Talk:Paroxetine/Archive 2

Protection
I've protected the article due to edit-warring. I'm not going to block anybody this time, but if this edit warring resumes once the temporary protection ends, lengthy blocks will be entirely appropriate.

Some useful sources to guide your discussion of this point would be and, if anybody needs full-text of these articles please e-mail me. Tim Vickers (talk) 23:17, 15 January 2009 (UTC)

I appreciate the sharing of all of the current journal articles in this discussion. Statistical testing and proving significance is difficult in this field. "Selective serotonin reuptake inhibitors (SSRIs) for depressive disorders in children and adolescents" linked to by Tim above is essentially inconclusive, in part due to statistical complexity. An additional study by Gibbons concludes "Results of our analysis of the FDA AERS data revealed that newer antidepressants are associated with lower rates of suicide adverse event reports compared with older antidepressants. " although it does not appear that this study targetted adolescents. I may be wrong. Mwalla (talk) 21:54, 16 January 2009 (UTC)mwalla

I goofed with the lead
In trying to model the lead on sertraline, I introduced some wrong info, which an IP picked up. I have corrected - paroxetine has more anticholinergic activity than other SSRIs. Apart from that, I think it flows nicer now. Casliber (talk · contribs) 09:42, 16 January 2009 (UTC)


 * Sertraline's lead: "Due to the rarity of this side effect, statistically significant data is difficult to obtain, and suicidality continues to be a subject of controversy." is inappropriate for paroxetine. We should replace it with: “Paroxetine is associated with a statistically significant increase in the risk of suicidal tendencies in both adults and children.”


 * The fact that paroxetine increases suicidality of children is beyond any doubt, and there is a special report by MCRA on that matter (I am trying to find the ref).


 * In other respects, I consider a recent systematic analysis of published and unpublished trials of paroxetine in the Canadian Medical Association Journal (, free access) as definitive:


 * "The present analysis,which suggests that paroxetine is associated with a statistically significant increase in the risk of suicidal tendencies, expands the results of previous re-analyses of GlaxoSmithKline data.16,17 In particular, in the analysis carried out by Glaxo-

SmithKline of suicide attempts by adults with major depression, the frequency was higher among patients who received paroxetine than among those who received placebo (11/3455 v. 1/1978; OR 6.7, 95% CI 1.1–149.4).27 The recently released re-analysis by the US Food and Drug Administration of 372 placebo-controlled antidepressant trials involving almost 100 000 patients with any psychiatric disorders confirmed these figures by showing that, among the selective serotonin reuptake inhibitors and newer antidepressants, only paroxetine was significantly associated with an excess risk of suicidal behaviour (US Food and Drug Administration codes 1, 2 and 3) (OR 2.76, 95% CI 1.16–6.60).19" 71.244.121.113 (talk) 12:54, 17 January 2009 (UTC)


 * Interesting. There are still problems that are frustrating in trial design with selection of patients etc. - milder cases, not using people who voice suicidal ideation. etc. Anyway, as it is a metaanalysis and not a review article, we should word it to reflect the source. Casliber (talk · contribs) 14:08, 17 January 2009 (UTC)


 * Works for me. As you can see from the above excerpt, the results of CMAJ analysis simply confirm the results of another 3 analyses including the ones by the FDA and GSK itself. So they seem to be uncontroversial. 71.244.121.113 (talk) 15:20, 17 January 2009 (UTC)

Is the Canadian Medical Association Journal as prestigous a journal on the topic of psychiatry at the American Journal of Psychiatry? The CMAJ Besides, I think the conclusion of the paper is misquoted. Here is a quote from an editor of the journal about the reference paper: "I was disappointed and puzzled by the bizarre primary outcome measure selected by the authors: the proportion of patients who left a study early for any reason. Consider the ideal situation in which no one in either study arm drops out; it would be impossible for the active treatment to be better than placebo even if all treated subjects went into remission. How can this be a measure of effectiveness?

In my practice, the biggest challenge is persuading patients to persist with therapy through the first few days of unpleasant side effects until the beneficial effects become manifest. I consider early dropout to be a failure of my persuasive powers and not an indication that the therapy is ineffective. I believe that it is important to distinguish between dropout in the first days of treatment, which is a consequence of the predictable and often transitory unpleasant side effects, and delayed dropout, which may reflect treatment failure. The authors failed to stratify their analysis on the time of dropout and their analysis is thus not informative with respect to dropout for the important end points of treatment failure or persistent side effects.

As a clinician I am primarily interested in the effectiveness of a drug in those who actually take it. It is thus disappointing that the authors gave short shrift to their secondary outcome measures, all of which showed a significant benefit of active treatment."

This sentence "A large systematic review of published and unpublished randomised trials showed that paroxetine is associated with a statistically significant increase in the risk of suicidal tendencies in both adults and children." should be vetted in this discussion before it goes in the main page. Instead of putting words in the mouths of the authors, why not use the actual conclusion from the study:

Paroxetine was more effective than placebo, with fewer patients who did not experience improvement in symptoms of at least 50% —Preceding unsigned comment added by 69.243.189.111 (talk) 00:59, 18 January 2009 (UTC)


 * No the analysis was not misquoted."The present analysis,which suggests that paroxetine is associated with a statistically significant increase in the risk of suicidal tendencies, expands the results of previous re-analyses of GlaxoSmithKline data." Nobody doubts that paroxetine is better than placebo, and that is reflected in its indications and in the main text. However, suicidality is a troublesome and serious side effect and should be also reflected in the lead. No "vetting" is required for writing in WP. 71.244.121.113 (talk) 01:39, 18 January 2009 (UTC)

Paroxetine is associated with a clinically significant weight gain
I would like to change the sentence "however, it has a more pronounced anitcholinergic effect and may cause weight gain." to "Paroxetine is associated with a clinically significant weight gain".

I would like to skip mentioning the mechanism. While the anticholinergic effect of paroxetine may be responsible for the weight gain, it is hypothetical, since the M3 inhibitory properties of paroxetine are relatively weak.

The studies on paroxetine and weight gain are nicely summed up by Papakostas in his recent review in Journal of Clinical Psychiatry (. Sorry, folks, no public access to this one.)

"34-week clinical trial27 comparing the SNRI duloxetine (40–120 mg/day) with the SSRI paroxetine (20 mg/day) and placebo reported that both paroxetine and duloxetine treatment were more likely to result in weight gain than placebo. The incidence of >7% body weight gain was 3.1% among placebo-treated patients, 10.8% among duloxetine-treated patients, and 13.8% among paroxetine-treated patients. Thus, it appears that there may be differences among SSRIs in the risk for weight gain during long-term treatment. In fact, a study28 involving the treatment of MDD with the SSRIs paroxetine, fluoxetine, or sertraline reported weight gain of 7% or greater by 25.5% of patients taking paroxetine for 26 to 32 weeks. Conversely, only 6.8% of patients taking fluoxetine and 4.2% of patients taking sertraline experienced clinically significant weight gain in that study." A note of explanation - 7% is generally considered to be a cut-off for a clinically significant weight gain. Over a year or two 3-7% of patients would show such a weight gain on placebo. So 4% on sertraline and 7% on fluoxetine are not much different from normal, while 25% on paroxetine is huge. 71.244.121.113 (talk) 15:37, 17 January 2009 (UTC)
 * I don't think the wording was meant to imply that anticholinergic effects cause weight gain, just note that a) paroxetine has relatively pronounced anticholinergic effects, and b) may cause weight gain. I agree it should be reworded. Fvasconcellos (t·c) 17:10, 17 January 2009 (UTC)

Is efficacy of paroxetine for depression similar to that of older tricyclic antidepressants?
Is efficacy of paroxetine for depression similar to that of older tricyclic antidepressants? - It may well be. However, one study I know about - indicates that it is not so. Are there other studies, more positive for paroxetine? Until we have those refs, maybe we need to remove this equal efficacy statement. 71.244.121.113 (talk) 16:03, 17 January 2009 (UTC)
 * A Cochrane Review is in the works. Meanwhile, this is an excellent source for comparisons with other newer antidepressants.  may be useful but I'd have to see the full text (or perhaps a kindly soul will have a look themselves? [waves at Cas, Tim) Fvasconcellos (t·c) 17:28, 17 January 2009 (UTC)
 * Thank you. According to the paper you recommended, tricyclic antidepressants showed trend (close to statistical significance p=0.07) to be more efficacious than paroxetine. The questionable introductory sentence should be re-written. 71.244.121.113 (talk) 01:19, 18 January 2009 (UTC)

An anonymous URL changed that sentence from "less effective" to "more effective;" I followed the links and ended up compromising on "comparable.".TVC 15 (talk) 08:07, 26 January 2009 (UTC)

CMAJ study
An anonymous URL (161.150.2.55) removed a quote from the introductory paragraph stating, "The CMAJ study is essentially discredited by the editor of the CMAJ in the same issue." However, the edit left in place a summary of the study's pro-paroxetine findings. If the study is discredited, then it should not be quoted at all. I will remove the sentence here so the issue can be sorted out.


 * For major depression, a large systematic review of published and unpublished randomised trials showed that paroxetine was more effective than placebo, with fewer patients who did not experience improvement in symptoms of at least 50%; however, the same review found, "Among adults with moderate to severe major depression in the clinical trials we reviewed, paroxetine was not superior to placebo in terms of overall treatment effectiveness and acceptability."

And, please remember to sign in, and discuss here rather than anonymously changing the article.TVC 15 (talk) 20:44, 26 January 2009 (UTC)

Ends vs. Means
This article and accompanying discussion seem to have evolved through three stages. The article seems to have started out with pro-GSK bias that many above suspected was written by GSK and/or its public relations firm(s). Predictably, that led to a backlash, and some people seem to have a negative view of drugs generally and everyone who makes or sells them. Now, we seem to have entered a third stage, perhaps related to the increase in suicides during the years after the FDA's black box warning; some users from WP's pharmacology project and some anonymous URLs seem to be trying to rehabilitate SSRIs generally including paroxetine specifically. Sadly, some of these users (Mwalla and at least one of the anonymous URLs) act as if "the ends justify the means," i.e. if you put paroxetine in the water supply, the total number of suicides might be reduced, thus lives would be saved, and we should silence any contrary information to save lives. In trying to defend reliably sourced information, I personally have been falsely accused of all kinds of things, when in fact I am just trying to maintain WP:NPOV balancing positive and negative reliably sourced information. Every drug has side effects, and every FDA-approved drug has some proven benefits. Wikipedia is not a sales tool for drug companies or clinicians, no matter how well intentioned, to shepherd customers into making what someone else thinks is 'the right decision' or what's best for them. Wikipedia is about accurate, reliably sourced information, which readers can use in making their own decisions. Please, let WP report objectively.TVC 15 (talk) 21:07, 26 January 2009 (UTC)


 * The increase of suicides in the US in 2004 was nothing but a statistical blip. In the following 2005 the number of suicides decreased (see doi: 10.1176/appi.ajp.2007.07091467). The MHRA report on the safety of SSRI's issued in December 2004 was skeptical about the attempts to interpret the suicide rates in any way: "A range of factors influence population suicide rates. It is therefore challenging to distinguish the discrete effects of increased antidepressant prescribing from changes in other risk factors. Furthermore, declining overall suicide trends may mask rises in some age/sex groups. In Australia, recent rises in antidepressant prescribing were associated with declines in suicide in some age/sex groups but with increases in others. In Britain, declines in suicide preceded increases in prescribing, and rises in antidepressant prescribing since 1991 in different age/sex groups do not consistently coincide with clear changes in previous suicide trends." (see www.mhra.gov.uk/home/groups/pl-p/documents/drugsafetymessage/con019472.pdf, p. 71). 71.244.121.113 (talk) 02:49, 27 January 2009 (UTC)

161.150.2.55 is back, and illustrates what I mean. Previously, 161.150.2.55 deleted the CMAJ review's conclusion that overall effectiveness and acceptability was no better than placebo, but tried to lead the article with a pro-paroxetine finding cherry-picked from the same review. 161.150.2.55 expressly argues in favor of bias because reliably sourced facts are "scary" and might complicate clinicians' sales pitches, so the information must be deleted to save lives. 161.150.2.55's factual assumptions are false as demonstrated by 71.244.121.113 above, and the (il)logic is totally misplaced on Wikipedia. WP is not about hiding information in order to collude in manipulating people into what 161.150.2.55 claims is best for them. WP is about making reliably sourced information freely accessible.TVC 15 (talk) 01:07, 28 January 2009 (UTC)
 * TVC, just because someone disagrees with you does not make them pro-bias, manipulative, or unobjective. Unless you can stop imputing sinister motives to your fellow editors I will initiate dispute resolution.  Skinwalker (talk) 17:21, 31 January 2009 (UTC)


 * Responsibilities elsewhere have delayed my reply to the comment immediately above, but the reply belongs here. I continue to assume good faith, and Skinwalker is sometimes right, but the 'straw man' argument above strains credulity.  I never suggested that merely disagreeing with me about something means any of the things Skinwalker listed.  Rather, MWalla and 161.150.2.55 have _stated_ their _goal_ is to bias the article in order to manipulate people: specifically, to suppress reliably sourced information because such "scary" facts might discourage people from taking the drug.  As Skinwalker has acknowledged, and as linked studies and reviews have found, there are better anti-depressants (and non-drug therapies that are also better than paroxetine).  As has been widely reported elsewhere, the sales pitches of prescribers are frequently influenced by drug makers' marketing, including sometimes direct payment.  WP provides a rare forum for objective, reliably sourced information without financial bias, and honest debate may lead to canceling out other biases as well.  To move the discussion forward, I would be very interested if someone can find reports on what paroxetine is actually prescribed for.  I suspect that paroxetine may be prescribed primarily as an anxiolytic, which would make MWalla and 161.150.2.55's arguments (which have already been disproved by 1.244.121.113), and references to depression, less relevant.TVC 15 (talk) 00:25, 10 February 2009 (UTC)

