The serotonin theory of depression: a systematic umbrella review of the evidence, Moncrieff et al, 2022

Joanna Moncrieff, Ruth E. Cooper, Tom Stockmann, Simone Amendola, Michael P. Hengartner & Mark A. Horowitz

The serotonin hypothesis of depression is still influential. We aimed to synthesise and evaluate evidence on whether depression is associated with lowered serotonin concentration or activity in a systematic umbrella review of the principal relevant areas of research.

PubMed, EMBASE and PsycINFO were searched using terms appropriate to each area of research, from their inception until December 2020. Systematic reviews, meta-analyses and large data-set analyses in the following areas were identified: serotonin and serotonin metabolite, 5-HIAA, concentrations in body fluids; serotonin 5-HT1A receptor binding; serotonin transporter (SERT) levels measured by imaging or at post-mortem; tryptophan depletion studies; SERT gene associations and SERT gene-environment interactions.

Studies of depression associated with physical conditions and specific subtypes of depression (e.g. bipolar depression) were excluded.

Two independent reviewers extracted the data and assessed the quality of included studies using the AMSTAR-2, an adapted AMSTAR-2, or the STREGA for a large genetic study. The certainty of study results was assessed using a modified version of the GRADE. We did not synthesise results of individual meta-analyses because they included overlapping studies. The review was registered with PROSPERO (CRD42020207203).

17 studies were included: 12 systematic reviews and meta-analyses, 1 collaborative meta-analysis, 1 meta-analysis of large cohort studies, 1 systematic review and narrative synthesis, 1 genetic association study and 1 umbrella review. Quality of reviews was variable with some genetic studies of high quality.

Two meta-analyses of overlapping studies examining the serotonin metabolite, 5-HIAA, showed no association with depression (largest n = 1002).

One meta-analysis of cohort studies of plasma serotonin showed no relationship with depression, and evidence that lowered serotonin concentration was associated with antidepressant use (n = 1869).

Two meta-analyses of overlapping studies examining the 5-HT1A receptor (largest n = 561), and three meta-analyses of overlapping studies examining SERT binding (largest n = 1845) showed weak and inconsistent evidence of reduced binding in some areas, which would be consistent with increased synaptic availability of serotonin in people with depression, if this was the original, causal abnormaly. However, effects of prior antidepressant use were not reliably excluded.

One meta-analysis of tryptophan depletion studies found no effect in most healthy volunteers (n = 566), but weak evidence of an effect in those with a family history of depression (n = 75). Another systematic review (n = 342) and a sample of ten subsequent studies (n = 407) found no effect in volunteers. No systematic review of tryptophan depletion studies has been performed since 2007.

The two largest and highest quality studies of the SERT gene, one genetic association study (n = 115,257) and one collaborative meta-analysis (n = 43,165), revealed no evidence of an association with depression, or of an interaction between genotype, stress and depression.

The main areas of serotonin research provide no consistent evidence of there being an association between serotonin and depression, and no support for the hypothesis that depression is caused by lowered serotonin activity or concentrations. Some evidence was consistent with the possibility that long-term antidepressant use reduces serotonin concentration.

https://www.nature.com/articles/s41380-022-01661-0

https://twitter.com/user/status/1549697303778320384

 

The idea that low serotonin caused depression always seemed to me to be based on someone's bright idea that had potential as a money-making scheme. And, boy! it must have made billions and billions and billions since the 1980s. Have patents started to run out?

I have wondered what percentage of people recovered from depression on SSRIs and were able to come off them, cured, what percentage got better but had to stay on them for life, and what percentage got no benefit whatsoever on SSRIs.

 

Expert reactions gathered by the Science Media Centre, mostly critical of the study, pointing out:

- even though the exact mechanisms by which SSRIs function, evidence from clinical trials supports their use for depression

- the methodological limitations of the study, i.e. looking at proxy measurements of brain serotonin levels and not taking into account evidence from clinical trials of SSRIs

- psychiatrists have long stopped believing serotonin is the sole contributor to depression

- recent evidence with better technology provides some support to this hypothesis

https://www.sciencemediacentre.org/...-paper-on-the-serotonin-theory-of-depression/

Edit: Dr Mark Horowitz, senior author, addresses some of these claims in his Twitter thread.
https://twitter.com/user/status/1549692299025813504

 

It's wrong but it's still correct? Damn. What an ideology, uh? Especially given the very dubious evidence from trials, all of which ignore the serious side effects. Psychiatry seriously needs massive auditing and a firm smack down, having basically made medicine worse for decades with almost nothing to show for it.

