The Case of the Disappearing Journal Article
15 CommentsBy Ed Silverman // October 21st, 2008 // 2:42 pm
Last week, Kevin Huang, the editor of the Harvard Health Policy Review, a student-run journal with a tiny circulation, received an e-mail from Richard Frank, a Harvard Medical School health economics professor, to say that a recently published article about medical journals and ethical standards was unfair to Frank and two of his colleagues. As a result, he would no longer serve as an advisor to the publication. In response, Huang pulled the site down for further review.
Why the fuss? The objectionable article was written by Donald Light and Rebecca Warburton, who chronicled an episode that took place four years ago concerning an earlier article they wrote for the Journal of Health Economics. In that piece, they chastised a widely cited study published in 2003 in the same journal that claimed it cost $802 million to develop a new drug. The lead author, by the way, works for the Tufts Center for the Study of Drug Development.
They questioned the estimate, in part, because the Tufts Center receives industry funding, a point not noted in the original JHE article (although the authors did not receive industry funding for the study). And there was heated debate involving three JHE editors - Harvard’s Frank, Tom McGuire and Joe Newhouse - over various accusations contained in the critique. Thus ensued a year-long battle in which passages were fought over, responses and counter-responses were written, bad blood boiled over, and a lawsuit was even threatened.
Ultimately, their critique was published. But their latest article in the Harvard journal, which cites the episode as a case study of ethical standards at medical journals, appears to have revived the ruckus. Some academics began complaining that Frank used his influence to squelch the article after Light, who is a professor of comparative health systems and policy at the University of Medicine & Dentristy-New Jersey, alerted others to Huang’s decision to pull down the web site. “I think it’s very peculiar the whole web site was shut down and I’m wondering why it really was done,” he tells us.
We contacted Huang, who wrote: “The site was temporarily taken down because we panicked. We thought that we had might have inadvertently published something that was potentially very biased and/or unsubstantiated. Our articles are closely reviewed by the editors immediately in charge of them, but not all the articles are closely scrutinized by all members of the senior staff. More specifically, I felt completely uninformed about what potential controversy, if any, the article would raise as it had not been flagged for my attention before and I didn’t want us to continue to publish something that I didn’t fully understand.”
For his part, Frank tells us: “I sent an e-mail to Huang saying that I felt the way they dealt with the article - by not fact checking or offering the ability to respond- I felt it violated my sense of fair play. That’s their call, but I just didn’t want to be associated with the publication anymore…My e-mail never indicated any action on my part, other than not wanting to be affiliated with the publication. It was a personal decision…Whatever action they took was on their own…Given they had not sought my advice in a long time, I didn’t think it was anything but an opportunity for me to sever my ties…I saw it as a means to not influence them through my continued position. If anything, it puts less pressure on them. I didn’t want to use my position to influence them.”
Epiloque: Huang writes us a short while ago to say the site is back up and the article by Light and Warburton is accessible once again - along with a profuse apology. As of now, there is no accompanying response from Frank, Newhouse or McGuire, but he indicated that a written reply may yet appear.
Joseph DiMasi
Ed,
Let me correct two points made in your story. The “widely cited article” did not quote a Tufts University study claiming it cost $802 million to develop a new drug. The article IS the so-called Tufts study. Secondly, while the Tufts Center receives unrestricted funding from industry sources, we (and my coauthors are not part of the Tufts Center) neither received nor sought funding for this study. The fact that we did not receive funding for this study was noted in the article.
Joe
Ed Silverman
Hi Joe,
Thanks for the note. I’ve fixed the first reference so its now clear the ‘paper’ and the ’study’ are one in the same.
However, the paper identifies you as being with the Tufts Center, so its not as if the authors - as a group - have no connection to the center. Also, please forgive me, but I can’t locate any mention - one way or the other - of funding. If you can please direct me to the correct section or page, I’d gladly make any needed fix.
Thanks
ed
Joseph DiMasi
Hi Ed,
I don’t really follow your point. My coauthors do not work for the Tufts Center or receive funding from it. In the same way, I am not part of or connected to the Duke University economics department (Grabowski) or the University of Rochester Simon School of Business adminstration (Hansen).