Skinwalker has made a number of changes to the article. On the whole, I think these changes are good. I am sure some will disagree, but I think his approach was balanced. We need to inform people of facts and let the decide for themselves. Paroxetine is a chemical. It helps some people. It does not help others. People with clinical depression and anxiety disorders need help. Do not scare them away from following their doctor's advice. Until mental health care improves, paroxetine is a valuable tool. It is not a cure all but then again, the clinically depressed used to be treated with much harsher "solutions" such as imprisonment or torture. —Preceding unsigned comment added by 24.15.179.168 (talk) 03:07, 1 February 2009 (UTC)
 * Thank you. Let me make my intentions clear:
 * This article should not give the impression that paroxetine is an uncontroversial cure-all for depression. It is not.  Like tricyclic antidepressants, it is no longer considered a first-line treatment for clinical depression.  It can have serious side effects.  There has been controversy over its side effects and marketing, and the article should neutrally represent this.
 * On the other hand, the fact that there are negative side effects is not a reason to use "scare quotes", nor to represent SSRI discontinuation syndrome as a sign of drug addiction when addiction specialists have determined it is not, nor to give undue weight to the point of view that some evil corporation intentionally released a hysterically dangerous drug onto an unwitting populace. I mean, come on, SSRIs are not thalidomide.
 * Most SSRIs have side effect profiles similar to paroxetine. Most other articles on pharmaceutical drugs neutrally present both the positive and negative consequences, and do not attempt to give readers medical advice.
 * This article should rely on the best available sources, namely metaanalysis articles and reviews published in medical journals. In the near future I will review the sources of this article and prune those that do not meet these standards.
 * I invite review and discussion of my contributions. However, I will not put up with accusations that I am shilling for GlaxoSmithKline or trying to whitewash the article of criticism.  Yes, paroxetine has been correlated with an increased incidence of suicidal ideation.  So have pretty much all modern antidepressants.  Without sounding heartless, sufferers of clinical depression (of which I am one) tend to think about offing themselves.  Yes, it can cause a nasty discontinuation syndrome.  These items should be discussed in the article, but should not take prevalence over a neutral discussion of the drug's benefits and risks.  Skinwalker (talk) 03:40, 1 February 2009 (UTC)


 * Skinwalker, thank you for your very nice summation of what is wrong with the article and how it should be put right. I support almost every point you bring up. All the changes you made to the article so far have been excellent. I hope you also agree with a little cleanup I did with the side effects part. On the other hand, TVC15 is also right, and so far has not made any bad changes to the article.


 * Until very recently I would have also subscribed to your ridicule of the "point of view that some evil corporation intentionally released a hysterically dangerous drug onto an unwitting populace. I mean, come on, SSRIs are not thalidomide." However, the revelations that came in 2008 about the misconduct of leading medical researchers and of the best pharmaceutical companies left me aghast. For example, the editorial in a neutral "Journal of Neurology" states that GSK and the researchers may have suppressed the suicide risk data in paroxetine trials.
 * Nemeroff got $2 million and never reported it. How would you treat the reviews and meta-analyses he conducted? What about the work of Thase and Montgomery who are also known for having tight relationships with the industry?
 * An editorial in JAMA in April 2008 entitled "Impugning the Integrity of Medical Science. The Adverse Effects of Industry Influence." goes on to describe "how Merck may have misrepresented the risk-benefit profile of rofecoxib in clinical trials involving patients with Alzheimer disease or dementia. The authors show that the company’s report to the FDA appears to have attempted to minimize the mortality risk by using an “as-treated” analysis, whereas an internal analysis conducted by the company several months earlier and using the correct intention-to-treat analysis provided evidence of a significantly increased mortality risk among patients assigned to receive rofecoxib." 71.244.121.113 (talk) 12:21, 1 February 2009 (UTC)
 * The editorial in the Journal of Neurology is a reliable source, and we should use it. Your paraphrase above is a good summation.  The situation with Nemeroff is somewhat complicated.  On one hand, I am personally appalled at his lack of disclosure, and I expect the NIH to yank any remaining grant funding he retains.  On the other, it is really, really hard to find any pharmacological research that does not have some relationship with industry.  I would not use Nemeroff's papers in this article, but I would judge others on whether or not there is evidence of a "smoking gun", i.e. are there allegations in a reliable source that a researchers' judgement is compromised?  Can you provide a similar source for Thase and Montgomery?  I don't think simply taking research money is a cause for concern, but proven lack of disclosure (extreme in Nemeroff's case) or similar shenanigans on the part of the authors do impeach sources IMO.  As for rofecoxib (better known as Vioxx), that's an altogether different drug.  However, our article on rofecoxib neutrally reports what reliable sources, mostly medical journals, have said on the topic (which is pretty damning) without resorting to polemics.  This article should meet that standard as well.  Skinwalker (talk) 13:13, 1 February 2009 (UTC)

Arbitrary break. How to treat conflicts of interests
My main point is that, unfortunately, evil corporations and unscrupulous researchers do exist, and this has practical implications for this article. For example, a meta-analysis of paroxetine vs tricyclics conducted by Montgomery finds that they are equal (http://cat.inist.fr/?aModele=afficheN&cpsidt=972015). A meta-analysis by Anderson finds that tricylics are likely to be more efficacious than paroxetine. Say, I can find a disclosure from Montgomery stating that he has financial ties with GSK. Anderson does not have any ties with GSK. Should we treat the Anderson's analysis as more valid? Should we leave Anderson's conclusion in the lead and mention Montgomery's analysis only in the body of the article as unreliable? 71.244.121.113 (talk) 14:16, 1 February 2009 (UTC)
 * We should report both, e.g. "One metaanalysis says X, the other says Y.". Let the reader decide.  I don't think either belong in the lead.  As I said above, financial ties are not sufficient to disqualify a source if there has not been significant criticism of that specific researcher, as there has been in Nemoff's case.  If you like, we can ask for a third opinon on how to treat sources that may or may not have financial conflicts.  Skinwalker (talk) 19:44, 1 February 2009 (UTC)
 * There have been significant criticism of Stuart Montgomery. For example, The Guardian (http://www.guardian.co.uk/society/2004/oct/04/health.businessofresearch) describes how Montgomery was sitting on the Committee on the Safety of Medicines, which was deciding whether to approve sertraline. And at the same time Montgomery was advising Pfizer (the sertraline manufacturer) on the best course of action to overcome the committee doubts. Does that disqualify him? If not, I can probably find more. 71.244.121.113 (talk) 22:34, 1 February 2009 (UTC)
 * In another example, in 1995 Stuart Montgomery signed an article, an analysis of paroxetine trials, which concluded that paroxetine decreases suicides 6-fold (see ). Later, Montgomery's co-author David Dunner, admitted that the paper was ghostwritten by Glaxo. (see http://www.guardian.co.uk/Archive/Article/0,4273,4351264,00.html) 71.244.121.113 (talk) 23:13, 1 February 2009 (UTC)

Who cares who writes the research? If you do not trust the research, replicate the expirement yourself. If it is sceintific research, use it. Everyone has a bias. Psychologists are biased against medication. Liberals are biased against "evil corporations". It is not likely that millions of patients, doctors, and scientists are in a mssive conspiracy to trick people into taking paroxetine. What is likely, is that drug companies try to find a product which people are willing to pay for. It is called capitalism. Sometimes fraud happens... but there is no scientific evidence of that in the case of paroxetine. —Preceding unsigned comment added by 161.150.2.55 (talk) 15:11, 2 February 2009 (UTC)


 * Is the US Senator Chuck Grassley, Republican from Iowa, a liberal? He posted on his official website a report detailing Glaxo misdeeds in concealing paroxetine side effects (see http://finance.senate.gov/press/Gpress/2008/prg061208a.pdf). 71.244.121.113 (talk) 01:04, 3 February 2009 (UTC)

Last I checked, Chuck Grassley is not a medical doctor. Would you take George Bush's opinion on embryonic stem cells? —Preceding unsigned comment added by 161.150.2.55 (talk) 14:33, 3 February 2009 (UTC)


 * Grassley is a conservative. He is one of the senior members of the Senate. His attention to the problem of the conflict of interests means that it is serious. The actual report (http://finance.senate.gov/press/Gpress/2008/prg061208a.pdf) was written by Joseph Glenmullen, a professor of psychiatry at Harvard University. Here is an excerpt from Grassley's press-release: "Senator Chuck Grassley has asked the Food and Drug Administration to carefully scrutinize information it received from drug maker GlaxoSmithKline about the anxiety disorder drug Paxil, based on the contents of a newly available report about the drug’s risk for suicide among adults. Grassley also asked the FDA to review findings released earlier this year by the British drug-safety agency which charged that the drug maker has known about suicide risk with pediatric use of Paxil since 1998. The report cited by Grassley was prepared by Dr. Joseph Glenmullen, a professor of psychiatry at Harvard University. The report asserts that GlaxoSmithKline had to know of Paxil’s suicide risk when it sought FDA approval for the drug. The Glenmullen report was recently released from under court seal by a Kansas judge. It is posted with this news release at finance.senate.gov. Grassley asked GlaxoSmithKline about the Glenmullen report last February. Weeks later, the British Medicines and Healthcare products Regulatory Agency released its own report that was four years in the making.“The British counterpart to our country’s FDA found that GlaxoSmithKline withheld important safety data on Paxil," Grassley said. "If the company engaged in this behavior in the U.K., then I want to make sure that the same didn't happen here in the U.S. The FDA should investigate this question thoroughly and be forthcoming about its findings." (http://finance.senate.gov/press/Gpress/2008/prg061208.pdf) 71.244.121.113 (talk) 01:07, 5 February 2009 (UTC)

The Glenmullen report was commissioned by lawyers pursuing cases against GSK. But nobody ever reports that conflict of interest! —Preceding unsigned comment added by 86.154.37.130 (talk) 00:10, 12 February 2009 (UTC)

Are side effects of paroxetine treatment frequent?
The side effects of paroxetine treatment are frequent. There is nothing alarmist about that, and in that respect paroxetine is not much different from other SSRIs. The following information is taken from the Paxil Prescribing Information. 20% of patients treated with paroxetine in the clinical trials for major depressive disorder discontinued due to the side effects. Among the most frequent treatment-emergent side effects in the trials for MDD were: nausea (26% on paroxetine vs 9% on placebo), somnolence (23% on paroxetine vs 9% on placebo, ejaculatory disturbance (13% on paroxetine vs 0% on placebo), other male genital disorders (10% on paroxetine vs 0% on placebo), asthenia (15% on paroxetine vs 6% on placebo) and sweating (11% on paroxetine vs 2% on placebo). 71.244.121.113 (talk) 15:04, 31 January 2009 (UTC)

Is 10% frequent? I would say frequent is >50%. Suicide is "common" amongst the clinically depressed. Paroxetine reduces this risk. When it comes to side effects, the good side effect of no suicidie trumps the bad side effect of nausea. Doctors are in the best position to weigh the risk of suicide vs. the risk of diarrhea. In the US, you can't get paroxetine without a consultation with a doctor.

Also, what is an ejaculatory disturbance? Is that like a nocturnal emission?


 * In response to the vandal:
 * "In efficacy trials, on average, 61 percent of patients experienced at least one adverse event during treatment. Nausea, headache, diarrhea, fatigue, dizziness, sweating, sexual dysfunction, tremor, dry mouth, and weight gain were the commonly reported adverse events. Overall, second-generation antidepressants led to similar adverse events." (Report by The US Agency for Healthcare Research and Quality on Comparative Effectiveness of Second-Generation Antidepressants in the Pharmacologic Treatment of Adult Depression at http://effectivehealthcare.ahrq.gov/repFiles/Antidepressants_Final_Report.pdf)71.244.121.113 (talk) 00:52, 3 February 2009 (UTC)

Lets be clear, "side effects" such as nausea are insignificant compared to the "side effects" of clinical depression, for which paroxetine has been known to reduce. Individually, most patients find these side effects tolerable. Naturally, if you are not clinically depressed or suffer from severe anxiety, do not take paroxetine. As always consult your doctor. As for teenagers, there is no statistically accurate proof that suicide is caused by the drug. "suicide ideation" is not easily quantified, therefore there is as yet no proof that paroxetine causes suicide nor suicide ideation. In fact, paroxetine reduced the incidence of actual suicides. http://www.cleveland.com/nation/index.ssf/2008/11/after_2_decade_decline_teen_su.html

Statistically significant risk of suicidality with paroxetine

 * "Significantly more patients in the paroxetine group than in the placebo group left their respective studies because of side effects (random effect RR 1.77, 95% CI 1.44–2.18) or experienced suicidal tendencies (odds ratio 2.55, 95% CI 1.17–5.54)." "The present analysis, which suggests that paroxetine is associated with a statistically significant increase in the risk of suicidal tendencies, expands the results of previous re-analyses of GlaxoSmithKline data.16,17 In particular, in the analysis carried out by GlaxoSmithKline of suicide attempts by adults with major depression, the frequency was higher among patients who received paroxetine than among those who received placebo (11/3455 v. 1/1978; OR 6.7, 95% CI 1.1–149.4).27 The recently released re-analysis by the US Food and Drug Administration of 372 placebo-controlled antidepressant trials involving almost 100 000 patients with any psychiatric disorders confirmed these figures by showing that, among the selective serotonin reuptake inhibitors and newer antidepressants, only paroxetine was significantly associated with an excess risk of suicidal behaviour (US Food and Drug Administration codes 1, 2 and 3) (OR 2.76, 95% CI 1.16–6.60).19" 71.244.121.113 (talk) 01:50, 5 February 2009 (UTC)

