Someone reduces or stops an antidepressant. A few days later, they feel dizzy, agitated, sleepless, emotionally raw, or suddenly unlike themselves.
What happened?
That sounds like a straightforward question until you notice how much rides on the answer. If it is withdrawal, the experience may say something about the medication change. If it is relapse, it may be understood as the return of the original problem. And if it is both, or neither cleanly, those two labels can make a complicated human experience sound far more settled than it is.
I wondered why we so often talk as if the distinction should be obvious.
The labels are doing a lot of work
“Withdrawal” is sometimes treated as a narrow physical event: a few uncomfortable days, perhaps dizziness or nausea, after a drug is stopped. “Relapse,” meanwhile, can become a catch-all for feeling worse after treatment ends.
But people do not experience their lives in diagnostic categories. They experience a disrupted sleep pattern. A racing mind in a meeting. A sudden inability to concentrate on a book. Irritability that changes how they speak to their partner. Fear about whether they can keep working, parenting, driving, or simply trusting their own judgment.
Those are meaningful outcomes. The name we give them matters partly because it affects what gets noticed next.
A systematic review of SSRI withdrawal found that symptoms can be varied and may be mistaken for signs of relapse. That is not the same as saying every difficult experience after stopping an antidepressant is withdrawal. It is saying that the timing alone does not automatically settle the question.
Why the research does not give us a tidy answer
The obvious response is to ask how often withdrawal happens. Unfortunately, even that question turns out to be less tidy than it sounds.
A 2024 systematic review and meta-analysis in The Lancet Psychiatry estimated that about 31% of people reported at least one discontinuation symptom after stopping an antidepressant, compared with 17% after stopping placebo. The authors estimated that, after accounting for symptoms also seen in placebo groups, withdrawal symptoms affected roughly one in six to seven people.
That figure is useful, but it is not a universal forecast. The studies included different medications, different ways of stopping, different lengths of prior treatment, and different methods for asking about symptoms. Some studies observed people only briefly. Some relied on symptoms reported during trials that were designed to study antidepressant effectiveness, not withdrawal in the depth people may want when making sense of their own experience.
Another 2025 meta-analysis of randomized trials found an increase in discontinuation symptoms during the first week after stopping, particularly dizziness, nausea, vertigo, and nervousness. It did not find an association with depression symptoms in the trials that measured them. That is an interesting result, but not a final verdict. A week is a short window, and trials cannot capture every version of a real-world medication change.
Then there is the other end of the question. A 2025 systematic review of persistent symptoms found that the research on longer-lasting withdrawal experiences was sparse and mostly low certainty. That does not prove prolonged symptoms are common, rare, or impossible to distinguish from other problems. It tells us that the confidence people may want from the research is not yet available.
Similarity is the problem
Withdrawal and relapse can look alike in ways that matter. Low mood, anxiety, insomnia, irritability, trouble concentrating, and changes in appetite are not owned by one explanation.
And a person’s life does not pause while clinicians and researchers sort that out. A return of insomnia can make work harder. Anxiety can shrink someone’s willingness to leave the house or make plans. Poor concentration can affect school, caregiving, finances, and the ordinary decisions that hold a week together.
That is why “Were the symptoms counted?” is not enough. I want to know: Did someone regain the ability to participate in their actual life? Could they think clearly enough to make decisions? Did a change that looked successful on a symptom checklist come with a cost in energy, emotional range, sexual functioning, or connection?
Those questions apply whether a person continues a medication, reduces it, stops it, or restarts it. They are not arguments for one path. They are arguments against reducing a person’s experience to a single word too quickly.
What makes the distinction especially hard
There is a real risk of oversimplifying in both directions.
Calling every return of distress “withdrawal” could obscure a recurrence of depression or anxiety that deserves careful attention. Calling every return of distress “relapse” could obscure the effects of a medication change and leave a person feeling that their experience was never fully heard.
The Royal College of Psychiatrists acknowledges this overlap in its guidance on stopping antidepressants: some symptoms can be difficult to distinguish from the return of the condition for which the medication was prescribed. That may sound unsatisfying. In a way, it is. But false certainty would be worse.
The useful question may not be, “Can we prove the label immediately?” It may be, “What changed, when did it change, and what did that change do to this person’s ability to live their life?”
That shifts the conversation away from winning an argument about terminology. It brings us back to the person in front of us: their history, their medication, their timing, their symptoms, and the parts of life that became easier or harder.
Sometimes the honest answer will be withdrawal. Sometimes it will be relapse. Sometimes it will be uncertainty that needs to remain uncertainty for a while.
That is not a failure of judgment. It may be the beginning of better judgment.