Antidepressant Withdrawal vs. Relapse: Symptoms, Risks & Safe Taper Guide

Medication Management

Stopping Antidepressants: What Withdrawal Actually Feels Like, Why It Gets Mistaken for Relapse, and How to Taper Safely

About one in six people who stop an antidepressant will have withdrawal symptoms as a direct result. A large share of them will be told their depression came back. Those are two completely different problems with two completely different solutions — and telling them apart may be the most consequential conversation you have with your prescriber.

Reviewed by the clinical team at East Coast Telepsychiatry  |  Evidence-based patient education  |  Reading time: 14 minutes

Here is a scenario that plays out in primary care offices thousands of times a week. Someone has been on an antidepressant for three years, feels stable, and decides they'd like to stop. They taper over two weeks, or stop outright. Ten days later they're dizzy, nauseated, irritable, sleeping badly, and getting strange electrical sensations in their head. They go back to their doctor, who says: this is your depression returning, you should go back on the medication. They do. They feel better within forty-eight hours. Everyone concludes the illness was still there.

Sometimes that's exactly what happened. But often it isn't — and the clue is in that forty-eight hours. Depression does not lift in two days. Withdrawal does.

For most of the last three decades, this distinction got very little attention. Antidepressants were described as "non-habit-forming," which is true in the sense that they don't produce craving or compulsive use, and was widely misread as meaning the brain doesn't adapt to them. It does. Stopping a drug the nervous system has adjusted to over years can produce a real, physical, sometimes miserable set of symptoms — and the last two years have produced the best evidence we've ever had about how common that is, along with a genuine scientific fight about what the evidence means.

This article is about what's actually known, what's still contested, and what a careful taper looks like. It is not an argument that you should come off your medication, and it is emphatically not instructions for doing it yourself.

11.4%U.S. adults who took prescription medication for depression in 2023 (CDC) 1 in 6Will have withdrawal symptoms as a direct result of stopping ~1 in 35Will have symptoms rated severe 5 yearsMedian duration of antidepressant use in the U.S.

Is Antidepressant Withdrawal Real?

Short answer

Yes. The largest analysis to date — 79 studies and 21,002 patients, published in The Lancet Psychiatry in June 2024 — found that 31% of people stopping an antidepressant reported at least one discontinuation symptom, compared with 17% of people stopping placebo. The difference between those figures, roughly 15%, is the share attributable to the drug itself: about one in six people. Around 3% had symptoms rated severe.

Those numbers are worth sitting with in both directions, because they are routinely misquoted by both sides of this debate.

The 17% placebo figure is the part people forget. A meaningful chunk of what gets called "withdrawal" is nocebo effect, anxiety about stopping, the ordinary background noise of being a human with a body, or the depression itself. Any honest account has to subtract that. The study's lead author, Dr. Jonathan Henssler of Charité – Universitätsmedizin Berlin, was explicit that severe discontinuation symptoms are not the typical experience.

But "one in six" is also not a rounding error. Applied to the roughly 11.4% of American adults the CDC reports took prescription medication for depression in 2023 — with women at 15.3% and men at 7.4% — it describes millions of people. And the analysis found the risk was not evenly distributed: imipramine, paroxetine, and venlafaxine/desvenlafaxine carried the highest rates of severe symptoms.

What Does Antidepressant Withdrawal Feel Like?

Short answer

Withdrawal typically shows up in three clusters: neurological symptoms including dizziness, "brain zaps," and tinnitus; physical symptoms resembling a mild flu, with nausea, fatigue, headache, and sweating; and psychological symptoms including anxiety, irritability, vivid dreams, insomnia, and sudden emotional swings. Dizziness is the single most common symptom. Onset is usually within two to four days of a dose reduction.

Neurological

Dizziness and unsteadiness (the most frequently reported symptom), "brain zaps" — brief electrical-shock sensations in the head, often triggered by moving the eyes — tinnitus, tremor, numbness, and difficulty concentrating.

Flu-like and physical

Nausea, stomach upset, headache, muscle aches, fatigue, chills and sweating. This cluster is why withdrawal is so often mistaken for a virus, particularly when the timing coincides with a dose change nobody connected it to.

Psychological

Anxiety, agitation, irritability, emotional lability, crying spells, insomnia, and unusually vivid or disturbing dreams. This is the cluster that gets misread as relapse, because it overlaps with depression and anxiety symptoms.