Review and evaluation of clinical data: relationship between psychotropic drugs and pediatric suicidality, p 30, author Hammad TA, date 2004-08-16, work Joint Meeting of the Psychopharmacologic Drugs Advisory Committee and Pediatric Advisory Committee. September 13 - 14, 2004. Briefing Information.; publisher = FDA. (http://www.fda.gov/ohrms/dockets/ac/04/briefing/2004-4065b1-10-TAB08-Hammads-Review.pdf): "Paxil, all indications (Fixed effect model), Suicide Behavior or Ideation [codes 1, 2, & 6], Risk ratio (95% CI) = 2.65 (1.00,7.02). Note that the signal is suggested in both the MDD and the non-MDD trials." 71.244.121.113 (talk) 02:10, 5 February 2009 (UTC)

There is a difference between suicide and suicide ideation. Paroxtine lowers the risk of suicide in adults. There is not proof for the effects on teenages, but studies suggest teen suicide has increased since the addition of blackbox warnings. —Preceding unsigned comment added by 161.150.2.55 (talk) 14:19, 5 February 2009 (UTC)
 * That is exactly what the paragraph you keep vandalizing used to say: "increase in suicidal tendencies". You must not change the information based on several references and replace is by your own unsupported opinion. 71.244.121.113 (talk) 11:21, 6 February 2009 (UTC)

Placebo v antidepressants
What are people's views on this paper? Initial Severity and Antidepressant Benefits: A Meta-Analysis of Data Submitted to the Food and Drug Administration-- Literature geek |  T@1k?  15:58, 5 February 2009 (UTC)

Thanks for the link. I think it underscores the fact that anti-depressants are medcations intended to treat serious illnesses, such as clinical depression. They are not intended to treat a case of "the Mondays". :) To quote the article " The difference in improvement between the antidepressant and placebo reached clinical significance, however, in patients with initial HRSD scores of more than 28—that is, in the most severely depressed patients. " —Preceding unsigned comment added by 161.150.2.55 (talk) 21:52, 5 February 2009 (UTC)
 * Since the lead author has received income from Squibb and Pfizer, who manufacture drugs that compete with the four antidepressants discussed in the article, I'd say it's clearly biased and unusable. Skinwalker (talk) 23:53, 5 February 2009 (UTC)
 * Good point Skinwalker! I read this study by Kirsch previously, and although I noticed that antidepressant sertraline (Pfizer) was not included into the analysis for spurious reasons, I never linked that with the lead author getting money from Pfizer. Some secondary conclusions of the study are bizarre. On the other hand, we may go easy on Kirsch: the conclusions of the study are broadly anti-antidepressant so they are harmful for all manufacturers. In addition, another, broader and more careful analysis by Turner (.), reached the same main conclusion, although Turner's interpretation is much more objective. 71.244.121.113 (talk) 03:15, 6 February 2009 (UTC)
 * I was being somewhat facetious about not using this source. I find your desire to include this source - which favors the view you've been advocating - while deprecating other sources to be inconsistent with WP:NPOV.  Each of the sources we've discussed have financial connections with industry.  Why should we favor one over the other?  We are likely not going to be able to find medical references whose authors have not taken funding from pharmaceutical companies.  We have a duty, per NPOV, to report all notable views on a topic.  Each source should be discussed in the article.  Each source should be linked externally so that readers can see the financial disclosures on the papers.
 * I strongly suggest you reconsider your approach to editing this article. To paraphrase what you've said above, is is inappropriate to push conclusions that "are harmful for all manufacturers".  Advocacy is frowned upon here, as is the exclusion of reliable and verifiable sources.  You're going to have a tough time here if you can't follow these policies.  Skinwalker (talk) 18:19, 7 February 2009 (UTC)
 * Also, Turner states: "In contrast to Kirsch and colleagues’ conclusion that antidepressants are ineffective, we concluded that each drug was superior to placebo."  I can't see how this represents the same main conclusion as Kirsch, as you state above.  Turner found that certain antidepressants were less effective than is apparent from journal articles, not that these drugs were inferior to placebo.  Skinwalker (talk) 18:44, 7 February 2009 (UTC)
 * Why did you decide that I want the Kirsch paper included? I was actually critical about it. As I said, Turner's paper is much better. Regardless, neither one belongs to this article because paroxetine was combined with other antidepressants within those meta-analyses. Both papers have been objectively reflected in Treatment for depression. Let me repeat myself: antidepressants are more effective than placebo. The devil is in the details, that is, in the balance of benefits and side effects. It is a general consensus in the psychiatry, and I believe you agreed with it in one of your posts, that escitalopram, sertraline and fluoxetine offer a better balance of risks and benefits. Accordingly, they are prescribed more often, as well. I really have no disagreements with you. 71.244.121.113 (talk) 21:30, 7 February 2009 (UTC)
 * On an unrelated note, what is your opinion about 161.150.2.55 edits. Do they constitute vandalism? 71.244.121.113 (talk) 03:15, 6 February 2009 (UTC)
 * Well, vandals don't usually stick around on the talk page to discuss their edits. The paper he or she is concerned about is poorly written (from a grammatical standpoint, at least) and puts forth contradictory conclusions at times.  I myself read it and couldn't figure out if it was saying suicidal tendencies increase or decrease, and whether it was with the control or test groups.  Probably just need to read it closer. I'm uncomfortable describing this as vandalism.  Response to your other point later.  Skinwalker (talk) 11:34, 6 February 2009 (UTC)
 * I checked the definition of vandalism on WP:VANDAL, it is someone who deliberately compromises the integrity of Wikipedia. In this case 161.150.2.55 changes "statistically significant increase in suicide" to "statistically significant decrease", without changing references. This creates a false impression that the references support "decrease" while they actually support "increase". Another type of vandalism includes addition of obscenities or crude humor, which 161.150.2.55 also did here . 71.244.121.113 (talk) 00:58, 7 February 2009 (UTC)
 * Please stop referring to this editor as a vandal, and discuss instead whether the reference supports his conclusion or yours. As for the "crude humor", SSRIs have sexual side effects and can be used to treat premature ejaculation, so his comment to that respect was not off-topic or out of line.  Skinwalker (talk) 18:19, 7 February 2009 (UTC)

Skinwalker, I would agree that the paper is not the best. However, I would not conclude that because someone received money from a drug company, that their research is biased. We should not assume bias or lack of bias. We as readers need to be unbiased in our assesment, make sure they follow scientific method, and look to see if their results can be confirmed with evidence from others sources. We know that in mass media, sensationalism sells newspapers, therefore are all news stories sensationalist? If the BBC does an investigation into Paroxetine, are they likely to report that "it works great, no news here"? —Preceding unsigned comment added by 161.150.2.55 (talk) 15:37, 6 February 2009 (UTC)
 * With the edit block on the page, that user is a certified vandal now. 71.244.121.113 (talk) 21:30, 7 February 2009 (UTC)
 * See my response above. Each source should be discussed in the article so that the reader can come to their own conclusion.  Skinwalker (talk) 18:19, 7 February 2009 (UTC)

Edit warring
There are some edit wars going on here. I've full protected the page 1 week, rather than semi-protecting, as both IP users and logged in users are involved--semi would be an unfair advantage to the logged in users, and that's not what protection is for in a content dispute. Please hash this stuff out on this talk page, or else the protections will likely grow in duration, and then no one will be able to unfortunately edit this page. rootology ( C )( T ) 17:14, 7 February 2009 (UTC)

Paroxetine has no dependence potential

 * I would like to question the subchapter "=Dependence" inserted recently. It is based on a poor quality evidence. The study quoted interviewed 54 "antidepressant users". It does not specify what antidepressants were used, and paroxetine is not mentioned in the article even a single time. The study participants were not blinded or even randomized.
 * There is a general consensus in psychiatry that SSRIs do not cause dependence, and have no abuse potential. For example, a careful study of post-marketing reports (open access at http://www.pubmedcentral.nih.gov/articlerender.fcgi?tool=pubmed&pubmedid=8971432) by the experts from the UK government Committee on Safety of Medicines and Medicines Control Agency concluded: "They [withdrawal reactions] have been reported more often with paroxetine (0.3 reports per thousand prescriptions) than with sertraline and fluvoxamine (0.03), and least often with fluoxetine (0.002)." However, "there was no evidence of habituation or tendency to escalate the dose, and no other evidence of physical dependence." The Sceptical Chymist (talk) 00:38, 15 February 2009 (UTC)

The consensus is that antidepressants are not associated with the symptoms of psychological dependency/addiction such as dosage escalation, abuse and craving. There is a consensus that they can cause a dependence in terms of a withdrawal/discontinuation syndrome. That paper that I cited confirms that. I included in the data that I added that antidepressants are not associated with abuse. I think that the problem is what an author of a paper or review means by dependency. Some when saying no dependency are meaning little to no addiction/habituation (in terms of craving and dose escalation) and abuse. The no evidence of physical dependence I personally think is scientific fraud and the result of conflict of interest and lobbying and whatever. There is plenty of evidence in the peer reviewed literature of upregulation and down regulation of transmitter systems and receptor sites in animal studies and it undoubtly occurs in humans as well and is why you get physical withdrawal syndromes when discontinuing antidepressants. If there is no physical changes or adaptions in the CNS how else do you explain the withdrawal syndrome which can be severe and quite prolonged in some people from SSRIs scientifically?-- Literature geek |  T@1k?  00:55, 15 February 2009 (UTC)

Hmmm, do you really believe that 0.03% of people per thousand experience withdrawal syndromes from paroxetine? Here is what the CSM initially said about benzodiazepines in 1980. "The number dependent on the benzodiazepines in the UK from 1960 to 1977 has been estimated to be 28 persons." I thought you were the sceptical chemist. :=)-- Literature geek |  T@1k?  01:01, 15 February 2009 (UTC)

I have a feeling that I have opened up the can of worms which I try to avoid, which is the unwinnable debate of just what is dependence. Are dependence and addiction the same thing? I still think though the statement that antidepressants don't induce physical dependence to be scientific fraud but I admit that many review bodies have made this statement in their reviews and am aware of it. SSRIs like other non-GABAergic psychotropic drugs almost certainly don't induce a physical dependence to such a severe extent as GABAergic drugs such as alcohol, benzos, ether etc which science has well established produce some of the most profound physiological adaptions (physical dependence) of all psychotropic drugs.-- Literature geek  |  T@1k?  01:09, 15 February 2009 (UTC)


 * I applaud your decision keep to the published facts as WP guidelines recommend. We may disagree with it, but the commonly accepted definition of physical dependence does include tolerance, which is not observed with paroxetine and other SSRIs. Even if we accept the conclusions of the article you quote, the fact that paroxetine was not mentioned in it makes it unsuitable for the Paroxetine article. Before starting this section, I wanted to find real examples of paroxetine dependence, so I ran a PubMed search on "paroxetine and dependence". The single relevant article out of 59 hits was the CSM study I quoted above, which argues against physical dependence. I would welcome any real citation indicating paroxetine dependence. Please remember that withdrawal syndrome, even severe, does not constitute dependence. Caffeine also gives a withdrawal syndrome (see Caffeine. Potentially lethal withdrawal syndrome can be caused by such uncontroversial drugs as propranolol, see and prednisolone, see . The Sceptical Chymist (talk) 01:44, 15 February 2009 (UTC)

Hmmm, true and physical adaptions is evidence of a degree of tolerance developing, but it may only be partial tolerance and therapeutic and pharmacological properties may still be retained on a constant dose (if partial tolerance and not full tolerance develops). A problem with that definition is an alcoholic never becomes "tolerant" to alcohol. They will always get drunk. Even if they drink the roughly the same amount say a litre of vodka per day, they will still notice its effects and get drunk/intoxicated on a litre of vodka. So by that definition one could argue that alcohol does not induce physical dependence. Opiates also generally don't lose their pain relieving properties completely and pain patients can be kept on a maintenance does with some analgesia, although partial tolerance may still occur. I think that SSRIs retain some of their therapeutic properties although i have heard from one psychologist who said that after 5 years SSRIs stop working. He also added most of the time they don't work but when they do work they work quite well but stop working after 5 years or so (due to tolerance? dunno annecdotal word of mouth research :)). Anyway so tolerance doesn't develop to alcohol's intoxicating effects or if it does it is at best slight to moderate, so can alcohol be categorised along with paroxetine as being a drug which doesn't induce physical dependence? Do you see what I mean by the unwinnable argument I started? :+-) I am not looking for an argument, I mean no harm.-- Literature geek |  T@1k?  01:53, 15 February 2009 (UTC)

Propranolol and prednisone I know can cause withdrawal type reactions if discontinued. Infact prednisone, modulates the GABAA receptor much like barbiturates, benzodiazepines do, increasing chloride ion flow via modulation of the neurosteroid binding site. So prednisone probably invokes similar adaptions to alcohol, benzos and barbs, although I am not completely up to scratch on the effects of neurosteroids on the brain and not prednisone in particular. Maybe it modulates other sites in addition or more strongly so I could be wrong but am fairly sure.-- Literature geek |  T@1k?  02:02, 15 February 2009 (UTC)

No, the alcohol is not a good exmple. Drunkards drink amounts regularly, which could kill the naive person. This is even more the case with opiates. A naive person could swiftly be killed by the normal "maintenance" dose of an addict. 70.137.151.133 (talk) 02:07, 15 February 2009 (UTC)

Propranolol is a good example and has nothing to do with GABA or addiction 70.137.151.133 (talk) 02:07, 15 February 2009 (UTC)

In propranolol the guy just gets such cardiovascular effects that his heart explodes or he pops the cork. :) 70.137.151.133 (talk) 02:15, 15 February 2009 (UTC)

I guess my point is physiological adaptions (receptor upregulation, down regulation, desensisation etc) occur with almost any psychotropic drug, even proton pump inhibitors or nasal decongestants can cause withdrawal or rebound symptoms but these adaptions are made by the body for no other reason than to overcome (produce tolerance) the drugs effects. With many drugs this is only partial tolerance and therapeutic effect is only partially lost or in the case of alcohol intoxicating effects are only partially lost due to partial tolerance. So if a person cannot discontinue or has great difficulty discontinuing a drug that their body is physiological tolerant to and getting physiological withdrawal symptoms, it is not physical dependence because the tolerance is not complete? And the question returns what about alcohol? Why is it classed as a drug of physical dependence if tolerance is only partial?-- Literature geek |  T@1k?  02:18, 15 February 2009 (UTC)

