Patients treated with an antidepressant are often told they need to continue taking the drug indefinitely, as otherwise they may relapse (maintenance treatment). Yet there are many reasons people want to stop taking the drugs, such as weight gain, loss of sexual functioning, and emotional numbing.
In randomized relapse studies, patients who are doing well on antidepressants are randomized either to drug maintenance or to a drug discontinuation protocol. Now, a new meta-analysis in Lancet Psychiatry demonstrates that slow antidepressant tapering with psychological support prevented relapse of depression just as well as antidepressant continuation. Fast tapering and slow tapering without support did poorly by comparison, with relapse rates equivalent to abruptly discontinuing the drug.
“This network meta-analysis supports slow tapering plus psychological support as an effective and tolerable strategy for antidepressant discontinuation in people with remitted depression, and shows that abrupt or fast discontinuation should be avoided.,” the researchers write.
The issue of antidepressants withdrawal has burst into the larger public discussion about the merits of antidepressants. Research has found that antidepressant withdrawal is common in those trying to get off the drugs, and is severe in about half of those who experience it. The UK’s NICE guidelines now acknowledge the potential for severe and long-lasting antidepressant withdrawal.
However, this study does not provide much insight into many of the withdrawal-related questions. For instance, in this study, “slow tapering” was defined as tapering that lasted more than four weeks. Yet this is a very broad way to define “slow tapering.” Some people are tapering for a month or two, while others take six months, and still others take a year or even longer. The study was not able to provide data on whether any of these lengths provided superior results. (One exception is that the study did include a secondary analysis on tapering >12 weeks and found no difference between that and tapering >4 weeks.)
In addition to tapering length, there are various schools of thought on how to taper, with hyperbolic tapering increasing in popularity recently. Hyperbolic tapering has theoretical support, since it is based on the dose-response curve for serotonin receptor occupancy. Preliminary studies have found that hyperbolic tapering and very slow tapering may limit withdrawal symptoms, too. Yet it remains unclear if outcomes for hyperbolic tapering are better than for other tapering strategies, or if some people might be better served with a quicker discontinuation. Unfortunately, the current study couldn’t address this nuance.
There are also many antidepressants of various classes, which affect various brain chemicals differently and have different half-lives. The study didn’t look into whether tapering needed to be different for different drugs. For instance, is the tapering strategy for fluoxetine (Prozac, an SSRI) the same as for venlafaxine (Effexor, an SNRI) or for bupropion (Wellbutrin, an NDRI)? Nor did this study account for years of exposure to antidepressants as a factor when trying to taper from antidepressants.
The study was led by Debora Zaccoletti at the WHO Collaborating Centre for Research and Training in Mental Health and Service Evaluation at the University of Verona, Italy. The study was a meta-analysis, including 76 studies with 17,379 participants. Studies were included if they were randomized controlled trials of adults with remitted symptoms who engaged in antidepressant discontinuation.
The primary outcome was that a number of strategies were better than abrupt discontinuation for preventing relapse: slow tapering (>4 weeks) with support, continuing the drug, continuing the drug with support, and continuing the drug at a reduced dose. Slow tapering without support, abrupt discontinuation with support, and fast tapering (≤4 weeks) with or without support were all no better than abrupt discontinuation without support.
There are obvious limitations and confounding factors with this meta-analysis of relapse studies. One problem is that withdrawal symptoms are often mistaken for relapse (return of depression) or misdiagnosed as “functional disorders.” In one study, more than two-thirds of people in withdrawal had their symptoms misdiagnosed as other psychiatric disorders. Yet those in withdrawal also commonly exhibit somatic symptoms that clearly separate this from psychiatric problems, such as pain, fatigue, arrhythmia, diarrhea, blurred vision, numbness, brain zaps, amnesia, and stroke-like symptoms, among others. So it’s not that drug maintenance is preventing the return of depression, but rather the exposure to antidepressants creates the risk of withdrawal symptoms upon discontinuation.
Beyond that, some of the analyses in this study are far from definitive, since the included studies were heterogeneous and some had high risk of bias.
The analysis specifically focused on those who no longer met the criteria for depression but continued taking antidepressants as “maintenance” treatment, ostensibly to prevent relapse. The study included patients with anxiety as well, but the researchers had less confidence in that analysis, since they were a smaller proportion of the sample and several of the discontinuation strategies were not tested for anxiety in the included studies. Still, they write that it would make sense for the effects to be similar for anxiety.
“Although data on anxiety were scarce and several strategies were under-represented, population characteristics and effect estimates were broadly consistent with those in depression, supporting plausible generalisability to anxiety of our findings for depression,” the researchers write.
While the study did isolate support as a critical factor for successful discontinuation, various forms of “support” were only tested in 10 studies, and they included bespoke combinations of various types of support (e.g., lifestyle change planning, CBT, mindfulness, etc.). The researchers then combined these 10 studies into one group on “support” in order to do the analysis. While this was necessary to do a meta-analysis, it therefore provides no information about which elements of support are most important, or if there are other forms of support as yet untested that might be more effective at assisting discontinuation.
The duration of the included studies ranged from six weeks to one year, with none providing actual long-term information about discontinuation.
Another issue: the study focused only on those who remitted (no longer met criteria for depression), so it only included people for whom the drugs worked extremely well. Yet antidepressants are ineffective for many, beating placebo in only half of clinical trials (by such a slight margin it has been deemed clinically insignificant). And in real-life studies, less than a quarter of depressed patients even respond to treatment. Those who don’t experience success with an antidepressant are even more likely to want to discontinue the drug, but this study provided no data on how they should proceed.
This study does point to the importance of providing support for people who are seeking to taper from antidepressants, and that tapering speed is an important factor to consider. But it also reveals that existing relapse literature is woefully inadequate when it comes to addressing the many questions present with withdrawal from antidepressants.
****
Zaccoletti, D., Mosconi, C., Gastaldon, C., Benedetti, L., Gottardi, C., Papola, D., . . . & Ostuzzi, G. (2025). Comparison of antidepressant deprescribing strategies in individuals with clinically remitted depression: a systematic review and network meta-analysis. Lancet Psychiatry, 13, 24-36. (Link)
Source link
