Photo of Michelle Yeary

This case is from the non-Butler Snow side of the blog.

It’s a relatively new term (at least for us Gen Xers) – but we have all been ghosted. A second date that never gets confirmed. A contractor who quotes the job and then vanishes. A group text that just stops. It stings, but usually the only casualty is your ego. In Sarmiento-Tenecela v. Teva Pharmaceuticals USA, Inc., No. ESX-L-002653-20, slip op. (N.J. Super. Ct. Law Div. Aug. 31, 2026), the casualty was the entire case.

Plaintiff sued over a Paragard IUD that fractured on removal a decade after it was placed, and she planned to prove her claims with two experts — David Kessler, the former FDA Commissioner and go-to plaintiff’s expert in pharmaceutical mass torts, and Dr. Jimmy Mays, a polymer chemistry PhD who has done general causation work in the Paragard MDL on the device’s materials and design. These are not fly-by-night witnesses pulled off a for-hire circuit. They are the same experts other Paragard plaintiffs have relied on nationwide. Small problem — neither one ever sat for a deposition in this case. Defendants asked for dates. Plaintiff did not respond. The court ordered depositions to happen by a date certain and warned that noncompliance could mean preclusion of expert testimony or worse. Plaintiff still did not produce them. When the motion for sanctions finally landed, plaintiff’s excuse amounted to “we could not reach them” and “counsel had a family medical situation.” The court was sympathetic to the latter as a human matter and unmoved by it as a legal one. You cannot rely on an expert to beat summary judgment while simultaneously keeping that expert unreachable for the other side. That is not litigation, that is hide and seek.

Plaintiff’s defense was that Kessler and Mays are general experts in the Paragard MDL and that plaintiff’s counsel does not control their calendars. That much may be true. These experts likely are booked across dozens of cases and answer to MDL leadership, not to any one plaintiff’s firm, but it cuts against plaintiff, not for her. Nationally prominent, frequently deposed experts are not hard to find or schedule; they have existing deposition protocols, standing availability windows, and counsel who negotiate their calendars as a matter of routine. Plaintiff also cited counsel’s mother’s hospitalization for cardiac procedures as a reason the deadline slipped. The court did not dismiss those circumstances, but it also did not let them do the work plaintiff needed them to do. Personal hardship might explain a missed phone call. It does not explain months of silence on a court-ordered deposition schedule, and it certainly does not explain why they could not be produced even after the court compelled compliance.

Defendants asked for dismissal outright as a sanction. The court declined to go that far, noting that dismissal as a discovery sanction is reserved for cases where the misconduct goes to the very foundation of the claim or is deliberate and contumacious. Id. at 5. Instead, the court went with the lesser remedy of preclusion. Id. at 6-7. Kessler and Mays were barred from offering testimony at trial. But benching the experts turned out to be the ballgame anyway because plaintiff needed them to survive summary judgment.

To defeat preemption of a failure to warn claim, a plaintiff has to show the manufacturer could have made a unilateral label change under the FDA’s changes being effected regulation (“CBE”). Which requires plaintiff to produce evidence of “newly acquired information” that reveals risks of a “different type or greater severity or frequency” than what was available to the FDA at the time of approval.  Id. at 7.  In this case that meant, plaintiff needed to show defendant possessed qualifying newly acquired information before plaintiff’s IUD was implanted in April 2008. She did not. The adverse event reports she pointed to were already reflected in the FDA-approved label before her IUD was placed, and a study done in 2015 was simply seven years too late to support a CBE label change before 2008.  The court noted, tellingly, that it did not even know what Kessler would have said about newly acquired information as of 2008, because his MDL report never addressed the question. Funny how that works when you will not produce your expert for a deposition. The court found no genuine dispute of material fact and held the warning claims preempted. Id. at 18.

Plaintiff also argued the IUD’s breakage warning did not adequately cover fracture during routine removal without preexisting embedment. The court did not need to resolve that dispute because plaintiff’s only medical expert testified the device was in fact embedded at the time of removal. The precise risk plaintiff says the label failed to disclose is not the risk that happened to her. And having conceded she would not offer her case-specific medical expert on label adequacy at all, plaintiff was left with no warning expert whatsoever. Id. at 19.

