A Hopelessly Flawed Seminar in “The Lancet” About Suicide

On 14 May 2022, The Lancet published the seminar, “Suicide and self-harm.” Lancet seminars are “Disease-oriented clinically focused overviews for the generalist, covering epidemiology, pathophysiology, diagnosis, management, and prevention. These clinical overviews are always externally peer reviewed.”

The seminar was very long, 14 printed pages, with 142 references. Many people consider Lancet a highly prestigious journal and, whatever you think of the journal, it is highly influential. It is therefore important that what gets published is honest, trustworthy and based on the best available evidence. This was not the case.

Lancet has immunized itself against criticism. A journal that, in its instructions to authors, writes that “Letters for publication in the print journal must reach us within 2 weeks of publication of the original item and should be no longer than 250 words” does not invite criticism and a sound scientific debate. Many people will not know that an article has been published before it is too late to criticize it, and 250 words are much too little if an article is flawed in several ways. If journals were really interested in serving the progress of science, there should be no time limit for letters that point out fatal flaws in an article.

The Lancet seminar is one of the most misleading articles about suicide I have ever seen, but I shall only mention a few issues. The authors write:

“Research has also identified associations between suicidal behaviour and dysregulation of the hypothalamic–pituitary–adrenal axis and serotonergic neural transmission.26,27

The authors try to resurrect the stone dead myth about a chemical imbalance in the brain being the cause of psychiatric disorders, which prominent psychiatrists have called an “urban legend.”

The two references the authors provide in support of this myth are untrustworthy. The first alludes to epigenetic modification of genes, alterations in key neurotransmitter systems, inflammatory changes, and glial dysfunction in the brain as causal factors.

The second reference is similar. Its authors suggest hypothalamic-pituitary-adrenal (HPA) axis dysfunction, which “in turn can be traced back to genetic predisposition” and “early life stress-related epigenetic mechanisms.”

Please consider this: If a house burns down and we find ashes, it doesn’t mean that it was the ashes that set the house on fire. Similarly, if a lion attacks us, we get terribly frightened and produce stress hormones, but this doesn’t prove that it was the stress hormones that made us scared. It was the lion. No genetic predisposition or “chemical imbalance” is needed for this.

The total failure of the much touted biological psychiatry has been documented numerous times, but the psychiatrist won’t give up. They continue to produce misleading brain scan studies, brain chemistry studies, and genetic studies, and psychiatric textbooks for students of medicine, psychology and psychiatry are full of this. I have read the five most used textbooks in Denmark and will soon publish my own where I discuss what is wrong with the official textbooks in relation to these issues and a lot more. The tentative title is “The truthful textbook of psychiatry.”

Among risk factors for suicide, the seminar authors mention “harmful substance use.” They do not mention depression pills even though they double the risk of suicide, both in children and adults. This is taboo. It is also taboo to mention that other psychiatric drugs, e.g. antiepileptics, which misleadingly are called “mood stabilizers,” double the risk of suicide. And it is taboo to mention that the psychiatric profession itself increases the risk of suicide markedly, much more than what can be explained by its use of harmful drugs.

A Danish register study found that admission to a psychiatric ward increased the suicide risk 44 times, and, surprisingly, the potential biases in the study were conservative, i.e. favoured the null hypothesis of there being no relationship. An accompanying editorial noted that there is little doubt that suicide is related to both stigma and trauma, and that it is entirely plausible that the stigma and trauma inherent in psychiatric treatment – particularly if involuntary – might cause suicide. The editorialists believed that a proportion of people who commit suicide during or after an admission to hospital do so because of conditions inherent in the hospitalisation.

The seminar authors write that “The use of medication to prevent suicide is controversial” and that there is a “possibility of exacerbating suicidal thoughts, particularly in young people.11

This is dishonest. There is a load of meta-analyses, including the one made by the FDA in 2004, that show that depression pills double the risk of suicide in children and adolescents, the age group we want to protect the most against suicide. But FDA, which is much too industry-friendly, downplayed what they found. When FDA employees published their results, they also reported a doubling of the suicide risk but concluded in their abstract: “Use of antidepressant drugs in pediatric patients is associated with a modestly increased risk of suicidality.”

The seminar authors speak about a possibility of exacerbating suicidal thoughts, and only suicidal thoughts, not suicidal behaviour or attempts or suicide, and the FDA employees use the term associated with. All of this is wrong. These drugs not only increase suicidal thoughts, but also suicidal behaviour and suicide attempts. Worst of all, they also increase suicides. We use the term associated with in observational studies where we are uncertain about a cause-effect relationship. We do randomised trials because they can prove cause-effect relationships. And only modestly increased? No. The FDA employees should have concluded that depression pills may cause children to kill themselves. They write that “The overall risk difference for all drugs across all indications was 0.02 (95% CI, 0.01-0.03).” Thus, for every 100 children treated with a depression pill, two become suicidal. This is anything but modest for drugs that do not even have clinically relevant benefits.

The seminar authors do not quote any of the many meta-analyses that have shown that depression pills increase the suicide risk. Not one. Instead, they quote a book (their reference 11) written by the last author of the seminar and by Robert D Goldney who has published a totally unreliable review about depression pills and the risk of suicide. His paper is a classic example of how one should not do a review. He cherry-picked those observational studies that supported his idea that antidepressants protect against suicide.

