The presence of discontinuation symptoms when patients stop taking antidepressants has sparked heated discussion, both on social media and in the general public press. In the scientific literature, there is a more restrained debate about the frequency and severity of discontinuation symptoms and the best way to go about deprescribing antidepressants.
In The Lancet Psychiatry, Jonathan Henssler and colleagues1 have provided some important numbers that will help part of this debate. In their systematic review, they found that 31% of patients (95% CI 27–35) reported at least one antidepressant discontinuation symptom when stopping an antidepressant, compared with 17% (14–21) who reported symptoms after stopping a placebo, a difference of about 14%. There is the potential for selection bias when comparing these two populations, so the authors also looked at the subset where individuals were randomly allocated to an antidepressant or placebo so that a comparison could be made with more validity. The difference was then 8% (4–12) between the two groups. They also estimated that 2·8% of people (1·4–5·7) reported severe withdrawal symptoms after stopping antidepressants, compared with 0·6% of people (0·2–1·3) stopping placebo.
There are some important implications here. First, withdrawal symptoms are reported after stopping a placebo, so in clinical practice, we must be careful when we infer that any symptom is causally related to stopping the medication. It is also important to reassure patients about the issue and discuss the possibility that their symptoms might not be due to the discontinuation of antidepressants. Second, the difference between the active and placebo groups is an important one from a scientific point of view. A rough estimate of the true prevalence of discontinuation symptoms is about 8–14% and of severe withdrawal syndromes about 2%. Reports of withdrawal symptoms that are not compared with a placebo will give a large overestimate of the frequency of such symptoms.2 Future studies of withdrawal symptoms should ensure that comparisons are made with a placebo when possible.
These results can explain some of the controversy. Given that antidepressants are prescribed to many millions of people, the relatively uncommon severe withdrawal symptoms will still affect a substantial number of people. Organisations that help people stop prescription drugs have many members who have difficulty in stopping antidepressants.3 However, for individual clinicians, severe withdrawal symptoms will seem uncommon and most patients will likely not be troubled by antidepressant withdrawal, especially when medication is tapered over a few weeks.
All systematic reviews are limited by the strength of the constituent papers. In this case, a major limitation is how discontinuation symptoms were defined and measured. The authors relied upon the definitions provided by the studies themselves, and this probably contributed to the high levels of heterogeneity. As expected, structured assessments provided higher estimates than studies that relied upon self-reporting. We should also note that many of the studies were small, often using antidepressants not commonly used now, and studying people who had not taken the antidepressants for a very long time. Despite these limitations, the results here are a substantial improvement on anything that has been published before.
The paper does not inform the ongoing debate about the relationship between discontinuation symptoms and the depressive and anxious symptoms of relapse, but this debate underpins any interpretation of these data.4 The evidence that long-term maintenance antidepressants prevent relapse of depression is very strong.5 The more recent ANTLER study6 has strengthened this evidence by examining long-term maintenance and reported a doubling of relapse rate in those who discontinue their medication. Cessation of antidepressants will therefore lead, on average, to an increase in depressive and anxious symptoms. Discontinuation symptom scales such, as the Discontinuation-Emergent Signs and Symptoms (DESS) scale,7 include symptoms that are identical to those expected in depression and anxiety. For example, some of the most common symptoms reported using the DESS are fatigue, nervousness or anxiety, irritability, sudden worsening of mood, bouts of crying, dizziness and trouble sleeping.4,7 Distinguishing between relapsing symptoms and withdrawal is difficult to do.
For the patient and clinician, the main issue is how to manage the discontinuation of medication, and to decide if the medication was or, more importantly, will be a net benefit. Short-term symptoms that reduce quickly, without intervention, are best thought of as a form of withdrawal, even if those symptoms might be similar or identical to the symptoms of depression and anxiety. More serious and longer-term symptoms might best be managed by tapering more slowly,8 or even deciding to remain on the antidepressant. Shared decision-making is essential, and psychotherapy might help.9 Arguing about whether these should be called relapses or withdrawals will help no-one.
GlL acknowledges the support of National Institute for Health and Care Research (NIHR), the Wellcome Trust, UK Research and Innovation (UKRI), and University College London Biomedical Research Centre (UCLH BRC); grants from NIHR, the Wellcome Trust, and UKRI; and travel and accommodation to attend the European College of Neuropsychopharmacology 2023. GeL acknowledges the support of the Wellcome Trust and UCLH BRC, and is a Wellcome Trust
Henry Dale and Royal Society fellow.
*Glyn Lewis, Gemma Lewis
glyn.lewis@ucl.ac.uk
UCL Division of Psychiatry, University College London, London, W1T 7NF, UK (GlL, GeL)
1 Henssler J, Schmidt Y, Schmidt U, Schwarzer G, Bschor T, Baethge C. Incidence of antidepressant discontinuation symptoms: a systematic review and meta-analysis. Lancet Psychiatry 2024; published online June 5. https://doi.org/10.1016/S2215-0366(24)00133-0.
2 Davies J, Read J. A systematic review into the incidence, severity and duration of antidepressant withdrawal effects: are guidelines evidence- based? Addict Behav 2019; 97: 111–21.
3 Coe A, Abid N, Kaylor-Hughes C. Social media group support for antidepressant deprescribing: a mixed-methods survey of patient experiences. Aust J Prim Health 2024; 30: PY23046.
4 Horowitz MA, Taylor D. Distinguishing relapse from antidepressant withdrawal: clinical practice and antidepressant discontinuation studies. BJPsych Adv 2022; 28: 297–311.
5 Kishi T, Ikuta T, Sakuma K, et al. Antidepressants for the treatment of adults with major depressive disorder in the maintenance phase:
a systematic review and network meta-analysis. Mol Psychiatry 2023;
28: 402–09.
6 Lewis G, Marston L, Duffy L, et al. Maintenance or discontinuation of antidepressants in primary care. N Engl J Med 2021; 385: 1257–67.
7 Rosenbaum JF, Fava M, Hoog SL, Ascroft RC, Krebs WB. Selective serotonin reuptake inhibitor discontinuation syndrome: a randomized clinical trial. Biol Psychiatry 1998; 44: 77–87.
8 Horowitz MA, Taylor D. Tapering of SSRI treatment to mitigate withdrawal symptoms. Lancet Psychiatry 2019; 6: 538–46.
9 Breedvelt JJF, Warren FC, Segal Z, Kuyken W, Bockting CL. Continuation of antidepressants vs sequential psychological interventions to prevent relapse in depression: an individual participant data meta-analysis.
JAMA Psychiatry 2021; 78: 868–75.
Lancet Psychiatry 2024
Published Online
June 5, 2024 https://doi.org/10.1016/ S2215-0366(24)00174-3
See Online/Articles https://doi.org/10.1016/ S2215-0366(24)00133-0
Incidence of antidepressant discontinuation symptoms: a systematic review and meta-analysis
Jonathan Henssler, MD Yannick Schmidt, MD, Urszula Schmidt, MD, Guido Schwarzer, PhD, Prof Tom Bschor, MD, Prof Christopher Baethge, MD, published June 05, 2024