Photo of Eric Alexander

For as long as we have been representing drug and device companies in product liability litigation, the plaintiffs have accused our clients of “buying the science.”  Sometimes, this has allegedly been by funding studies or offering support to outside researchers such as free product or access to administrative support.  Sometimes, this has allegedly been by not initiating or funding research that might have supported (or, as they rarely offer, debunked) plaintiffs’ causation theory.  Because causation is a plaintiff’s burden and, as the revisions to Fed. R. Evid. Evid. 702 have emphasized, it is the burden of the proponent of a novel causation opinion to establish its reliability, the lack of studies supporting a causal relationship is not excused by such attacks.  The Seventh Circuit famously recognized this in Rosen v. Ciba-Geigy Corp., 78 F.3d 316, 319 (7th Cir. 1996). Yet, plaintiffs love to pursue evidence of the defendant’s role in shaping the science and, if they get to trial, attack established drugs or devices with often fairly tangential evidence about the defendant’s relationship with authors of studies the plaintiffs do not like.  By fairly tangential, we mean that someone who comes up with a study, conducts it on her own patients, submits the study to a peer-reviewed journal for publication, and later starts consulting for the defendant on a different product is supposedly incapable of producing legitimate science.  Maybe we are just defense hacks, but we cannot think of many instances where the plaintiff’s own experts could say the data in a “defense” study was bogus.

On the other side, these pages are replete with attempts by plaintiff lawyers and their fellow travelers to buy the science.  Of late, many courts have called out the role of questionable and over-interpreted data from plaintiff-affiliated labs in driving unsupported litigation. (See, e.g., here.) It has also been the case for a long time that the use of citizen’s petitions to FDA has been a tool by the plaintiff bar to shape, and sometimes sidestep, the science. (See, e.g., here.)  We have also seen direct funding of studies by plaintiff lawyers, occasionally even disclosed in publications.

While it may be that we just do not cover discovery disputes very often, we have not written much on the issue of from whom and how much discovery can be sought about studies or other publications that one side contends are biased.    We have often lamented that discovery in product liability litigation, particularly litigations with MDL and/or other coordinated proceedings, is uneven.  Part of that is because the plaintiffs’ lawyers, consortiums, funders, jobbers, etc., are not parties and manufacturers are.  If John Doe sues Company X and he alleges Company X improperly influenced Study Y, then discovery can be sought from the defendant.  If John Doe sues Company X and Company X alleges that Doe’s lawyers and their affiliates improperly influenced Study Z, then good luck getting discovery from the lawyers and their typically unknown intermediaries.  If an author on one of these studies signs up as an expert for either side, then there is another obvious avenue for discovery.  However, direct subpoenas from either side on non-party (and non-retained) study authors or non-party journals are fairly unusual.  The decision in In re: Paraquat Prods. Liab. Litig., MDL No. 3004, 2023 WL 8372819 (S.D. Ill. Dec. 4, 2023), provides a look at motions practice on a subpoena served on a study author.  (This post is not on behalf of counsel for any party in the litigation.)

The Paraquat litigation is just outside our usual bailiwick, as it relates to allegations that an herbicide caused plaintiffs to develop Parkinson’s Disease.  We have touched on the MDL’s rejection of public nuisance claims popularized by, and largely rejected by appellate courts in, the opioid litigation.  The dynamic leading to the motions practice on a subpoena on a study author, though, rings familiar.  We checked the docket to flesh out some of the timing discussed here.  In December 2022, the court set a schedule for expert discovery and Rule 702 motions, which were then due on April 17, 2023.  A few weeks before that deadline, but after plaintiff’s causation experts had been deposed and it was obvious that a strong Rule 702 motion would be coming, the first publication to claim the herbicide caused Parkinson’s Disease was published.  Not only that, but the article claimed the manufacturers had been trying to buy the science.  The two authors of the article were not retained experts, but at least one had been approached about serving as a plaintiff expert and had shared a draft of the article with a retained plaintiff expert right around the time the plaintiffs would have been finalizing their rebuttal reports.  “Coincidence,” you say?  “Bias,” you say?  One of the defendants subpoenaed both authors, one of whom complied, and one of whom refused to produce some of the requested materials.  Cross-motions followed.

