After the publication of our umbrella review of serotonin last summer, several psychiatrists wrote letters to the journal, Molecular Psychiatry, as usually occurs after the publication of a major finding. We were invited by the editor of the journal to respond to the points raised in the letters, again a routine procedure in scientific literature.

One letter was written by a large group of authors headed by Sameer Jauhar from King’s College London. Jauhar and colleagues made various criticisms, and we responded in detail to all of these and to points raised in other letters in our ‘authors’ response’ . However, when it was published, the letter by Jauhar and colleagues was referred to not as a letter, but as a ‘comment’. King’s College issued a press release about this ‘comment’, in which there was no mention of our response. Anyone who read the press release would have had the impression that this was a new article raising criticisms that we had not, and may be could not, respond to. 

In our ‘authors’ response’ we highlight how, contrary to the accusations, we had used approved and well-accepted methods for the umbrella review, including pre-registering the protocol, using recommended search methods and quality assessments, and we had not, in fact, missed out studies as was claimed. We also pointed out how, even if every point of criticism was correct (which it was not), it would not affect the conclusions of the review, and would not establish a link between low serotonin and depression.

Some of the letters claimed the serotonin theory of depression had long since been abandoned. Jauhar and colleagues say they do not subscribe to the simple theory that depression is caused by low serotonin, but still maintain that serotonin is abnormal in people with depression, and that this can explain how antidepressants work. To justify this claim they cite evidence from tryptophan depletion studies published before 2007, involving 180 people with depression or fewer (there is evidence of possible duplication of studies), most of whom were using or had used antidepressants (which may affect results). There has been no systematic review since 2007, but these findings are contradicted by later studies. In any case, the evidence suggests and everyone agrees that tryptophan depletion does not induce depression in people who do not have depression, which is the real test of the hypothesis that low serotonin causes depression.

Several of the letters were critical of us for suggesting that the lack of evidence for a relationship between serotonin and depression has relevance for the understanding and use of antidepressants. We maintain it is highly relevant. First it suggests that the marginal differences between antidepressants and placebo that are apparent in clinical trials are likely to be produced by alternative, more plausible mechanisms like the emotional blunting effects of the drugs or by amplified placebo effects, rather than by targeting underlying biological mechanisms (since these have not been demonstrated). And second it highlights how we don’t know what antidepressants do to the brain exactly, which is a cause for concern. The public need to know these facts because if they had all this information, they might make different decisions about whether to take antidepressants.

It remains the case that no reasonable scientist could conclude that the link between serotonin and depression has been established.

Author’s response in Molecular Psychiatry: https://www.nature.com/articles/s41380-023-02094-z