Yea but people who are physically dependent on benzodiazepines or barbiturates say like anywhere from 100 mg of diazepam or up to 2000 mg of diazepam equivalent can walk about and appear sober but an alcoholic could not appear to be sober after drinking 1 or 1.5 litres of vodka but would be very drunk. A normal person would be comatosed on 2000 mg of diazepam or possibly be dead but certainly would be hospitalised, likely in ICU. There is a difference.-- Literature geek |  T@1k?  02:18, 15 February 2009 (UTC)

Anyway to end the discussion, I don't mind deleting the reference and data I added if you feel it is necessary. The article wasn't on paroxetine specifically anyway but as antidepressants as a class.-- Literature geek |  T@1k?  02:18, 15 February 2009 (UTC)

Paroxetine in children. Not better than placebo. Suicidality and side effects - worse than placebo.
According to the MHRA report (see Report of the CSM expert working group on the safety of selective serotonin reuptake inhibitor antidepressants, read here )
 * p.45: "Efficacy [of paroxetine] in major depressive disorder. Efficacy was evaluated in three randomised, double-blind, placebo-controlled trials (one of which also had an imipramine arm) involving a total of 767 randomised patients aged seven to 18, treated for eight to 12 weeks. Of these 378 received paroxetine. Dose range of paroxetine was 10mg-50mg/day. Efficacy was not demonstrated."
 * p.45: "General safety profile [of paroxetine in children]. Data from controlled clinical trials were available for 378 patients treated with paroxetine at doses of 10mg–50mg for 12 weeks. Of these, 263 completed eight to 12 weeks of treatment. No deaths occurred in the trials. There are no controlled data on long-term safety. Emotional lability, hostility, insomnia, tremor, dizziness and somnolence were reported more often by paroxetine-treated patients than by placebo-treated patients. Discontinuation due to adverse events occurred in 38 (10%) patients treated with paroxetine and in 15 (5%) of patients treated with placebo."
 * p.58. In a cohort study conducted by GSK "Among children and adolescents, rates of suicidal behaviour were significantly higher in ... paroxetine users compared with other SSRI users (HR=1.6, 95%CI=1.2-2.1)."
 * p.60. Conclusion: "Data on the safety and efficacy of paroxetine in MDD in children and adolescents under the age of 18 did not demonstrate efficacy in depressive illness in this age group, and showed an increase in the risk of harmful outcomes, including episodes of self-harm and potentially suicidal behaviour in the paroxetine group compared to placebo. The balance of risks and benefits in this population was negative."

The opinion that "the risk of inaction may be greater" was expressed in http://content.nejm.org/cgi/content/extract/351/16/1598 about a general issue of treatment with antidepressants, not specifically about treatment with paroxetine. In the light of the above data, this opinion does not apply to paroxetine. The Sceptical Chymist (talk) 02:23, 18 February 2009 (UTC)

Thanks for the link and the breakdown including page references. The problem with this study, is that it excludes patients at the highest risk of suicide, which are the ones most likely to respond positively to the drug. It then must resort to subjective measures such as suicidality. Although this study is not specifically on paroxetine, it includes a much broader population. http://www.cmaj.ca/cgi/content/abstract/178/8/1005
 * This paper is irrelevant. This is an ecological study on prescription of antidepressants in general - not on paroxetine.The Sceptical Chymist (talk) 00:11, 19 February 2009 (UTC)

If you look at the risk ratio for paroxetine in this study, they overlap 1, indicating that there was no significant increase in risk. The data just do not support the conclusion that paroxetine increases the risk of suicide. http://archpsyc.ama-assn.org/cgi/content/full/63/3/332?ijkey=4336f950cf7e456762f36d974baea9212b8e2657 —Preceding unsigned comment added by 161.150.2.55 (talk) 13:41, 18 February 2009 (UTC)
 * The study you are quoting showed 2.65-fold increase in suicidal tendencies for paroxetine for all indications with the 95% confidence interval from 1.00 to 7.02-fold increase. It is statistically significant in my book. The FDA review, on which this study is based also came to the same conclusion:(http://www.fda.gov/ohrms/dockets/ac/04/briefing/2004-4065b1-10-TAB08-Hammads-Review.pdf): "Paxil, all indications (Fixed effect model), Suicide Behavior or Ideation [codes 1, 2, & 6], Risk ratio (95% CI) = 2.65 (1.00,7.02). Note that the signal is suggested in both the MDD and the non-MDD trials." The Sceptical Chymist (talk) 00:21, 19 February 2009 (UTC)

Additionally, SSRIs and Paroxetine are still recommended for treatment in adolescents with MDD. http://www.ncbi.nlm.nih.gov/pubmed/17974724?dopt=Abstract http://www.ncbi.nlm.nih.gov/pubmed/11437014 —Preceding unsigned comment added by 161.150.2.55 (talk) 14:07, 18 February 2009 (UTC)


 * You misrepresent the literature. PID 17974724 states :"Note that only fluoxetine has been approved by the FDA for use in children and adolescents with depression...Overall, fluoxetine has had the largest number of studies with positive results, whereas paroxetine has had the largest number of studies with negative results.34–37,39,42,43". is the only positive trial of paroxetine. When combined with other negative trials as in the FDA or MHRA analyses, the results are negative. That is why paroxetine was never approved for the use in minors.The Sceptical Chymist (talk) 00:46, 19 February 2009 (UTC)

If it includes 1, it is not statistically significant. Careful, your "book" sounds like original research.


 * You 161.150.2.55 say "The problem with this study is that it excludes patients at the highest risk of suicide, which are the ones most likely to respond positively to the drug." That is your own interpretation, which is called original research and is forbidden on Wikipedia, see WP:NOR. You are trying to delete the conclusions of the systematic and comprehensive review by the UK drug authority MHRA: Data on the safety and efficacy of paroxetine in MDD in children and adolescents under the age of 18 did not demonstrate efficacy in depressive illness in this age group, and showed an increase in the risk of harmful outcomes, including episodes of self-harm and potentially suicidal behaviour in the paroxetine group compared to placebo. The balance of risks and benefits in this population was negative. You are trying to replace it by your own OR and a broken quotation from an irrelevant editorial - "the risk of inaction may be greater". The Sceptical Chymist (talk) 00:05, 19 February 2009 (UTC)

When a journal such as the CMAJ publishes an eidotorial about a paper in that journal, it usually means that the paper is controversial or there was disagreement among the editors.

I agree Sceptical and they were using a statement from an article on the general use (risk of inaction in depression in kids) of antidepressants in children and not paroxetine itself but are using their original research to apply it to paroxetine.-- Literature geek |  T@1k?  00:35, 19 February 2009 (UTC)

The sample is not on paroxetine exclusively. Although it may not be recommended, it is up to the doctors discretion and paroxetine is widely prescribed among both adults and adolescents.d

Paroxetine - more completed suicides in adults
The recent insertion in the lead associates paroxetine "with fewer suicides in adults". The supporting reference http://ajp.psychiatryonline.org/cgi/content/abstract/164/7/1044, does not really support it. The reference is not about paroxetine at all, but about the "relationship between antidepressant treatment and suicide attempts in adult patients in the Veterans Administration health care system." Although the veterans prescribed antidepressants had fewer suicides, it is a fallacy to conclude from these data that paroxetine decreases suicides in adults.

To the contrary, according to the FDA analysis of the clinical trials of paroxetine, there was 0 completed suicides in placebo groups vs. 1 completed suicide in the paroxetine group. (see Clinical review: relationship between antidepressant drugs and suicidality in adults by Stone MB, Jones ML in the Overview for December 13 Meeting of Psychopharmacologic Drugs Advisory Committee (PDAC), p.42 available at http://www.fda.gov/ohrms/dockets/ac/06/briefing/2006-4272b1-01-FDA.pdf)

According to the internal GSK data made public in the Glenmullen's report, in clinical trials worldwide there were 5 suicides on paroxetine vs. 0 suicides on placebo. (posted at the US Senate Committee on Finance website at http://finance.senate.gov/press/Gpress/2008/prg061208a.pdf) The Sceptical Chymist (talk) 04:05, 21 February 2009 (UTC)

That is not true. The vast majority of clinical trials of paroxetine show statistically significant reduction in adult suicide. The same is true for teens, but that evidence has not been published. There is an ethical restriction against including suicidal teens in these studies. Think about it, would you give a suicidal teen a sugar pill and tell him "let me know how this works? Paroxetine is one of the top 50 most widely perscribed drugs.  I guess you know better than everyone else. —Preceding unsigned comment added by 24.15.179.168 (talk)

So, you are saying that both the FDA and GSK are lying? :) The Sceptical Chymist (talk) 00:20, 22 February 2009 (UTC)

In reply to the anonymous URL above ("Paroxetine is one of the top 50 most widely perscribed drugs."), the undeserved popularity of the drug is one of the reasons why it is so controversial. Vioxx was a huge seller despite causing serious cardiovascular problems. If you are going to claim that a "vast majority of clinical trials" show a particular outcome, please cite at least one; WP article contents are limited to published reliable sources. Likewise, I am restoring the linked quotes from the prescribing information (which Skinwalker had deleted as "original research and scare quotes"), because they are published (not original research) and reliably sourced. I try to assume good faith, but Skinwalker, please explain why you call published quotes from GSK's prescribing information "original research" and "trial attorney puffery," and why you think quotation marks are scary?TVC 15 (talk) 04:26, 22 February 2009 (UTC)

Paroxetine. Dangerous during pregnancy.
The theratogenic effect of some antidepressants is a relatively new finding. It appears in reviews since 2006. However, it is uncontroversial and well-established. The prevailing consensus is that paroxetine is likely to cause congenital malformations.

Review : "Some reports indicate that paroxetine is more commonly associated with neonatal withdrawal than other SSRIs. Recently, paroxetine was associated with a 1.82-fold (95% CI 1.17 to 2.82) increased risk of congenital malformations compared with other antidepressants...Paroxetine may cause adverse outcomes in the neonate when used during pregnancy and should be discontinued in women who are pregnant or trying to become pregnant."

Review : "Selective serotonin reuptake inhibitors (SSRIs) are not generally thought to be major teratogens. Some recent studies, however, have suggested that paroxetine may be associated with a small increase in risk of congenital abnormalities, particularly cardiac defects."

Review : "For paroxetine, recent data call for caution in prescribing such a drug in early pregnancy."

Review : "It is unlikely that any marked teratogenic effect occurs with the possible exception of an increased risk for cardiovascular defects after maternal use of clomipramine or paroxetine."

Review : "Based on the studies analyzed, first-trimester paroxetine exposure was associated with a significant increase in the risk for cardiac malformation (odds ratio [OR], 1.72; 95% CI, 1.22-2.42)...Based on the results of this metaanalysis, first-trimester exposure to paroxetine appears to be associated with a significant increase in the risk for cardiac malformation."

Review : "Recent data suggest that paroxetine treatment leads to increased rates of congenital malformations, although this evidence is not conclusive. Paroxetine and paroxetine CR are not indicated for use in the paediatric population and are categorised as Pregnancy Class D."

FDA/GSK prescribing information (http://us.gsk.com/products/assets/us_paxil.pdf): "Usage in Pregnancy: Teratogenic Effects: Epidemiological studies have shown that infants born to women who had first trimester paroxetine exposure had an increased risk of cardiovascular malformations, primarily ventricular and atrial septal defects (VSDs and ASDs). In general, septal defects range from those that are symptomatic and may require surgery to those that are asymptomatic and may resolve spontaneously. If a patient becomes pregnant while taking paroxetine, she should be advised of the potential harm to the fetus. Unless the benefits of paroxetine to the mother justify continuing treatment, consideration should be given to either discontinuing paroxetine therapy or switching to another antidepressant. For women who intend to become pregnant or are in their first trimester of pregnancy, paroxetine should only be initiated after consideration of the other available treatment options."

American College of Obstetricians and Gynecologists recommends that "paroxetine use among pregnant women or women planning to become pregnant be avoided, if possible."

The only review that denies the teratogenic potential of paroxetine is [], a non-systematic review, which did not conduct meta-analysis: "Given the inconsistency of the findings and limitations of the methodology of the published studies, the teratogenic potential of paroxetine that has been reported in some studies remains unproven."