"Brain zaps" deserve a specific mention, because people who experience them often don't report them — the sensation is strange enough that it sounds like something you'd be embarrassed to describe. It's a recognized phenomenon with published case literature, it is not dangerous, and it is not a sign you're losing your mind. It is a well-documented feature of antidepressant discontinuation, most commonly with shorter-half-life drugs.

Which antidepressant you're on changes the withdrawal picture substantially — largely because of how quickly the drug clears your system.

How Do You Tell Withdrawal Apart From Relapse?

Short answer

Four features usually separate them. Withdrawal starts within days of a dose change; relapse typically takes weeks to months. Withdrawal includes physical symptoms like dizziness and brain zaps that depression does not cause. Withdrawal improves within hours to days if the dose is restored; depression takes weeks to respond to medication. And withdrawal tends to fade on its own, while untreated relapse tends to deepen.

This is the single most useful table on this page. It is also the distinction that the February 2026 analysis in Psychiatric Times argues gets missed constantly — leaving people on medication for years on the basis of a symptom episode that was never relapse at all.

	Withdrawal (discontinuation symptoms)	Relapse of depression
Timing after stopping	Usually 2–4 days; nearly always within the first 1–2 weeks	Typically weeks to months later, rarely in the first few days
Physical symptoms	Prominent — dizziness, nausea, brain zaps, flu-like aches, sweating	Not characteristic. Depression causes fatigue and appetite change, not brain zaps or vertigo
Response to restoring the dose	Improves within hours to a few days	Takes 2–6 weeks to respond, like any new course of medication
Trajectory if nothing changes	Usually peaks and then fades over days to weeks	Tends to persist or worsen without treatment
Symptom quality	Often described as "wired," unstable, electric, physically unwell	Often described as heavy, flat, slowed, with loss of interest and hopelessness

The overlap is real and the table is not a diagnostic test. Anxiety, insomnia, and low mood appear in both columns, which is exactly why this gets confused. But the combination of fast onset, physical symptoms, and fast response to reinstatement is a pattern depression simply does not produce. If all three are present, withdrawal is the far more likely explanation — and the right response is a slower taper, not a conclusion that you need the medication indefinitely.

Why Is Withdrawal Worse From Some Antidepressants Than Others?

Short answer

Mostly half-life — how long the drug takes to clear your system. Drugs that leave quickly produce an abrupt drop in brain levels, which is what the nervous system reacts to. Fluoxetine has a very long half-life and effectively tapers itself, making withdrawal uncommon. Paroxetine, venlafaxine, and desvenlafaxine clear fast and are consistently associated with the hardest discontinuations.

Relative risk	Examples	Why
Higher	Paroxetine (Paxil), venlafaxine (Effexor), desvenlafaxine (Pristiq), duloxetine (Cymbalta), imipramine (Tofranil)	Short half-lives mean blood and brain levels fall steeply between doses and after stopping. Missing a single dose can be noticeable
Moderate	Sertraline (Zoloft), escitalopram (Lexapro), citalopram (Celexa), mirtazapine (Remeron)	Intermediate half-lives, roughly a day or so. Withdrawal is common enough to plan for but usually more manageable
Lower	Fluoxetine (Prozac), bupropion (Wellbutrin)	Fluoxetine and its active metabolite persist for days to weeks, creating a built-in taper. Bupropion acts on different receptors and is less associated with discontinuation symptoms

This is clinically useful rather than trivia. One established strategy for someone struggling to come off a short-half-life drug is to switch to a longer-acting one first and taper from there — a decision that belongs entirely with a prescriber, but one worth knowing exists if you've been told your only options are "stay on it" or "push through."

What Is Hyperbolic Tapering, and Why Does It Matter?

Short answer

Because the relationship between dose and brain-receptor effect is curved rather than straight, equal-sized dose cuts do not produce equal-sized biological changes. The last reductions near the bottom of the dose range cause by far the biggest drop in effect. Hyperbolic tapering means making progressively smaller cuts as the dose gets lower, so each step feels about the same.

This is the most important technical idea in the entire field, and it's surprisingly intuitive once you see it.

