Prescriptions for antidepressant medication continue to rise in the UK and globally, partly driven by increasing long-term treatment (Mars et al., 2017). Recent data suggest that, in England, around 1 in 15 people are now receiving a prescription for antidepressant medication (NHSBSA, 2025).
Some individuals report meaningful alleviation of their symptoms from these medications and enter recovery or become “remitted” (Eveleigh et al., 2019). At this point, some may feel ready to reduce their dose or stop treatment altogether but may be unsure of the safest way to go about discontinuing their antidepressants, often worrying about relapse or withdrawal symptoms if they attempt to do so (Eveleigh et al., 2019; Meißner et al., 2024). A double-blind randomised controlled trial (ANTLER) conducted by Lewis et al. (2021) found relapse risk was higher for patients who were discontinuing treatment compared with patients maintaining their usual dose (as summarised in Raphael and Sameer’s Mental Elf blog), highlighting potential concerns with discontinuation.
Clinical guidelines offer broad practical advice on how to manage antidepressant discontinuation. Tapering plans (where the patient gradually reduces their dose over time) are usually created on a case-by-case basis between the patient and their GP, but the optimal speed and structure of tapering remains uncertain (van Leeuwen et al., 2024). Patients and clinicians alike would benefit from knowing which deprescribing strategies minimise relapse risk and the likelihood of withdrawal effects, which is exactly what this systematic review and network analysis by Zaccoletti and colleagues (2025) aimed to do.
Individuals that have recovered following antidepressant treatment often wish to reduce or stop their treatment, but are unsure about the best way to do this.
Methods
Researchers searched medical databases and trial registries up to April 2025 for randomised controlled trials (RCTs) involving adults with depressive or anxiety disorders, who were fully or partially remitted at randomisation and taking an antidepressant. They categorised the discontinuation strategies as:
- Abrupt discontinuation
- Fast tapering (gradual antidepressant discontinuation over ≤ 4 weeks) with or without psychological support
- Slow tapering (>4 weeks) with or without psychological support
They also included continuation treatment with or without psychological support. Two researchers independently screened titles, abstracts, and full texts and extracted data using a standardised form. Outcomes were analysed using random-effects network meta-analysis, with relapse rate at the end of the trial as the primary endpoint. Individuals with depression and anxiety were analysed together after separate analyses showed the results were broadly similar. They also compared the outcomes of different strategies across different classes of antidepressants and showed that these effects were also consistent.
The team also assessed risk of bias, heterogeneity, inconsistency, and conducted multiple sensitivity analyses. Risk of bias was assessed using the Cochrane Risk-of-Bias tool 2 and deemed to be low in 53% of studies, high in 28%, and raised some concerns in 20%.
Results
Study characteristics
Seventy-six RCTs involving 17,379 participants were included (mean age = 45.9 years; mean proportion female = 67.5%). Across studies, there was an average follow-up of 46 weeks, with most trials involving individuals with major depressive disorder (79%).
Of the total sample, 81.7% were fully remitted and 18.3% partially remitted at the baseline. Most of the participants were continuing antidepressant treatment at the standard dose (51%) followed by abrupt stopping (20%), and fast tapering (19%). Selective Serotonin Reuptake Inhibitors (SSRIs; 40%) and Serotonin Noradrenergic Reuptake Inhibitors (SNRIs; 23%) were the most frequently used antidepressants. Only 5% received psychological support in the form of mindfulness-based cognitive therapy, preventive cognitive therapy, or cognitive behavioural therapy (CBT).
Network meta-analysis
Using abrupt stopping as the reference deprescribing strategy, the most effective strategy for preventing the risk of relapse was continuing the antidepressant at standard dose with psychological support (relative risk [RR] = 0.40, 95% confidence interval [CI] [0.26 to 0.61]), followed by continuing at standard dose without support (RR = 0.51, 95% CI [0.46 to 0.58]), and finally slow tapering with psychological support (RR = 0.52, 95% CI [0.38 to 0.72]). Continuing at a reduced dose also showed benefit but with lower certainty (RR = 0.62, 95% CI [0.42 to 0.92]). In contrast, fast tapering with psychological support, abrupt stopping with psychological support, and slow tapering alone did not differ significantly from abrupt discontinuation.
Treatment rankings placed continuation plus psychological support as the top-performing approach, closely followed by slow tapering plus psychological support. The results appeared to be consistent across the different sensitivity analyses. The RR estimates of each strategy were less precise after excluding high-risk or industry-sponsored studies, but the order of most to least effective strategies remained the same.
