Photo of Steven Boranian

A federal judge in California filed an important OTC preemption decision earlier this month, and it’s important because it applied federal preemption to shut down (for now) one of the many recent benzene-related consumer class actions.  The case is Daugherty v. Padagis US LLC, No. 24-cv-02066, 2025 WL 2243622 (N.D. Cal. Aug. 6, 2025), and the products at issue are OTC benzoyl peroxide acne treatments. 

Pre-pubescents and gangly teens have been rubbing benzoyl peroxide creams on their faces to treat blemishes since the early 1960s, which we would call a pretty good sample size.  So what’s their beef now?  Well, it seems our friends at the “independent” lab Valisure got their hands on various benzoyl peroxide products and purportedly found that they “contain or systematically degrade” into benzene.  If the plaintiffs in Daughtery had known, they allegedly would have paid less for the products or would not have purchased them at all.  So give them their money back. 

We have seen this play before.  Moreover, as we reported here, the FDA tested “95 acne products containing benzoyl peroxide for possible benzene contamination,” and “more than 90% of tested products had undetectable or extremely low levels of benzene.”  FDA, Statement ¶1 (March 11, 2025). 

The plaintiffs in Daugherty sued anyway, and their lawsuit, like others, ran aground on federal preemption.  They had standing to sue, based on their allegation that they experienced some economic harm, i.e., injury in fact.  But with regard to OTC products, the FDCA expressly prohibits states from establishing “any requirement” that “is different from or in addition to, or that is otherwise not identical with” federal requirements.  That is express preemption, and the FDA’s Acne Monograph expressly permits the use of benzoyl peroxide as an active ingredient and imposes specific labeling and warning requirements.  Daugherty, at *3. 

To the extent the plaintiffs were seeking anything different, including different warnings about benzene, the FDCA expressly preempted their claims.  As the district court held, “This Court joins its sister courts in holding that the FDCA preempts Plaintiffs’ state law claims which challenge the omission of warnings on BPO products that are different from and not required by the FDCA . . . .”  Id. at *8. 

Many states, however, incorporate federal requirements—such as federal current Good Manufacturing Practices (“cGMP”) and federal adulteration standards—into state law, which allowed plaintiffs to claim that they were enforcing state requirements that echoed federal requirements.  In other words, federal law did not expressly preempt their claims because they were not “different from or in addition to” federal requirements.  They were the same.

The district court, however, was inclined to find those “parallel” claims impliedly preempted, too.  First, the plaintiffs’ state law claims appeared to require substantial interpretation of the FDCA and federal requirements.  Good Manufacturing Practices, for example, often call for “appropriate” determinations, or “satisfactory” conformance to specifications, or “thorough investigation” of nonconforming products.  Allowing courts from around the nation to interpret these provisions would (1) create an obstacle to the federal objective of national uniformity in OTC drug quality standards and (2) intrude on Congress’s delegation of exclusive enforcement of the FDCA to the FDA.  If the plaintiffs’ allegations required interpretation of federal requirements, implied preemption would apply. 

Second, the plaintiff failed to plead that the state laws they cited imposed requirements that are “parallel” to federal requirements.  The plaintiffs provided a “laundry list” of state statutes, but did not establish “that these state regulations actually impose requirements identical to federal ones.”  Daugherty at *12.  The plaintiffs alleged purported violations of cGMP, but they were not specific and did not elaborate.  They basically alleged that the defendants must have violated cGMP because they allegedly allowed benzene-contaminated products to reach consumers, sort of a res ipsa loquitur theory.  That, however, was not enough because “Plaintiffs do not assert how Defendants violated or deviated from specific cGMP standards and how such deviations led to the adulteration of [their] BPO products.”  Id.  In addition, to the extent the plaintiffs were alleging an omission-based theory of consumer fraud, their non-specific allegations failed to meet the standard under Rule 9(b). 

The court therefore was inclined to dismiss the plaintiffs’ claims, but allowed them leave to amend to allege, with specificity, the parallel state laws they are asserting and specific facts establishing that the defendants violated particular cGMPs.  Moreover, to avoid implied preemption, the asserted violations must be “limited to those which can be proven without interpretation” of federal requirements, which is a tall order.  Id. at *13.  The court could be throwing the plaintiffs a lifeline here, or it could be allowing them an opportunity to improve the record before it lowers the boom.  Given the court’s reasoning and its stated “inclination” to dismiss the case, we hope and expect the latter. 

Photo of Lisa Baird

Not that long ago (last time we posted), the Sixth Circuit issued an unpublished opinion that affirmed the exclusion of an expert (yay!) but missed the boat on the amendments to Federal Rule of Evidence 702 and its reinforcement of the expert exclusion test (boo!). 

Today, we get to tell you about a published opinion from the Ninth Circuit that affirmed the district court’s exclusion of expert testimony and then summary judgment for lack of evidence of causation (yay!) and very much got on board with the Rule 702 amendments Don’t Say Daubert approach (double yay!). 

The opinion is Engilis v. Monsanto Co., No. 23-4201, __ F.4th __, 2025 U.S. App. LEXIS 20377, 2025 WL 2315898 (9th Cir. Aug. 12, 2025), and it comes out of the Roundup MDL.  Engilis underscores the scrutiny courts must give experts under Rule 702, the necessity for expert opinions to be grounded in sufficient facts and reliable methodology, the importance of scientific experts acting like experts instead of paid partisans, and—most significantly—is a thorough endorsement of the rigorous Rule 702 approach we long have advocated. 

Given that the Ninth was one of the circuits that led the drafters to seek the 2023 Rule 702 amendments, this is cause for celebration indeed.

The Background

Engilis involved a plaintiff who alleged that use of Roundup caused chronic lymphocytic leukemia, a form of non-Hodgkin’s lymphoma.  In support of this claim, he relied on a board-certified oncologist for both general causation (can the product cause the condition?) and specific causation (did it cause the condition in this plaintiff?). 

The district court initially granted the defense motion to exclude on the papers, then granted reconsideration of the specific causation issue and held a Don’t Say Daubert hearing at which the expert was cross-examined.