Receptor occupancy saturates. At higher doses, most of the relevant receptors are already occupied, so cutting the dose changes the biological effect only modestly. Near the bottom of the range, each milligram is doing a disproportionate amount of work. Which means the standard approach — halving the dose, halving it again, then stopping — front-loads the easy cuts and saves the hardest one for last, at the exact moment the patient has been told they're almost done.

Linear taper — feels uneven 20 → 15 → 10 → 5 → 0 Equal milligram steps. The final steps produce the largest drop in receptor effect, so withdrawal often hits hardest at the end — when it's least expected. Hyperbolic taper — feels even Progressively smaller cuts Each reduction is a proportion of the current dose rather than a fixed amount, so the biological change per step stays roughly constant all the way down.

In practice this often requires liquid formulations or compounded doses to hit the small increments, and it can take months — sometimes much longer than either patient or prescriber initially expects. The UK's NICE guidance on medicines associated with withdrawal symptoms now reflects this logic directly: reduce in stages, individualize the pace, and — critically — wait until symptoms from the last reduction have resolved or become tolerable before making the next one. That last clause is the whole game. A taper schedule set in advance on a calendar, without reference to how the person actually feels, is not really a taper.

A related strategy sometimes called "cut and hold" follows from the same principle: reduce, then stay at the new dose until you've genuinely stabilized, however long that takes, rather than treating the schedule as a deadline.

The Evidence Here Is Genuinely Contested — And You Should Know That

Most patient education on this topic picks a side and doesn't tell you there is one. There are roughly three camps, and the disagreement between them is substantive rather than political.

"Real but usually mild" The Lancet Psychiatry 2024 meta-analysis: about one in six affected, around 3% severe. Withdrawal is real and under-discussed, but severe cases are the exception. "Smaller than claimed" A July 2025 JAMA Psychiatry review found average symptom scores at one week well below the clinical threshold, and no increase in depressive symptoms — suggesting later low mood reflects relapse, not withdrawal. "Badly underestimated" Critics counter that these trials mostly studied people on antidepressants for weeks, while real patients take them for a median of five years — so the hardest cases may be largely absent from the data.

That third objection is the one to take most seriously, because it's a structural point rather than a quibble. If the evidence base is built on short-duration trials, it may simply not describe someone five years in. Meanwhile the second camp's finding — that depressive symptoms specifically don't rise on discontinuation — is a genuinely useful piece of information, because it supports the central practical claim of this article: low mood appearing weeks later is more likely relapse, while the physical, fast-onset stuff is more likely withdrawal.

You don't have to resolve this debate. You do have to know it exists, because a prescriber who tells you withdrawal is nothing and a wellness influencer who tells you antidepressants are chemical dependency are both overstating a case the literature doesn't settle.

Please Don't Stop On Your Own

Everything above is an argument for a slower, better-supervised taper — not for stopping without one. Abrupt discontinuation is the single biggest predictor of a difficult withdrawal, and the people who have the worst time of it are overwhelmingly those who stopped suddenly, ran out of a prescription, or tapered on a schedule nobody adjusted when symptoms appeared.

There is also a real risk of genuine relapse. Depression and anxiety disorders are recurrent conditions, and for some people ongoing medication is the thing keeping a serious illness in remission. A plan that accounts for both risks is better than a plan that only takes one seriously.

If you've already stopped and feel awful: contact your prescriber rather than waiting it out alone. Restoring the previous dose and then tapering more gradually is a standard, reversible, entirely unembarrassing step — not a failure, and not evidence that you'll be on medication forever.

What Does a Well-Run Taper Actually Involve?

Short answer

A conversation about why you're stopping and whether now is the right time; a realistic timeline measured in months rather than weeks; progressively smaller reductions; symptom tracking so you and your prescriber can distinguish withdrawal from relapse in real time; permission to pause or step back up; and a plan for what happens if things go badly.

• Timing that isn't accidental. Coming off during a divorce, a job loss, or a move stacks the deck. Stable periods are better launch points, and "not right now" is a legitimate answer.
• An honest reason. Side effects, pregnancy planning, feeling well for a long stretch, and simply not wanting to take it anymore are all valid. Naming the reason shapes the plan.
• A written schedule you can revise. Not a fixed countdown. The next reduction happens when the last one has settled.
• Symptom tracking. Even a few notes a day on a phone gives you and your prescriber the pattern — onset timing, symptom type, trajectory — that distinguishes withdrawal from relapse far better than memory does.
• Explicit permission to pause. Holding at a dose for a month, or going back up a step, should be a planned-for option rather than a crisis.
• Someone who knows. A partner, friend, or family member who can tell you what they're noticing. Mood changes are easier to see from outside.
• Non-medication support in place. Therapy, sleep, exercise, and structure matter more during a taper than at almost any other time.

