New Study: Symptoms Are Common but Often Not Clinically Significant When Stopping Antidepressants

There are long-standing concerns about symptoms that can occur when antidepressant medications are discontinued, and what those symptoms mean. A new systematic review and meta-analysis published in the July 2025 issue of JAMA Psychiatry found that symptoms including dizziness and nervousness are common when antidepressants are discontinued, even after adjusting for a placebo effect.[1]

The analysis further showed that mood worsening was not associated with antidepressant withdrawal, leading the researchers to conclude “that later presentation of depression after discontinuation is indicative of depression relapse.”

Discontinuing antidepressants

Abruptly discontinuing antidepressants without medical supervision is ill-advised and can lead to symptoms including irritability, agitation, anxiety, insomnia, dizziness, lack of coordination, nausea, fatigue, muscle pain or chills.[2] However, discontinuing antidepressants is clinically indicated in many situations, including: (1) when treatment is deemed successful and completed, (2) in preparation for or in response to pregnancy or surgery, (3) to switch to another medication or psychotherapy, (4) to avoid intolerable adverse effects, or (5) because the drug is not working.[3]

The diagnosis of discontinuation or withdrawal syndrome is not standardized. The threshold for such a diagnosis is low and variable; one or more of the symptoms listed above may be present or become worse. Other symptoms may include headache, tremor and vivid dreams, or the rarer symptoms of anorexia, loss of balance, cognitive impairment, crying spells, dry mouth, diarrhea, high blood pressure, mania, psychosis, sexual dysfunction and suicidal ideation.

According to the medical reference UpToDate, antidepressant discontinuation syndrome occurs in about one-quarter of patients who stop their antidepressants, and the syndrome “is usually mild and dissipates spontaneously over one to three weeks.” Severe discontinuation symptoms occur about 3% of the time. However, even when participants in clinical trials stop a placebo antidepressant, the estimated incidence of discontinuation syndrome is about 17%. Thus, discontinuation symptoms often represent a placebo effect.

Psychiatrists have thought that the risk of discontinuation syndrome is elevated by abrupt rather than gradual discontinuation of medications, especially for antidepressants with shorter half-lives in the body. The causes are uncertain, but because the most commonly used antidepressants are selective serotonin reuptake inhibitors, the neurotransmitter serotonin may play a role.

Importantly, antidepressant withdrawal is different from addiction withdrawal (from substances like opioids or nicotine) because antidepressant cessation does not disrupt euphoria or other reinforcing effects, and patients do not crave antidepressants in increasing doses.

The new study

The systematic review and meta-analysis aimed to update and refine estimates of antidepressant discontinuation symptoms using randomized clinical trials and standardized or individual symptom scales. Previous work often relied on categorical endpoints (for example, any discontinuation symptoms versus no symptoms) or did not include an assessment of symptoms after placebo discontinuation.

The meta-analysis identified 50 studies including 17,828 total participants, who were predominantly White. All were adults (mean age 44 years); 67% were women. Their diagnoses included major depression, generalized anxiety, panic disorder, fibromyalgia, premenstrual dysphoria, posttraumatic stress, social anxiety and compulsive shopping. The outcomes were either the presence of an individual symptom or a count of total symptoms.

After compiling data from multiple studies with sufficiently similar data elements, the researchers developed statistically adjusted estimates of discontinuation symptoms. The methods enabled between-group comparisons (for example, antidepressant versus placebo changes or abrupt versus gradual discontinuation of drugs).

Compared with placebo discontinuation, antidepressant discontinuation was associated with approximately one additional symptom within a week of stopping the medication. In individual studies, the mean count of symptoms was approximately 2.4 for antidepressants and 1.5 for placebo. According to the researchers, this difference was below the threshold for clinically important discontinuation syndrome and lower than the rates found in prior reviews. Importantly, because symptom intensity was not assessed, the increased symptom count was an uncertain indicator of increased withdrawal severity. In general, the study found that symptom counts declined between post-discontinuation weeks 1 and 2.

Antidepressant discontinuation was associated with significantly increased odds of dizziness, nausea, vertigo and nervousness compared with placebo. Within two weeks after discontinuation, dizziness or light-headedness were reported as the single most common symptoms. Moreover, dizziness or light-headedness showed the largest difference in frequency between participants receiving an antidepressant and those receiving a placebo: approximately 7.5% and 1.8%, respectively.

Nervousness or irritability was evident in 3.0% of participants receiving antidepressants and 0.85% of those receiving placebo, a statistically significant difference. Depression or mood worsening was evident in 1.3% of participants receiving antidepressants and 1.5% of those receiving placebo, which was not a statistically significant difference. These results suggest that symptoms soon after discontinuing antidepressants may reveal the emergence or worsening of anxiety, not depression. Later presentation of depressive symptoms, however, may indicate depression relapse.

Sufficient data were available to show that the antidepressants desvenlafaxine (PRISTIQ and generics), duloxetine (DRIZALMA SPRINKLE and generics), escitalopram (LEXAPRO and generics), paroxetine (PAXIL and generics) and vortioxetine (TRINTELLIX) were more likely than placebo to cause discontinuation symptoms. By comparison, agomelatine (not approved in the United States) was the only antidepressant that did not show a discontinuation effect.

Consistent with previous meta-analyses, the new study did not find an association between treatment duration and discontinuation symptoms. It also concluded that abrupt discontinuation, compared with tapering, was not associated with more withdrawal symptoms at week 1. However, available data were insufficient to determine whether tapering over a longer time might reduce symptoms or prevent illness relapse.

What You Can Do

If you are taking antidepressants and think it may be time for you to stop taking them, work with a clinician to formulate and implement a tapering and monitoring plan to minimize the potential for adverse effects or illness relapse.

Although discontinuing antidepressants was associated with more symptoms than discontinuing a placebo antidepressant in clinical trials, the symptoms typically are not clinically significant and do not persist. If depressive symptoms develop later, seek medical attention because they may indicate a depression relapse.
 

References

[1] Kalfas M, Tsapekos D, Butler M, et al. Incidence and nature of antidepressant discontinuation symptoms: A systematic review and meta-analysis. JAMA Psychiatry. 2025;82(9):896-904.

[2] Drug profile for selective serotonin reuptake inhibitors (SSRIs). Worst Pills, Best Pills News. Last reviewed July 31, 2025. https://www.worstpills.org/monographs/view/53. Accessed October 3, 2025.

[3] Hirsch M, Birnbaum RJ. Antidepressant discontinuation syndrome and discontinuing antidepressants in adults. UpToDate. July 14, 2025.