Specific causation was the question, and the expert had conducted a differential etiology in reaching his opinion that the plaintiff’s cancer was caused by the product, a process which should involve listing all potential causes of the alleged medical condition, then “ruling out” those deemed implausible.

As the author of Engilis (Circuit Judge Morgan Christen) took the time to explain, “differential etiology” is not exactly the same as a “differential diagnosis”: 

In the clinical context, “differential diagnosis” refers to a process for “identifying a set of diseases or illnesses responsible for the patient’s symptoms,” whereas “‘differential etiology’ refers to identifying the causal factors involved in an individual’s disease or illness.”  Put another way, “differential diagnosis actually refers to a method of diagnosing an ailment, not determining its cause,” and “differential etiology . . . is a causation-determining methodology.”  

One potential cause that should have been on the expert’s list was obesity.  The expert had promptly crossed this potential cause off his list based on the “Plaintiff’s Fact Sheet”—the provenance of which was “not clear from the record”.  The fact sheet said plaintiff was “negative” for obesity.  On cross-examination, however, it became clear the expert had no knowledge of whether the plaintiff had obesity; he had never checked the plaintiff’s medical records, and he had not examined him to determine his physical condition either. 

That was an issue—a potential cause was ruled out without sufficient basis for doing so—but the expert’s fallback was that obesity should not have been on the list in the first place.  Contrary to medical literature, the expert said he believed obesity was not a risk factor for non-Hodgkin’s lymphoma, based solely on his “clinical experience”. 

Not good enough. The district court excluded his specific causation opinion, then entered summary judgment for lack of evidence to support the essential element of causation. 

The Ninth Circuit’s Masterpiece

We already gave away the ending of Engilis:  The Ninth Circuit affirms.  In doing so, the court noted that this ultimate result—exclusion of the expert and affirmance of summary judgment—would have been the same under the pre-2023 version of Rule 702 as well as the current version. 

But what makes Engilis such a special case is the work the Ninth Circuit put in to explain the history of Rule 702 and the rigor it requires.  It really is a contrast to the recent Hill decision from the Sixth Circuit, which likewise agreed its expert exclusion affirmance would be the same under either version of Rule 702 but then blindly recited the type of authorities the 2023 Rule 702 amendments sought to move away from.

First, the Ninth Circuit started with a little Rule 702 history lesson.  It recounted the original adoption of Rule 702 in 1975; the Supreme Court’s interpretations of it in Daubert v. Merrell Dow Pharmaceuticals, 509 U.S. 579 (1993) as well as the refinements in General Electric Co. v. Joiner, 522 U.S. 136 (1997), and Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999); and the codification of the Supreme Court’s Daubert trilogy through the 2000 amendments to Rule 702 which “clearly envisioned a more rigorous and structured approach” to Rule 702 than “some courts were employing” but was not entirely successful in getting all courts to fully embrace that “more rigorous and structured approach”. 

Then, they got to the really good stuff:  the 2023 Amendments to Rule 702, which sought to bolster the rigor of Rule 702 analyses in two main ways: (1) by expressly requiring a proponent of expert testimony to “demonstrate[] to the court that it is more likely than not that” the four admissibility requirements are satisfied; and (2) by also requiring that “the expert’s opinion reflects a reliable application of the principles and methods to the facts of the case.”

Quoting the 2023 Rule 702 Advisory Committee:

This [2023] amendment sought to “clarify and emphasize” that proffered expert testimony must meet the admissibility requirements of Rule 702 by a preponderance of the evidence. … Before the amendment, “many courts” had erroneously held “that the critical questions of the sufficiency of an expert’s basis, and the application of the expert’s methodology, are questions of weight and not admissibility.” Properly applied, Rule 702 requires that challenges to an expert’s opinion go to the weight of the  evidence only if a court first finds it more likely than not that an expert has a sufficient basis to support an opinion. The amendment also aimed “to emphasize that each expert opinion must stay within the bounds of what can be concluded from a reliable application of the expert’s basis and methodology.” “Judicial gatekeeping is essential” to ensure that an expert’s conclusions do not “go beyond what the expert’s basis and methodology may reliably support.”

 Some of the key principles recognized by the Ninth Circuit in Engilis include:

  • Burden of Proof Under Rule 702 Does Not Favor Admission: The court reiterated that the proponent of expert testimony always must establish admissibility by a preponderance of the evidence.  Say it with us, and the court:  “[T]here is no presumption in favor of admission.”  Thus, anytime plaintiffs cite earlier Ninth Circuit authorities at us, like Messick v. Novartis Pharms. Corp., 747 F.3d 1193, 1197 (9th Cir. 2014), and Wendell v. GlaxoSmithKline LLC, 858 F.3d 1227, 1232 (9th Cir. 2017), we will cite Engilis back at them, with its rejection of the notion that Rule 702 should be applied with a “’liberal thrust’” favoring admission.
  • Reliability and Sufficient Facts Requires Grappling With the Evidence: The Engilis expert goofed in his reliance on his clinical experience to write off published literature.  “Sure there are studies, but I don’t believe them” is not a reasoned, scientifically sound explanation. Clinical experience may be more necessary when a rare disease is at issue, but if studies have dealt with the issue, the expert has to account for them in a reasoned way.  Rule 702 authorities do not stand for the “proposition that an expert’s mere talismanic invocation of ‘clinical experience’ suffices to establish that a differential etiology passes muster under Rule 702.”
  • Gatekeeping Function: The Ninth Circuit reaffirmed the district court’s “gatekeeping” responsibility under Rule 702 to ensure that expert testimony is both relevant and reliable before it is presented to a jury.  The court rejected the “let it in and let the jury sort it out” and “goes to weight not admissibility” nonsense.  Here too, it disavowed prior authority like Wendell, which had stated that, where experts’ opinions “are not the ‘junk science’ Rule 702 was meant to exclude,” then, “the interests of justice favor leaving difficult issues in the hands of the jury and relying on the safeguards of the adversary system . . . to ‘attack[] shaky but admissible evidence.'”
  • Failure to Rule Out All Causes Must Be Based In Scientifically Sound Reasons:  The court did allow that an expert might reach a causation opinion without ruling out absolutely every possible potential cause, but there has to be a solid evidentiary foundation for the opinion they do reach and a scientifically sound rationale for ruling causes in or out.
  • Don’t Say Daubert Hearings Are Permissible, But Not Required.  One expert exclusion area where district courts retain a lot of discretion is “the appropriate form” of an inquiry into the expert’s opinions.  The Ninth Circuit seems to believe that hearings in which the expert is cross-examined are “commonly used” (which to us seems like a bit of an overstatement) but “not required.”  Sometimes an expert’s deficiencies are plainly obvious on paper, but we like that the court seems to approve of hauling a challenged expert before the judge to test their theories.  When scientific issues are complex, it certainly can be helpful to the judge exercising the gatekeeping function to be able to assess the issues in real time.
  • If It Isn’t In The Expert’s Report, The Expert Is Asking For Trouble.  The expert in Engilis had prepared a report that assumed the plaintiff was not obese, and when cross-examination revealed that assumption to be flawed, he took issue with medical records reflecting the plaintiff’s BMI by arguing BMI was not a reliable measure of obesity.  Regardless of whether that BMI criticism was accurate, that part of his “opinion was inadmissible for a separate reason: it was not disclosed in his expert report”.  We love this part of the opinion, because all too often we have seen courts readily allow plaintiffs to backfill deficient expert opinions once they are called out.  That is not how it is supposed to work.  Engilis notes:  “The federal rules generally ‘forbid[] the use of any information not properly disclosed.’… . Thus, ‘when a party fails to provide information required by Rule 26, such party “is not allowed to use that information or witness to supply evidence on a motion, at a hearing, or at a trial, unless the failure was substantially justified or is harmless.”’”

Engilis is going to be our go-to on the Rule 702 changes for some time, and we have little doubt the case name is going to become a short-hand for good expert exclusion standards amongst the drug-and-device defense bar.  Now we just have to come to agreement on how to pronounce it. Ang-eh-less?  Ing-liss? An-jill-es?

It might be Dow-burt / Dough-bear all over again…

Photo of Stephen McConnell

We recently returned from our summer vacation in a small European country with a tiny but charming coastline, formidable mountains, abundant vineyards, and relentlessly friendly service.  That last bit serves as a clumsy segue into today’s case, Aguila v. RQM+LLC, 2025 U.S. Dist. LEXIS 155232 (S.D. Fla. Aug. 12, 2025), which is mostly about service of process.

Aguila continues this particular author’s run of Sunshine State cases (three in a row).  That is not on purpose, unless you want to chalk it up to our predilection for weird cases.  Call it a confounding factor.

This dismissal of a case would not ordinarily be bloggable, except that the plaintiff’s bizarre argument culminated in a judicial decision on a service of process issue that might be of general interest to our clients, particularly those based in other countries.  

The plaintiffs in Aguila included Mr. Aguila and the sole proprietorship that he owned. Both were Florida citizens. The sole proprietorship was “in the business of medical device evaluations.”  It operated as a third party review organization accredited by the Center for Devices and Radiological Health (CDRH), a branch of the Food and Drug Administration (FDA). In that role, the plaintiff entity was primarily responsible for reviewing new medical device applications, including the review of 510(k) submissions, and recommending to CDRH whether the subject medical devices should be cleared/approved or not. 

The defendants included Australian corporations and a Pennsylvania limited liability company.  We will get to the Pennsylvania LLC in a bit.  The Australian companies developed a medical device used for electroencephalogram analysis. The plaintiffs alleged that they contracted with the defendants to provide third-party medical device review services in connection with CDRH’s review of a 510(k) submission for the device. 

Things apparently did not go well.  The plaintiff sued the defendants, alleging that, during the course of the regulatory review process, the defendants made false representations to CDRH about the plaintiffs’ business conduct. According to the plaintiffs, such misrepresentations “were a significant factor in the CDRH’s eventual decision to withdraw [the defendants’] accreditation as a third-party review organization.”  We presume that loss of such accreditation was injurious to the plaintiffs’ business.

The plaintiff ended up suing the manufacturer on a variety of legal theories, including racial discrimination in medical device safety testing in violation of a couple of Florida statutes, fraudulent misrepresentation, breach of contract, tortious interference with business relations, and violation of the Florida Security of Communications Act.  The Aguila decision does not address the substantive merit of any of these causes of action, which is sort of a pity because we all could have been treated to more Florida weirdness. Did we mention that the plaintiffs were proceeding pro se?

The case was initially filed in Florida state court, and then the defendants removed the case to federal court.  Rather than follow the foreign service rules, (Federal Rule of Civil Procedure 4), the plaintiff purported to serve that company’s “U.S. agent designated for FDA purposes” with process.  That U.S. agent was the Pennsylvania LLC. The defendants moved to quash service of process per Fed. R. Civ. P. 12(b)(5) and to dismiss for failure to serve properly and for lack of personal jurisdiction per Rule 12(b)(2). 

The court granted the motion to quash service of process. The procedural requirements for service of process on individuals and corporations not located in the United States are set forth in Fed. R. Civ. P. 4(f) and 4(h).  The defendants did a good job of providing evidence showing that the Australian defendants were, well, Australian, and that the Pennsylvania LLC was an agent only for FDA purposes (see 21 U.S.C. 360(i)(1)) and was not empowered to accept service of process. The burden was squarely on the plaintiffs to demonstrate proper service. (Yes, Rule 4 is “flexible” and yes, some leniency is afforded to pro se litigants, but the relevant law and rules of court still actually apply.) The plaintiffs  could not meet their burden.