Because this is basically another in a long line of confident claims that turn out to be completely bunk, initially based on nothing at all. It's basically the most common tradition in psychiatry, and it won't change until this whole disaster is reformed. I would be very surprised if, say in 30 years, anything current in psychiatry actually holds up, most of it is too superficial to hold up to actual scrutiny.

 

I hope you all enjoy the comments in the SMC release that their experts Bloomfield and Nutt have "No declarations of interest". That may also have been the opinion of their readers. That is not a statement that they have no conflicting interests.

 

I'm reading this, and trying to reconcile the conclusion to focus on psychosocial issues, with a paper I recently saw on psilocybin being used to successfully treat depression in fruit flies, and it's really hard not to conclude that depression has simply been used as a catch-all for a number of different things, at least as badly as ME/CFS, or fatigue in general. Not that it's not a thing, but that it's clearly not a single thing, it's probably just some kitchen drawer of tens, if not hundreds, of different things that are simply poorly lumped together because of tradition, mostly depending on which bias applies.

There was a reddit thread where some of the authors replied to questions, and basically what they defined as depression in this case is essentially sickness behavior: lethargy, not eating, etc. It says that they used the same "swim stress test" that they use to torture rats to create a model of chronic fatigue. I have no idea how any of this relates to depression at all, or how acute overexertion has anything to do with a chronic condition.

Now I'm no expert in fruit flies but flies don't even have a cortex, let alone a neocortex, just about the most basic nervous system you can think of. It has no conscious thoughts or beliefs, certainly don't have complex moods relating to anything psychological or social the way they're typically defined for humans: mostly well-being and overall community belonging. They don't have career goals, aspirations, or anything related to high-order thinking, which is the only place when those thoughts and beliefs, cyclical or not, can happen.

But so many people who became "depressed" met every single opposite criterion, they had everything psychosocial going for them and still became "depressed", so it clearly can't be that. It's clearly not simple life satisfaction, or lack thereof.

Now when it comes to clinical depression, I have so often read people saying they suffered from it and how they were overwhelmed with negative thoughts, excessive self-criticism, thinking of themselves as failures, unloved, useless, etc. Which has nothing to do with those psychosocial factors, this is clearly a different thing.

And then of course the vast majority of physicians are taught and believe that chronic fatigue is basically depression, even though for most of us all the conditions fail: so many of us who lived happy fulfilling lives in good health and never did at any point have such intense... let's say self-persecution. So clearly nothing to do with the new shiny version of the philosopher's stone.

What I'm seeing basically is old ideas that are still being carried to this day, and those traditional definitions are seriously holding back overall progress in medicine. Medicine has essentially completely stagnated on an entire half of the discipline, illness, and is unable to move the whole field forward because it's basically leaving out most of what matters, now that we can effectively handle most threats to life.

Seems that most of all, medicine is simply incapable of self-reflection, especially of self-blame. Which is ironic, considering how much self-blame there is in depression, at least in some, and in most of the models. It's just unthinkable that old ideas could be wrong, there is a clear tyranny of dead people here.

 

I hope the author's next paper is about POTS in fruit flies, I want to see the tilt tables.
Sorry for being off-topic. I just... can't anymore with those studies.

 

The problem has always been that depression has multiple causes. A minor subset may be due to serotonin pathway issues, but if a majority are not, we may still see the sort of results from the studies cited above.

 

Yes, fatigue, pain and depression seem to be broad categories requiring deep investigation and subcategories. This can play havoc with drug design.

I have not taken the serotonin hypothesis of depression seriously since about 2000. I was talking to a serotonin pharmacologist, and the best evidence against it was not even published. Drug companies worked on this and the better the drugs got restoring serotonin the less they worked on depression. SSRIs are selective in the marketing sense, not the biochemical sense, because prior drugs were less selective. High drug specificity causes them to utterly fail despite raising serotonin. I could be wrong but I suspect we still do not know why SSRIs can work. What was not ruled out back then is whether or not its serotonin plus something else that works, and SSRIs did both.

