[subhead] How a UW professor lent his name to drug companies research findings--without bothering to check whether they were true

Dr. David Dunner of the University of Washington's Department of Psychiatry admitted he "ghostwrote" an article that appeared in the March 1995 issue of the journal European Neuropsychopharmacology (EN) on behalf of pharmaceutical manufacturer SmithKline-Beecham (which has since merged with GlaxoWellome to become Glaxo-Smith-Kline, or GSK). Dunner had purportedly analyzed the data of clinical studies involving GSK's antidepressant and anti-anxiety drug Paxil and concluded that it is less likely to lead to suicidal thoughts than the older antidepressant imipramine or a placebo(sugar pill). Dunner never looked at any of the data but he was still listed as an "author" of the article.

Meanwhile, Dr. David Healy of the University of Wales Department of Psychological Medicine presented a different analysis of this same data during the Paxil trial last year. The family of Wyoming resident Donald Schell, 60, sued GSK in federal court after Schell shot his wife, daughter, granddaughter and then himself to death in 1998 after two days on Paxil. Healy testified on behalf of the plaintiffs. He argued GSK's internal records demonstrated that there is a substantially increased suicide risk for patients put on Paxil. The jury agreed with the plaintiff's position that Paxil was primarily responsible for Schell's actions and awarded them $6.4 million in June of last year. The judge in the case rejected GSK's challenge of Healy's testimony and sent them packing to a federal appeals court in Denver in August. A suicide warning has since been put on the label for the drug, not here in the United States, but in Britain, where it is known as Seroxat.

Where did Dunner's stunning admission of having ghostwritten such an important article appear? In either of the local dailies the Seattle Times or Post-Intelligencer? No, not even close. A Seattle resident would have to purchase a ticket on the next British Airways flight from SeaTac to London to meet the author of the article Dunner was quoted in: Sarah Boseley, health editor of the Guardian. Boseley's February 7 article confirmed what I have suspected all along about Dunner: he has had little or nothing to do with the actual research and writing behind many of the articles where his name, D.L. Dunner, appears.

The current situation in peer-reviewed publication in academic and clinical medical research is starting to resemble the Catholic Church's control over publication of anything in, say, 1300. The pharmaceutical industry funds most of the research into the crucial Phase III clinical trials that lead to FDA approval for marketing drugs (the truth is most of the cost of bringing a drug onto the market is put up by taxpayers). This gives the drug companies an enormous amount of influence over study design and, most crucially, they typically retain control over study data as proprietary information. The end result is that it's difficult to find a published article in peer reviewed medical journals dealing with clinical drug studies that weren't supported by funding from the drug's manufacturer. Most medical journals are now dependent on pharmaceutical industry advertising to stay in business. This didn't stop 13 prestigious journals, including the Journal of the American Medical Association, New England Journal of Medicine, and Britain's Lancet, from publishing an editorial in September of last year condemning the conditions put on researchers by the drug companies as, "draconian for self-respecting scientists, but many have accepted them because they know if they do not, the sponsor will find someone who will." The aim of the journals was to implement editorial policies to authenticate authorship of articles.

"The pharmaceutical companies usually retain the right to withhold the information their researchers find that they don't want the public to see," said Keith Hoeller, community college instructor in psychology in Seattle and editor of Review of Existential Psychiatry and Psychology (who also brought the Boseley article to the author's attention). "These kinds of contractual stipulations also provide researchers with incentives to find the kind of results the drug companies are looking for in the studies they fund."

"It would be a simple matter to say that the data is inherently unscientific while it remains proprietary," said David Healy. "There is no other branch of science in which the raw data remains inaccessible to investigators generally and indeed essentially to the public." We can all see what happened when the clinical data in the Paxil trials became a scientific matter in a court case last year. This tells us that there may be something to what many critics of the marketing of psychiatric drugs during the '90s suspected all along: The public has been fed a self-serving party line about the efficacy and safety of Paxil, Prozac, Zoloft, and all of the other newer antidepressant drugs.

Actually Dunner is anything but friendly and approachable on the topics being reviewed in this article. Dunner won't talk to me about what he knows about the side effects of the newer antidepressant drugs and his conflicts of interest with the pharmaceutical industry. I don't mind Healy's willingness to discuss these matters with me. It's just that I help pay Dunner's salary and help fund his department every time I pay Washington state's regressive sales tax. None of this is a problem in the area of financial conflict of interest, though, as I have recently received copies of Dunner's honoraria and conflict of interest records from the UW.