A U.S. agent for FDA purposes “shall assist the FDA in communicating with the foreign establishment, respond to questions concerning the foreign establishment’s products that are imported or offered for import into the United States, and assist FDA in scheduling inspections of the foreign establishment.”  21 CFR 807.40(b)(2)) The Aguila court read the regulation to “make clear that designation of the U.S. agent pursuant to 21 U.S.C. 360(i)(1) is for FDA’s purposes only.”  The plaintiffs, unsurprisingly, argued that the FDA “U.S. agent” designation is not limited to facilitating communications with the FDA.  But the plaintiffs “cite no authority for their argument that a U.S. agent designated for FDA purposes is also authorized by law to receive service of process under Rule 4(h).”

The Aguila court generally referred to cases cited by the defendants in their motion to quash, but did not cite such case law in its opinion.  As far as we can tell, Aguila looks like the first decision definitively holding that the FDA agent of an overseas entity regulated by the FDA and required to designate a US agent for regulatory purposes is not also an agent for service of process. Further, even if the defendants could be said to have actual notice of the lawsuit, such actual notice is not sufficient to cure defectively executed service.   

The Aguila court did not need to address the personal jurisdiction argument, but did go on to make the point that the defendants’ notice of removal was neither a responsive pleading nor a motion made under Rule 12, and thus did not operate as a waiver of the defense of lack of personal jurisdiction under Rule 12(h).   

Photo of Eric Hudson

Today’s decision, Croci v. Zoll Medical Corp., 2025 WL 2307728 (S.D.N.Y. Aug. 11, 2025), is a straightforward, preemption-based 12(b)(6) dismissal of a complaint involving a Class III medical device.  The case involves claims about a Life Vest, which is worn externally and monitors the wearer’s heart rate. The device is able to detect ventricular tachycardia and ventricular fibrillation, and it can deliver a therapeutic shock to restore the wearer’s heart rate to a normal rhythm.  The complaint alleged that the decedent experienced problems with the Life Vest, and that a representative of the Life Vest’s manufacturer visited the decedent’s home to “desensitize the device.”  Id. at *1.  Shortly after the alleged visit from the representative, the decedent experienced a cardiac arrest and died.

Continue Reading PMA Preemption in the Southern District of New York
Photo of Bexis

In his latest off-label use-related law review article in 2021, Bexis pointed out:

[A]lthough states have traditionally been recognized to have broad authority to regulate the practice of medicine, aside from occasional attempts to restrict use of abortifacients, state-law interventions restricting the off-label uses of prescription medical products have been infrequent.

James M. Beck, “Off-Label Use in the Twenty-First Century:  Most Myths and Misconceptions Mitigated,” 54 UIC J. Marshall L. Rev. 1, 10-11 (2021) (quotation marks and multiple footnotes omitted).  Indeed, at that point many more states had pro-off-label use statutes on the books (mostly concerning insurance coverage) than legislation seeking to restrict any such uses.  Id. at 11.

Any doubt about whether states have the authority to ban off-label uses was recently eliminated by United States v. Skrmetti, 145 S.Ct. 1816 (2025), which upheld a state ban on (among other things) the off-label use of FDA-approved drugs to treat gender dysphoria.  The state could rationally consider such treatment to be “experimental,” and some studies had found “the evidence on treatment efficacy and safety is still insufficient and inconclusive” and the “long-term risks . . . are largely unknown.”  Id. at 1825-26 (citations omitted).  But only Justice Thomas’ concurrence, of all the opinions, addressed off-label use as such.

For purposes of treating gender dysphoria . . ., puberty blockers generally are administered “off-label,” meaning without FDA authorization for the specific use.  Although it is neither unusual nor unlawful for drugs to be used off-label, the FDA has recognized that “just because a drug has been approved for one class of patients doesn’t mean it’s safe for another.”  That admonition is important here. . . .  The use of drugs to suppress normal puberty has multiple organ system effects whose long-term consequences have not been investigated.  This absence of evidence is a major drawback in assessing the effects of puberty blockers on children with gender dysphoria.

Id. at 1841-42 (citations and quotation marks omitted) (Thomas, J., concurring).

The basis for the Skrmetti decision thus was not the power of the states to ban this or that off-label use, but rather whether this particular ban violated the Equal Protection Clause by discriminating against those being treated off-label as a protected class subject to “intermediate scrutiny.”  Id. at 1829.  The court found no violation, as “[c]lassifications that turn on age or medical use are subject to only rational basis review,” id., and under that level of scrutiny defendants virtually always win (as was the case in Skrmetti).  See Id. at 1835-36.  Ironically, the on-label uses of these drugs were what saved the state’s off-label use ban:

[The statute] does not mask sex-based classifications. . . .  Under [the statute], no minor may be administered puberty blockers or hormones to treat gender dysphoria, gender identity disorder, or gender incongruence [the off-label uses]; minors of any sex may be administered puberty blockers or hormones for other purposes [those being the on-label uses].

Id. at 1831 (emphasis original).  Thus, the majority reasoned, “[a] law prohibiting the administration of specific drugs for particular medical uses does not” discriminate on the basis of sex – because the FDA-approved uses are available to all.  Id.

But just because states can ban off-label uses, doesn’t mean they should – particularly where, as in Skrmetti, the ban was motivated by political rather than medical reasons.  Justice Thomas celebrated the triumph of politics over medicine in his concurrence.  “[S]o-called experts have no license to countermand the wisdom, fairness, or logic of legislative choices.”  Id. at 1840.

[T]his case serves as a useful reminder that the American people and their representatives are entitled to disagree with those who hold themselves out as experts. . . .  The views of self-proclaimed experts do not shed light on the meaning of the Constitution.  Thus, whether major medical organizations agree with the result of Tennessee’s democratic process is irrelevant.  To hold otherwise would permit elite sentiment to distort and stifle democratic debate under the guise of scientific judgment.

Id. (citations and quotation marks omitted).  We disagree with this sort of populist anti-intellectualism, particularly when directed against the medical community.  “Scientific judgment” – especially medical science −  not politics, was responsible for average life expectancy in the United States rising from around 48 years in 1900 to over 78 years in 2023.  CSC, National Center for Health Statistics, “Life Expectancy,” & Figure 1.  Justice Thomas (age 77) is certainly a beneficiary of that trend.

