Withdrawal Symptoms Are Common After Stopping Antidepressants, Studies Show

• The effect of PAWS on your brain health — The researchers noted that PAWS involves a multitude of symptoms, including “dizziness, vertigo, tremor, nausea, insomnia, fatigue, mood dysregulation, anxiety, panic, irritability and agitation.” In severe cases, those in withdrawal report suicidal thoughts and behavior.³

• Symptoms can last for a long time — According to the studies reviewed, lingering symptoms lasted anywhere from 1.5 to 166 months (around 13.8 years). That upper figure is an outlier at the far end of a wide range, and it comes largely from online self-reports rather than controlled follow-up. Doctors usually tell patients that stopping antidepressants will be uncomfortable for a week or two, but this data tells a very different story.

• Paroxetine was linked to the most problems — The most consistent risk factor identified across these studies was long-term use of paroxetine, though the authors caution that most of the included studies could not speak to this question. According to the researchers:

“Two of the included studies indicate that in particular long-term paroxetine use may carry an increased risk of PAWS, but most studies were uninformative on this subject. One analysis of online self-reports further showed that the duration of tapering was positively correlated with the duration of withdrawal symptoms, suggesting that patients experiencing more persistent withdrawal symptoms try to taper more slowly.”

• PAWS is not a relapse — It’s not your original depression coming back. PAWS has its own distinct pattern, and the symptoms often feel different than the ones that led you to take antidepressants in the first place. That distinction is key because it helps explain why people may be misdiagnosed as having a “relapse,” when in fact they’re going through withdrawal:⁴

“[I]t appears that in both clinical research and practice, protracted and persistent withdrawal symptoms are frequently misdiagnosed as a relapse of the primary mental health condition (typically depression) or new emergent mental disorders, the latter being specifically embedded in the concept of persistent post-withdrawal disorders,” the study authors said.

• Stopping medication versus continuing usage — One noteworthy finding is that people who tapered off their antidepressant usage still had more recorded withdrawal symptoms than those who stayed on the medication.

The researchers cited a 2021 randomized controlled trial (RCT) published in The New England Journal of Medicine. Here, 478 patients were split into two groups — maintenance and discontinuation. By the end of 52 weeks, 135 people in the discontinuation group experienced withdrawal symptoms, and 92 in the maintenance group.⁵ Going deeper into the findings, the researchers noted:⁶

“In this study, 39 weeks after patients had started tapering citalopram, fluoxetine, sertraline or mirtazapine, the number of recorded withdrawal symptoms was still significantly increased compared to patients maintained on their antidepressant medication.

Of note, in the design of this RCT, the popular antidepressant drugs paroxetine and venlafaxine were deliberately excluded, because, as written by the authors, both are known to cause marked withdrawal symptoms when treatment is discontinued.”

• Effective treatment of PAWS is needed — After analysis, the researchers suggest that more RCTs are needed regarding PAWS. However, they recommend that the focus need to be on effective treatments, since there is little information regarding this topic:

“[R]igorous long-term RCTs are required to test the efficacy of treatment and management strategies. These studies would inform clinicians about effective interventions to mitigate the severity and duration of this impairing syndrome, as to date not a single clinical intervention has been formally evaluated.”

• Many are affected once medication is stopped — The team found that 31% of people who stop antidepressants experience some kind of withdrawal symptom. Severe ones were less common — 2.8% of everyone who discontinued an antidepressant, compared with 0.6% of those who discontinued placebo.

According to the analysis, symptoms typically began within a few days of stopping the drug, but in many cases, they didn’t fully resolve for weeks. In addition, the presence of symptoms wasn’t necessarily linked to how long someone had taken the medication — a meta-regression found no significant association between withdrawal incidence and duration of antidepressant treatment, which ranged across the included studies from one to 156 weeks.

• What the 31% means once placebo is accounted for — The authors’ bottom-line estimate is lower than the raw figure. Subtracting the symptoms reported after placebo discontinuation, they put the incidence of genuine antidepressant withdrawal at “approximately 15%, affecting 1 in 6 to 7 patients who discontinue their medication” — a figure they say should inform patients “without causing undue alarm.”⁸

• Expectation alone can produce symptoms — People who stopped a placebo still reported symptoms at a rate of 17%, or roughly 1 in 6. The authors describe these as non-specific or “discontinuation-like” symptoms, consistent with a nocebo effect in which the expectation of discomfort produces real symptoms.

• Severity and type of symptoms weren’t uniform — Specifically, desvenlafaxine, venlafaxine, imipramine, and escitalopram were associated with the highest frequency of symptoms, while imipramine, paroxetine, and desvenlafaxine or venlafaxine were associated with greater symptom severity.

In contrast, drugs like fluoxetine (Prozac), which linger longer in the bloodstream, showed a much lower rate of withdrawal complaints. The researchers also suggested that the half-life of these drugs (how long they remain in the body) plausibly affects not just the severity of the symptoms, but their time of appearance as well.

• What the evidence says about tapering — Notably, this featured meta-analysis itself found no difference between studies that tapered the drug and studies that stopped it abruptly (31% versus 29%). The authors nonetheless point to separate research on how tapering is done:⁹

“Tapering of antidepressants is recommended in most guidelines, and there is research suggesting that prolonged and hyperbolic tapering of antidepressants will substantially reduce (although not completely exclude) withdrawal effects and increase the likelihood of successful discontinuation of antidepressants.”

The same authors add that hyperbolic tapering has also been criticized and, because its pragmatic feasibility is limited, may not be indicated for every patient.