The design defect claim fared no better. Plaintiff’s two design theories, the type of resin used and the barium sulfate concentration, required evidentiary support plaintiff simply did not have. She offered no evidence her own device used the disfavored resin, and no expert testimony that a different barium sulfate concentration would have prevented the fracture. Nor did she offer any expert evidence of a feasible alternative design.  Under New Jersey law, a design defect claim without an alternative design is not a design defect claim. Id. at 18.

This is a defense win on preemption, but our focus is on the preclusion ruling. Discovery gamesmanship around expert witnesses has consequences, even when a court is not willing to nuke the whole case as an outright sanction. A plaintiff who will not make her experts available for deposition simply does not get to use them, full stop, and courts should not need to hear the phrase “we could not reach them” more than once before losing patience.  Ghosting has consequences. In this case, plaintiff’s experts skipped the date, and the whole case got left standing at the altar.

Photo of Bexis

Last month we published a thought piece, Medical Monitoring and the Limits to Equity, pointing out how no-injury medical monitoring claims asserted in federal court could well fall outside the foundational limits on those courts’ equitable powers, given the holding in Grupo Mexicano de Desarrollo S.A. v. Alliance Bond Fund, Inc., 527 U.S. 308 (1999), limiting federal equity powers to the “the jurisdiction in equity exercised . . . in England at the time of the adoption of the Constitution and the enactment of the original Judiciary Act,” that being, of course, 1789.  Id. at 318.  We also concluded that the Judicial Act’s limits to federal equitable powers likely applied in diversity cases, given, Guaranty Trust Co. v. York, 326 U.S. 99, 104-06 (1945), and its progeny.

We cautioned, however, “that this argument seems never to have been tried before in the medical monitoring context,” given that we found no “opinion in any court citing to either Grupo Mexicano or Guaranty Trust that also included the phrase ‘medical monitoring.’”

That’s no longer true.

In Bio-Lab, Inc. v. Tartt, ___ S.E.2d ____, 2026 WL 2395789 (Ga. Aug. 18, 2026) – decided less than two weeks after our post – the Georgia Supreme Court came to the same result as the second part of our analysis – that the limits of equitable jurisdiction exercised by federal courts in diversity cases are federal in nature, given the Guaranty Trust line of cases.  Id. at *5-6.  Bio-Lab involved a medical monitoring claim, the state high court having accepted certified questions from a federal district court:

The Defendants moved to dismiss that request for equitable relief, arguing that, under Georgia law, the requested remedy is unavailable in the absence of allegations of present physical injury. . . .  [G]iven the dearth of Georgia case law on medical monitoring in connection with mass torts, [the federal court] was “left with considerable uncertainty as to whether Georgia law permits medical monitoring as a tort remedy absent a current physical injury.”

Discerning no clear answer to that question, the district court opted to seek our input and certified the following questions to this Court:

(1) Under Georgia law, can a plaintiff exposed to toxic substances which may cause future disease obtain equitable relief in the form of medical monitoring costs absent an allegation of present physical injury; and

(2) If so, what standard must the plaintiff meet to obtain such relief?

Id. at *2 (emphasis added).

The Georgia Supreme Court interpreted the first question “as implicating two broader legal issues with relevance beyond the specific facts,” the first being “under what circumstances is equitable relief available.”  Id. at *3.  That was a question governed by state law.  Id.  To be entitled to equitable relief, a plaintiff “must first identify some legally cognizable ‘wrong’ or ‘injury’ that needs to be remedied.”  Id. (citation and quotation marks omitted).  That is a “separate question whether that plaintiff is entitled to some form of relief.”  Id.

Bare fears of injury will not authorize equitable relief.  Rather, where a plaintiff seeks an equitable remedy for a threatened legal injury, that injury must be imminent, not remote, before equitable relief is authorized.

