Jack Goulder is an Academic Clinical Fellow at King’s College London and a psychiatry resident at South London and the Maudsley Hospital. He writes about mental health and the history of psychiatry.
Shortly after I qualified as a doctor, I started working in a child and adolescent psychiatry clinic. It was in the months after the COVID-19 pandemic, but in its wake young people started coming to see us in ever greater numbers. Their lives, perhaps more than any of us, had been irreversibly altered. As the restrictions eased, we began to see a second, silent epidemic among them.
Our clients appeared more complex, more severely distressed, than those we had seen before. The service was staffed mainly by therapists and social workers. As a result, medications like antidepressants—called “antidepressants” but used for a variety of conditions including anxiety disorders—had always been sparingly used, after everything else had been tried. More and more, therapists would come to see one of the doctors and suggest we try an antidepressant.
To me, it seemed like an act of therapeutic hope, to think a drug could do much for the uncertainties they were facing. We knew that antidepressants were far from miracle drugs.
For some, though, they did meaningfully help.
Obsessive-compulsive symptoms were especially common after COVID-19, perhaps inevitably after our collective anxieties around infection and contamination. I remember seeing a fourteen-year-old who every day bandaged their hands, arms and legs with thick plaster-tape, the skin red and irate underneath them. I could see this bright, curious person sitting in front of me, but they were obscured beneath a deep sense of threat that shaped almost every thought.
We started an antidepressant. After a few months, and with the help of therapy too, the tightness of that fear had begun to slip. Each appointment, more skin appeared from beneath the bandages.
I thought of them, and other people I’d known helped by medication—patients, loved ones—when I read this passage by the Canadian writer, Sheila Heti.
“I was in the grocery store when it began. I felt suddenly unafraid. I had never before realized I had always been so afraid. The people around me doing their shopping seemed less menacing than before, when I would try to avoid them and avert my eyes… That night, getting ready for bed, Miles sang a sweet and funny song he made up on the spot, about me, and I said to him, suspiciously, Why are you being so nice to me? Then I said, Are you always this nice to me? He said, Yes.”
In Motherhood (2018), Heti describes her experience of depression, her many attempts to feel well, “by old-fashioned means – delving into my past, religions, spirituality, dreams,” and eventually with antidepressants. Like other people I knew, they helped her not by making her happy. Rather, they reduced an enveloping sense of dread that colored her inner life. In that small space left open by constant rumination, easier thoughts began to emerge.
Antidepressants work for some people but not for others. We don’t know why. A 2022 study led by Marc B. Stone at the FDA looked at the data of 73,000 trial participants and showed that about 25% of those prescribed an antidepressant would get a large clinical benefit from them (most no longer met criteria for depression). But then, 10% of people given a placebo experienced a large improvement as well.
We often compare treatments using the “Number Needed to Treat”: roughly, the number of people who need to receive a treatment for one additional person to benefit compared with placebo. For antidepressants, this seems to be about 7 people. That is not great, but it’s not that bad either, and it’s the same as many commonly used drugs, like those that treat back pain or migraines.
But the idea that antidepressants might ever be helpful or that a subset of patients might experience substantial benefit from them is still disputed by some. In 2022, whilst I was still working in the clinic, I noticed growing negative discussion online and in the media about these drugs. Some of it seemed for good reasons, like discussion around side-effects when medications are stopped. But I also noticed that people increasingly voiced the view that antidepressants had no use at all.
Having worked through a desperate time in mental health, I thought there was all the difference in the world between having a mediocre treatment and no treatment at all.
Newspaper articles kept citing the same group of British “critical” psychiatrists. One of the most prominent was Joanna Moncrieff, an NHS psychiatrist and a professor at UCL. In July 2022, she and others had published an umbrella review of evidence around serotonin’s role in depression that had found there was no evidence to support the view that low serotonin causes depression. When quoted in the media, these researchers often argued that this meant antidepressants could not be, and are not, effective treatments for depression. But that study never asked the question of whether they worked.
The same group of researchers has a long history of challenging the evidence supporting their use. They believe that trials showing their benefits are full of irremediable biases, like unblinding, and that harms have been underplayed. Disagreement among researchers, even about the same evidence, is very common. What interested me was that their critiques didn’t seem to settle as new evidence emerged. Instead, it appeared to grow louder—and more public.