The Sceptical Chymist (talk) 00:00, 19 March 2009 (UTC)


 * Thanks for review links, although "prevailing consensus is that paroxetine is likely to cause congenital malformations" is perhaps not quite what you meant as that implies (to me) a greater than 50% chance of problem. I agree the papers agree a relative increased risk, so perhaps you meant "prevailing consensus is that paroxetine is likely to cause an increased risk of congenital malformations" ? Risk of continuing treatment has to be weighed against risk of stopping and it is interesting that the papers have slightly different takes on this. I've done a little rephrasing, but I agree paroxetine would not be my ideal choice in pregnancy (but "Dangerous during pregnancy" implies risk to mother rather than a neonatal risk). David Ruben Talk 04:08, 19 March 2009 (UTC)


 * David, thank you for rephrasing! Indeed the paragraph now looks much better. And you are correct, I meant "increased risk of". The only thing I do not quite agree with you about is the Gentile review. You note in your comment that the review "states further studies need (ie observing outcomes in those so treated, not quite same as deliberately choosing to treat with this drug". In fact, the review advocates large controlled prospective studies of paroxetine in pregnancy, which means exactly deliberately choosing pregnant women to treat with this drug. Here is a quotation from the full text of the review conclusions: "Large, epidemiologic, prospective, controlled studies should be designed to include a control group of untreated women diagnosed with the same disorder as the mothers who accept taking paroxetine during pregnancy." Suggesting such a study is not only ethically questionable in the light of the numerous warnings, it is also disingenuous. Such a study is all but impossible to run because you will never be able to recruit a sufficient number of pregnant women to include into the paroxetine group. The women will read the consent form that lists the teratogenicity suspicions and the FDA recommendations to switch to another antidepressant—and they will choose to be in the control group.The Sceptical Chymist (talk) 09:58, 19 March 2009 (UTC)
 * Not being able see full article, I can't directly comment: but I wondered if a prospective study needs to involve starting someone de novo on a drug, as they might already be taking it and for a variety of reasons choose/unable to discontinue once they become pregnant, and only these select group of patients followed for outcomes. Hence akin to monitoring say smokers for effects on bladder cancer, which does not need any non-smoker to start and clearly will exclude those who sensibly discontinue. But perhaps I'm misundertanding prospective cohort study, or in this specific example of what the proposed study design was ? Anyway the intriguing thing, for me, is that my local psychiatric service seems to dislike use of citalopram in pregnancy vs fluoxetine, yet above papers suggest no/little cause concern between these two.David Ruben Talk 13:06, 19 March 2009 (UTC)
 * RE: controlled studies. You are correct, that is how it is usually done. Perhaps, that is what Gentile meant. What puts me off is the wording "mothers who accept taking paroxetine". Not "who accepted", or "who have been using". Regardless, such a study is not going to happen. How will they find enough pregnant women being treated with paroxetine? With paroxetine now generic, who is going to sponsor the study? With other, possibly safer, antidepressants, who needs to run a study with paroxetine? Gentile should know that, nevertheless he "suggests" further studies and that is disingenuous. The Sceptical Chymist (talk) 23:38, 19 March 2009 (UTC)
 * RE: citalopram vs. fluoxetine. There was a study that indicated that fluoxetine may result in a higher frequency of congenital heart malformations than even paroxetine. Then, there were other studies that did not find any signal for fluoxetine. I have not read anything bad about citalopram, though... This is just another example that prescribing habits are often not evidence-based. ;) The Sceptical Chymist (talk) 23:43, 19 March 2009 (UTC)

I've added a study with link that was supplied to me by Literaturegeek; I tried to cite using the tool that The Sceptical Chymist linked to, but was unable to get the tool to work. Also, I clarified the topic sentence of the section to reflect what the last linked source actually says, i.e. distinguishing between antidepressants generally and paroxetine specifically.TVC 15 (talk) 02:10, 8 April 2009 (UTC)

You put a tick/check mark in the fields "url" and "ref" and then copy and paste the PMID number into the field beside the drop down menu and then click on the submit button. Sometimes you need to manually copy and paste the full text URL if it is available into the URL field. This is the tool that I am talking about and use.-- Literature geek |  T@1k?  02:28, 8 April 2009 (UTC)

The PMID number is at the bottom of abstracts on pubmed and also in the URL of all the publications that they archive.-- Literature geek |  T@1k?  02:31, 8 April 2009 (UTC)

anxiety disorders vs. depression
I found one of the articles addressing the marketing history of paroxetine. It suggests paroxetine may be prescribed primarily for anxiety:


 * "In 1992, when Paxil hit the market, it faced a seemingly uphill battle to wrest customers from older SSRIs such as Prozac and Zoloft. The U.S. Food and Drug Administration had approved Paxil for the treatment of depression, like those drugs. But its manufacturer, SmithKline Beecham (because of a 2001 merger, now GlaxoSmithKline), was more interested in positioning Paxil as a remedy for anxiety disorders...
 * The company quickly secured permission to market Paxil for the treatment of panic disorder and obsessive-compulsive disorder. By mid-1995, Paxil had become the fastest-growing SSRI in the United States. In 1996, sales of the drug had climbed 54 percent to $291 million.
 * Impressive though those numbers may be, SmithKline was on the verge of a much bigger marketing coup: The company had been working to win approval from the U.S. Food and Drug Administration to market Paxil as a treatment for the first of a series of little-known mental health ills. In 1999, the FDA agreed to allow Paxil's prescription for the previously rare "social anxiety disorder."
 * SmithKline hired a New York public relations firm to raise awareness about the syndrome. According to the trade journal PR News, Cohn & Wolfe "developed a plan to educate reporters, consumers, and, in some cases, physicians, in an effort to encourage diagnosis and treatment."

The article continues:
 * SmithKline's so-called public-awareness campaigns were just one half of a coordinated strategy, though. In 1997, the FDA relaxed its rules on pharmaceutical advertising to let the pharmaceutical industry bypass healthcare providers to market its wares "direct-to-consumer" (DTC in marketing shorthand). In 1996, drug companies spent $595 million on advertising. Within a year, spending rose to $843 million. By 2000, the amount had shot up to nearly $2.5 billion.
 * Paxil was the first central-nervous-system drug to be advertised by name on television, according to Advertising Age. With such tag lines as "Your life is waiting" and "What if you were allergic to people?", the spots targeted 18-to-34-year-old professionals.
 * In the weeks following the attack on the World Trade Center, Glaxo positioned Paxil as the perfect antidote to post-9/11 anxiety. "Your worst fears," agonized one woman, seated at a kitchen table, "the what-ifs... I can't control it." "I'm always thinking something terrible is going to happen," another woman fretted. "It's like a tape in my mind," a third confessed. "It just goes over and over and over."

If paroxetine is prescribed primarily as an anxiolytic, then the depression studies seem less relevant. Mwalla and Skinwalker argued that depressed patients are at risk of suicide anyway, so suicidality among paroxetine patients might not be caused by the drug. However, anxiety patients are not usually considered a suicide risk, and suicidality is an extraordinarily serious side effect for an anxiolytic. Generic benzodiazepines can in rare instances cause death (e.g. Heath Ledger), but they may actually be safer (and less addictive) than the "not habit forming" paroxetine that made so much $$$ for GSK. Since the major (possibly only) selling point for paroxetine was being supposedly safer than older drugs, the previously concealed withdrawal syndrome seems one of the most important facts about paroxetine.TVC 15 (talk) 00:07, 23 March 2009 (UTC)
 * I'm afraid I don't understand your point. Are you trying to refocus this article towards discussing anti-anxiety applications of paroxetine?  Or are you just on your "Paxil is evil, GSK has infiltrated Wikipedia" soapbox again?  Skinwalker (talk) 00:17, 23 March 2009 (UTC)


 * Skinwalker, before you falsely accuse me of soapboxing, please explain why you call published quotes from GSK's prescribing information "original research" and "trial attorney puffery," and why you call quotation marks scary? My point in the comment above was as stated above, i.e. your defenses of paroxetine based on depression may be largely misplaced because paroxetine's market position seems primarily as an anxiolytic.  As for whether paroxetine is intrinsically evil, we have already discussed that above, so please re-read what I said there instead of falsely attributing the opposite to me.  As for whether GSK has infiltrated Wikipedia, I have said repeatedly that if you and the other pro-paroxetine users worked for GSK, you would respect its trademark instead of carelessly confusing generic paroxetine with brand name Paxil.  However, some on this page have expressly argued for a pro-paroxetine bias, in order to avoid 'scaring' depressed patients away from the drug.  As I have said all along, WP:NPOV calls for neutral presentation of facts (favorable and unfavorable), not deliberate bias in favor of someone's idea of the right medical decision for people they haven't even met.TVC 15 (talk) 00:55, 23 March 2009 (UTC)
 * To respond to your points one by one:
 * You'll need to provide diffs where I referred to such information as original research or puffery. I don't recall the context.  Furthermore, you have a habit of using scare quotes to denigrate material that you disagree with.  Please stop it.  It's not neutral, and it conflicts with the Manual of Style.


 * As far as the anioxlytic vs. antidepressant angle goes, it's nonsense. Our sources largely discuss the antidepressant applications of paroxetine, so that is what the article will focus on.  A non-scholarly article in an alternative weekly does not trump peer-reviewed sources.


 * (Ignoring the bit about paroxetine vs Paxil trademark issues, which is just way too offtopic).


 * As far as a pro-paroxetine bias goes, you're really off base here. The article currently says that paroxetine is not effective vs. placebo, causes a discontinuation syndrome, and may be a teratogen.  Do you know why I don't complain about that?  Because that's what peer-reviewed sources, written by medical professionals, say.  Further piling on with scare quotes, crap sources from alternative newspapers, and endless talkpage soapboxing is unhelpful at best, and disruptive at worst.  Skinwalker (talk) 01:21, 23 March 2009 (UTC)


 * Here are two of the diffs: and .  I used quotation marks to indicate that the phrases were quoted from the linked source.  Your replacement of common terms with the manufacturer's preferred terms is not neutral, so please stop that.  Encyclopedias (including especially WP) are written for a general audience, and WP:RS does not limit sources to manufacturer-sponsored studies.  As for your other accusations, including "crap sources," you need to tone down your language and provide some diffs.TVC 15 (talk) 02:10, 23 March 2009 (UTC)


 * Skinwalker and TVC 15, please, let's be civil! The truth lies, perhaps, somewhere in between your opposing points of view. Since both of your points are legitimate and fact-based, it should be possible to find a compromise, and/or include both of them into the article. The Sceptical Chymist (talk) 10:47, 23 March 2009 (UTC)


 * The article _did_ say that paroxetine is not better than placebo for moderate to severe major depression, but that fact was deleted by one of MWalla's sock puppets (see, ). That was one of the examples of disingenuous bias that I cited above.  Since the conclusion comes from a peer-reviewed medical journal, and Skinwalker thinks it's still there anyway, does anyone mind if I restore it?TVC 15 (talk) 07:17, 27 March 2009 (UTC)


 * No, it would be very confusing. The deleted quotation is formally correct ("Among adults with moderate to severe major depression in the clinical trials we reviewed, paroxetine was not superior to placebo in terms of overall treatment effectiveness and acceptability."). However, for some mysterious reason the authors decided "effectiveness and acceptability" to be equal to "the proportion of patients who left a study early for any reason". The conventional measure of treatment efficacy - the number of patients who improved is significantly higher for paroxetine - "Data extracted from 22 trials ... showed a statistically significant positive effect of paroxetine in terms of the proportion of patients who did not show an improvement of at least 50% on depression measures." To summarize: "In this systematic review of published and unpublished studies comparing paroxetine with placebo in adults with major depression, we found that the drug was not superior to placebo in terms of the proportion of patients who discontinued treatment for any reason. However, when we examined the results using specific measures of depression, we found that paroxetine was significantly superior to placebo." The Sceptical Chymist (talk) 22:40, 27 March 2009 (UTC)


 * Thanks for the explanation. Perhaps you might be willing to put a similarly clear summary into the article, to prevent the possible confusion that concerns you?  Also, if I may point out, comparing the number of people who _left_ the study early, even "for any reason," would still under-report the side effects if withdrawal symptoms don't show up until _after_ the study.  Also, the authors' reasoning does not look very mysterious to me.  When evaluating a central nervous system drug, where side effects may be perceived very differently by different patients, it makes sense to look at overall results rather than listed reasons.  Suicidal ideation may be under-reported, for example, as well as any other reason that might cause later embarrassment (e.g. male sexual side effects) or complicate future job applications (e.g. hallucinations).  Major media have reported episodes where "confidential" data leaked out somehow, e.g. .  If you plan to apply for a sensitive job, e.g. in the military, it probably seems safer to say family commitments require you to drop out of the study rather than risk someone finding out about your history of drug-induced hallucinations.  Ditto suicidal ideation for aspiring airline pilots.  Study participants are often students, for whom teenage social insecurities may be either a recent memory or a continuing fact of life, in addition to career anxiety.TVC 15 (talk) 00:16, 28 March 2009 (UTC)
 * I do not know about putting the additional explanations into the article. That paroxetine is efficacious for depression goes without saying. Withdrawal syndrome with paroxetine is bad and some other side effects are pronounced. At at the same time, other similar antidepressants with milder side effects are available. That suggests, IMHO, that paroxetine should have only a limited place in the treatment of depression and anxiety. However, judging by the number of prescriptions, there are thousands of doctors who prescribe paroxetine as a first line treatment. What can you do? The Sceptical Chymist (talk) 00:44, 28 March 2009 (UTC)


 * What we can do is, counter misinformation with accurate information that is reliably sourced. That, as I understand it, is why we volunteer to write for Wikipedia.  Widespread outdated sources including the "Complete Guide to Prescription Drugs" on my bookshelf echo GSK's false claim that paroxetine is "not habit forming," omit the black box warning, and omit the drug's proven effects in minors (i.e. all bad, no benefit, possible suicide).  I suspect many practitioners (perhaps Mwalla) are not as diligent as they should be in reading the prescribing information for the drugs they prescribe, which is why I added a link to the prescribing information and was disappointed to see Skinwalker delete it twice.( and .)  I agree with your opinion that the merits of the drug do not justify the sales, but I draw a different conclusion: large numbers of patients are suffering needlessly when better therapies (drug and non-drug) are available, and providing accurate information can help.TVC 15 (talk) 01:01, 28 March 2009 (UTC)
 * I edited the article again. Constructive comments are welcome.  Skinwalker (talk) 01:35, 28 March 2009 (UTC)

Your edits appear fine to me.-- Literature geek |  T@1k?  01:50, 28 March 2009 (UTC)

Well, changing quotes to take out the word "guilty" when that is what the reliable sources say is not fine as far as I am concerned. Also, although MWalla inserted sales into the lead paragraph to show the drug's notable popularity, sales appear to be dropping fast in the wake of all the controversy: down 20% from 2006 to 2007; so, I put that in its own section. I will review the other edits when time permits.TVC 15 (talk) 06:58, 28 March 2009 (UTC)

P.S. From OJ's WP bio, "A jury found Simpson guilty...." The bio uses the words "found...guilty," even though he might think "accused" sounds more neutral.TVC 15 (talk) 07:01, 28 March 2009 (UTC)