Id. at *4 (citations omitted).  Because “equity follows law,” equitable relief cannot be afforded “contrary to the law.”  Id. (citation and quotation marks omitted).

Given the pleadings-based status of the case, the Tartt court refused to go any further, and did not directly address medical monitoring.

[R]esolving the issue of whether equitable relief is available will first involve a thorough and intensive evaluation of whether the Plaintiffs’ complaint includes sufficient factual allegations to demonstrate both a legally cognizable injury and the necessity of equitable relief to remedy that injury.  But this Court, when answering questions certified to us by a federal court, does not reach the merits of the underlying case; instead, we answer the questions posed to us only in a general sense. . . .  Thus, we respectfully decline to answer the district court’s questions to the extent that it calls on us to decide the ultimate issues raised by the Defendants’ motion to dismiss.

Id. (citation and quotation marks omitted).

If the court had stopped there, Tartt would have been a non-event, since it did not reach the issue of whether no-injury medical monitoring was an available equitable remedy under Georgia law.

What made Tartt bloggable was the court’s treatment of the second question – “when equitable relief is available, what is the appropriate form and scope of that relief?”  Id. at *5.  That question, Tartt recognized, was “largely” a federal question under Guaranty Trust:

Unlike the question of entitlement to equitable relief (something we agree with the district court that state law governs), the form of equitable relief that a federal court sitting in diversity may grant may (we believe) be largely a question of federal law.  [citing Guaranty Trust and progeny decisions]  We acknowledge that this is a surprisingly thorny issue.  But we think the best reading of the relevant decisional law − which is not exactly a beacon of clarity − is that the appropriate form and scope of equitable relief must be determined as a matter of federal law.

Id. (other citations omitted) (emphasis added).  Why?  “[S]tate law clearly cannot commit the federal courts to grant equitable relief [in a diversity case] contrary to restraints imposed by the Constitution or an Act of Congress.”  Id. (Wright & Miller citation omitted).  “[C]rucially, some have interpreted Guaranty Trust to mean that state law does not govern the scope of the equity powers of the federal court, even when state law supplies the rule of decision.”  Id. at *6 (citation and quotation marks omitted) (emphasis original).

The referenced “some” cases include those that we had collected in the Guaranty Trust portion of our post not two weeks earlier.  Don’t worry, we’re not getting too full of ourselves – Tartt did not cite the same actual cases we did; ours were considerably more recent, so there’s no reason to believe that anybody involved in Tartt had ever read our post.

But, as to the second issue in our post, that federal law governed the scope of available equitable relief in diversity cases, Tartt reached the same conclusion we did:

So − as we understand the current state of the admittedly ambiguous law − a federal court, although limited by state jurisprudence on the question of whether the injunction remedy may be granted, is free to decide for itself what the scope and content of that injunction should be as befits the historically discretionary character of equity jurisprudence.  State law provides the basis for relief; it does not necessarily constrain the scope of that relief.  And that seems especially true where, as here, state law supplies no clear answer as to the availability of a specific form of injunctive relief.

2026 WL 2395789, at *6 (citation and quotation marks omitted).  Tartt thus “declined to opine” on the availability of any “specific form of equitable relief,” such as medical monitoring.  Id. at *7 (“We therefore respectfully decline to address whether Plaintiffs are entitled to a medical monitoring injunction.”).  Before that could happen, the federal court hearing the case would have to “determine[] that equitable relief is appropriate (or properly asserted).”  Id.

And that brings us back to the first, and most intriguing, point of our prior “Limits to Equity” blogpost – that “equitable” medical monitoring claims in federal court are subject to the United States Supreme Court’s Grupo Mexicano holding that federal equitable relief is limited to those powers recognized by the English courts of equity back in 1789.  We have no reason to believe that any English chancellor in equity during the reign of King George III (or before) had permitted uninjured persons to recover for medical testing expenses based on mere exposure to a purportedly toxic substance.  That’s something that we hope that the defendants in Tartt (and defendants in any medical monitoring litigation in federal court) will raise as a defense in the future.