In September 2024, I took a break from clinical work and moved to Boston to start a year-long fellowship in the history of science at Harvard. I wanted to research the history of medical ethics: how doctors responded to the many crises in public confidence they had faced during the twentieth century and how this had changed their practices. Within months of arriving, the country I had moved to had shifted dramatically.
In November, Trump won re-election and soon nominated Robert F. Kennedy Jr. for Secretary of Health and Human Services. During his Senate hearings, he made clear his skepticism about many established scientific positions, including the safety of common vaccines. He also made some other—to me, more surprising—claims about psychiatric medications, including that there was evidence that antidepressants raised rates of ‘homicidal behavior.’ He repeated this view after the Minneapolis shooting tragedy a few months later, suggesting that SSRIs might be behind the rising rates of these acts of violence.
In July, the FDA, under Kennedy, hosted a panel on SSRIs and pregnancy. After various members made unevidenced claims about the harms they posed to pregnant women and neglected much of the evidence that showed they were safe and helpful, professional bodies voiced concern. The American College of Obstetrics and Gynecology called the panel ‘alarmingly unbalanced.’
One of the panel members was Moncrieff. Joining it on zoom, she told the audience, ‘there is very little evidence that antidepressants have actual benefits in depression. It is completely understandable that they will have a variety of harmful effects because they are changing the normal state of our brain chemistry… And of course, as we’ve heard, these effects can be passed onto the offspring as well.’
A few months later, the HHS announced its new efforts to curb psychiatric prescribing at a “Make America Healthy Again” summit. Moncrieff posted on X, saying ‘I don’t agree with all RFK Jr’s views, but am grateful that he is calling attention to the enormous problem of overmedicalisation and epidemic levels of use of psychiatric drugs.’ She was later quoted in an HHS press release.
I didn’t understand why a British psychiatrist and researcher would be willing to lend support to a political movement that seemed so uninterested in balanced scientific inquiry. This was unlike any academic debate I had seen; it no longer felt like a scholarly disagreement among scientists about data. It felt more motivated, but by what I didn’t understand.
I spoke to my supervisor and told her I wanted to change my research question. Over the following months, I began looking at the long history of antidepressants, retracing their controversies, the arguments of their critics, and interviewing some of the people involved.
That work would eventually become an article, “Beyond reasonable doubt: Antidepressants and the limits of scientific critique,” published recently in the academic journal, Medical History.
Like much of psychiatry, antidepressants have had a turbulent history. After their mood lifting effects were first discovered in the 1950s, they were first given the name “psychic energizers” but were later rechristened as the idea of depression changed around them. Epidemiological studies had begun to suggest depressive symptoms were much more common than previously thought. This, and the discovery of the effects of the drugs on neurotransmitters, made space for a more biological understanding of depression to emerge.
These were often tied to—what seem to us now—simplistic stories of chemical imbalances which were seized upon by drug companies when they advertised a new wave of drugs, the SSRIs, that were licensed in the 1980s. It was a hugely successful strategy. Prozac became the most prescribed medication in the USA by the 1990s. But, before long, a backlash had emerged.
Some grew increasingly concerned about the toxic effects of financial interests on the supposedly objective trials used to win approval for these drugs. Negative trials were suppressed and only positive ones published. In some instances, industry executives had ghost-written academic articles. Many researchers had serious doubts about the science that underpinned their use.
These practices were not unique to psychiatry, but some of the worst examples came from antidepressant trials. They were referenced in US Senate committee hearings leading up to the passing of the 2007 FDA Amendments Act that forced companies to pre-register all trials, so negative ones could be less easily suppressed.
Others made more outlandish claims. The psychiatrist Peter Breggin was one of the first to push back against the rising use of medication, but drew controversy after he argued SSRIs contributed to the 1999 Columbine school shooting. This claim would be repeated after future tragedies by other figures, including eventually by RFK Jr.