Moreover, once governments start down this slippery slope, who knows where it will end.  Perhaps not coincidentally, in the same week that Skrmetti was decided, so was Culture of Life Family Services, Inc. v. Bonta, 2025 WL 1687929 (S.D. Cal. June 13, 2025).  California brought suit over another off-label use “abortion pill reversal (‘APR’) treatment,” which “consists of taking the hormone progesterone in order to counteract mifepristone.”  Id. at *1.  When the state instituted enforcement actions against various persons who promote that off-label use, those targeted sought “pre-enforcement relief.”  Id. at *2.  They claimed a First Amendment “Free Speech” right to engage in this particular off-label use.  Id. at *4.

They lost.  Their asserted right to engage in off-label use “failed to make a colorable claim and even if it has, the government has justified its restriction on speech.”  Id.  Once again, a court found that off-label use was not constitutionally protected.  The speech in question was commercial in nature:

Although [plaintiff] alleges that, as a religious nonprofit, it provides numerous free services, including “free APR treatment,” this belies the fact that it still accepts insurance and payment for APR treatments from women who do have the ability to pay. . . .  Thus, it seems apparent that [defendant] has an economic motivation behind its APR speech.

Id. at *6.

But why is that any different than our (and our clients’) position that truthful off-label speech is First Amendment protected under Central Hudson Gas & Electric Corp. v. Public Service Comm’n, 447 U.S. 557 (1980)?

Because the plaintiffs’ speech was rife with misrepresentations.  After a lengthy discussion of relevant, and competing, medical testimony on several points, id. at *7-9, 10-11, 12, 13 Culture of Life concluded:

  • “[I]t would be at the very least potentially misleading to state that supplemental progesterone can ‘reverse’ an abortion.”  Id. at *10.
  • “[G]iven the lack of robust scientific study on this issue, statements on the effectiveness of APR are potentially misleading.”  Id. at *12.
  • “Due to the dearth of medical evidence on the effects of [the off-label use], the Court finds it potentially misleading to say that APR can only cause non-life-threatening side effects.”  Id. at *13.
  • “Given the lack of scientific evidence on this specific question, the Court finds the statements on birth defects following APR to be potentially false and misleading.”  Id.
  • It is “unclear what the actual aggregate success rate would be.  Statements purporting a 64-68% success rate are therefore inherently false and misleading.”  Id.
  • Because “the 64% success rate is not a reliable statistic . . . and there is no reason to generalize this statistic to all women who started APR, . . . these kinds of statements [large number of “lives saved”] potentially false and misleading.”  Id. at *14.
  • “[S]tatements that APR treatment could work in non-standard situations are inherently misleading.”  Id.

The court therefore concluded that plaintiffs were not entitled to First Amendment protection because:  (1) many of the statements were false and misleading, either inherently or potentially; (2) enforcement “directly advance[d] a substantial government interest” of “protecting consumers from misleading advertising by medical professionals,” particularly given “the special interest that states have in regulating professions”; and (3) “women, in exercising their reproductive rights, are also consumers who must be given the correct information.”  Id. at *13.

As Culture of Life points out, California had not actually banned the off-label use itself.  Id. at *14.  However, Skrmetti simultaneously established that states (unlike the FDA) have that power, and if California had done so the First Amendment outcome in Culture of Life would not have changed, although the Central Hudson analysis would have been much shorter.  “As a threshold question, Central Hudson specifies that if the regulated commercial speech concerns illegal activity or is misleading, the First Amendment extends no protection and the analysis ends.”  Culture of Life, 2025 WL 1687929, at *6 (citations and quotation marks omitted).  Next time maybe California will.

And the list could go on.  Off-label use of Ivermectin was controversial during the recent COVID-19 pandemic.  No state did at that time, but any state that wanted to – or still wants to – could ban it, either for the safety reasons that the FDA cited, or for purely political reasons.

Banning off-label uses is a slippery slope, and we don’t think states should start down it, particularly for the populist, anti-science motives that Justice Thomas espoused in Skrmetti.

Photo of Eric Hudson

Today’s post is our second installment about a case in the District of Massachusetts alleging injury from a Class III medical device. We blogged about it this spring when the court dismissed plaintiff’s claim that an implanted defibrillator and associated leads caused her numerous, unnecessary shocks. The court dismissed that complaint based on straightforward, premarket approval preemption. But the court granted plaintiff leave to amend, and she moved to file an amended complaint.  Today’s decision, Summers v. Medtronic, Inc., 2025 WL 2201110 (D. Mass. Aug. 1, 2025), denied plaintiff’s motion for leave based on the learned intermediary rule in Massachusetts. It’s a helpful decision because it applies the learned intermediary rule at the 12(b)(6) stage in the context of a device recall.

Continue Reading Summers Round 2: Learned Intermediary Rule Applies to Claim Based on Device Recall
Photo of Eric Alexander

In law school, many students struggle with the distinctions between personal jurisdiction, venue, and forum non conveniens (aka, FNC).  In the real world of product liability, FNC is definitely viewed as a less effective tool in the defense litigation toolbox than the other two.  Part of the reason is that the showing required for a dismissal is that the forum the plaintiff chose, which is typically accorded deference, is not a convenient forum for the case to proceed.  It is not simply that there might be a slightly more convenient forum somewhere else, a forum minus commodus motion, if you will.  Thus, the bar for winning an FNC motion is high and depends heavily on the discretion of the trial judge.  If you compare the number of our surveys and other posts on FNC to the number of our surveys and other posts on personal jurisdiction, then you will get an idea of how rarely we see a decision that is worth discussing. 

Back when we were a 1L in law school, long before the Blog was twinkle in the eye of Bexis, we know that one of the Supreme Court cases featured in the discussion of FNC in our Civil Procedure class was Piper Aircraft Co. v. Reyno, 454 U.S. 235 (1981).  Piper is the primary Supreme Court case cited in the Eighth Circuit’s decision in Dibble v. Torax Med., Inc., — F.4d –, 2025 WL 2250561 (8th Cir. Aug. 7, 2025), that we are discussing today, so we figure it is worth a bit of a recap. 