Goldney cited studies in the Nordic countries that linked antidepressant prescribing with a reduction of suicide, but these studies are totally unreliable. Nordic researchers have shown that

there is no statistical association between the increase in sales of SSRIs and the decline in suicide rates in the Nordic countries. These authors reported that the decline in suicides in Denmark and Sweden pre-dated the introduction of SSRIs by ten years or more!

The Nordic researchers had no conflicts of interest while Goldney had “received honoraria and research grants from a number of pharmaceutical companies.” Of course he had. With such flawed reviews, of which there are numerous, Goldney must be worth far more than his weight in gold for the drug industry.

What Goldney did is what I call the UFO trick. It is very common in science to mislead your readers this way, and it is about not losing power or giving up on your wrong ideas. If you use a fuzzy photo to “prove” you have seen a UFO when a photo taken with a strong telephoto lens has clearly shown that the object is an airplane or a bird, you are a cheat. When randomised trials have documented that depression pills double the suicide risk, we can assign all the observational studies that tell the opposite story to the graveyard of untrustworthy research, if there is any room left there.

The seminar authors write that “treatment of underlying psychiatric conditions through medication can reduce suicidal behaviour.”

The authors give no references to this dishonest information. Which are the miraculous drugs that can reduce suicides? All that I know is that psychiatric drugs increase suicides.

A little later, the authors write: “Evidence from several studies, most of which were observational, suggests that antidepressants might reduce the risk of suicide.91” But suggest and might reduce is not the certainty in the sentence just above: can reduce. They use the UFO trick again and quote a review that reported that meta-analysis had found that “antidepressants prevent suicide attempts, but individual randomized controlled trials appear to be underpowered.” These meta-analyses were of observational studies, and all meta-analyses of randomised trials have shown the opposite.

In the next sentence, they write: “However, some research has found an association with increased risk of suicide-related outcomes in young people.”

The is blatantly false. When the FDA looked at all relevant research, not just some research, and indeed the best we have, all the randomised placebo-controlled trials, it was clearly a causal relation and not just an “association”.

In the ensuing sentence, they write: “The evidence base is far from complete, since many randomised trials exclude people at heightened risk of self-harm or suicide.11,91

This is utter nonsense. We have all the data we need to conclude that depression pills double suicides. The authors use the trick philosopher Arthur Schopenhauer in his book, The Art of Always Being Right, calls diversion:

“If you are being worsted, you can make a diversion – that is, you can suddenly begin to talk of something else, as though it had a bearing on the matter in dispute and afforded an argument against your opponent … it is a piece of impudence if it has nothing to do with the case, and is only brought in by way of attacking your opponent.”

The authors claim that “Lithium has been associated with reduced suicide rates in people with bipolar disorder and depression, which might be a specific effect not seen with other drugs designed to stabilise mood.92–94

So, this seems to be the wonder drug that reduces suicides. However, most lithium trials are highly unreliable because of withdrawal effects. Patients were on lithium before they were randomised and some of those who got placebo experienced a cold turkey, which increases the risk of suicide.

A Swedish psychiatrist and I reviewed the placebo-controlled trials and included only those where the doctors had not harmed the patients in the placebo group, i.e. lithium naïve patients. We found 45 eligible studies but only four studies reported any suicides or other deaths. Some suicides and deaths were likely missing. A systematic review of psychiatric drug trials found that only half of the suicides and deaths are being reported. There were only three suicides, all on placebo. So much for the widely touted idea that lithium reduces suicides. We simply do not know.

The latest fad in psychiatry is ketamine. The authors write that “Ketamine has shown promise.” Of course it has. Give the patients hallucinogenic agents so they will forget about their troubles. Ketamine is commonly used as a street drug. Why not LSD, which some psychiatrists are trying to revive?

Ketamine seems to work mainly through stimulation of opioid receptors. My colleagues and I have explained why this drug and its S-enantiomer, esketamine, should not be used.

There is a glimpse of light in the darkness of the seminar. The authors write that “cognitive behavioural therapy and related treatments have the strongest evidence base for reducing suicidal ideation and repeat self-harm compared with treatment as usual.”

This is correct, but the authors quoted a review that included self-harm even though self-harm does not always imply a suicidal intent. My research group therefore did a review where we excluded self-harm studies. We found that psychotherapy halves the risk of a new suicide attempt in people acutely admitted after a suicide attempt. Our systematic review was published in 2017 but was not among the seminar authors’ 142 references even though it sends a very strong message: Do not use pills but psychotherapy if you want to prevent suicide in patients at very high risk of suicide, those who have already tried to kill themselves.

It is disappointing that Lancet publishes highly misleading articles like this seminar but it has happened numerous times before, also in relation to psychiatry. I shall only mention the two recent network meta-analyses by Andrea Cipriani et al., one in children and one in adults.

Cipriani et al. concluded that fluoxetine was significantly better than placebo for depression in children and adolescents and was also better tolerated than duloxetine and imipramine. However, they mostly used published trial reports. When David Healy and I restored the two pivotal fluoxetine trials by analysing the clinical study reports submitted to the FDA, we found that fluoxetine was ineffective and unsafe (Int J Risk Saf Med, in press).

About Cipriani et al.’s second network meta-analysis, I published the article, Rewarding the companies that cheated the most in antidepressant trials, on Mad in America. Later, my research group showed that the outcome data reported in Lancet differed from the clinical study reports in 12 of the 19 trials they examined.

Lancet is not the source to go to if one wants reliable information about depression pills. It is the extended marketing arm of the pharmaceutical industry, just like the New England Journal of Medicine is, also in relation to denying that depression pills cause suicide.


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.

Leave a Comment