
344 newly admitted patients with schizophrenia were randomised to phenothiazines such as chlorpromazine or to placebo.155 The investigators reported, without offering any numerical data, that the drugs reduced apathy and made movements less retarded, the exact opposite of what these drugs do to people, which the psychiatrists had admitted a decade earlier.5:49,5:61
The investigators claimed a huge benefit for social participation (effect size 1.02) and that the drugs make the patients less indifferent to the environment (effect size 0.50). The drugs do the opposite. The authors also claimed, with no data, that 75% versus 23% were markedly or moderately improved and suggested that the drugs should no longer be called tranquillisers but antischizophrenic drugs.
Their study contributed to shaping the erroneous beliefs that schizophrenia can be cured with drugs and that psychosis pills should be taken indefinitely.1
The truth is that psychosis pills do not have clinically relevant effects on psychosis. Despite the formidable biases—cold turkey, lack of blinding, and industry funding that often involves torturing the data till they confess6,7—the published outcomes have been very poor.4 The least clinically relevant effect corresponds to about 15 points on the PANSS scale156 commonly used in the trials. Yet, what was reported in the placebo-controlled trials of recent drugs submitted to the FDA was only 6 points, or 3% of the maximum score of 210 on this scale.152
A textbook claimed that the effects on the dopamine system can restore homeostasis in brain signal transmission.18:97 This assumes that there is a defect in the dopamine system to begin with, which has never been documented and is unlikely (see Chapter 4). We are also told that the treatment response is related to dopamine activity.16:220 This is not possible for drugs that don’t work.
There were case stories in one of the textbooks and they were always positive in relation to the drugs used, but most of them were misleading. Here are some examples.
A patient improved within a few weeks on a psychosis pill and no longer heard voices or experienced persecution.18:87 The pills do not have such effects.
A patient improved a lot on a psychosis pill and had relapses when he did not want to continue with the drug.18:89 It is highly likely that the psychiatrists confused withdrawal symptoms with relapse. And there is no reliable evidence that the pills can prevent relapse (see below).
A patient got a small dose of a psychosis pills and support, and improved.18:89 It was more likely the support that helped the patient, or the patient would have improved anyway, without treatment or support.
An increased dose of a psychosis pill affected the time to relapse.18:105 These pills do not have increased effect with increased dosage.157
It would be an eye-opener if the psychiatrists tried a psychosis pill on themselves. Two physicians have described how a single dose of haloperidol knocked them down.158 They experienced a marked slowing of thinking and movement, profound inner restlessness, a paralysis of volition and a lack of physical and psychic energy, being unable to read or work.
David Healy found the same in 20 staff from his hospital who received droperidol.4:116 Everyone felt anxious, restless, disengaged, and demotivated to do anything; a volunteer found it too complicated just to obtain a sandwich from a sandwich machine. Some felt irritable and belligerent and many were unable to recognise the altered mental state they were in and to judge their own behaviour. Peter Breggin calls this “medication spellbinding.”135,159
The predominant subjective effects reported by patients on the Internet when they take psychosis drugs are sedation, cognitive impairment, and emotional flattening or indifference.160 We also know from telephone help lines that what medicated persons miss the most are themselves.1:179
Psychosis pills were hailed as a great advance, but this was because they kept the patients docile and quiet, which was very popular with the staff in psychiatric wards.148 It was a huge conflict of interest that the same staff evaluated whether patients had improved or not, and this conflict of interest clouds psychiatric practice and research even today.
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.