Whilst consistent effects were found in subgroup analyses for SSRIs and for depressive disorders, the evidence for SNRIs, tricyclics, and anxiety disorders was less certain due to the limited data available for these drug classes. An alternative classification of “very slow tapering” (>12 weeks) was also considered and found this strategy was only effective when the tapering was combined with psychological support. No meaningful differences emerged between discontinuation strategies on adverse events, serious adverse events, or rates of completing discontinuation. Surprisingly, withdrawal-related effects were slightly more common in those continuing antidepressants than individuals discontinuing treatment.
Beyond antidepressant continuation, the most effective strategy for preventing risk of relapse in depression is slow tapering in combination with psychological support.
Conclusions
This network meta-analysis shows that slow tapering combined with psychological support is an effective and well-tolerated strategy for discontinuing antidepressants in remitted individuals. Its protection against relapse is comparable to continuing medication, while abrupt stopping and fast tapering substantially increase relapse risk and should be avoided. Psychological support enhances the effectiveness of tapering, whereas support alone does not improve outcomes when continuing or abruptly stopping. Although evidence for anxiety disorders was limited, overall patterns suggest results may generalise cautiously beyond depression.
This review underscores the importance of gradually reducing antidepressant medication and strongly discourages stopping abruptly or quickly (in 4 weeks or less).
Strengths and limitations
A major strength of the study was its comprehensive evidence base, synthesising 76 RCTs and more than 17,000 participants; substantially larger than many previous meta-analyses on antidepressant discontinuation. This means the findings rest on a large, well-powered dataset that strengthens the statistical reliability of the conclusions, while the inclusion of diverse study designs and patient groups also makes the results more robust. Using network meta-analysis allowed the research team to compare across multiple deprescribing strategies too, even when trials did not directly compare them, enabling a clearer ranking of the different clinical approaches, which can inform clinical decision-making. The authors also applied rigorous methodology, including duplicate screening and data extraction, formal risk-of-bias assessment, evaluation of transitivity and inconsistency, and extensive sensitivity, subgroup, and meta-regression analyses, which collectively strengthen the reliability of the findings.
Further, presenting certainty ratings alongside effect sizes allows for a more reliable interpretation of the effectiveness of each strategy that takes into account the certainty of each effect, continuing to strengthen our trust in the findings. Importantly, incorporating trials with psychological support enabled the first robust, head-to-head comparison of tapering strategies with and without psychotherapy. This matters because it helps identify the deprescribing conditions under which psychological support is most likely to reduce relapse risk. Such insights can guide clinicians in deciding when and how to prescribe additional therapy during discontinuation. Finally, they also consulted a group of individuals with relevant lived experiences that informed the interpretation and dissemination of their analyses. Engaging with members of the public on the study’s key findings is important, to place the results in context and ideally contribute to the meaningful implementation of findings. It is a shame that this feedback was not reported, but this may have been beyond the scope of the study’s report.
However, some limitations must also be considered:
- The classification of tapering strategies (fast vs slow taper) was somewhat arbitrary and not grounded within previous literature or pre-existing clinical definitions. Although this distinction may have obscured meaningful differences in tapering, the authors did test an alternative definition of slow tapering (‘very slow tapering’, >12 weeks) in a sensitivity analysis and found similar results.
- The studies of patients suffering anxiety disorders were underrepresented (~ 20%) and therefore only included evidence for fewer strategies, limiting generalisability to these disorders.
- Evidence on the additional support offered by psychological therapy was also relatively sparse, meaning that any conclusions about the effectiveness of prescribing psychotherapy alongside a slow taper, although promising, should be considered preliminary.
Findings indicate that psychological support lowers risk of relapse during slow tapering. However, limited evidence means it isn’t possible to draw firm conclusions on their combined effectiveness.
Implications for practice
These results provide valuable insight for patients thinking about stopping antidepressants, as well as clinicians deciding on the best method to manage antidepressant withdrawal. The main message is clear: avoid abrupt discontinuation or tapering too quickly, as both approaches raise the chances of depression returning. Instead, when considering how best to deprescribe, it seems to be safest to follow a slow tapering plan with additional support, which aligns with guidance from the National Institute for Health and Care Excellence (NICE, 2022) to:
slowly reduce the dose to zero in a stepwise fashion, at each step prescribing a proportion of the previous dose (for example, 50% of previous dose).
The findings of this review reinforce the importance of clinicians working with patients to create a personalised tapering plan that allows them to gradually reduce their antidepressant dose over time. The results also suggest that current clinical guidelines could go further by recommending a minimum tapering period of at least one month, particularly where the patient is worried about relapse or has experienced difficulty stopping treatment previously. The potential added benefits of psychological support when combined with a slow tapering regime is also highlighted, although much further research is needed disentangle its specific contribution to reducing relapse risk. Previous research has shown that psychological therapy alongside antidepressant medication seems to be more effective at treating severe depression than medication alone (Cuijpers et al., 2023), so psychological therapy during withdrawal may offer similarly meaningful benefits.
