August 7, 2008
Speaking of Funding Bias...
David Michaels of SKAPP forwarded me an interesting order I missed the first time around in the welding rod litigation in the Northern District of Ohio. In it, the judge ordered (Welding_rod_funding_order.pdf [PDF]) that both sides disclose the extent of funding provided to the authors of articles, treatises, etc., relied upon in litigation. In total, in this particular sample, the defendants' funding totaled something over $10 million, while the plaintiffs' around $500,000. A Mother Jones article describes the case more.
I'm not sure, incidentally, that the proportionate amounts spent means as much as it might -- i.e., it's not self-evident that spending 20 times more necessarily means more 20 times more wicked conduct or 20 times more intent to buy the science. I would hope that industry is spending a lot more money studying the dangers of their products than plaintiffs' lawyers are.
Now, I'd prefer it to be consistently good work (and the article notes some pretty sketchy sounding stuff), but the fact of spending money doesn't necessarily mean bad work or bad science.
In any event, the outcome of the order -- disclosure all around -- is sound.
TrackBack URL for this entry:
Listed below are links to weblogs that reference Speaking of Funding Bias...:
» Around the web, August 22 from PointOfLaw Forum
All-toxic-tort edition: Representative class action plaintiff of the day: "For five years, her diet consisted almost exclusively of Chicken-of-the-Sea tuna..." [WSJ law blog, Legal Intelligencer, Beck & Herrmann] Defendants said to be way outspending p... [Read More]
Tracked on Aug 21, 2008 7:35:50 PM
Of course, the plaintiffs would complain the failure to study demonstrates egregious corporate indifference meriting punitive damages.
Separately, how much have plaintiffs spent on litigation-driven diagnoses, given the number of fraudulent diagnoses in these cases?
Posted by: Ted | Aug 7, 2008 5:35:04 AM
The biased left wing article does not disclose the denominator. Were defense article authors paid more than plaintiff article authors each, not totally?
Does more funding mean higher quality or lower quality, or nothing? Is a Bentley a better car than a Corolla? Is the court qualified to answer such a question on the quality or objectivity of scientific articles?
What if the defense has no Bentley, but 20 Corollas on its list, and the plaintiff has 2? Doesn't the total imply, the overwhelming majority of the articles favor the defense, and not just their expenses?
Unless the author of an article is a testifying expert, aren't articles hearsay? Isn't the funding information of the article hearsay about hearsay?
Let's have mini-trials about each article's credibility, depose the authors. I would be interested in a mini-trial about the funding and background of the judge.
Posted by: Supremacy Claus | Aug 7, 2008 6:11:18 AM
The math is as follows:
Total Number of Experts: 65
Total Spent: $12.7 Million
Average Payment: $195,000 per expert
Total Number of Experts: 12
Total Spent: $520,000
Average Payment: $43,000
The standard disclaimer about how misleading averages can be applies.
Note that the Defendants tried to conceal some of the expert spending with a "work product" argument the court rejected.
Posted by: Justinian Lane | Aug 7, 2008 11:02:34 AM
It's worth observing that the payments weren't all directly to experts ("Payments by the parties were made both to authors directly and also to affiliated entities, such as corporations the authors control or educational institutions for which they work."). I wonder how much of it, if any, relates to expenses. I expect it costs a lot less to re-evaluate extant data than it does to collect and create the data in the first place.
It may well be that those are apples-to-apples numbers, but I'm not sure they are.
Posted by: Bill Childs | Aug 7, 2008 11:12:04 AM
The number of studies on each side should get the case dismissed under a Frye standard. If one takes the biggest, and best study, it should get it dismissed under a Daubert standard.
Is the funding a measure of quality or of bias? The judge is trying to invent an unauthorized rule of evidence without any competence to do so.
Again, the court has no competence to settle a scientific controversy. The variability of an effect suggests the defendant is being asked to pay for pre-existing conditions or the effects of aging.
This matter may be taken up again after the science has a better consensus.
Posted by: Supremacy Claus | Aug 10, 2008 7:39:35 AM