Paxil Withdrawal
A Review of Literature and Foreign Labels

Going through a severe Paxil withdrawal is like being dragged into the event horizon of a "black hole." Once withdrawal takes hold no amount of resistance or determination can stop what's next: total personal annihilation. It's like being murdered, shredded to pieces from the inside out, but you're still alive....

—Rob Robinson
Paxil withdrawal survivor

GSK's "Do More, Feel Better, Live Longer" video
(Windows Media Player or RealOne Player required to view)

Physicians in the U.S. are bombarded with marketing messages designed to minimize concerns surrounding withdrawal. For instance, GSK uses the term “discontinuation syndrome” rather than “withdrawal” because it infers “addictive” qualities. In fact, the term “discontinuation” is not even a legitimate medical term — it is not listed in any recognized medical dictionary. GSK’s contrived distinction between “discontinuation” and “withdrawal” is wholly inconsistent with the language found in foreign labels for Paxil, wherein GSK acknowledges that Paxil can cause withdrawal symptoms:

Italy: “Withdrawal symptoms may occur if treatment is discontinued abruptly. Such symptoms ... include: insomnia, dizziness, sweating, palpitations, nausea, anxiety, irritability, parasthesia and headache;”

UK:“[W]ithdrawal symptoms have been reported on stopping treatment. ... Dizziness, sensory disturbance (e.g. parasthesia), anxiety, sleep disturbances (including intense dreams), agitation, tremor, nausea, sweating and confusion have been reported following abrupt withdrawal of ‘Seroxat’ [Paxil];”

Ireland: “Withdrawal reactions have been reported following discontinuation of ‘Seroxat’ [Paxil], these include dizziness, sensory disturbance (e.g. paraesthesia), anxiety, sleep disturbances (including intense dreams), agitation, tremor, nausea, sweating and confusion;”

Netherlands: “[A]brupt discontinuation of Seroxat therapy must be avoided as this may result in withdrawal symptoms such as sleep disturbances, sensory disturbances, dizziness, agitation or anxiety, sweating and nausea;”

Spain: “Withdrawal symptoms[.] Discontinuation of paroxetine administration (especially if it is abrupt) may lead to withdrawal symptoms such as dizziness, sensory disturbances (including paraesthesia and sensation of cramps), headache, sleep disturbances, agitation or anxiety, nausea and sweating;” and

France: “Abrupt withdrawal of the treatment may cause, within one week, symptoms such as dizziness, sensory disorders, sleep disturbances, agitation and anxiety, asthenia, digestive disorders and sweating. These signs may persist for 1-2 weeks.”

GSK continues to assert that “discontinuation symptoms” happen with all antidepressants and are “generally mild” and “self-limiting.” And at one time GSK stated in its promotional material: “Is Paxil addictive? No. Paxil is not a controlled substance. Paxil belongs to a class of medications called SSRIs, which have not been shown to be associated with addiction.” However, according to the World Health Organization publication, “WHO Drug Information,” Volume 15, No. 1, 2002, concerning the December 14, 2001 label change: “The Food and Drug Administration has published a product warning for paroxetine regarding severe withdrawal symptoms of the kind that could lead to dependence.”

GSK also continues to make statements such as “in most instances” withdrawal effects are simply “relapse.” (Glamour Magazine, April 2003.)

This is entirely inconsistent with the Paxil label in Switzerland, which states:

“Discontinuation of Seroxat may lead to discontinuation symptoms. Typical symptoms include nausea, anxiety, dizziness, sensory disturbances (including paraesthesia), headache, sweating, agitation, fatigue and tremor. These symptoms usually start abruptly within a few days of discontinuation and can be distinguished from relapse symptoms, which occur later and build up gradually. To prevent such discontinuation symptoms paroxetine should be discontinued slowly and with a gradual reduction (maximum 10mg/week).”