Piper built on the principles around since at least Gulf Oil Corp. v. Gilbert, 330 U.S. 501 (1947), that dismissal on FNC imposed a heavy burden on the movant and that the decision based on the trial judge’s discretion in applying a number of public and private interest factors.  Piper, though, involved an early version of litigation tourism, so it took a look at what might be different in the sort of case where our clients might give FNC a shot.  Scottish plaintiffs sued over an accident in Scotland of a plane that was owned and operated by U.K. companies and registered in the U.K.  They sued the Pennsylvania manufacturer of the plane and an Ohio manufacturer of the propeller in California because their lawyers were in California and liked California law.  The case was removed and transferred to the Middle District of Pennsylvania, which dismissed the case on forum non conveniens.  (Under current standards, there would not have been general or specific personal jurisdiction over the defendants in California in the first place.)  In the Supreme Court, plaintiffs offered the expected arguments for why U.S. courts should entertain suits from foreign plaintiffs against U.S. companies based on sales and alleged injuries abroad.  The Piper court did not see ensuring that U.S. courts decide cases involving U.S. defendants as a reason to allow foreign plaintiff to sue:

At least where the foreign plaintiff named an American manufacturer as defendant, a court could not dismiss the case on grounds of forum non conveniens where dismissal might lead to an unfavorable change in law. The American courts, which are already extremely attractive to foreign plaintiffs, would become even more attractive. The flow of litigation into the United States would increase and further congest already crowded courts.

454 U.S. at 251-52.  Piper also rejected the idea that U.S. courts have to entertain suits from foreign plaintiffs to deter bad acts by U.S. companies:

Respondent argues that American citizens have an interest in ensuring that American manufacturers are deterred from producing defective products, and that additional deterrence might be obtained if [the manufacturers] were tried in the United States, where they could be sued on the basis of both negligence and strict liability.

Id. at 60.  After this discussion, the Supreme Court reinstated the FNC dismissal.  This walk down memory lane was brought to you by the letter “D,” as in “discretion” and “deference.”  Perhaps the most lasting influence of Piper was its emphasis on the deference that is supposed to be shown to the district court’s decision:

The forum non conveniens determination is committed to the sound discretion of the trial court. It may be reversed only when there has been a clear abuse of discretion; where the court has considered all relevant public and private interest factors, and where its balancing of these factors is reasonable, its decision deserves substantial deference.

Id. at 257.

The Eighth Circuit dabbled with discretion in Dibble, but did not show much deference to the district judge’s decision.  Indeed, the only references in its opinion to deference related to the deference afforded to a plaintiff’s forum choice and how much less there should be with a foreign plaintiff.  2025 WL 2250561, *3.  There was no discussion of the reasonableness of the district judge’s balancing or a finding that there had been a failure to consider all the relevant factors.  Even without saying what either court held—save in our title, which we know most of you blow right by—we can relay that this appellate treatment looked much more like a de novo review than the type of review dictated by the Supreme Court in Piper.

After all that, we will be brief with what actually happened in Dibble per the opinion.  An unspecified medical device was “installed” for an unspecified purpose—three analysis flags already—in a Japanese citizen in the United Kingdom.  An unspecified “failure” led the plaintiff to have the first device explanted in Colorado, at which time an unspecified second device—perhaps the same kind as the first—was implanted.  The plaintiff found the second device somehow inadequate, had some unspecified treatment in Thailand, and later sued the manufacturers of the first device in Minnesota federal court.  One of the manufacturers used to be based on Minnesota perhaps at a time relevant to the design, manufacture, or labeling of the first device, but had apparently moved to Ohio, and the other had always been in New Jersey.  Dibble was pretty loose in its recounting of the alleged facts, for sure.  The district court had dismissed on FNC because “most if not all of the facts underlying this case occurred outside in the U.K. or elsewhere outside Minnesota” and most of the documents and witnesses were also elsewhere.  Id.  By all appearances, the district court considered the appropriate public and private interest factors.

However, the Eighth Circuit disagreed with the weight the district court—in its discretion—afforded the various factors and concluded that it “erred when it viewed all evidence outside of Minnesota as weighing in favor of the United Kingdom.”  Id. (emphasis added).  That does not sound like substantial deference to us.  For us, we think it is a little strange to put all the contacts with Colorado, New Jersey, or Ohio in Minnesota’s column, which is what the Eighth Circuit did in Dibble.  Id. at *4.  After all, there could be an FNC decision about whether a case should be dismissed from Minnesota state court in favor of proceeding in Colorado state court.  But that is the difference between a de novo review of the recognized factors on appeal and affording substantial deference to the weighing of those same factors already done by the district court.  In the implanted medical device product liability space, where the plaintiff resides, where the implant occurred, where the injury occurred, or where the defendant is currently based tend to be the options for where a case sticks.  Where one of manufacturers used to be based rarely is.  Of course, as was generally rejected in Piper, the desire to avoid less plaintiff-friendly law in the jurisdiction where the plaintiff resides, etc., is not a good reason to invite a foreign plaintiff into U.S. court.

Photo of Bexis

Today’s guest post is from Jamie Lanphear and Daniel Kadar, both of Reed Smith, who follow product liability events in Europe closely. They are discussing the implications of recent changes on the availability of attorney/client and work product privileges—called “legal professional privilege” in Europe—not only in Europe itself, but how European restrictions might find their way back across the pond to parallel litigation in the United States. As always our guest posters are 100% deserving of all praise (and any blame) from their posts.

**********

If you’ve ever found yourself grumbling about the breadth of U.S. discovery, you’re not alone. For U.S. lawyers, turning over reams of company documents is a familiar—if unwelcome—part of litigation. For most of our European colleagues outside of the UK, however, the idea of broad, adversarial discovery has generally not been a concern. That’s likely about to change. The EU’s new Product Liability Directive (PLD), which the blog has previously covered here, here, here, and here will expand disclosure obligations in product liability matters across the EU, and the implications for legal privilege—especially for in-house and U.S. counsel—are significant.

