One in six people who stop taking antidepressants experience symptoms after stopping them

The Lancet Psychiatry publishes the first meta-analysis of the incidence of antidepressant treatment discontinuation symptoms that includes data from more than 20,000 patients collected from 79 randomised controlled trials and observational studies. The study sought to distinguish between symptoms directly caused by medication discontinuation and other ‘non-specific’ symptoms that may be associated with patient or professional expectations (the nocebo effect). The study concludes that one in six to seven patients will experience one or more symptoms directly caused by stopping medication, and one in 35 are likely to experience severe symptoms.  

06/06/2024 - 00:30 CEST
Expert reactions

Eduard Vieta - abandono antidepresivos EN

Eduard Vieta

Researcher of the Centre for Biomedical Research in Mental Health Network (CIBERSAM). Head of the Psychiatry and Psychology Department at the Hospital Clínic de Barcelona and lecturer at the University of Barcelona

Science Media Centre Spain

This is a high-quality meta-analysis that examines the clinically relevant phenomenon of rebound symptoms and discomfort at the time of discontinuation of antidepressant treatment, which occurs with some frequency (in the study data, in 15% of patients). 

The study confirms what we already knew, i.e. that this is a real phenomenon and that it occurs with some frequency with some drugs, but not all. 

The methodology is excellent but not free of biases and limitations, mostly derived from those of the studies analysed. Among them, it should be noted that the study has focused on controlled clinical trials, which is a strength as it has a placebo control group, but also a limitation, as the data may not faithfully reflect what happens in clinical samples outside the scope of clinical trials and that the discontinuation phase of treatment is usually open-label (without masking). 

The implications of the study are that one in six patients may experience symptoms following discontinuation of antidepressant treatment that may appear to be relapse symptoms, but are actually caused by more or less abrupt discontinuation of the drug, and that venlafaxine, desvenlafaxine, paroxetine and imipramine are the drugs most frequently associated with this phenomenon. Both patients and psychiatrists should avoid abruptly discontinuing these drugs or opt for others that are not associated with this problem. 

The author has declared they have no conflicts of interest

Alberto Ortiz - abandono antidepresivos

Alberto Ortiz Lobo

Doctor of Medicine and Psychiatrist at the Carlos III Day Hospital - La Paz University Hospital (Madrid)

Science Media Centre Spain

The extraordinary increase in the prescription of antidepressants should alert us to how we are transforming human and everyday discomfort into a medical condition that we treat with psychotropic drugs. A treatment that, logically, causes adverse effects during its consumption, but also upon withdrawal, a problem that has barely been given visibility until a few years ago.

This meta-analysis aims to measure the incidence of withdrawal syndrome when antidepressants are discontinued. It analyzes the results of 79 previous studies characterized by heterogeneity in design and data collection, and mostly aimed at measuring the efficacy of antidepressants and, secondarily and variably, their adverse effects. The meta-analysis does not provide data on crucial aspects that influence the results, such as the method of discontinuing antidepressants, previous treatment durations, or their doses.

With all these biases, the results are underestimated, especially when compared to previous reviews indicating that more than half of the people who take antidepressants experience withdrawal syndrome when trying to stop them.

It is noteworthy that throughout the article this phenomenon is referred to as "discontinuation symptoms" (a euphemism commonly used by the pharmaceutical companies that market them) and not as "withdrawal syndrome," which is the usual term to describe the physical and psychological response after reducing or stopping substances that act on the central nervous system, as is the case with antidepressants.

The author has declared they have no conflicts of interest

Diego Hidalgo - abandono antidepresivos

Diego Hidalgo

Psychiatrist and researcher at the Bipolar and Depressive Disorders Unit of the Hospital Clínic de Barcelona

Science Media Centre Spain

The discontinuation syndrome at the moment of stopping antidepressants has always been a topic of discussion among psychiatrists, from their training to their practice, generating debate between theoretical understanding and day-to-day clinical practice. Positions within the discipline vary, ranging from those who deny its existence to those who take extreme measures to mitigate it due to its very high frequency. More recently, the topic has been addressed by patients themselves, thanks to social media, offering diverse perspectives, which has added value to the discussion and highlighted the need for further study.

In this sense, this study comes at a good time and in good form: it scientifically rigorously covers the topic, aiming to objectify this potential problem by including all clinical trials and studies that have reported this issue. The study compares the frequency in patients who received the active antidepressant with those who received a placebo, a design that aims to clarify the prevalence of this discontinuation syndrome in a straightforward manner, leaving little room for error.

Although it has the main limitation that the definition of discontinuation syndrome has varied among different studies, the study clearly concludes its prevalence in approximately 15% of patients who discontinue an antidepressant. Additionally, it identifies those antidepressants that are more likely to produce it, especially with severe symptoms.

In some way, this study validates what has been done from training to clinical practice in psychiatry and has been a subject of discussion in recent years. However, it still does not address one of the main concerns in this regard, which is to have strong and clinically valuable arguments to discern whether the anxious-depressive symptoms that frequently arise upon discontinuing antidepressants are due to discontinuation syndrome or the reappearance of the underlying disorder's symptoms.

The authors and commentators of this article point precisely in this direction as a future line of research to be followed by studies to try to find an answer that can help psychiatrists and patients make joint decisions about starting and discontinuing the medication with a greater degree of certainty.

The author has not responded to our request to declare conflicts of interest
Incidence of antidepressant discontinuation symptoms: a systematic review and meta-analysis
  • Research article
  • Peer reviewed
  • Systematic review
  • Meta-analysis
The Lancet Psychiatry
Publication date

Henssler et al. 

Study types:
  • Research article
  • Peer reviewed
  • Systematic review
  • Meta-analysis
The 5Ws +1
Publish it