Today, I review, link to, and excerpt from Psychiatry Online‘s “Antidepressant Withdrawal“. Publication Date: 27 August 2025. Psychiatric News
Volume 60, Number 9 https://doi.org/10.1176/appi.pn.2025.09.9
All that follows is from the above resource.
Even as one in eight adults in the U.S. population is prescribed an antidepressant, there is little formal guidance on how to safely, effectively taper off these medications. Here’s what psychiatrists need to know about withdrawal, including how to avoid it—and what can be done to improve its detection, prevention, and management.
Antidepressant withdrawal, recognized as antidepressant discontinuation syndrome in the DSM-5-TR, is a distressing condition that occurs after the abrupt or rapid discontinuation of antidepressant drugs. Once considered mild and short-lived (Rosenbaum & Zajecka, 1997), it is increasingly recognized as a significant clinical concern often associated with substantial morbidity and functional impairment (Horowitz & Taylor, 2019).This report highlights the epidemiology and clinical features of antidepressant withdrawal, associated risk factors, principles of safely tapering antidepressants, and current gaps in care resulting from lack of formalized guidance and research.Epidemiology and Clinical Features
Systematic reviews estimate the overall incidence of antidepressant withdrawal as ranging from 33% to 56% (Henssler, et al., 2024; Davies & Read, 2019). The estimated incidence of severe forms of withdrawal varies considerably. A 2019 meta-analysis reported that severe symptoms occur in 25% of patients, but this finding was likely inflated due to selection bias associated with the inclusion of online surveys (Davies & Read, 2019). A more recent meta-analysis estimated the rate to be as low as roughly 3% of patients, but this analysis included mostly studies that relied on spontaneous reporting of symptoms, and also included many studies of shorter duration: Nearly half of included studies were less than 12 weeks, and the weighted average of all included studies was 25 weeks (Henssler, et al., 2024). Given that the risk of withdrawal appears to increase significantly with longer-term use and that the median duration for antidepressant treatment in the United States is five years (Ward, et al., 2025), the inclusion of studies with such short treatment periods likely led to an underestimation of incidence in typical clinical populations. Thus, incidence—particularly of severe withdrawal—has not yet been definitively characterized and should be a research priority moving forward.Risk Factors
Serotonin reuptake inhibitors (SRIs) such as SSRIs and serotonin-norepinephrine reuptake inhibitors (SNRIs) are more strongly linked to withdrawal reactions than non-serotonergic antidepressants such as mirtazapine (an atypical teracyclic) and—especially—bupropion, although comparative studies are lacking. Paroxetine is considered the highest-risk SSRI likely due to a short elimination half-life and additional anticholinergic and noradrenergic effects that may be independently associated with withdrawal. Like paroxetine, SNRIs may be more likely to cause withdrawal than non-paroxetine SSRIs due to short elimination half-lives and potent noradrenergic effects. Two large pharmacovigilance analyses found that paroxetine, venlafaxine, desvenlafaxine, and duloxetine had the highest rates of self-reported withdrawal issues among antidepressants (Gastaldon, et al., 2022; Zhang, et al., 2024), and these four agents should be considered highest risk. As elimination half-life is inversely proportional to withdrawal risk, fluoxetine may be associated with lower incidence and severity of withdrawal symptoms and is occasionally used to help taper off SRIs (“fluoxetine substitution”), although this strategy has not been empirically validated (Rosenbaum, et al., 1998; Shapiro & Cohrs, 2025). The other SSRIs (sertraline, citalopram, escitalopram, and fluvoxamine) have elimination half-lives of about one day and should also be considered high-risk for withdrawal reactions.
Table 1. Risk factors for antidepressant withdrawal.
•Potent serotonin reuptake inhibition (SSRIs, SNRIs, vilazodone, vortioxetine, clomipramine)•Short elimination half-life (e.g., paroxetine, venlafaxine, duloxetine)•Limited drug formulary (i.e., limited options for achieving subtherapeutic dosages)•Long-term treatment (≥6 months)•Higher dose (SNRIs)Although not yet been corroborated by controlled clinical studies, observational data reliably indicate greater overall withdrawal risk with prolonged (e.g., >6 months) use and even greater risk—particularly of severe or protracted forms—with longer-term (e.g., >2 years) use (Horowitz, et al., 2025; Read, et al., 2018).Ecological studies suggest that withdrawal risk is dose-dependent for SNRIs, while findings for SSRIs are mixed (Horowitz, et al., 2023). This is consistent with linear increases in noradrenergic effects of SNRIs within the therapeutic dosage range that contributes to withdrawal risk when treatment is discontinued. To date, no genetic factors (e.g., SERT polymorphisms) have been identified that increase vulnerability to antidepressant withdrawal.Tapering Antidepressants to Mitigate Withdrawal Symptoms: Lessons From Radioligand Imaging Studies
Positron emission technology (PET) radioligand imaging studies allow for direct measurement of serotonin transporter (SERT) in response to antidepressant treatment in human subjects. These studies have provided key insights that explain perplexing findings of little benefit of “gradual” tapers compared with abrupt discontinuation and have led to a rethinking of how to optimally taper SRI antidepressants. These data consistently indicate that minimum therapeutic doses of SRIs (other than vilazodone, vortioxetine, and levomilnacipran) occupy about 80% of the serotonin transporter, with marginal increases in occupancy at higher therapeutic doses (Table 2) (Shapiro, 2018).
Table 2. Maintenance doses of SSRIs (mg/day) and associated serotonin transporter (SERT) occupancy in humans.
SERT Occupancy (%)* Fluoxetine Paroxetine Citalopram Sertraline 20 0.6 1.3 0.8 2.2 40 1.7 3.4 2.2 5.9 50 2.7 5.0 3.4 9.1 60 4.5 7.4 5.4 14.1 80 25.9 18.9 19.0 45.8 *Data are based on logarithmic curves of the form f(x)=a(x/[b+x]) reported in Meyer, et al., 2004.