The truth is, from its inception onto the U.S. market in December 1992 to the present, medical researchers throughout the world have documented Paxil's propensity to induce dependence/ withdrawal symptoms when patients attempt to reduce or discontinue taking Paxil. These observations have been published in an abundance of peer reviewed medical journal articles. In fact, after a very short time on the market (in 1993), the Committee on Safety of Medicines ("CSM"), the U.K.’s counterpart to the FDA, reported 78 cases of withdrawal after discontinuation of Paxil and noted that "such reactions have been reported more often with [Paxil] than with other SSRI's." Although GSK has been aware for the past decade that Paxil can cause withdrawal, this vital information was not included in the drug’s label. Although GSK was required by the FDA to strengthen its warning about Paxil withdrawal in December 2001, many in the public allege that it remains inadequate.

Our effort to get the Paxil warning label revised to reflect the truth about the withdrawal problems with Paxil began back in August 2000. That is when we joined attorney Don Farber in filing a lawsuit in an effort to force SmithKline Beecham to revise the label for Paxil. Although the language in the revised label, in our view, should be stronger, we feel the new language is a definite step in the right direction and we feel vindicated in our efforts.

—Karen Barth-Menzies
Baum Hedlund Law Firm

The medical literature also points out that Paxil has the highest incidence of withdrawal reactions, “significantly more frequent than with other SSRIs.” According to one study involving four SSRIs (Paxil, Prozac, Zoloft, and Celexa), wherein Paxil was by far the worst in causing withdrawal effects, the authors stated that the withdrawal effects seen with Paxil were "not evident in patients receiving (Prozac), (Zoloft), and (Celexa), suggesting they are not an SSRI class phenomenon."

If discontinuation reactions occur in patients stopping (Paxil), the majority will experience symptoms that are mild to moderate in intensity, and are usually limited to two weeks.

—Mary Anne Rhyne
GSK spokesperson


A review of the literature shows that, on average, over one-third of people taking Paxil for any extended period of time experience withdrawal symptoms, and of those, 21% experience severe withdrawal symptoms.

"Discontinuation of Treatment"
The side effects are things like dizziness, nausea, headache, um, and are clearly labeled in the information made available to doctors and patients."

Dr. Alastair Benbow
GSK's European Medical Director

from GSK's web site

In the clinical trials of Paxil, a significant percentage of patients (up to 50% according to some studies) experienced withdrawal, despite the fact that a tapering regimen was utilized in those trials. Notwithstanding these alarming figures, according to a survey conducted in 1997 and published in the Journal of Clinical Psychiatry, the vast majority of physicians are not aware of this problem. According to the authors, “education about discontinuation reactions … is needed for both psychiatrists and family practice physicians." These concerns were echoed in an article published in December 2000, wherein the author stated that there is widespread "misdiagnosis of antidepressant discontinuation symptoms" and "increased professional awareness of discontinuation symptoms is necessary to prevent misdiagnosis and inappropriate treatment." A few years ago, the Harvard Mental Health Letter disclosed that as many as 75% of non-psychiatric physicians were unaware that Paxil can cause withdrawal reactions or how to properly diagnose the associated symptoms.

Despite the abundance of data pointing to a problem, GSK has never prospectively studied the issue of Paxil withdrawal. This is particularly disconcerting given Paxil’s remarkably short half life. According to one of the few non-confidential documents obtained by Plaintiffs in this litigation, titled “Draft Guidelines For Abuse Liability Assessment,” compiled by the Subcommittee on Guidelines for Abuse Liability Assessment, Drug Abuse Advisory Committee, Food and Drug Administration, U.S. Public Health Service states: “In general, rapid onset, short duration drugs appear to have greater abuse liability than do pharmacologically similar drugs with slow onset and long duration of actions.”

The Slippery Slope of Semantics: "Discontinuance" vs. "Withdrawal"

A perfect storm is a situation where, by the confluence of specific events, what might have been a minor issue ends up being magnified to proportions that are out of control. In such a situation, it is clear that if one element was removed from the mix, the whole matter would have collapsed, but because just the right things were in the mix, the situation balloons.