• Using science to give an air of legitimacy — In a report by investigative journalist Maryanne Demasi, Ph.D., she criticizes a 2025 JAMA Psychiatry meta-analysis of 50 studies and 17,828 participants¹² which concluded that the mean number of discontinuation symptoms one week after stopping was “below the threshold for clinically significant discontinuation syndrome” — a finding widely reported as showing that withdrawal is only “mild.”¹³

One of the reasons Demasi did her analysis is because of the prestige behind the journal. When it releases an influential study, consumers and medical practitioners tend to accept that the findings are true and accurate.

• Media is brought in to help — Demasi argues that industry-aligned researchers mobilized mainstream media to push a narrative that antidepressants can be slowly stopped and, as mentioned above, only produce mild side effects in doing so:

“The authors mobilised a rapid media campaign to shape the public narrative, with the Science Media Centre issuing expert commentary to ‘reassure both patients and prescribers’ that most withdrawal symptoms were ‘not clinically significant.’”

• The methodology is flawed — Digging deeper into the issue, Demasi broke down the logic of the JAMA Psychiatry meta-analysis. According to her findings, the studies used in the review are either biased or poorly designed. For example, she pointed out that the study followed patients for just two weeks — this isn’t applicable to antidepressant users in America, as half of them have already been taking these medications for more than five years. Demasi continues:

“Worse, many trials enrolled patients already taking antidepressants — then abruptly withdrew them before randomisation. As a result, those assigned to placebo experienced withdrawal symptoms that blurred the difference between treatment and control groups, artificially minimising the harms.”

• Industry funded most of the included studies — Demasi noted that most of the studies included in the meta-analysis were funded by the pharmaceutical industry. Furthermore, the researchers excluded medications such as paroxetine and escitalopram, which are already strongly linked to severe PAWS.

For context, the Lancet Psychiatry meta-analysis cited earlier ran its own funding subgroup analysis and found little difference between industry-funded and independently funded studies (31% versus 28%), so funding alone does not account for the range of estimates in this literature.¹⁴

• A firsthand account of PAWS — In an interview with National Public Radio (NPR),¹⁵ a Canadian patient named Phillipa Munari described taking Effexor (venlafaxine) for 10 years. When she decided to stop taking it, problems started to appear even with the help of a doctor to taper it off.

According to Munari, she developed nerve pain, as well as chronic pain in her neck and shoulders. To make matters worse, she also had severe anxiety — all of which she didn’t have before. In an effort to manage her health issues, she went back to taking Effexor and then weaned off it again, much more slowly the second time.

• The effects of PAWS can be lasting — Munari shared that after her second round of Effexor (with a better tapering plan), her nerve pain and fatigue improved. However, the anxiety worsened. She describes about two years of severe symptoms and roughly four more years beyond that before she felt fully recovered.

• • Exercise compares favorably with standard care — In an umbrella review of 97 reviews covering 1,039 trials and 128,119 participants, published in the British Journal of Sports Medicine, researchers reported that physical activity was extremely beneficial for symptoms of depression, anxiety, and distress — about 1.5 times more effective than counseling or the leading medications. As noted by the lead author Ben Singh, Ph.D.:¹⁶

“Physical activity is known to help improve mental health. Yet despite the evidence, it has not been widely adopted as a first-choice treatment … Higher-intensity exercise had greater improvements for depression and anxiety, while longer durations had smaller effects when compared to short and mid-duration bursts.”

• Any exercise is better than nothing — Singh noted that moving your body, no matter how you do it, is beneficial. He explains:¹⁷

“We also found that all types of physical activity and exercise were beneficial, including aerobic exercise such as walking, resistance training, Pilates, and yoga. Importantly, the research shows that it doesn’t take much for exercise to make a positive change to your mental health.”

• Start with a walk — While it’s tempting to start off with an intense session at the gym, you can already gain plenty of benefits by going for a walk. In my interview with Dr. James O’Keefe, he noted that it’s a great way to boost your fitness — an average of 10,000 steps a day is associated with meaningful benefits. It’s also a moderate-intensity activity that is difficult to overdo for most people, which makes it sustainable as a daily habit. Aim for about an hour of walking a day.

• Strength training complements walking — Lifting weights is another beneficial strategy that goes hand in hand with your daily walks, however, it’s important that you don’t overdo it. According to O’Keefe, observational data suggest the additional mortality benefit appears to flatten,¹⁸ and may reverse, beyond roughly 130 to 140 minutes of total strength training per week:

“I’ve always been a fan of strength training … But again, the devil is in the details about the dosing. When you look at people who do strength training, it adds another 19% reduction in all-cause mortality on top of the 45% reduction that you get from one hour of moderate exercise per day.

When I strength train, I go to the gym and spend anywhere from 20 to 40 minutes, and … I try to use weights that I can do 10 reps with … After that, you’re feeling sort of like spent and … it takes a couple of days to recover. If you do that two, at the most three, times a week, that looks like the sweet spot for conferring longevity.”

• There are other useful strategies to support your mood — While exercise is effective, it’s not the only option available. I recommend supporting your gut with fermented foods (preferably homemade), given how closely gut health is intertwined with your brain function.

In addition, getting deep, restorative sleep is important for your mental health. Managing stress, as well as resetting your body’s internal clock by getting sun exposure within about 20 minutes of waking up, will also help support your mental and physical well-being. For an in-depth explanation of these tips, read “Depression Accelerates Physical Illness and Increases Disease Risk.”