As with other disputes on third-party subpoenas, this came down to the intersection of Fed. R. Civ. 26(b)(1) and Fed. R. Civ. P. 45(d).  For those who have not memorized all the subparts of the Rules, the former has the current—and somewhat narrowed compared to the longstanding provision it replaced—definition of what is discoverable in general, as determined by relevance, privilege, and proportionality.  The second relates to protecting third parties against “undue burden or expense.”  Because each author had produced documents, the focus in Paraquat was a weighing of the incremental value and burden of documents responsive to the three categories of the subpoena that were contested by one of the authors:  1) drafts and sources of information for the article, 2) documents relied upon in the article, and 3) documents related to the peer review process for the article.  The prior production of documents related to communications with plaintiffs’ counsel and their experts limited the value of the additional documents to “expose the article as an advocacy piece that was not the product of a ‘genuine scientific effort.’”  Id. at *4 (quoting a brief).  By contrast, there was a “potential chilling effect that such disclosures could have on [the resisting author] himself and the scientific community overall.”  Id. at *3.  The court continued, “[t]hese are precisely the types of documents that researchers and scientific journals ordinarily keep confidential to ensure the integrity of the peer-review process.”  Id.  This was enough to quash the requests at issue.

Somewhat surprisingly, the court did not cite 45(d)(3)(B)(i), which provides a court the discretion to quash a subpoena that requires “disclosing a trade secret or other confidential research, development, or commercial information.”  Nor did it cite 45(d)(3)(B)(ii), which provides the same discretion as to requests requiring “disclosing an unretained expert’s opinion or information that does not describe specific occurrences in dispute and results from the expert’s study that was not requested by a party.”  The only part of Rule 45 that the court cited was 45(d)(3)(A)(iv), which provides for mandatory quashing when the court finds there is “undue burden,” which is itself a discretionary and multi-factorial consideration.  Another mandatory provision not cited is 45(d)(3)(A)(iii), which applies to “disclosure of privileged or other protected matter, if no exception or waiver applies.”  We raise this bit of bingo board blather because of our concern that plaintiffs would get more latitude in seeking discovery from or concerning authors of “defense” studies or articles.  As we said above, discovery could normally be obtained from the party, which does not have the protections in 45(d)(3).  Probably not from the journal that published it.  See In re Bextra, 249 F.R.D. 8, 14 (D. Mass. 2008) (cited in Paraquat, 2023 WL 8372819, *3 n.3).  It is unclear, though, if an author plaintiffs could say was tainted by an alleged affiliation with the defendants would get the same deference in resisting a plaintiff subpoena.  The author would presumably have at least as strong of arguments as to the mandatory and discretionary bases for quashing set out above.

As the Paraquat court said:

[The issuing defendant’s] experts are perfectly capable and will have every opportunity to explain why they consider the article to be scientifically unsound when they testify.  But analyzing [the author’s] drafts, his reliance documents, and his peer-reviewers’ comments to critique the scientific rigor of his article would likely devolve into a collateral inquiry over what does and does not constitute a genuine scientific analysis.

2023 WL 8372819, *4.  This may be so, but we would like to be sure that the conclusion would be the same if the parties were flipped.  When plaintiffs’ lawyers start a litigation without any published study support—which was surely the case when the Paraquat MDL petition was filed in March 2021, two years before the paper at issue—attacking the science and fabricating arguments about why the absence of supporting science is really the defendant’s fault are sure to follow.  A corollary of that Rosen saw we mentioned up front (“But the courtroom is not the place for scientific guesswork, even of the inspired sort. Law lags science; it does not lead it.”) is that plaintiff should not get more leeway in discovery to deflect from the lack of science behind their case.