Few practices in mental health are debated more than the long-term use of antidepressant medications.
A reassessment began in 2019, when two British researchers published a study that found that 56% of patients suffered from withdrawal symptoms when they stopped antidepressant medications and that 46% of those described their symptoms as severe, reports The New York Times.
The findings made headlines in Britain and had a powerful ripple effect, forcing changes to psychiatric training and prescribing guidelines. And they fed a growing grassroots movement calling to rein in the prescription of psychotropic drugs – that has, in recent months, gained new influence in America with the rise of Robert F Kennedy Jr as Health Secretary.
Now, a study published in JAMA Psychiatry makes the case that these warnings were overblown. The authors of the paper found that a week after quitting antidepressants, patients reported symptoms like dizziness, nausea and vertigo, but that they remained, on average, “below the threshold for clinically significant” withdrawal.
Dr Sameer Jauhar, one of the authors, said the analysis should reassure both patients and prescribers.
“The messaging that came out in 2019 was all antidepressants can cause this and this can happen in this proportion of people – but that just doesn’t survive any scientific scrutiny,” said Jauhar, a Professor of Psychiatry at Imperial College London.
He criticised the earlier study for including data from online surveys as a quantitative measure, for failing to control for the placebo effect, and for failing to distinguish between various types of antidepressants. These methodologies, he said, led to inflated estimates of withdrawal.
“The public concern was based on the data that were presented,” he said. “And maybe if the evidence had been this evidence, you might not have had things escalate to that degree.”
The latest analysis, based on 50 studies with more than 17 000 subjects, is not an apples-to-apples comparison. The authors did not try to measure the prevalence of withdrawal symptoms or their severity. Instead, they used a research tool to track individual symptoms present in patients after stopping an antidepressant, and compared them with the symptoms of patients who had stopped taking a placebo.
Patients who had quit antidepressants were 5.5 times as likely to report dizziness, 6.4 times as likely to report vertigo and 3.1 times as likely to report nausea compared with placebo. But on average, the number of withdrawal symptoms had risen modestly, by only one compared with placebo, which is not considered clinically significant.
The report is unlikely to settle the debate over withdrawal from antidepressants.
Even before the JAMA paper had been published, critics online were laying out their arguments against it – principally, that too many of the trials the team included were short-term efficacy trials, in which subjects had been taking the medications for no more than eight weeks.
“If you are looking at people on the drugs for eight weeks, you are not going to find withdrawal,” said James Davies, one of the authors of the 2019 study that found high rates of withdrawal symptoms. “It’s like saying cocaine isn’t addictive because we did a study on people who had only been taking it for eight weeks. “
Davies, an associate Professor of Psychology at the University of Roehampton in England, said he worried that the findings, “if read uncritically”, would “cause considerable harm by significantly downplaying the effects of real-world antidepressant use.”
“According to their conclusions, the tens of thousands of people online who are struggling with severe and protracted withdrawal aren’t really in withdrawal,” he said. “There are real people out there … who aren’t making up the fact that they are in a lot of pain coming off these drugs.”
One reason for the confusion is that for decades, researchers showed little interest in antidepressant withdrawal and did not conduct rigorous long-term trials. Physicians recognised withdrawal symptoms in some patients, but viewed them as transient, usually resolving after a few days. But attention to the subject is rising along with the controversy.
In 2024, a team of German researchers published a review of 79 studies in The Lancet that concluded that one in three patients quitting antidepressants reported withdrawal symptoms – and so did one in six patients quitting placebo.
Accounting for the placebo effect, 15% of patients experienced withdrawal symptoms like dizziness, headache, nausea and insomnia. Three percent of patients experienced severe withdrawal, they found.
Dr Jonathan Henssler, one of the authors of the 2024 study, said that he had hoped the work would contribute balance and calm to the discussion, but that had not happened.
“Unfortunately, the emotional charge of the discourse and the strong polarisation of reception both in the public media and in scientific discourse have not substantially diminished,” he said. He added that he worried, particularly, about the way findings would be conveyed to the public.