Researchers also began to look for ways to overcome the flaws in earlier studies. They pooled the results from many different trials, which are then added together to give larger, more objective results. In 2018, a group from the University of Oxford published the largest of such studies, including data from over 500 trials. It concluded that most antidepressants seemed modestly effective. Not the wonder drugs they were sold as, but useful for some. The scale of the study seemed to quell the doubt for many clinicians and researchers. But not all of them.
Moncrieff had been an early critic of antidepressants during their years of controversy and remained unconvinced by this new evidence. Along with colleagues, she published a response criticising the study, suggesting that the paper showed antidepressants should be used ‘far less or perhaps not at all.’ They argued positive findings were better explained by the fact that participants knew they were taking these medications because they had side-effects. (The evidence for this is surprisingly sparse, but Stone’s paper showed most studies that had looked at this found participants were no better than chance at guessing whether they were taking an antidepressant or not.)
Shortly afterwards, she assembled a team of like-minded researchers to start work on the umbrella review of the ‘serotonin hypothesis,’ an early theory of how antidepressants worked, which had been disproven but remained a common understanding among many people.
When it was published in 2022, it was quickly picked up by media outlets across the globe and spread rapidly online. Some scientists were puzzled why a paper that seemed to say something they already knew was getting so much attention. Moncrieff and her co-authors argued that it was a sign of how much public misunderstanding there was.
But others also noted that the media often seemed to confuse what the paper actually said with the claim that the drugs didn’t work at all. This wasn’t just a misunderstanding by journalists. Whilst the paper itself was careful to avoid making claims about efficacy, a press release written by Moncrieff and co-author Mark Horowitz explicitly linked the umbrella review to a lack of efficacy in antidepressants.
Press releases, unlike academic articles, do not go through peer review. Nevertheless, staff in the UCL media office were uneasy about including comments on efficacy as this diverged from the article’s content. They tried to work with the authors on a revised version. However, Moncrieff and Horowitz pre-emptively distributed their own version to the press that included comments on efficacy without the press team’s knowledge.
In later media appearances, Moncrieff frequently suggested that the paper discredited the efficacy of antidepressants. To be fair, Moncrieff and colleagues have made a case for antidepressants being ineffective in various prior academic papers, but as their views became increasingly fringe, these publications neither convinced the scientific community nor generated significant attention from the media. Moncrieff used the attention generated by the umbrella review to publicly promote her view that the inefficacy of antidepressants was conclusively demonstrated. Even as the paper itself faded from view, it left a widespread public perception that scientists disagreed among themselves about whether they worked at all.
The authors are not the first researchers to misrepresent their findings in a press release. However, this small act of obfuscation seemed to mirror a pattern I had seen among figures critical of antidepressants, where media coverage was used to create doubt among the public, even when that doubt seemed resolved among the majority of scientists. This doubt would be seized upon by the “Make America Healthy Again” movement in the years that followed.
My question remained, however: what drove this small group of researchers to diverge from their peers, and do so even more loudly as the science appeared more settled?
In 2004, the historian of science, Naomi Oreskes, published an essay in Science analysing 900 peer-reviewed climate abstracts and found that none rejected the consensus that human action has contributed to global warming. While much of the response was warm, it drew a small number of personal, hostile attacks. One of them was from the respected physicist, Fred Singer.
Talking to another historian, Erik Conway, at a conference a few months later, she told him about the attack. Conway, who was researching ozone hole science, recognised Singer as the same man who had attacked the work of the pioneer scientist, Sherry Rowland. Oreskes and Conway realised there was a much larger pattern at work.
I read their seminal book, Merchants of Doubt, whilst taking a seminar with Naomi at Harvard. In it, they look at how, during the twentieth century, a small group of contrarian scientists consistently worked to undermine public understanding about the scientific consensus on issues like climate change. They did this by repeatedly challenging new findings, by publicly insisting on doubt when there was none.
What fascinated me was what drove people like Singer to stick to positions that seemed to be disproven by a balanced view of the evidence, and to use their public platform to undermine those who disagreed with them. Some of them had received money from fossil fuel companies, but many others hadn’t. And their disagreement often started before any industry support.
Instead, most were driven by something more personal, and harder to declare on ethics disclosures. They were, at heart, cold-war free-market libertarians. They hated the idea of government regulation in industry and any science that seemed to justify it. This isn’t a guess. They wrote about this worry repeatedly in correspondence and published material.