Tapering well depends less on any single decision than on frequent check-ins — which is one of the few things telepsychiatry does better than a twice-yearly office visit.

Should You Come Off At All?

Short answer

That depends on your history, not on a general principle. Someone with a single episode who has been well for a year or more is in a very different position from someone with three prior episodes, a family history, or a past hospitalization. Long-term medication is a legitimate, evidence-supported choice — and so is a carefully supervised attempt at stopping. Neither is the virtuous option.

It's worth naming the cultural pressure here, because it runs in both directions and neither version serves patients well. One narrative treats staying on medication as a failure of willpower or a pharmaceutical trap. The other treats any interest in stopping as symptom-driven non-compliance. Both skip the actual clinical question, which is specific to you: how many episodes have you had, how severe were they, how long have you been well, what else is in place to support you, and what would the cost of a relapse be right now?

Those questions have answers. They're just answers that require someone to know your history.

Where to Find Reliable Information

• NIMH — Mental Health Medications, for plain-language information on antidepressant classes and what to expect.
• NICE NG215 recommendations, the clearest published clinical guidance on staged withdrawal and individualized tapering pace.
• The 2024 Lancet Psychiatry meta-analysis, if you want to read the primary incidence data rather than a summary of it.
• CDC/NCHS Data Brief 528, for current U.S. antidepressant use figures.
• SAMHSA National Helpline — free, confidential, 24/7 treatment referral at 1-800-662-4357.

Frequently Asked Questions

How long does antidepressant withdrawal last?

For most people, symptoms appear within two to four days of a dose reduction, peak over the following week or two, and resolve within a few weeks. A minority report a much longer course, and that longer-duration experience is one of the most contested areas in the research — partly because the trials underpinning the main estimates studied people who had been on medication for weeks rather than years. If symptoms are dragging on, that's a reason to involve your prescriber rather than to assume you have to endure it.

Are antidepressants addictive?

No, not in the clinical sense. Addiction involves craving, escalating use, and continued use despite harm, and antidepressants don't produce that pattern. What they do produce is physiological adaptation — the nervous system adjusts to the drug's presence and needs time to readjust when it's removed. That's dependence in the pharmacological sense, the same thing that makes blood pressure medication unwise to stop abruptly. It isn't addiction, and conflating the two has made this conversation much harder than it needs to be.

Which antidepressant is hardest to stop?

Paroxetine (Paxil) and venlafaxine (Effexor) are the two most consistently identified as difficult, both because of short half-lives and because the 2024 meta-analysis found them among the drugs with the highest rates of severe discontinuation symptoms, alongside desvenlafaxine and imipramine. Fluoxetine (Prozac) is generally the easiest, because its long half-life means it tapers itself to some degree.

Can I just take my medication every other day to taper?

This is a common suggestion and, for short-half-life drugs, usually a poor one. Alternate-day dosing creates a sawtooth pattern of rising and falling brain levels — which is precisely the kind of fluctuation that provokes withdrawal symptoms. Smooth, gradual reduction using liquid or smaller solid doses is generally preferred. Fluoxetine is a partial exception because of how slowly it clears, but this is a decision for your prescriber, not a workaround to adopt on your own.

My doctor says withdrawal isn't a real thing. What now?

You're describing a view that's increasingly out of step with the published evidence, including a 79-study meta-analysis in a major journal and explicit national clinical guidance on staged withdrawal. You're entitled to bring those to a conversation, and you're also entitled to a second opinion from a clinician who specializes in psychiatric medication. Our providers handle medication management as their primary work rather than as one item among many.

If I come off successfully, can I go back on later if I need to?

Generally yes, and knowing that tends to make the whole process less frightening. Most people who respond to a medication respond again if they restart it. Treating a taper as a reversible experiment rather than a one-way door is both more accurate and considerably easier on your nerves.

If You're in Crisis Right Now

Withdrawal can include agitation, severe insomnia, and sudden emotional instability, and for a small number of people it includes thoughts of self-harm. If you're having thoughts of harming yourself, call or text 988 to reach the Suicide & Crisis Lifeline, available 24/7 across the United States, or chat at 988lifeline.org. If you're in immediate danger, call 911 or go to your nearest emergency department.