 

After giving it some more thought, I am somewhat struck by the expert reactions (those of eminent psychiatrists) gathered by the SMC, in that they highlight what has been called the inability of the biopsychosocial model to define saliency.

On the one hand, psychiatrists say they adhere to the BPS model of depression (and anxiety, for that matter) but very much seem to focus on the biological part, probably because they believe antidepressants are an effective treatment and because genetic and neuroimaging studies provide evidence in favour of this view.

On the other hand, they reject biomedical explanations for ME/CFS despite the wealth of data (even though most lack in quality), save for it being triggered by an infection in most cases, and instead insist that the disease must be maintained solely by psychosocial factors.

In a sense, the lack of a clear delineation between each component of the biopsychosocial model seems like a “get out of jail free” card. With it, psychiatry always ends up being correct even when it is wrong because the BPS model covers so much ground (too much, in fact).

It is certainly true that multiple conditions have ended up being lumped under what is called depression. A work-related burnout is not the same as a familial history of depression, nor is depression from adverse childhood events the same as grieving someone’s passing or a romantic relationship. Needless to say, this lumping has served the interests of pharmaceutical companies well, but it is very unhelpful when it comes to understanding what depression is — or rather, what depressions are. It is past time for psychiatry to make the difference between them.

 

Isn't that the whole point of the biopsychosocial model? It is nothing but a series of "category errors" which includes everything, excludes nothing, and makes differentiation impossible. it was not an accident that Wesssely was stating in 1989, or whenever it was, that aetiology is not mportant. You don't know what is and what is not important until the problem is solved.

 

Rather than anything substantive to me this paper feels like a Reformation era intra-sectarian counterblast where one group of theologians seeks to lay waste to another group that has suggested a particular scriptural interpretation is/is not heretical.

Here's a more substantive counter counterblast than given by the SMC:

 

It seems to me like it's more of a symptom, like fatigue or pain, which can be broad in range and impact, and constitutes one possible outcome of sickness behavior, but will vary wildly depending on how it actually works. But since it's a symptom that rarely comes alone, it's almost impossible to study with current medical systems, which are simply unable to deal with this level of complexity, especially given how specialties work in isolation and symptoms tend to be multisystemic. It's probably part of the sickness response in a mostly generic sense, with a root cause that could be identified but is simply beyond our level of technology.

Medicine still knows very little about the brain. It's the confident assertions going back decades that ruin everything, that once a fashionable explanation has been latched on, it will have to be tyrannically enforced as if it's the very word of all gods and anyone going against the fashion will be swiftly punished for getting out of line. Medicine is way too comfortable with being confidently wrong, making all the emphasis on confidence and paying very little effort on what's actually true, because for most medical knowledge it's a social construct, it's not about what's real because we don't know that yet, but what people accept as real, especially in relation to what's already known, how many egos it will bruise.

Something I've been thinking about for a while is that medicine needs to adopt an uncertainty principle, the overarching idea that no matter what they know about a patient or about the human body in general, most things are impossible to know because they have to be systematically and reliably checked before being sure. Physicians should never think they've checked everything or that a single cause can explain all the outcomes, everything is uncertain and however much that's awkward to work with, it's just the damn truth: ignorance is the default state in healthcare, if only for the fact that all medical knowledge is built on averages of human bodies, and no one is perfectly average.

There is simply far too much confidence in old ideas that completely lack in validity. This is by far what's holding back medicine the most, even above technology, because all the technology that we could have already built wasn't simply because of the need to be confidently incorrect above accepting uncertainty. It's an indirect form of Luddism, the idea that current knowledge is superior to future knowledge.

If it means changing the culture of medicine in a way that many practicing physicians would simply not be able to operate, then so be it, it's the equivalent of leaving out from law enforcement authoritarian personalities because they are most likely to use unnecessary violence.

 

Uh, never noticed that. They really do eat their cake and have it, too.