In her personal and philosophical writings, Moncrieff has written about the things that motivate her. She expresses a worry about the medicalizing impulses of psychiatry, the tendency to numb normal suffering with psychiatric drugs, the malign impact of industry and economic forces both on mental health and society at large.
Reading through these articles during my research, I found myself sympathetic to much of it. Like her, I was a reluctant medical student, more interested in philosophy, and resistant to the easy reductionism of biological psychiatry. And we aren’t alone. In fact, many psychiatrists I meet have critiques to make of it and have been influenced by rebellious figures who have called it into question.
For Moncrieff, that figure was Thomas Szasz, the libertarian psychiatrist who saw psychiatric diagnoses as tools society used to suppress and control its citizens. Szasz’s argument rested on the fact that mental illnesses seemed to have no biological underpinning—a view which was much more plausible in the 1950s, before the emergence of modern neuroscience.
All scientists bring their prior beliefs to how they look at evidence. I can see how coming up through a time where scandals around antidepressants were mounting in the 1990s would push you further towards an entrenched position against them, especially if your priors are already pointed that way.
However, as I conclude in my article, some prior beliefs appear to impede the ability to accept new evidence. Moncrieff inherits Szasz’s commitment that any attempt to link human experience with the biology of the brain is fundamentally flawed. By extension, medication can never really treat mental illnesses, they can only generate the appearance of doing so.
Leaving aside the strengths and flaws of this Szaszian line of thinking, committing to this position too rigidly means that one can never really accept new evidence that suggests these drugs can have a meaningful benefit. Instead, as the historical record shows, that evidence only provokes ever louder critique.
By increasingly voicing that critique outside scientific arenas—like peer-reviewed journals—in public venues and online, these critics appear to replicate many of the tactics Oreskes and Conway describe in their work, what they term “manufactured doubt”: repeatedly reopening questions that appear to be answered and presenting the scientific consensus as if it remains unsettled.
From the perspective of these critics, this isn’t as cynical as it sounds. They sincerely believe in the truth of their views and the epistemic force of their doubts. But motivated reasoning rarely registers as motivated reasoning when one is in the middle of it.
I started practicing in the wake of another scandal, one where scientific knowledge has become far more contested—online, in the media, and increasingly among the politicians who supposedly safeguard the structures that produce that knowledge.
Scholars like Oreskes and the late Bruno Latour have drawn attention to the fact that the “game” of philosophical critique can go too far: it has at times undermined attempts to obtain verified knowledge about the world—knowledge that can be used for the benefit of people in that world.
Looking back at the history of antidepressants over the last eighty years, my view is a little different. I think that psychiatry needs careful external scrutiny and critique. It has often benefited from it—by exposing harms, forcing changes in practice, and correcting overconfident claims.
But I think there is a difference between this kind of productive doubt, which demands further inquiry and leads to advances in collective knowledge, and an unproductive kind where further inquiry only reinforces further doubt.
I think a lot about the patients I met in the adolescent clinic, even years after leaving it. It was my first lesson in professional humility. But it also taught me the value of therapeutic hope. Psychiatrists can be both sceptical of the tools they use but be ready to use them when there is need.
It should be possible to hold both sides of the debate. We can recognise converging lines of evidence from decades of research and clinical practice that support the use of our treatments. We can also have the epistemic humility to recognize the limitations of our knowledge and to listen to patients when they tell us about the harms as well as the benefits of our treatments.
See also:
Dr Peter Kramer wrote about this in his book ‘ Listening to Prozac “Worth reading .
This is a beautifully written, brilliantly argued analysis. What stands out is Goulder's open-mindedness, fair-mindedness, and even-handedness in explaining the positions of those holding positions lacking a sound evidentiary (to say nothing of logical) foundation — indeed, in many cases, positions with "supporting" evidence that actually undermines them. This is not to suggest that psychiatry's critics do not have valid critiques. They do. But that psychiatry has harmed some (even many) over the past two centuries does not mean it has helped none. It has certainly helped me and hundreds of thousands of others. Both can be true. Both are true. Sadly, nuance is not a way to build a brand, monetize a movement, or become a talking head. Provocation may sell well; it does not heal.