“Anyone who communicates in an exaggerated and alarmist manner, in the worst case, could leave patients braced for a negative outcome,” he said, “with the risk that they then close themselves off from potentially helpful treatment or maintain antidepressant treatment only out of fear.”
Dr Allan Young, one of the authors of the JAMA paper, said he had seen “a flare-up in concern about this from many patients” in his clinic after the 2019 analysis received widespread press coverage.
“I’ve certainly had patients at my clinic thinking they shouldn’t go on antidepressants when that was what I would recommend,” he said, adding that patients should be reassured by the new analysis.
“It’s quite clear that you can have some effects when stopping antidepressants,” he said, “but these are a limited number, and they do seem to decline over time.”
Study details
Incidence and Nature of Antidepressant Discontinuation Symptoms: A Systematic Review and Meta-Analysis
Michail Kalfas, Dimosthenis Tsapekos, Matthew Butler et al.
Published in JAMA Psychiatry on 9 July 2025
Abstract
Importance
The incidence and nature of discontinuation symptoms following antidepressant cessation remain unclear.
Objective
To examine the presence of discontinuation symptoms using standardized scales (eg, Discontinuation-Emergent Signs and Symptoms [DESS]) and the incidence of individual discontinuation symptoms in individuals who stop taking antidepressants.
Data
Sources
The databases Embase, PsycINFO, Ovid MEDLINE, and Cochrane Library were systematically searched from inception until November 7, 2023.
Study Selection
Randomised clinical trials (RCTs) reporting discontinuation symptoms using a standardized scale or individual symptoms (eg, adverse events) following antidepressant cessation were included.
Data Extraction and Synthesis
Data extracted were cross-checked by two reviewers. Additional unpublished data from 11 RCTs were included. A random-effects meta-analysis was conducted to calculate standardised mean difference between individuals who discontinued an antidepressant vs those who continued an antidepressant or discontinued placebo. A proportion and odds ratio (OR) meta-analysis was performed to assess incidence of individual discontinuation symptoms compared to placebo. Subgroup analyses were conducted to compare different antidepressants. Data analysis was conducted between September 2024 and December 2024.
Main Outcomes and Measures
The primary outcomes were incidence and nature of antidepressant discontinuation symptoms measured using standardised or unstandardised scales.
Results
A total of 50 studies were included, 49 of which were included in meta-analyses. The 50 studies included 17 828 participants in total, with 66.9% female participants and mean participant age of 44 years. Follow-up was between 1 day and 52 weeks. The DESS meta-analysis indicated increased discontinuation symptoms at 1 week in participants stopping antidepressants (standardised mean difference, 0.31; 95% CI, 0.23-0.39; number of studies [k] = 11; n = 3915 participants) compared to those taking placebo or continuing antidepressants. The effect size was equivalent to 1 more symptom on the DESS. Discontinuation of antidepressants was associated with increased odds of dizziness (OR, 5.52; 95% CI, 3.81-8.01), nausea (OR, 3.16; 95% CI, 2.01-4.96), vertigo (OR, 6.40; 95% CI, 1.20-34.19), and nervousness (OR, 3.15; 95% CI, 1.29-7.64) compared to placebo discontinuation. Dizziness was the most prevalent discontinuation symptom (risk difference, 6.24%). Discontinuation was not associated with depression symptoms, despite being measured in people with major depressive disorder (k = 5).
Conclusions and Relevance
This systematic review and meta-analysis indicated that the mean number of discontinuation symptoms at week 1 after stopping antidepressants was below the threshold for clinically significant discontinuation syndrome. Mood worsening was not associated with discontinuation; therefore, later presentation of depression after discontinuation is indicative of depression relapse.
See more from MedicalBrief archives:
Antidepressant withdrawal symptoms common – meta-analysis
Global depression guidelines lack withdrawal advice – Danish review
UK lawmakers’ research finds antidepressant withdrawal symptoms severe
The challenge of going off psychiatric drugs