 * Skinwalker’s changes are mostly fine, except for removing "guilty" (see TVC's comment above). That is what the reference actually states. The Sceptical Chymist (talk) 12:24, 28 March 2009 (UTC)
 * I am not sure about the one-sided replacement of "withdrawal" with "discontinuation" throughout the article. That may not reflect the original references correctly, at least in some cases. "Withdrawal" and "discontinuation" are used in the literature interchangeably. For example, the title of the BMJ editorial quoted in the article is "Withdrawal from paroxetine can be severe, warns FDA". Even in the J.Clin.Psychiatry, 1997, 58 (supplement 7) ground-breaking Supplement on "discontinuation syndrome", which was the main force for the introduction of the "discontinuation syndrome" you can see "withdrawal" all the time. For example,


 * p.7 "Barr et al.17 studied the effects of a 7- to 14-day paroxetine taper in six patients. Three developed a withdrawal syndrome despite the slow taper. The authors noted that symptoms of withdrawal may occur despite progressive dose reduction of paroxetine."
 * p.7 "The authors26 of one report of paroxetine discontinuation symptoms in five young women in whom paroxetine was being tapered over a month noted that withdrawal symptomatology occurred despite conservative tapering and suggested that the paroxetine should be reduced by 5 mg/week—to below the minimum effective dose—to avoid a discontinuation syndrome. To reduce the risk of withdrawal symptoms, tapering of the shorter half-life SRIs fluvoxamine,paroxetine, and venlafaxine thus may have to continue for up to several weeks."
 * p.8 "Most of the symptoms associated with SRI withdrawal are physical rather than psychological."
 * p.9 "Fava and Grandi7 also reported that withdrawal syndromes tend to occur only after 3 to 4 months of paroxetine treatment."
 * p.9 "Withdrawal symptoms are rapidly reversed by the reintroduction of the original medication or one that is pharmacologically similar and can be minimized by slow tapering or by using a drug with an extended half-life."
 * p.14 "Of the 352 patients, 171 (48.6%) discontinued treatment under supervision, and at least one new symptom (dizziness, paresthesia, or, in one patient, nightmares) emerged in 21 patients despite slowly tapered withdrawal. In this analysis, no adverse events occurred after fluoxetine discontinuation. Dizziness and headaches were most frequently reported after paroxetine discontinuation."
 * p.17 "For example, Pacheco et al.9 described five young women who experienced vertigo, light-headedness, or gait instability during tapered withdrawal from paroxetine."
 * p.19. "Table 1. Studies of SSRI Discontinuation Reactions. Withdrawal Symptoms."
 * p.26 "Published data suggest that withdrawal symptoms are more likely to occur after paroxetine discontinuation."
 * p.28 The whole article is entitled "Physicians’ Knowledge of Antidepressant Withdrawal Effects: A Survey."
 * And about a hundred of other examples. The Sceptical Chymist (talk) 12:24, 28 March 2009 (UTC)

Discontinuation or withdrawal syndrome? - Both! - Straight from the horse's mouth
Now I remember where I read that! Straight from the horse's mouth, that is from the recent review by Haddad on the discontinuation/withdrawal syndrome : "The terms ‘antidepressant discontinuation symptom’ and ‘antidepressant withdrawal symptom’ are used interchangeably in the literature. ‘Discontinuation’ is preferred by some authorities, as it does not imply that antidepressants are addictive or cause a dependence syndrome, whereas the term ‘withdrawal’ may imply this. Both terms are likely to remain in use and it is more important to be clear about what they refer to rather than which is preferable." The Sceptical Chymist (talk) 11:32, 29 March 2009 (UTC)

Thanks for remembering that :) I've added some etymology to the history section of the SSRI discontinuation article; according to the World Health Organization and the New York Times, withdrawal was renamed "discontinuation syndrome" after a symposium sponsored by Eli Lilly and Company, maker of fluoxetine.TVC 15 (talk) 08:06, 31 March 2009 (UTC)

I downgraded the article to Class C
This article is seriously lacking in several areas, most notibly, an interactions section, a contraindications section. I recommend that these sections be added before putting the article back to a B class article. I also noted that mania and hypomania had been neglected from the side effect list. These are one of the most important side effects of antidepressants. I have added these side effects tonight.-- Literature geek |  T@1k?  22:21, 23 March 2009 (UTC)

I've restored information on preventing / managing withdrawal symptoms that had previously been deleted (by whom I dare not say, lest someone take offense). The link was graciously provided to me by Literaturegeek, and I think it helps the article a lot.TVC 15 (talk) 04:29, 27 March 2009 (UTC)
 * Please dare say. Stating a disagreement and then declining to identify the offending party is tendentious and passive-aggressive.  Skinwalker (talk) 01:38, 28 March 2009 (UTC)
 * Here you go again. Please, civility! The Sceptical Chymist (talk) 11:07, 28 March 2009 (UTC)

TVC 15 is referring to Mwalla who was blocked for 3 months for using sockpuppets and stalking or harrassing several members on their sockpuppets. See Suspected_Wikipedia_sockpuppets_of_Mwalla and Sockpuppet_investigations/Mwalla/Archive.-- Literature geek |  T@1k?  01:42, 28 March 2009 (UTC)

Actually, the information had been deleted by Skinwalker.TVC 15 (talk) 07:15, 28 March 2009 (UTC)

Well to be fair to Skinwalker, his deletion was justified. Wikipedia does not give medical advice, especially not uncited medical advice. If data like that is readded it will need to be cited and it would need to be reworded in a factual tone rather than instructional tone.-- Literature geek |  T@1k?  11:53, 28 March 2009 (UTC)

Umm, I cited and quoted the source that you gave me. Skinwalker then "chopped" the quote.TVC 15 (talk) 22:16, 28 March 2009 (UTC)

To address the remaining reasons why the article was downgraded to class C, I have added sections on contraindications and interactions. Combined with the improvements made by Literaturegeek and The Sceptical Chymist, I think the article has recovered from Mwalla and is actually better now. Is it ready to restore/upgrade?TVC 15 (talk) 09:14, 31 March 2009 (UTC)


 * Changed the rating to B. The article meets the formal definition for B and is close to an average pharmacology B-article. The difference between B and C is subjective to a significant degree, anyway. But please use this tool when adding references. Thank you. The Sceptical Chymist (talk) 10:39, 31 March 2009 (UTC)

Thanks Sceptical Chymist :) Looking ahead to possible future improvements and upgrades, I think this article (and other pharmacology articles) would benefit from more direct comparisons of the number of patients who experience efficacy.  You wrote above that paroxetine efficacy for depression "goes without saying," but the data seem to show efficacy in only around 20% of patients.  Most study subjects seem to get no benefit.  For example, a study on 'late-life depression' reported the following:


 * Response, defined as a score of 1 or 2 on the Clinical Global Impressions-global improvement scale, was achieved by 72% of paroxetine CR patients (LOCF; p < .002 vs. placebo), 65% of paroxetine IR [Immediate Release] patients (p = .06 vs. placebo), and 52% of placebo patients. Remission, defined as a HAM-D total score < or = 7, was achieved by 43% of paroxetine CR patients (LOCF; p = .009 vs. placebo), 44% of paroxetine IR patients (p = .01 vs. placebo), and 26% of placebo patients... CONCLUSION: Paroxetine CR and paroxetine IR are effective and well tolerated treatments for major depressive disorder in elderly patients, including those with chronic depression.

The response difference between paroxetine IR (the originally approved version) and placebo is only 13%; the new patented Paxil CR is only slightly better, 20%. The remission difference for both versions is less than 20%. In other words, four out of five patients taking the drug did not benefit from it. Yet, the conclusion seems nearly as enthusiastic as the TV ads. Similarly, a study on Generalized Anxiety Disorder reported:


 * reductions in total score on the Hamilton anxiety scale were significantly greater for both paroxetine groups. Response was achieved by 62% and 68% of the patients receiving 20 and 40 mg of paroxetine, respectively, compared with a 46% response rate in the placebo group. Remission was achieved by 30% and 36% of patients in the 20- and 40-mg paroxetine groups, respectively, compared with 20% given placebo... CONCLUSIONS: This study demonstrates that paroxetine is an efficacious and well-tolerated treatment for generalized anxiety disorder.

Again, only around 20% experienced "response," and even fewer experienced remission. The remaining 80% did not respond favorably to the drug. While I recognize even 20% response rates provide some support for the studies' conclusions, I think the WP article would benefit from comparisons to alternatives like increased exercise. For example, "A 2001 study by the Duke University in North Carolina found that exercise is a more effective treatment for depression than antidepressants, with fewer relapses and a higher recovery rate." In the USA, the TV advertisements exhort the entire audience to "ask your doctor" about the advertiser's pill, and doctors complain they are besieged by patients asking about pills they've seen on TV; the #1 complaint of general practitioners is patients lying to get prescriptions for drugs they don't need. (An almost equally widespread complaint is struggling with insurance companies to get payment for things the patients actually do need.) While I understand that some people can't go out for a daily walk, and that even some physically fit patients may still benefit from medication, I think a general-audience encyclopedia like WP should mention reliably sourced data comparing the subject drug to non-drug alternatives.TVC 15 (talk) 22:13, 31 March 2009 (UTC)


 * The results of single studies may vary widely. That is why we have to look at the results of meta-analyses, which reliably show that antidepressants perform better than placebo. However, it is widely acknowledged that effectiveness of all antidepressants is weak to moderate and that companies relentlessly spin the positive results (for more information see this open-access article Turner EH, Matthews AM, Linardatos E, Tell RA, Rosenthal R (January 2008). "Selective publication of antidepressant trials and its influence on apparent efficacy". N. Engl. J. Med. 358 (3): 252–60. doi:10.1056/NEJMsa065779. ). Nevertheless, in real-life situations, antidepressants are highly effective because the medication effect and placebo effect add up to 60-70% response. Psychotherapy has similar effectiveness (see here Treatment_for_depression). Thus, the comparison between treatments is based on side effects (favors psychotherapy) and cost-effectiveness (favors antidepressants). The direct advertisement to consumers is allowed only in Australia and the US, and it is not a problem worldwide. One can hope that the US Congress/Obama administration eventually get to that issue as a part of the health care reform. The Sceptical Chymist (talk) 10:55, 1 April 2009 (UTC)


 * Thanks for the reply and links :) Regarding direct to consumer (DTC) advertising, alas a side effect of the WWW may be to spread drug ads worldwide.  The prognosis for health care reform is difficult because once a revenue-driven system gets big enough to entrench powerful political lobbies, it metastasizes and spreads.  The U.S. does seem to have the most, where a symbiotic relationship has developed between the evening news ("if it bleeds, it leads;" "world to end at ten, details at 11") and the advertisers (anxiolytics, antidepressants, antihypertensives, antacids, headache relief, etc.), but others might catch up.  Most users of the English-language Wikipedia are either already within reach of DTC drug ads or soon may be, and may check here for more balanced information.  It seems notable that at least 80% would be better off going outside for a daily walk.  They won't hear that from drug companies, or any other revenue-driven enterprise, because there is no revenue involved.  Also I wonder if the improvement in the placebo group might simply reflect the inherently transient nature of most people's mental state.  For example, a common feature of both depression and happiness is the feeling that life has always been this way and always will be, just as someone traveling over hills and valleys sees only hilltops when on hilltops, and sees only valleys when in valleys.  Depression and anxiety study subjects may thus contain a high proportion of people whose condition would improve even with no therapy, and it seems inaccurate to count the placebo effect among the benefits of treatment.TVC 15 (talk) 22:44, 1 April 2009 (UTC)

Pregnancy: I Don't want an edit war!!
First of all, Ddave2425, current recommendations from both the FDA and GSK indicate that paroxetine can cause congenital defects during the first-trimester. Second, I'm not sure why Literaturegeek reverted all the way back to CliffC's edit since you sent me a message earlier commending those edits to this page!! I can only surmise it was because my name is in red and you assumed it was a vandalism without thinking or reading. My edit not only slightly reworded the issue for clarity, but also added reliable citations. Please indicate if there was any other reason for reverting my edits as well. DKqwerty (talk) 15:28, 4 May 2009 (UTC)

Sorry about that. I just reverted to the version before Ddave made any edits and unfortunately reverted one of your edits which was inbetween. It had nothing to do with your name being red. I assumed that your edits were just undoing Ddaves edits and didn't realise that I was actually losing good edits.-- Literature geek |  T@1k?  16:31, 4 May 2009 (UTC)

The FDA does not ban the use of paroxetine for pregnant or nursing mothers. Ddave2425 (talk) 21:13, 9 May 2009 (UTC)ddave2425
 * The article does not say it is "banned", only that they "should avoid using paroxetine" and "[consider]...either discontinuing paroxetine therapy or switching to another antidepressant". There is nothing stating it is forbidden. DKqwerty (talk) 21:50, 9 May 2009 (UTC)

I agree with qwerty.-- Literature geek |  T@1k?  10:45, 11 May 2009 (UTC)

That is not what the FDA reference says. Besides, the FDA advisory is from 2005 and based on unpublished studies. I included a published study from 2009. Ddave2425 (talk) 12:18, 16 May 2009 (UTC)ddave2425

It is current FDA guidance. The article does not rely soley on the current FDA guidance but also on citations to review articles which are secondary sources. Please read this page WP:MEDRS to understand why primary sources are of poorer quality than secondary sources. As FDA issues national guidelines a single small primary study is not valid enough to justify deletion of FDA guidance. This is not be being awkward but is the way wikipedia medicine related articles work. If you disagree with this policy you can challenge the policy on the talk page of this WP:MEDRS article.-- Literature geek |  T@1k?  14:47, 16 May 2009 (UTC)

What is "it"? Can you give a source before reverting a good faith edit? Ddave2425 (talk) 01:25, 17 May 2009 (UTC)ddave2425


 * "It" is the information you insist on removing despite both Literature geek 's and my advisement that a singular, isolated study does not constitute medical consensus. And the sources you're pretending don't exist are the "[10][11][12]" after the text you are continually removing. I seriously question your objectivity. Please, explain why you are so insistent that both the FDA and the manufacturer are incorrect in their assessment of the pregnancy risks and why you feel that your source is valid enough to supersede Wikipedia's criteria for reliable sources in medicine-related articles. Otherwise, we will continue to revert your edits in observance of these policies and you will most likely ultimately be blocked. DKqwerty (talk)  —Preceding undated comment added 01:53, 17 May 2009 (UTC).