Future research should focus on making detailed comparisons of the effectiveness of different methods of deprescribing, instead of comparing maintenance with a single method of antidepressant discontinuation. Importantly, data on withdrawal symptoms were scarce and often poorly measured in the reviewed studies, limiting the ability to assess their role in relapse and the impact of each deprescribing strategy on withdrawal-related effects (read Hannah’s Mental Elf blog to learn more about people’s lived experiences of withdrawal symptoms). This review highlights the ongoing gaps in knowledge on withdrawal symptoms and the need for more longitudinal assessments of discontinuation effects whilst patients are gradually reducing their medication.
The findings of this review align with current NICE guidelines regarding the deprescription of antidepressants, but also emphasise the importance of clinicians working with patients to create tailored tapering plans that work best for them.
Statements of interest
None to declare.
Edited by
Dr Nina Higson-Sweeney.
Links
Primary paper
Zaccoletti, D., Mosconi, C., Gastaldon, C., Benedetti, L., Gottardi, C., Papola, D., Ponzi, O., Purgato, M., Naudet, F., Cristea, I. A., Barbui, C., & Ostuzzi, G. (2025). Comparison of antidepressant deprescribing strategies in individuals with clinically remitted depression: a systematic review and network meta-analysis. The Lancet Psychiatry, Volume 13, Issue 1, 24 – 36
Other references
Science Media Centre – Expert Reaction to this paper (Zaccoletti et al, 2025).
Bowers, H. (2024). Antidepressant withdrawal: Recommendations for support from people with lived experience. The Mental Elf.
Cuijpers, P., Miguel, C., Harrer, M., Plessen, C. Y., Ciharova, M., Papola, D., Ebert, D., & Karyotaki, E. (2023). Psychological treatment of depression: A systematic overview of a ‘Meta-Analytic Research Domain’. Journal of Affective Disorders, 335, 141–151. https://doi.org/10.1016/j.jad.2023.05.011
Eveleigh, R., Speckens, A., van Weel, C., Oude Voshaar, R., & Lucassen, P. (2019). Patients’ attitudes to discontinuing not-indicated long-term antidepressant use: Barriers and facilitators. Therapeutic Advances in Psychopharmacology, 9, 2045125319872344. https://doi.org/10.1177/2045125319872344
Lewis, G., Marston, L., Duffy, L., Freemantle, N., Gilbody, S., Hunter, R., Kendrick, T., Kessler, D. S., Mangin, D., King, M., Lanham, P., Moore, M., Nazareth, I., Wiles, N. J., Bacon, F., Bird, M., Brabyn, S., Burns, A. J., Clarke, C. S., … Lewis, G. (2021). Maintenance or discontinuation of antidepressants in primary care. New England Journal of Medicine, 385(14), 1257-1267. https://doi.org/10.1056/NEJMoa2106356
Mars, B., Heron, J., Kessler, D., Davies, N. M., Martin, R. M., Thomas, K. H., & Gunnell, D. (2017). Influences on antidepressant prescribing trends in the UK: 1995–2011. Social Psychiatry and Psychiatric Epidemiology, 52(2), 193–200. https://doi.org/10.1007/s00127-016-1306-4
Meißner, C., Warren, C., Fadai, T., Müller, A., Zapf, A., Lezius, S., Ozga, A.-K., Falkenberg, I., Kircher, T., & Nestoriuc, Y. (2023). Disentangling pharmacological and expectation effects in antidepressant discontinuation among patients with fully remitted major depressive disorder: Study protocol of a randomized, open-hidden discontinuation trial. BMC Psychiatry, 23(1), 457. https://doi.org/10.1186/s12888-023-04941-3
National Institute for Health and Care Excellence. (2022). Depression in adults: Treatment and management (NICE guideline NG222)—Stopping antidepressant medication. https://www.nice.org.uk/guidance/ng222/chapter/recommendations#stopping-antidepressant-medication
NHS Business Services Authority. (2025, July 24). Medicines used in mental health – England – 2015/16 to 2024/25. https://www.nhsbsa.nhs.uk/statistical-collections/medicines-used-mental-health-england/medicines-used-mental-health-england-201516-202425
Rifkin-Zybutz, R., & Jauhar, S. (2021). Maintenance or discontinuation of antidepressants for depression? Findings from the ANTLER trial. The Mental Elf.
Van Leeuwen, E., Maund, E., Woods, C., Bowers, H., Christiaens, T., & Kendrick, T. (2024). Health care professional barriers and facilitators to discontinuing antidepressant use: A systematic review and thematic synthesis. Journal of Affective Disorders, 356, 616–627. https://doi.org/10.1016/j.jad.2024.04.060