Continue Reading Guest Post – The New EU Product Liability Directive: More Disclosure, More Risk for Privileged Communications
Photo of Michelle Yeary

This post comes from the non-Butler Snow side of the blog.

The Bair Hugger MDL has an up and down history. First, we lauded the district court’s Rule 702 rulings that led to summary judgment across the board for the defendant.  But then the Eighth Circuit reversed. The cases came back to the district court and the litigation has hobbled along since. That’s all about the substance of the cases. On procedure, the defendant is now batting two for two. Almost exactly one year ago to the day, we posted about the Bair Hugger MDL dismissal of multiple cases for failing to show any good cause for plaintiffs’ failure to file suggestions of death timely. At that time, we thought the Bair Hugger court sounded a little exasperated with the plaintiffs’ shenanigans. Well, plaintiffs are at again. Only this time, it’s Plaintiffs’ Fact Sheet shenanigans that are getting them in hot water.

The situation in Bair Hugger was not a single slacker who missed a deadline by 24 hours because their internet crashed during a thunderstorm. We’re talking about over 200 plaintiffs, who despite ample notice and time, were anywhere from 2 months to over 6 years overdue. In re Bair Hugger Forced Air Warming Products Liability Litigation, 2025 U.S. Dist. LEXIS 152331, *1, 4 (D. Minn. Jul. 24, 2025). That’s not a delay—that’s ghosting.  

While we know most of our audience is defense counsel, if anyone is reading this and thinks—what’s the big deal, it’s just a fact sheet? Why does that warrant dismissal?  Here’s the answer.  The whole point behind an MDL is streamlining pretrial proceedings, conserving judicial resources, and promoting efficient resolution of mass tort claims. A cornerstone of this process is the Plaintiff Fact Sheet (PFS)—a standardized discovery tool used in lieu of interrogatories that enables defendants to evaluate the merits of each claim, and courts to manage their dockets. In other words, MDLs are a bit like boot camps. They are here to whip thousands of lawsuits into shape, sort the wheat from the chaff, and figure out which claims actually have meat on the bone. Without a completed PFS, defendants are left to guess the basis of a claim, and courts are forced to manage a bloated docket with no ability to assess individual claims. 

Moreover, court orders are not polite suggestions. PFSs are not something to be filled out “at your earliest convenience.” Plaintiffs who ignore those orders, especially after multiple opportunities to comply, are negligent.  At that point, dismissal isn’t harsh. It’s just common sense. Which is the approach the Bair Hugger MDL court took. With limited exceptions, the court granted the motion to dismiss as to any plaintiff who was the subject of the defendant’s motion and who remained delinquent at the time the motion was heard. Id. at *3. First, by the time of the hearing, plaintiffs were on notice that failure to comply carried the risk of dismissal.  Second, the plaintiffs were well aware they were late. The 90-day deadline to serve a PFS had been in place for over eight years. Delinquent plaintiffs received at least two separate notices—they were placed on a list circulated by defendant’s counsel and then again made a subject of the motion. Third, there was no dispute that plaintiffs had ample time to cure their deficiencies (anywhere from 2 to 82 months). Fourth, each plaintiff had an opportunity to be heard. The court directed plaintiffs to respond to the motion and held a hearing. So, if defendant got it wrong for any reason, plaintiffs had the opportunity to correct the record. Id. at *4-5. Yet, “no plaintiff offered an explanation for their failure to comply with [the PFS] 90-day deadline.” Id. at *5-6.

Instead, plaintiffs argued that the motion to dismiss should be denied because the defendant failed to comply with the local rule requiring the parties meet and confer before filing a motion. The defendant relied on provisions in the PFS order that allowed certain status conference procedures to substitute for the meet and confer requirement. As written, the order was somewhat ambiguous as to whether that procedure applied to PFS deficiencies or complete delinquencies, so the court found defendant’s interpretation to be reasonable. But, more importantly, courts are not required to deny motions because the moving party failed to meet and confer. There are times when reaching the merits notwithstanding the failure to meet and confer is “common sense.” The court found this was such a case. In fact, defendants conferred with individual plaintiffs after the motion was filed which resulted in the defendant withdrawing the motion as to more than 50 cases. Those “conversations and withdrawals were consistent with the spirit, if not the letter, of [the local rule].” Id. at *8.

Plaintiffs also tried to make some noise that the defendant should not be allowed to dismiss cases en masse based on an attorney declaration and a spreadsheet. But the court could not fathom what “other evidence” would be needed to demonstrate plaintiff’s failure to serve a PFS.  Id. at *8-9. If plaintiffs found error with the defendant’s spreadsheet, they had the opportunity to speak up and correct it. Therefore, the court dismissed over 200 cases pursuant to Federal Rule 41(b), for failure to comply with a court order. Most of those dismissals were for never serving a PFS, but the same logic and conclusion was applied by the court to dismiss plaintiffs who failed to serve a verified response as the order required. 

In a case that has been around as long as the Bair Hugger MDL, this dismissal is like cleaning out the closet. The ghost plaintiffs—the ones who don’t fill out the PFS—are like the mystery single shoe wedged into the darkest corner of the closet. No one knows where it came from, and it definitely isn’t pulling its weight. When a plaintiff still fails to act—despite the generous accommodation we see here—it becomes apparent that the claim is either abandoned, frivolous, or asserted in bad faith. Making it perfectly reasonable for the court to show them the door—with a polite but firm, “You’re dismissed.”

Photo of Bexis

While updating the preemption chapter of his drug and device product liability treatise, Bexis came across a proposition he had not thought about in a long time.  In two opinions in the Birmingham Hip (“BHR”) MDL, the court, under the aegis of Fed. R. Evid. 702, excluded expert testimony because it solely concerned a preempted topic, and was therefore irrelevant:

It is “the district judge’s responsibility . . . to ensur[e] that an expert’s testimony both rests on a reliable foundation and is relevant to the task at hand.”  Relevant evidence is of course that which helps the trier of fact to understand the evidence or determine a fact in issue.  Testimony that relates only to claims that are preempted is not relevant to the remaining claims.