The statement about no funding for this paper follows the text section and is prior to the References section in a separate Acknowledgments section on page 183.
Thanks,
Joe
Ed Silverman
Hi Joe,
Thanks for directing me toward that sentence. I’ve massaged the language to reflect that point.
As to the point I was trying to make before, I think I misunderstand your first comment concerning the authors not being part of the Tufts Center. I mistakenly thought you were referring to yourself as well. Sorry about that.
So are we squared away now?
Cheers
ed
Joseph DiMasi
Hi Ed,
Almost. You write that the fact that the Tufts Center receives some industry funding was not noted in the article. However, that paints an incomplete picture. As far as I recall, there was no requirement or request by the journal that it be noted. Also, it is fair to say that it is quite well-known that the Tufts Center receives some of its funding from industry sources (unrestricted grants). Information about the nature of the Center’s funding is provided openly on the Tufts Center website.
Thanks,
Joe
Ed Silverman
Hi Joe,
There may not have been a request, but the connection to industry was as an issue, at least far as I understand it. That’s why I noted it.
Regards
ed
Joseph DiMasi
Hi Ed,
It wasn’t the connection-to-industry-funding part of the sentence to which I was referring. I was concerned about inferences that readers might draw from the “a point not noted in the original JHE article” part. That could be read by some that something untoward was done that Light and Warburton wanted point out.
Firstly, my comment fills out the picture here by noting that nothing of the kind was required. Secondly, as far as I know, while Light and Warburton wanted to raise industry funding as an issue, they did not criticize us or the journal regarding acknowledgments (at least not in any version that I saw).
Their comment was not fundamentally about author associations (again, at least not any version I saw). It was about data and methodology, and we had rebuttals for each and every one of their points. Those interested can read the entire set of published exchanges and make up their own minds.
Thanks,
Joe
Ed Silverman
Hi Joe,
Point taken. From my discussion with Light the other day, I understood this to have been an issue. I will attempt to clarify.
Regards
ed
Joe Gerstein
Marcia Angell in her book THE TRUTH ABOUT THE DRUG COMPANIES analyzed the original publication by DiMasi and colleagues at the Tufts Center for the Study of Drug Development that came up with the ever-growing average cost of drug development — at the time, $802 million. If you actually read their paper and not just the title, you find that $802 is only for new molecular entities discovered and developed entirely in-house (a vanishingly small fraction of the total), includes opportunity costs of not investing the money in the stock market (that would now subtract considerably from the outlay!), and does not account for tax deductions and credits. In short, she considered it a bogus number.
Dan A.
To add to Joe’s comment above:
Marcia’s book, released in 2004, The 802 million to bring a drug to market was stated in 2001, as determined by a group of economists, which was headed by Joseph DiMasi at the Tufts Center. To no suprise, it was reinforced often by the industry’s lobbying group PhRMA. The figure was announced at a press conference in Philadelphia in November of that same year. Marcia also states that The Tufts Center is largely supported by the pharma industry. The 802 million figure, as she references the NYT, is based on a very select and small amount of certain drugs that were the most expensive, and this was done deliberately. (page 42)
Other groups such as Public Citizen, as well as independent analysts, more thoroughly determined that the cost to bring a drug to market was around 100 million dollars, after taxes and so forth, only a few years ago.
Needless to say, the wide disparity between the two figures is likely because the industry has to justify the high cost of thier medications to avoid price controls.
atlex
Joe,
Last I checked, Marcia Angell is not a financial analyst. When investors look to make decisions, they consider opportunity costs when determining if the likely return justifies the risk of the investment. Pharmaceutical products are not only costly to bring to market, they take years from inception to approval, and come at an incredibly high risk. The Angell’s of the world may not like how investors view this, but it’s hardly a bogus view.
Atlex
Joseph DiMasi
Joe and Dan,
Atlex has it exactly right. By the way, Joe, $802 million is not in the title of the paper, and the abstract notes the difference between cash outlays and full opportunity costs, that the numbers are pre-tax, and the drugs are self-originated (which are not a “vanishingly small” portion of these companies portfolios).