 * Wikipedia is not a place for original research. My edits added neutrality.  You are misinterpretting the FDA.  You should quote it directly instead of spinning it. Ddave2425 (talk) 13:34, 13 June 2009 (UTC)ddave2425

I have explained that your source is a primary source and worse still discussing antidepressants in general. You are using it to delete FDA guidance. I gave you a link to WP:MEDRS. You aren't a banned user by the way? You are mwalla, great, you're back to sockpuppets again. How many sock puppets is this now? 11 or 12 are we up to now? Should have figured it out earlier, I did wonder how a brand new user knows about the admin notice board and knows how to use lingo like "good faith" etc to manipulate conversations. I hope I don't have to have another check user done again to verify this allegation.-- Literature geek |  T@1k?  01:47, 17 May 2009 (UTC)

Yup and it is you following my contribs around and you coincidently showed up immediately below my discussions regarding scurio on admin notice board.-- Literature geek |  T@1k?  01:57, 17 May 2009 (UTC)

DKqwerty, I would not engage anymore in any debates regarding sources. The user you are talking to is a vandal and sock puppeteer who is meant to be banned for 3 months. I am 99% certain of this. Please see sock puppet investigation for more information. Sockpuppet_investigations/Mwalla/Archive Just revert their vandalism. They are an experienced editor and playing games.-- Literature geek |  T@1k?  02:00, 17 May 2009 (UTC)


 * Duly noted. DKqwerty (talk) 02:16, 17 May 2009 (UTC)


 * I bounced here from AN/I. I'm not interested in addressing the socking, but the content. I note that back in April, Mwalla provided, in edits, multiple gov't studies demonstrating no ill effects, and now provides an NIH citation. Wouldn't it be the most responsible thing to report that different parts of the government have, in fact, provided conflicting reports about the safety of the drug, presenting both 'pro' and 'con' sections? from the size of the material, it looks like it should be structured Con/pro in this case, but we can neutrally report that different governing bodies say differing things. ThuranX (talk) 17:24, 13 June 2009 (UTC)

I don't know what gov ref he added in April. If a recent gov report exists contradicting the FDA findings by all means add it. I would Thurax avoid jumping to conclusions. If you read the sockpuppet investigation one of the problems raised was that Mwalla faked and made references say the opposite of what they said etc and then created arguments using sockpuppets to fight over his faked data basically to be disruptive etc. I honestly do not care, like really really do not care about what the article says about paroxetine in pregnancy. It is a matter of not allowing a sockpuppet to intentionally delete referenced data distort info and create arguments to get revenge for the sockpuppet investigation which I was involved in which got him banned. I would urge you not to feed the troll.-- Literature geek |  T@1k?  17:47, 13 June 2009 (UTC)

Also pubmed indexes every medical article ever published. So an medical abstract archived on NIH pubmed does not make it a differing "gov opinion". Please also read WP:MEDRS regarding using priumary sources to delete secondary sources.-- Literature geek |  T@1k?  18:03, 13 June 2009 (UTC)


 * Mm. Hostile, defensive, and offensive, all in one. link since you claim there was none. the NIH, as well as medical journals, are all WP:RS. I'm more disturbed by the amount of 'GlaxoSmithKline says XYZ' stuff in the version being defended. ThuranX (talk) 18:07, 13 June 2009 (UTC)

I appologise if it comes across that way but you accused me on ANI of persecutiing someone who is meant to be banned who are using sockpuppets to follow me around on several articles and cause arguments. It is immensely frustrating when a sockpuppet intentionally files a bogus reports and then to see the response belittling my response telling the truth and admins defending the banned sockpuppeteer. I shall request a checkuser to resolve this. I do not believe admin noticeboard is the right place to deal with sockpuppets.-- Literature geek |  T@1k?  18:15, 13 June 2009 (UTC)

I have done very little developing to this article. So please don't hold me accountable for the content of this article. This article only ended up on my watch list after I followed a request for opinions on the wiki medicine project. I have no big interest in the article. My problem is a sockpuppet causing me grief who is ban evading. thank you for listening.-- Literature geek |  T@1k?  18:15, 13 June 2009 (UTC)

I don't understand what you mean by showing the diff you gave as those refs that Mwalla added to the article are still in the article in pregnancy section. Like I say the issue is a ban evading user edit warring using sockpuppets. This convo should end as it is not related to the article but should be taken up at a checkuser request which I intend to do when I have the time. Thank you.-- Literature geek |  T@1k?  18:23, 13 June 2009 (UTC)


 * Dear Thuran. The consensus views on paroxetine in pregnancy are collected in one of the Talk topics above, see [] for your convenience. Unfortunately, one of the problems with Mwalla was that he would provide random links to support his views. Those links would look legitimate until you started checking them. The Sceptical Chymist (talk) 11:17, 14 June 2009 (UTC)


 * I second that observation, and am writing to express my continued appreciation for the contributions of Literature geek  and The Sceptical Chymist.  Thuran, mwalla's indefensible use of sockpuppets resulted in a 3-month block (due to expire soon).  To understand the history you might look at these diffs:, , , .  If you are unfamiliar with trolls, you might read  and.

As for the specific issue of paroxetine vs. pregnancy, risks include congenital defects and withdrawal syndrome (aka "discontinuation effects"). Reported discontinuation effects include miscarriage and neonatal convulsions, as noted in the article. Note that discontinuation may in some instances be involuntary, for example if paroxetine is first prescribed during pregnancy and produces intolerable side effects, or if the mother loses access to the drug. (In the U.S., the drug requires a prescription typically every 90 days, which can become prohibitively expensive if, for example, the customer had been getting insurance through an employer and then loses coverage.) One reason why reliable sources say that pregnant women, or women who may become pregnant, should avoid paroxetine is because commencing a habit-forming teratogen is an especially bad idea for them, especially when alternative therapies are available. (Alternatives include other SSRIs, other drugs, non-drug therapy, and the simplest and cheapest of all: increased physical exercise, which reportedly is more effective and doesn't cost anything.) I suspect another reason may be the extraordinary history of dishonesty promoting paroxetine; when a product is promoted so dishonestly, so widely and for so long, common sense suggests skepticism.TVC 15 (talk) 19:00, 14 June 2009 (UTC)


 * Got it. Get the fuck out, we've owned this page. I see perfectly good references being removed for saying a medicine is safe, not secretly deadly and being marketed by the evil corporations, and any challenge to that POV is unwelcome. I'm gone. ThuranX (talk) 23:29, 14 June 2009 (UTC)

What a rude dick ThuranX is. TVC I wouldn't take any offense. ThuranX knows he is wrong I feel but is one of these people that can't just say hey I made a mistake and has to get nasty etc. He is pissed because he was asked to apologise to me for calling me stupid on the admin noticeboard, which he wouldn't do and that he was then proven wrong about Mwalla. He is also aware of the sockpuppet investigation into Mwalla where multiple admins and checkusers agreed Mwalla was a disruptive sockpuppet and banned him for 3 months but I guess they are all wrong. The so called claim of well referenced stuff in the link being deleted that ThuranX gave above by Mwalla actually is not mwalla adding refs as all but one ref was already added. It was Mwalla deleting every single ref he didn't like LOL. Oh well I guess ThuranX is another victim of Mwalla's sockpuppet games.-- Literature geek |  T@1k?  00:11, 15 June 2009 (UTC)

The perfectly good references claim ThuranX makes is also bogus as I explained WP:MEDRS and gave him a link to it on admin noticeboard. He is aware that wiki guidelines say that primary sources should not be used to delete or debunk secondary sources and is aware the primary source itself was irrelevant to this article, (perhaps relevant to other antidepressant articles) as it only had a small number of paroxetine users and was on antidepressants in general.-- Literature geek |  T@1k?  00:32, 15 June 2009 (UTC)


 * Listen up, if you want to engage in personal attacks, I can play that game better than you ever could. I happen to have read through his citations, and they look like they provide a different viewpoint which, per WP:NPOV ought to be represented in the article. Had dave not reported on AN/I the insult you'd made, I wouldn't be back here at all. But I'm back this time, and hopefully not again, to state plainly :The sources he provides appear to me to represent a significant alternate opinion which needs Due Weight representation in this article, as I regard the NIH, and medical journals, to be significant reliable sources for Wikipedia articles. As I stated before, I do also find the amount of OWN going on to be onerous, and I'm not going to argue here anymore, because the reply to my previous comment about your OWN was to insult me for a paragraph. ThuranX (talk) 03:19, 15 June 2009 (UTC)


 * Okay, here's an article that states the opposite. Who is right? Given that they are both primary sources, it is appropriate to wait until a secondary source has weighed in on the research he keeps citing per WP:MEDRS. Ddave2425 continues to treat this primary source study as gospel, so far as to state, "Paroxetine is not associated with birth defects nor harmful effects in nursing newborns," which is flat out incorrect and misleading as it ignores long-standing associations to the contrary. He also misrepresents the FDA's position on prescribing Paroxetine to make it seem that the FDA doesn't regard it as a risk to the fetus, which it clearly does in the given citations, which he also removes. If he were to find reliable (secondary) sources, we'd welcome his adding this information to the article, but not substituting it for cited, reliable information. DKqwerty (talk) 03:38, 15 June 2009 (UTC)

That ref that you gave is a meta-analysis and a review article so it is a secondary source. In the case of Mwalla on his sockpuppets I suggest whenever possible not engaging in debating his sources and arguing over them as it feeds the troll and encourages them to create disruption which is why they do this. They are now a permanently banned abusive sockpuppeteer who has no right to edit/vandalise wikipedia. Next time they are suspected of editing/vandalising wikipedia they should be reported with a view to banning their sockpuppet.-- Literature geek |  T@1k?  11:37, 15 June 2009 (UTC)

Ddave2425 has been confirmed as a sockpuppet of Mwalla, and both have been blocked indefinitely.,,TVC 15 (talk) 08:37, 15 June 2009 (UTC)

You are more than welcome ThuranX to follow wikipedia medicine related article guidelines and find a good quality secondary source to dispute the FDA findings. If you feel that reliable sources for medicine articles guideline WP:MEDRS is inaccurate you are also more than welcome to go to the talk page there and challenge the guidelines. There is no ownership issues. Cite a recent secondary source and you will see that it won't get rreverted not by me anyway. Mwalla has now been blocked indefinitely as well as Ddave. You are welcome to engage in the personal attack game if you wish, you seem to enjoy it. You seem to think you have a right to insult people become aggressive with people but if they respond similarly in return you think their behaviour is out of order. Ddave and Mwalla have now been permanently banned from wikipedia so I think this issue is resolved. I would suggest ThuranX that you give this issue a rest and quit taking things so seriously. You do not even know me and you do not even have an interest in this article.-- Literature geek |  T@1k?  11:30, 15 June 2009 (UTC)

Where the positive spin on Paroxetine for pregnant women comes from
From the Wall Street Journal

More details are coming out about the relationship between Emory University psychiatrist Zachary Stowe and GlaxoSmithKline, which made payments to Stowe at the same time he was conducting federal research about the use of antidepressants, such as Glaxo’s Paxil, in pregnant women.

Emory has reprimanded Stowe, who was instructed to immediately eliminate conflicts related to current federal grants. In a statement, the school said Stowe had informed it of “previously unreported activities and has disclosed his failure to abide by Emory policies.” Stowe, through the university, declined an interview request. Here’s more on the story.

In a letter this month to Emory, Sen. Charles Grassley said records he obtained from Glaxo indicated Stowe was paid $154,400 by the drug company in 2007 and $99,300 during the first 10 months of 2008. Stowe is listed as the primary investigator on at least three National Institutes of Health grants, beginning in 2003 and continuing through last year, that involve antidepressant use in pregnant women and the effects on children delivered by those women.