In the end, popular and wrong ideas are popular, while unpopular and right ideas are unpopular. Right and wrong are determined according to what's in textbooks, which fails when textbooks are wrong, especially on ideas that are firmly cultural, never had any actual basis in evidence. This is one thing that's missing in how we deal with expertise: before textbooks are right, wrong ideas can actually make experts do worse than nothing at all. And textbooks are almost always wrong pre-breakthrough. But sometimes everything aligns so that the false truths in the textbooks will be used to completely block the actual truth from being known, it's too unpopular anyway, even when it comes out it's rejected. We're seeing it again and again with Long Covid.

The entire system of science needs reform, I don't think other fields are much better, but the impact in medicine is simply too big, too personal, it directly impacts the lives of millions at least as much as what legislatures do, and yet nothing is accountable.

 

@CRG Dr Harris’ thread seems to be a reiteration of some of the SMC’s expert reactions, specifically that the serotonin system is complex and that the areas of research investigated by the review do not cover it entirely / are simplistic. This seems like a valid criticism.

On the other hand, the authors are right to point out that the currently available evidence behind the serotonin theory of depression does not substantiate it. This was, after all, the basis for the development of SSRIs. Their conclusion that serotonin levels or activity in the brain may be incorrect if future evidence shows otherwise, but perhaps more plausibly, a better understanding of the biology of depression will tell us more in time.

Note that the effectiveness of antidepressants compared to placebo has been questioned by a more recent narrative review than the Cochrane systematic review (the one cited by Dr Harris, which only included 2 trials of SSRIs). It discusses the Hamilton depression scale and, importantly, its minimal clinically important difference: https://ebm.bmj.com/content/25/4/130

It is indeed a battle between two schools of thoughts. Dr Moncrieff and colleagues seem to support psychotherapy and behavioural / lifestyle changes rather than medication, as per Dr Horowitz’s thread, but these are modestly effective at best. They also place the onus of improving on the patient — this is particularly problematic their depression owes to systemic issues that are out of their control.

At the end of the day, depressed people are stuck between a rock and a hard place. They have to choose between taking antidepressants of uncertain efficacy and poor safety (including the risk of prolonged withdrawal symptoms), following psychotherapy — not free of harms either —, or doing both. The result is that we don’t have truly effective and safe treatments for depression, all because we need more biomedical research into it (as well as fewer financial conflicts of interests in psychiatry).

 

Well, the 2nd tweet is just strawmanning, and the 3rd misses the point that this paper doesn't prove anything, it just points out that the long-standing claims never had any basis. The burden of proof is on those affirming something, not those criticizing it. There simply was never any basis for the chemical imbalance, there is no need to prove anything, only to accept that an old "proof" was basically a giant mistake that needs to be corrected.

And to cherry-pick the "benefits" of SSRIs, even though many studies and reviews show the opposite while ignoring almost all issues and side-effects. That's the giant flaw at the center of EBM: you can cherry-pick anything that supports your view and ignore everything else, which is what this guy is doing.

The default position is that any affirmation has to be proven. The chemical imbalance never had any proof, it was confidently asserted in a way that diminishes the credibility of medicine, and will continue to do so until a major correction happens in the field.

And because it will be discussed, that something will definitely not be EBM+, the small tweaks to a fundamentally broken system that Greenhalgh is hyping these days.

 

Highly likely.

Very early science. Nothing we study is as complicated as the brain, and we still lack the tools we need.

The working title of the book I have not been able to work on in many years is "Embracing Uncertainty'. Part of the advice I wanted to suggest is the most important words out of a doctor's mouth are: "I don't know." Doctors are taught to draw quick conclusions (heuristic reasoning) and to present confidence (including the benevolent lie). From what I have seen not enough doctors balance heuristic reasoning with logic and hard evidence. Those who do are often researchers.

 

I vaguely recall this Ioannidis paper causing quite a bit of controversy at the time:

 

Freudian babble. Popper originally claimed Freudian psych was non-science because NO finding would contradict it. It was always right by definition, so no experiment would ever test it. Later this was relabelled pseudosience, but I do not recall by who. In I think 1970 we got the term psychobabble (from the book of the same name). I could be wrong about the date.

 

Absolutely.