Dan, I am not familiar with anyone other than Public Citizen putting the number at around $100 million. I would also argue that neither Public Citizen nor Marcia Angell can be called “independent analysts” in this area, given what seems like a fairly transparent anti-corporate bias. But we need not focus on idealogical or other biases here. One can make an objective assessment about their claims and estimations. The criticisms of a study deserve as much scrutiny as the study itself.
In an initial response to Light and Warburton’s comment, we wrote about the logical and methodological flaws in these writings. That was much too much space for the journal to absorb, so we wrote a full paper on the topic, referenced it in our journal reply, and posted it on the Tufts Center website for easy access to all. The Public Citizen data analyses are deeply methodologically flawed, and in ways that substantially bias downward the cost estimates. In addition, claims made by both Public Citizen and Angell about taxes, sample composition, and opportunity costs are conceptually flawed. You can find our critique of their critiques here: http://csdd.tufts.edu/_documents/www/Doc_231_45_735.pdf. I encourage you to read it with an open mind, and then draw your own conclusions.
Regards,
Joe DiMasi
Donald Light
How WAS that long, complex study, drawing on different data sets and involving many complex phases to estimate the high costs of R&D funded? Did DiMasi, Hansen, and Grabowski do it all on evenings and weekends?
What kinds of pressures were the editors of the Journal of Health Economics under when they suddenly pulled from production the entire critique, reply, and response which they had already accepted and set forward to Elsevier for page proofs? What pressures led them to go silent for over 2 months and then send back the set with nearly all of the response they had accepted now deleted? Very unusual.
Rebecca Warburton
Joe DiMasi probably never saw the issue of sponsorship raised in our work, because all such information was DELETED by the JHE editors. It’s also worth noting that the editors sent our first Comment to DiMasi and colleagues without telling us about it, or seeking our permission.
We’ve written short piece to let you decide for yourself whether you have ever been (or would like to be) treated as we were, read:
http://www.thejabberwock.org/blog/pdf/light001.pdf
On financial disclosure, at the Tufts Center website (http://csdd.tufts.edu/) under “About us” (top right) there is a drop-down that says “Financial Disclosure”. However I can’t open it because the whole drop-down menu disappears much too quickly to allow me to click anything but the first of five items. Previously, you could see the financial information, under “About us”.
Joseph DiMasi
Dr. Warburton,
I am stunned that you would think that the JHE editors would need your permission to offer us a chance to reply to your comment. Asking the authors of the original published article if they wish to reply to a comment or letter is standard editorial practice at academic journals (it is also the fair thing to do). That is also usually where it ends. Very few commenters are given any opportunity, as you were, for a rejoinder.
Perhaps, though, you are referring to your narrative in the piece published in the student-run journal mentioned above, in which in a fair reading of the text you make the claim that the editors sent us a draft of your comment sometime between March 2, 2004 and the end of April 2004 (when your initial comment was accepted) for our expert advice and fact-checking. You offer no evidence for that. I would like to know what you think is evidence for that assertion, because that claim is patently false. The first time that we received any version of anything you wrote, or even any indication that you were working on a comment, was in mid-May 2004. We were also not asked for expert advice or to fact-check the comment. We were simply asked if we wanted to reply to the comment.
With regard to the Tufts Center financial disclosure web page, I did not have any trouble getting it from the drop-down menu from my computer at work today, but I did have trouble with my laptop last evening. The problems that I had last evening were with all of the drop down pages for all of the top level sections.
Let me suggest that you hold the cursor over a top level area for a couple of seconds before you try to move down into the drop-down area. In any event, we clearly were not trying to withhold the financial disclosure page (if that is what you were insinuating). If it is an issue, then it is an issue for the drop-down pages generally. In addition, you can always get the financial disclosure page, or any other page, by clicking on the site map. There are links to all of the pages there. Nonetheless, I have suggested that the issue of a too-sensitive (if that is the way to put it) set of drop-down menus on the site be put to the Tufts Center’s website vendor for a fix. I am told, however, that the website is do for an overhaul soon anyway.