Here is a brief summary. Since July of 1997 and up until the end of last year, Dunner made at least 80 requests for outside professional work for compensation including 33 lectures sponsored by various pharmaceutical companies, 14 for various universities and 11 on behalf of various professional societies and other organizations. No exact dollar amount is ever given on any of the records I have obtained. He also spends at least a few days every year as a "reviewer" for various peer-reviewed academic journals.

Dunner's record also includes eight statements of significant financial interest. The most significant with regard to the Paxil clinical trial data is a disclosure he made in 1998. The disclosure was made regarding an application for a clinical trial dealing with Paxil in which Dunner would serve as an investigator at his Center for Anxiety and Depression. His Confidential Statement to the Vice Provost for Research states, "My involvement with SmithKlineBeecham (the pre-Glaxo merger owner of Paxil) involves being a member of the international advisory board related to paroxetine (Paxil)." In other words, Dunner's name appeared in the EN article not as an independent scientist but most likely as a part-time employee of the manufacturer of Paxil. This conflict of interest is not mentioned in the EN article.

Here is Dunner in his own words in Boseley's Guardian article: "I don't know who saw it (the Paxil clinical trial data). I did not. My role in the paper was that the data were presented to us and we analyzed it and wrote it up and wrote references." Dunner's co-author Stuart Montgomery, then of St. Mary's hospital medical school in London, declined comment for the Guardian article. His other co-author is Geoff Dunbar, a company employee (Dunbar's status as a SmithKline employee was mention in the EN article).

Healy figures about half of all articles appearing in medical journals today are ghostwritten jobs by pharmaceutical companies. What this tells us is that we should regard much of what we read in the published medical literature with a level of suspicion similar to that of a citizen of the former Soviet Union reading Pravda. In other words, take with a grain of salt any published article stating "studies show that drug X is a safer and more effective treatment than drug Y and placebo."

The lack of candor most of the psychiatric profession and the pharmaceutical industry has about these heavily marketed drugs has potentially deadly consequences for the patients taking them. According to Healy, "The evidence from across the board from all the companies producing SSRI's (i.e., Prozac, Zoloft, Paxil, etc.) is that their drugs can make 1 in 20 of us agitated to the extent that we drop out of trials." Healy figures that at least 250,000 people worldwide have attempted suicide because of Prozac alone and 25,000 have succeeded. That sounds like a huge number, but if you put 50,000,000 people on the drug, at least 2,500,000 will become extremely agitated. If ten percent of that population attempts suicide because of the extreme agitation, that gives you 250,000. If ten percent of that population succeeds in carrying out their suicide attempt, you end up with 25,000 dead bodies. Put another 100 million people on Zoloft (comedian Phil Hartman's wife Brynn), Paxil, Luvox (Eric Harris, one of the Columbine High School shooters), and the other more potent newer antidepressants and you might have another 50,000 dead bodies.

"The fact that (pharmaceutical) companies have chosen to market them as antidepressants rather than agents of agitation is a business decision rather than a scientific matter," said Healy. Yes, business has been great for the companies manufacturing these "antidepressant" drugs. The sale of these drugs is now a $10 billion-per-year industry. However, the outcomes for the patients put on these drugs have not necessarily been great. The four dead members of the Schell family are testimony to this fact. GSK's line in the Paxil trial that their drug didn't have enough time to "help" Donald Schell follows no coherent logic. What GSK is saying is that Schell needed to stay on the drug in his agitated state, and his dose possibly needed to be increased, a common occurrence in psychiatric practice. Increase the patient's dose and you get an even more agitated individual. You still end up with the deaths of three generations of Schells in Gillette, Wyoming on February 13, 1998, with Paxil as the suspected culprit, no matter what kind of semantic games GSK wishes to engage in with this case.

How much longer is David Dunner going to remain silent about the record of harm these "antidepressant" drugs have caused? How many articles in Dunner's publication record were ghostwritten? How much longer are the Times and P-I going to quote Dunner as an unbiased expert on mental health issues?