The study, conducted by Stowe, found Paxil is safe for breast-feeding mothers. The PR firm’s email to Glaxo reads: "Please review the attached press release and forward me any comments/edits. As you may know, Dr. Stowe is on board for publicity efforts and Sherri and I are coordinating time to meet with him next week to arm him with key messages for this announcement, which is slated for early February. We are sending the release for his review at the same time in efforts to secure distribution on Emory letterhead (as you know, would provide further credibility to data for the media)." The Sceptical Chymist (talk) 18:57, 20 June 2009 (UTC)


 * So Ddave2425 == Zachary Stowe?  ;) DKqwerty (talk) 19:24, 20 June 2009 (UTC)


 * No, Ddave2425 = Fndlytrucker talk ;(( The Sceptical Chymist (talk) 10:13, 24 June 2009 (UTC)

NPOV Noticeboard
I've created a summary of my concerns about this article at the NPOV Noticeboard here. I welcome outside opinions, comments, and ideas. blurredpeace ☮ 04:10, 24 June 2009 (UTC)

I have taken a break from editing the usual articles that I edit and decided to address these concerns that you have raised on my talk page regarding violations of NPOV on these two articles. I have started adding in some of the benefits of paroxetine as I felt it focuses too much on the negatives. As life is rather unpredictable I might end up giving up on doing this and revert back to my usual editing arena or I might remain active on the antidepressant articles bringing balance to them, time shall tell. Despite some side effects when on them or coming off of them they are very effective drugs for many people, even in the long term! The benefits outweigh the risks I feel for most people! Check out my recent edits to this article and let me know what you think. More work to be done? :)-- Literature geek |  T@1k?  00:44, 28 June 2009 (UTC)

I was away and missed that discussion, but I would like to read it. Is it archived somewhere? Meanwhile, I did find some others' edits overly enthusiastic about the drug. For example, singling out paroxetine as "highly effective" for severe PMS when in fact the linked study was of several SSRIs, and omitting paroxetine's unique teratogenicity. Likewise calling paroxetine a 'first-line treatment' for GAD seems biased when the linked source merely included paroxetine among other (possibly better) options like escitalopram and the WP article on GAD says CBT is more effective anyway. I assume good faith, but as the health care reform debate in the US heats up there seems to be a big push for PhRMA and the AMA (and the insurance and hospital corporations) at the expense of other priorities like education, exercise, nutrition, and basically everything else.TVC 15 (talk) 22:07, 15 August 2009 (UTC)
 * Yeah, well, you don't assume good faith. You assume that those who edit differently from you are shills for PhRMA/AMA/etc, even though the article has not been significantly edited during the health care debate.  We can work together on improving the article, but you've got to stop making these sort of commments.  Skinwalker (talk) 16:33, 16 August 2009 (UTC)


 * I look forward to working together to improve the article, but you've got to stop deleting reliably sourced information as you did today. I do try to assume good faith, but this article has drawn blatant misconduct including the now-blocked mwalla.  Also you make it difficult to assume good faith when, for example, you delete references to the manufacturer's own prescribing information as "trial attorney puffery;" I asked you to explain that, but you never do.  The recent edits are significant and are date-stamped, so please explain why you deny they are occurring "during the health care debate?"  There was consensus on retaining most of what you removed today, so I will restore those parts.  You used to make edits like that without even discussing them, until I drew you into the discussion page, so at least this is progress.TVC 15 (talk) 23:38, 16 August 2009 (UTC)


 * Let's start with something simple. The word "falsely" in the first sentence of the controversy section communicates an overt value judgement that is not present in the sources cited.  It therefore violates NPOV.  I've tagged the section accordingly.  Skinwalker (talk) 23:00, 18 August 2009 (UTC)


 * In this article, the word "falsely" communicates a clearly proven fact; GSK was ordered to stop advertising the drug as "not habit forming" because that statement was false. Therefore, it does not violate NPOV.  Many people do attach a negative value judgment to false advertising, especially when the false ads (1) are directly opposite to the facts and (2) continue until ordered to stop by courts and regulators.  Since that is what happened, and on an obviously notable scale, suppressing it or sanitizing it would not be NPOV.TVC 15 (talk) 23:57, 18 August 2009 (UTC)
 * No, it is not a "clearly proven fact". It is your opinion that you've formed from reading the two sources cited.  You did read the sources?  Neither use the word "false".  Look, it's not my responsibility to teach you the difference between an opinion and a fact.  The first source describes a warning about the addictive potential of seroxat.  The second describes a judicial ruling enjoining GSK from describing it as non-habit forming.  Why can't you just summarize what the sources say instead of bashing the reader over the head with an obvious value judgement that does not describe the specifics?  Skinwalker (talk) 00:09, 19 August 2009 (UTC)
 * Also, you've neglected to mention that the injunction preventing GSK from describing the drug as not habit forming was later lifted. This sort of cherry-picking of sources is endemic to this article.  Skinwalker (talk) 00:16, 19 August 2009 (UTC)
 * In fact, the judge reversed herself one week after ruling against GSK. This was in 2002.  Why doesn't the article say this?  A simple google news search must have turned this up along with the original injunction.  Why are you cherry-picking sources like this?  Skinwalker (talk) 01:34, 19 August 2009 (UTC)
 * In answer to your first question, yes I did read the sources. In answer to your second question, "stated falsely" summarizes accurately what the sources say.  Regarding the injunction in that specific case, (1) the subsequent technical ruling was based on federal pre-emption (a doctrine the US Supreme Court has since revisited in a different drug case); (2) the case settled on confidential terms; (3) publicly, the company stopped the ads and acknowledged the withdrawal symptoms (which it called "serious discontinuation symptoms").  As for the factual statement that the ads were false, there are many more sources (e.g., ).  In light of your edit history on this article, it is surprising that you try to invoke NPOV, since you previously removed citations to reliable sources (including even the prescribing information) and insisted on using the manufacturer's preferred term "discontinuation syndrome" and deleting every occurrence of the more commonly understood "withdrawal syndrome."  Lastly, there seems to be a pattern here: I make a comment that you don't like in the discussion section, and you retaliate against the article.  The section you're attacking currently had been stable for months, and was the consensus version after editing among several different editors; even you had accepted it, until I made a comment here about the ongoing debate about healthcare reform.  Frankly, your own comments are not particularly civil, but I don't respond by retaliating against the article.  Please try to refrain from passive-aggressive editing, and to improve your tone if you can.TVC 15 (talk) 02:04, 19 August 2009 (UTC)
 * You didn't answer my question about cherry-picking at all. Social Audit (a non-notable NGO) and citypages (an alternative weekly) are exceedingly poor sources for medical topics.  Look, it's clear that you are approaching this article as an activist and not as neutral editor.  I've tagged the entire article for an NPOV dispute, and I will seek outside intervention since it is clear that you are not open to collaborative editing.  Skinwalker (talk) 10:37, 19 August 2009 (UTC)


 * Since this topic causes such a heated discussion, the best is to cling to the sources as close as possible and attribute value judgments to a certain source. For example, "according to independent watchdog organization Social Audit, GSK knowingly mislead ...". "The XYZ court injunction noted ... This injunction was later lifted, and as a result of confidential settlement GSK stopped... and ..." Just my 2p The Sceptical Chymist (talk) 10:39, 19 August 2009 (UTC)
 * That's what I've been advocating. If we strictly report what reliable sources say, there is no need to add commentary about "falsely" or whatever.  TVC is insistent on presenting this material in the most negative possible light, and as I discovered above he is cherry-picking sources.  I intend to seek assistance at Wikiproject Medicine.  Skinwalker (talk) 10:47, 19 August 2009 (UTC)


 * I hope TVC will fix that (hint, hint). The evidence speaks for itself. Additional judgmental words like "falsely" could, actually, make reader doubt the overall objectivity of the article.
 * On the other hand, Skinwalker, it is unfair to tag the whole article only because you disagree with TVC about one part. You did not offer any concrete evidence as to why and how the rest of the article is skewed. You did not try to add any well regarded independent studies about paroxetine that would be more objective to it (which is what you should have done first). Overall, you decision to tag the whole article was not very well thought out. Please reconsider it. The Sceptical Chymist (talk) 10:54, 19 August 2009 (UTC)
 * I removed the tag over the article, but I have started a thread on WP:WQA about the comment below. Skinwalker (talk) 23:17, 19 August 2009 (UTC)
 * [outdent]Acting on Skeptical Chymist's hint, I've edited the relevant sentence and added two sources. Skinwalker, WP's sourcing standards can be found in WP:RS and you cannot limit the article to industry-sponsored publications.  Beth Hawkins is an award-winning investigative journalist and her article has stood the test of time.  In contrast, medical journal articles sometimes turn out to be ghost-written by PhRMA marketing,, and sometimes whole journals turn out to be advertisements in disguise, and of course there is the Emory story discussed above, where the underlying studies were conducted by a researcher who concealed drug-company payments.  Your MSc marks the beginning of an education, but not the end; if you insist on blinkering yourself and disregarding critical sources, then you deprive yourself of the faculty of reason, critical thinking, and comparative analysis.  WP is not so limited, and should not be.  Lastly, I did answer your accusation about cherry-picking, but perhaps I put it too politely so you didn't notice.  As I stated above, the subsequent reversal was related to federal preemption, i.e. the question of who has authority to decide a particular question.  The court did not decide that GSK's ads were true, but rather the court accepted FDA's argument that it was for FDA to decide.,  GSK then prudently settled.  In a subsequent case, Wyeth made a similar argument, and the Supreme Court ruled against them.  (That case involved a violinist who lost a hand and forearm due to an undisclosed side effect of a Wyeth product)  Your endless accusations (calling me an activist because I disagree with you, which is what Sarah Palin calls judges she disagrees with; accusing me of cherry-picking instead of reading why a decision was reversed) are becoming excessive.  Also, your refusal to explain your own behavior casts doubt on your motives, for example deleting references to the manufacturer's FDA-mandated prescribing information and calling it "trial attorney puffery.",  You wrote about your bad experience with venlafaxine, and I hope you've found something better since, or that you will soon.TVC 15 (talk) 19:28, 19 August 2009 (UTC)

Regarding the word "guilty," which Skinwalker and CliffC removed because IFPMA is not a court, please consult a dictionary, e.g.. The source uses the word guilty. Courts do not have a monopoly on that word; it is frequently used in academic and self-regulatory contexts, of which IFPMA is one. We can put "guilty" in quotes if you like, since it is a quote, but softening it into something GSK might prefer is not NPOV.TVC 15 (talk) 20:23, 19 August 2009 (UTC)

OK, sticking to third-person and content, here is what the BMJ reported in 2002: "The Food and Drug Administration in the United States published a new product warning about the drug, and in the same week the International Federation of Pharmaceutical Manufacturers Associations declared the company guilty of misleading the public about paroxetine on US television [and] announced that GlaxoSmithKline had breached two of the industry's codes of practice." The word "guilty" is frequently used in contexts other than courts, for example academic and self-regulatory organizations. According to its primary dictionary definition, the word means "responsible for a crime, wrong action, or error and deserving punishment, blame, or criticism." Deleting the word "guilty" is not neutral, in fact it leads to understatement, because it subtracts the element of DESERVING punishment, blame, or criticism. In line with the suggestion to use quotes, the quote would be more neutral.TVC 15 (talk) 01:44, 24 August 2009 (UTC)
 * Also sticking with third-person etc, at least two editors (myself and CliffC) disagree with this formulation. I'd be more comfortable with describing that the IFPMA announced that GSK breached two codes, as that is specific, it does not convey a value judgement, and does not give the impression that IFMPA is a judicial authority.  You're welcome to seek outside opinions at WP:3O or WP:RFC.  Skinwalker (talk) 21:50, 24 August 2009 (UTC)
 * Either of those might become necessary, although first it seems prudent to clarify the issues here if possible. It has been suggested that "If we strictly report what reliable sources say, there is no need to add commentary...."  The article section used to say "guilty," because that is what the source says.  Now, the word "guilty" has been replaced with commentary somewhat softening the controversy regarding GSK's marketing, and a NPOV tag has been added.  There also seems to be, in the comment above about the impression that the word "guilty" creates, a rejection of the dictionary definition of the word "guilty."  Is that a reasonable description of the disagreement?TVC 15 (talk) 06:22, 27 August 2009 (UTC)

Back on 11 July, the sentence read
 * "In 2002, the Food and Drug Administration of the United States published a new product warning about the drug, and the International Federation of Pharmaceutical Manufacturers Associations found GSK guilty of misleading the public about paroxetine and breaching two of the Federation's codes of practice. "

I'd be agreeable to couching the "guilty" within a quote, something like
 * "In 2002, the Food and Drug Administration of the United States published a new product warning about the drug, and the International Federation of Pharmaceutical Manufacturers Associations declared [not 'found'] GSK "guilty of misleading the public about paroxetine" and breaching two of the Federation's codes of practice. "

--CliffC (talk) 21:22, 27 August 2009 (UTC)


 * That would be good.TVC 15 (talk) 22:28, 27 August 2009 (UTC)


 * I am not sure that the word "guilty" used by BMJ writer is warranted. Secondly, I am not sure that the whole incident deserves such a detailed description. Compared with the concealment and misinterpretation of clinical trial data, it was a relatively minor infraction. Social Audit went after two sentences a GSK representative said on TV: "What we have seen in terms of the anecdotal reports [of withdrawal] is that it happens very rarely". In a voiceover, GlaxoWellcome's representative was reported as also having said that withdrawal or discontinuation syndrome " ...occurs in only 2 out of every 1,000 patients who are discontinued appropriately. Even then the symptoms are mild and short-lived."


 * In their conclusion IFPMA stated simply that "GlaxoSmithKline has been found to be in breach of Sections 1.3 and 1.7 of the IFPMA Code. The company has accepted this decision and has given assurance that they will take all possible steps to avoid similar breaches of the IFPMA Code occurring in the future." (see the full text here http://www.socialaudit.org.uk/461913.htm). By the way there was another breach of Code of Practice for the Pharmaceutical Industry (http://www.socialaudit.org.uk/4641001.htm#Dear) when another GSK spokesman said in an interview "There is no scientific evidence that Seroxat leads to addiction and dependency. There have been one or two reports of discontinuation symptoms with abrupt cessation, which is why our data sheets [doctor and patient information leaflets] reflect new advice to taper off the medication." (http://www.socialaudit.org.uk/4632-XSA.htm#Dear). But again, this is relatively minor. The Sceptical Chymist (talk) 00:57, 28 August 2009 (UTC)


 * I had looked for a statement on the IFPMA website, but could not find one, so went with the BMJ quote. Also, there seems to be a big difference between how paroxetine was presented in the USA compared to how it was presented in other countries.  Mostly that is the DTC TV ad campaign.  WP being a general audience encyclopedia, rather than a primarily professional reference, the DTC campaign seems particularly notable here.  Also, some studies have found tapering useless (except it helps the manufacturer sell more pills even when the customer wants to quit, and allows the manufacturer to blame the victim for the symptoms; the withdrawal syndrome reportedly kicks in around 10mg/day regardless of the speed of tapering).  This discussion has led me to read more sources, and they seem to point to a problem with the USA division of GSK, much more so than GSK worldwide.  It looks a bit like AIG, where one division brought down the whole company; alas, the head offices are too often happy to enable local chief executives as long as the money keeps rolling in.TVC 15 (talk) 04:39, 28 August 2009 (UTC)