— definition of "a perfect storm"

The following is an excerpt taken from Medicines Out Of Control? by Charles Medawar and Anita Hardon. This book is a must read for anyone interested in taking a "glass-bottomed boat tour through the Paxilian swamp;” it provides a spectacular "3-D view" of the “SSRI/SNRI denizens of the deep” darting about in murky depths below.

The word "discontinuation," as a euphemism for withdrawal, had entered the medical literature pretty much from the time of Eli Lilly's Christmas-in-Arizona "consensus panel" meeting, in late 1996. In the previous three years, Lilly had seen Paxil eroding Prozac's market share, and Lilly had decided to fight back. The point of the campaign was to draw attention to the withdrawal symptoms seen with Paxil.

Lilly needed to be sure the problem would not be perceived as "dependence" as that would have implicated Prozac as a member of the same class of drugs. Lilly, as it were, needed to throw Paxil overboard but without rocking the boat. So they threw out a lifeline: by promoting the word "discontinuation," Lilly kept the boat afloat. SmithKline Beecham (now GlaxoSmithKline) and the other SSRI Pharmas grabbed at the word and developed the concept. A detailed internal SKB briefing (revealed in 2003) advised employees that, "Terminology such as 'withdrawal symptoms' should be avoided as it implies dependence." The briefing emphasized that "discontinuation events are not a major clinical issue" and staff were advised: "Highlight the benign nature of the discontinuation symptoms, rather than quibble about their incidence."

Publicly, SKB (now GSK) had always denied that paroxetine (Paxil/Seroxat/Aropax) withdrawal was a problem but, behind the scenes, the company was telling its people that Prozac caused such problems too: "With fluoxetine, however, such symptoms may occur several weeks later and last several months." SKB (now GSK) also argued that the problem wasn't the drug, but the way it was used: "The severity of symptoms does not appear to be related to either dose or treatment duration, but to abrupt discontinuation." The company concluded that the issue, "has almost entirely been driven by marketeering and is often based on misrepresented or carefully selected data."

Over the years, and at the insistence of the Pharmas, this notion of "discontinuation" gained ground. It readily survived the merger, in 2002, that produced GlaxoSmithKline. GSK briefed its press and media staff to tell enquirers "Discontinuation symptoms are completely different to addiction or dependence..." (GlaxoSmithKline: "Reactive Key Messages and Issues Document" December 19, 2001).

In the same breath, GSK media people were told to say that the UK and European regulators had investigated thoroughly, and had "concluded that SSRIs do not cause dependency/addiction." ... "Seroxat is clearly shown as being neither addictive nor causing dependence." In fact, the European regulators had concluded that, "the lack of evidence for dependence does not prove the absence of a problem and any evidence, which will emerge or will be produced, should continue to be evaluated" (EMEA, April 2000).

To their credit, the European regulators had abandoned the NERO defense — no evidence of risk equal evidence of no risk — but then they seemed to go off the rails. They produced a policy paper that mused about the need to conduct batteries of studies on animals in order to confirm or deny the ample human experience of withdrawal and dependence problems (EMEA, December 2000). Nearly ten years downstream of paroxetine, the regulators had begun to think about requesting (but not requiring) both pre-clinical (animal) and clinical studies to establish what thousands of users already knew. This proposal to sacrifice some animals might best be understood as some gesture of faith or atonement, offerings to the Gods of Hope and Delay. There was simply no point in focusing on animal studies, when physical dependence and withdrawal problems were already clear from the experience of newborn infants, after exposure to SSRIs during pregnancy. Neonatal withdrawal problems had everything to do with the properties of the drug and nothing to do with relapse into depression. What better model of the risk of dependence could there be?

If you, or someone you know, suffered withdrawals as a consequence of taking Paxil, please report it immediately to the United States Food and Drug Administration. This is one way consumer leverage can be applied to GlaxoSmithKline; eventually this official documentation will compel the company to fully disclose the "adverse side effects" of Paxil. You can report your side effects at: MedWatch.