The Serotonin Surplus: Are We Medicating Sadness Out of Existence?
A patient stops her SSRI under medical supervision over four weeks. She has never previously experienced panic attacks, insomnia, or akathisia. All three now arrive at once, alongside brain “zaps,” agitation, and suicidality. Emergency services diagnose agitated depression, then revise to a psychotic disorder. She is involuntarily treated with antipsychotics. But a paper published in Psychotherapy and Psychosomatics in early 2025 by the psychiatrist Mark Horowitz and the psychotherapist James Davies documents precisely this pattern and names it for what it is: withdrawal symptoms misdiagnosed as relapse — or worse, as an entirely new illness. The diagnostic instruments used in most discontinuation trials cannot distinguish between the return of a condition and the pharmacological consequences of stopping a medication. The patient is not sick again. She is trapped.
[https://markhorowitz.org/]
That case sits at one end of a system that has, by any quantitative measure, lost all sense of proportion. In England, 92.6 million antidepressant items were dispensed to nearly nine million patients in the year to March 2025, a four per cent rise on the previous twelve months and part of a trajectory that has seen sertraline prescriptions alone climb 685 per cent between 2010 and 2023. In Australia, the proportion of the population filling an antidepressant script rose from twelve per cent to fourteen per cent over the decade to 2024–25 — 3.9 million people, with GPs writing ninety two per cent of the scripts. Iceland leads the OECD table at roughly 165 defined daily doses per thousand inhabitants as of 2023, with Portugal and Canada close behind. The global antidepressant market, already valued at some eighteen billion US dollars, is forecast to breach twenty billion by 2030, with the Middle East now the fastest growing regional segment — a striking detail for a region where, according to recent data, seventy two per cent of anxiety sufferers still rely on family or religious support rather than licensed clinicians. The pills are arriving before the psychiatrists.
None of this would matter if the prescribing were clinically justified. For moderate to severe major depressive disorder, the evidence base for antidepressants is solid and the drugs save lives. But the theoretical foundation has been quietly collapsing under the weight of the evidence. Professor Joanna Moncrieff’s 2022 umbrella review in Molecular Psychiatry — now downloaded over a million times — concluded that there is no consistent evidence that low serotonin levels cause depression. The chemical imbalance narrative that shaped four decades of prescribing practice has no robust empirical support. If antidepressants are not correcting a chemical deficit, what exactly are they doing? Meta-analyses of randomised controlled trials offer a sobering answer. As a recent appraisal in the Australian Journal of General Practice set out plainly, the average drug-placebo difference on a 52-point depression scale is roughly two points — below the seven point threshold that clinicians regard as the minimum for clinical importance. Even in severe depression, the gap is only about three points. These small effects are likely overestimated by unblinding, publication bias, and short follow up periods.
[https://pmc.ncbi.nlm.nih.gov/articles/PMC10618090/]
The Australian data was given further force by a paper co-authored by Moncrieff herself with Katharine Wallis and Anna King in the Medical Journal of Australia. Published in April 2025 under the title “Antidepressant prescribing in Australian primary care: time to reevaluate,” it documented a system in which one in seven Australians is now on antidepressants, rising to one in four among those aged seventy five and over, and in which long term use among young people aged ten to twenty four has more than doubled over the past decade. A University of South Australia study using PBS dispensing data confirmed that the prevalence of long term use rose from 66.1 to 84.6 per thousand population between 2014 and 2023, with apparent dose reductions — a proxy for deprescribing — showing minimal change over time.
[https://pmc.ncbi.nlm.nih.gov/articles/PMC12088320/]
The paediatric dimension deserves the hardest scrutiny. A Frontiers in Psychiatry editorial stated bluntly that although antidepressant use in children and adolescents has increased substantially over the past ten to fifteen years, convincing evidence that the benefits outweigh the risks is lacking, and treatment emergent suicidality remains a major concern. Robert F. Kennedy Jr.’s May 2026 federal action plan explicitly targets overprescription of psychiatric medications among children. Psychiatric professionals cautiously welcomed the intent while warning, correctly, that the emphasis on overprescribing risks obscuring the deeper problem of inadequate access to mental healthcare. Both things can be true simultaneously.
[https://pmc.ncbi.nlm.nih.gov/articles/PMC7661954/]
And that is precisely the structural tension: too many pills, too few therapists. The economics are unambiguous in the medium term, even if the short term picture is muddier. A decision analytic model published in the Annals of Internal Medicine found that over a five year horizon, CBT produced higher quality adjusted life years than second generation antidepressants and reduced total costs by approximately two thousand US dollars, though neither treatment demonstrated consistent superiority at one year. The UK CoBalT trial, published in the Lancet Psychiatry, was more striking still: CBT as an adjunct to usual care produced an incremental cost effectiveness ratio of £5,374 per QALY gained, with a ninety two per cent probability of being cost effective at the NICE threshold of £20,000. The treatment effect persisted at nearly four years’ follow up, long after the therapy sessions had ended. Antidepressants, by contrast, work only while you take them — and stopping them is harder than anyone acknowledged until recently.
The withdrawal problem is not peripheral. A 2025 reanalysis published in Psychological Medicine, using only studies that employed systematic measurement of withdrawal symptoms, found a pooled incidence of fifty five per cent. More than half of patients experience withdrawal. The symptoms — dizziness, insomnia, depressed mood, agitation, electric shock sensations — overlap extensively with the symptoms of depression itself, which means that clinicians using standard diagnostic instruments will frequently record a relapse that is in fact a drug effect. This creates a feedback loop: the patient stops the drug, develops withdrawal, is diagnosed as relapsing, is restarted, and becomes a long term user. The pharmaceutical companies need not conspire to keep people medicated. The pharmacology does it for them.
[https://pmc.ncbi.nlm.nih.gov/articles/PMC12315658/]
What would a genuinely reformed prescribing culture require? Enforcement of the guidelines that already exist: antidepressants for moderate to severe depression, structured psychological intervention for everything below that threshold. Investment in talking therapies at a scale commensurate with demand — England’s IAPT programme received 1.83 million referrals in 2023–24 but only 1.26 million people accessed the service, and the gap between referral and treatment is filled, almost by default, with a prescription. Mandatory long term outcome data, because most antidepressant trials run for six to eight weeks yet the average patient is on the medication for a year and a rising proportion for far longer. And honest public communication: the serotonin theory has, by the weight of the evidence, collapsed; the placebo gap is vanishingly small for mild depression; and the exit costs of these drugs have been significantly understated in clinical guidance.
[https://digital.nhs.uk/data-and-information/publications/statistical/nhs-talking-therapies-for-anxiety-and-depression-annual-reports/2024-25]
The serotonin surplus is not a conspiracy. It is a systems failure, born of a theoretical foundation the evidence no longer supports, underfunded mental health services, misaligned pharmaceutical incentives, time poor general practitioners, a withdrawal trap that the profession has only just begun to acknowledge, and a culture that has pathologised ordinary human suffering. Fixing it will require something that no pill can provide: the collective will to treat mental health as something more than a prescription to be filled.