Reaching out in the middle of a bad stretch is not an overreaction. It's how you keep options open until the stretch passes.

A Practical First Step

If you're considering a taper, or you've already stopped and aren't sure whether what you're feeling is withdrawal or returning illness, a structured baseline helps enormously. Our free, confidential screening tools include the PHQ-9 for depression severity, the GAD-7 for anxiety, and the ISI for insomnia — all three of which are useful precisely because repeating them over weeks turns a vague "I feel off" into a trend line you and a prescriber can read.

A screening isn't a diagnosis. What it gives you is a number from before, to compare against the number from now.

This Is Squarely What We Do

Medication management is East Coast Telepsychiatry's core work. That includes the part of it most often rushed: evaluating whether a taper makes sense for your history, designing one that reduces gradually rather than on a fixed countdown, seeing you often enough during it to catch problems early, and distinguishing withdrawal from relapse based on your actual symptom pattern rather than a guess.

We'll also tell you when we think the answer is no. If your history suggests a high relapse risk, or the timing is poor, we'd rather say so plainly and revisit it later than agree to a plan we don't think will hold. You can see how we approach the conditions we treat, what telepsychiatry visits involve, and what care costs before you book anything.

Related Reading

Conditions We TreatDepression, anxiety, bipolar disorder, ADHD, PTSD and more Free ScreeningsPHQ-9, GAD-7, ISI and other validated self-assessments Our ProvidersBoard-certified psychiatric clinicians across the East Coast Insurance & CostWhat's covered and what care actually costs Support OptionsWays to get help that fit your situation More ArticlesEvidence-based mental health education

A Taper Deserves a Plan, Not a Guess

Whether you're thinking about stopping, already struggling after stopping, or simply want someone to review whether your current medication is still the right one — that's a conversation worth having with a prescriber who has time for it.

Schedule an Appointment Browse Resources

Sources & Further Reading

1. Henssler J, et al. "Incidence of antidepressant discontinuation symptoms: a systematic review and meta-analysis." The Lancet Psychiatry, June 2024. 79 studies, 21,002 patients.
2. Healio. "15% of patients experience antidepressant discontinuation symptoms, which are mostly mild" — free-access clinical summary of the Henssler meta-analysis.
3. The Lancet Psychiatry / EurekAlert. "One in six people who stop antidepressants will experience discontinuation symptoms as a direct result" — summary of the above, 5 June 2024.
4. Kalfas M, et al. "Incidence and Nature of Antidepressant Discontinuation Symptoms: A Systematic Review and Meta-Analysis." JAMA Psychiatry, July 2025.
5. Badre N, Cohen D, Losada Strassle C. "Exiting Antidepressants: A Needed Spotlight on Withdrawal." Psychiatric Times, February 2026.
6. National Institute for Health and Care Excellence. NG215 — Medicines associated with dependence or withdrawal symptoms: safe prescribing and withdrawal management for adults.
7. NICE. "Adults with depression who want to quit antidepressants should be given support."
8. CDC / National Center for Health Statistics. Data Brief 528: Characteristics of Adults Age 18 and Older Who Took Prescription Medication for Depression, United States, 2023, April 2025.
9. National Institute of Mental Health. Mental Health Medications.
10. Horowitz MA, Taylor D. "Tapering of SSRI treatment to mitigate withdrawal symptoms." The Lancet Psychiatry — the original argument for hyperbolic tapering.
11. "Antidepressant Withdrawal Syndromes: Listening to the Patient and Taking It Slow." Psychiatric Times.
12. University of Queensland. RELEASE — Redressing Long-tErm Antidepressant uSE, an ongoing trial of supported deprescribing in primary care.
13. SAMHSA. National Helpline — 1-800-662-HELP (4357).

This article is for general educational purposes and does not constitute medical advice, diagnosis, or treatment, and nothing in it is a tapering protocol. Do not change or stop a prescribed medication without guidance from the clinician who prescribed it; abrupt discontinuation carries real risks, including withdrawal symptoms and relapse of the underlying condition. Drug names and relative risk groupings are included for general education and do not indicate that any particular medication is right or wrong for you. If you are experiencing a mental health emergency, call or text 988 or dial 911.