In re Smith & Nephew Birmingham Hip Resurfacing Hip Implant Products Liability Litigation, 2023 WL 6794318, at *2 (D. Md. Oct. 12, 2023) (quoting Daubert v. Merrell Dow Pharmaceuticals, 509 U.S. 579, 597 (1993)), other citations and quotation marks omitted).  As a result, p-side expert opinions concerning failure to recall, adequacy of PMA submissions, and inadequate warnings were excluded.  Id. at *6, 13-14.

An earlier BHR decision reached largely the same conclusion, excluding expert testimony “that relates to the PMA approval process” because “[t]he claims to which such testimony might be relevant are expressly or impliedly preempted by federal law.”  In re Smith & Nephew Birmingham Hip Resurfacing (BHR) Hip Implant Products Liability Litigation, 2021 WL 781682, at *5 (D. Md. March 1, 2021).  In addition:

[A]ny claim that [defendant] had a duty to communicate information to patients or the medical community should be preempted as an attempt to impose requirements that add to or differ from federal regulations.  Accordingly, the court will exclude the challenged testimony relating to Dear Doctor letters and other communications to the medical community or patients.

Id. at *8.  Further, since only parallel claims were left in the MDL, “[t]o the extent that any expert testimony seeks to rely exclusively on state law duties that are not pinned to federal requirements, they are irrelevant to the remaining claims in this case.“  Id. at *9.

That caused us to look for other decisions that had mixed preemption and the admissibility of expert testimony under Rule 702, and sure enough, they were out there if one looked hard enough.  In In re Baycol Products Litigation, 532 F. Supp.2d 1029 (D. Minn. 2007), one of the plaintiffs’ experts purported to offer opinions “that criticize [defendant’s] submissions to the FDA and the FDA’s approval of [the drug].“  Id. at 1052.  That, of course, is a no-no under Buckman Co. v. Plaintiffs Legal Committee, 531 U.S. 341 (2001).  The testimony was excluded for that reason.  “[T]o the extent [the expert’s] testimony is offered only to show that the FDA was misled, or that information was intentionally concealed from the FDA, the testimony must be excluded.”  Id. at 1053.  We note that a number of courts have excluded various kinds of fraud-on-the-FDA evidence on Buckman grounds (see our post here), but Baycol seems to be the only such decision that focused specifically on expert testimony.

Another ruling excluding expert testimony for preemption reasons occurred in the Lipitor MDL:

[The expert] opines that the [particular study’s] data did not establish efficacy of [the drug]. . . .  As explained above, any claims that the [drug’s] label was misleading based on [that study] are pre-empted.  Therefore, this testimony is irrelevant.  To the extent that it has any marginal relevance, it would be confusing and misleading to the jury to hear testimony on the allegedly misleading nature of the . . . label’s description of [the study] when such allegations cannot be the basis of Plaintiff’s claims.

In re Lipitor (Atorvastatin Calcium) Marketing, Sales Practices & Products Liability Litigation, 185 F. Supp.3d 761, 783 (D.S.C. 2016).

The same result occurred in Lowery v. Sanofi-Aventis LLC, 2021 WL 872620 (N.D. Ala. March 9, 2021), involving an injectable premarket approved medical device subject to extensive preemption under Riegel v. Medtronic, Inc., 552 U.S. 312 (2008).  The expert claimed that the defendant “should have conducted additional testing” in the course of manufacturing the device.  2021 WL 872620, at *21.  Preemption precluded those opinions from being admissible.  The “opinion that Defendant should have implemented different testing methods that are not compelled by a federal law goes, at best, to a theory of liability that is preempted and, for that reason, cannot be an issue put to the jury at trial.”  Id. (citing Daubert).

We also found a number of analogous decisions.  Preemption has led to expert exclusion in several railroad cases.  In Campbell v. Union Pacific Railroad Co., 616 S.W.3d 451 (Mo. App. 2020), expert opinions that railroad crossing safety equipment was inadequate were rejected.  “[O]pinions concerning the safety of the crossing and allegedly optimal warning devices have no bearing on the . . . preemption inquiry.  If a crossing is equipped with federally funded warning devices, a railroad does not have an additional duty to erect other warning devices.”  Id. at 475.  In Joiner v. National Railroad Passenger Corp., 2008 WL 11506272, at *3 (S.D. Miss. March 25, 2008), expert testimony by the plaintiff’s expert “that the subject crossing was ‘ultrahazardous [wa]s not relevant, reliable, or admissible under F.R.E. 702” because it attacked “the adequacy of the warning devices in place at the subject crossing,” which was preempted.  Id. at *3.  See Rasmusen v. White, 970 F. Supp.2d 807, 816 (N.D. Ill. 2013) (expert opinion about a train’s “excessive speed” precluded due to preemption because the expert was demanding a speed less than the applicable federal speed limit).

Finally, in a maritime personal injury case, expert testimony about certain regulations was excluded because those regulations were preempted in the maritime context.

The Court will grant Defendant’s motion and exclude [the] expert testimony in this case.  It is undisputed in this litigation that the [ship] is a Coast Guard-inspected vessel, and that the OSHA regulations [the expert] relies on as the bases for his opinion . .   are completely preempted in this case.  Those OSHA regulations therefore are irrelevant to the claims and defenses in this case.

Al Qari v. American Steamship Co., 2023 WL 5014782, at *5 (E.D. Mich. Aug. 7, 2023).

Given the above, defendants have a solid legal basis for including preemption (where factually appropriate) in their Daubert/Rule 702 motions as a basis for challenging expert testimony as irrelevant.  It’s one situation where defendants may decide to deviate from our general “Don’t say Daubert” admonition (since the reference to “relevance” is from Daubert rather than the rule).

Drug & Device Resources

Bexis’ Book

FDA-regulated products now account for an estimated one-fifth of overall economic activity in the U.S. They have also been the focus of a litigation explosion. This timely guide covers all aspects of litigation involving drugs, medical devices, vaccines and other FDA-regulated prescription products.
Learn more about this book