Photo of Eric Hudson

About four years ago we posted about a decision in which the Wisconsin Court of Appeals reversed a trial court’s injunction requiring the administration of a certain medication to treat a patient suffering from COVID-19. In that decision, the party seeking the injunction relied on an unsworn declaration from a physician who stated that he was “generally considered the foremost authority” on the proposed treatment. Gahl v. Aurora Health Care, Inc., 403 Wis. 2d 539, 554-55 (Wis. Ct. App. 2022). That decision rejected the assertions in the declaration and noted that the declaration was “conspicuously devoid of any discussion of the patient in [the] case.” Id.  Today’s decision, Kory v. Waters, 2026 WL 2507087 (2d. Cir. Aug. 26, 2026), comes full circle—the physician who offered the declaration in Gahl supporting what might be considered unorthodox treatment was named as a defendant in a medical malpractice claim related to his treatment of a patient with COVID-19.  The dismissal of the malpractice claims underscores the broad scope of immunity under the Public Readiness and Emergency Preparedness Act (the PREP Act), 42 U.S.C. § 247d-6d.  Even if the treatment might be considered “counter-culture,” if the Act’s requirements are satisfied, it is subject to immunity from suit.

Continue Reading PREP Act Comes Full Circle for “Counter-Culture” Expert in COVID-19 Treatment
Photo of Lisa Baird

Today’s guest post comes from three Reed Smith colleagues and fellow European Union Product Liability Directive nerds, Jamie Lanphear, Greg Speier, and Wim Vandenberghe. As always, our guest posters get all the cheers for their work along with any jeers, though those are hardly likely.

**********

The Blog has been sounding the alarm on the EU’s new Product Liability Directive (“PLD”) for almost two years now—and with good reason.

For those just tuning in, the PLD is the most sweeping reform of EU product liability law in four decades—expanding the definition of “product,” introducing claimant-friendly presumptions of defect and causation, broadening disclosure obligations, and extending the categories of recoverable damage. The Directive requires all 27 EU Member States to transpose it into national law by December 9, 2026. It applies to products placed on the market or put into service after December 8, 2026 (a typo was recent corrected in the original text, which had stated December 9).

Over the past year and a half, the Blog has covered the Directive’s overhaul of EU product liability law (see, e.g., here, here, here, and here, ), including its implications for software and digital products (here), its expanded disclosure obligations and the risks they pose to privileged communications (here), and what the European Commission itself has said about its goals for the new regime (here).  

This post is about implementation—where each Member State stands with fewer than four months to go, and where meaningful national divergences are emerging that will shape litigation risk on day one. The short version: three Member States have completed transposition, twelve more have draft legislation in the pipeline, and the rest are somewhere between initial steps and radio silence.

Where Things Stand

Member State transposition efforts fall into four categories.

Completed Transposition: Hungary, Croatia, and Lithuania have completed transposition. Hungary was first, with its parliament adopting the implementing law on December 16, 2025. Croatia followed on July 7, 2026, amending its Civil Obligations Act and notifying the Commission. Lithuania’s Seimas adopted its bill on June 25, 2026, as Act XV-1080, with an entry-into-force date of December 9, 2026.

Well Underway: Twelve Member States have prepared draft legislation: Belgium, Cyprus, Czech Republic, Denmark, Finland, Germany, Italy, the Netherlands, Poland, Slovakia, Slovenia, and Sweden. Germany, Slovakia, and the Netherlands are furthest along—each is in parliamentary process and tracking toward adoption this fall. Belgium, Italy, and Slovenia entered the pipeline more recently but are moving. The remainder have drafts in various stages of consultation and review.

Early Stages: Austria, Bulgaria, Estonia, France, the Republic of Ireland, Latvia, and Romania have taken initial steps toward transposition but have not yet published draft legislation. Austria expects to submit a draft to parliament this fall. Bulgaria is preparing draft legislation, with submission to the Council of Ministers expected in September. Estonia expects a government proposal to reach parliament in September 2026. France has confirmed an interministerial working group and stakeholder consultations, but no published text yet.

No Known Public Movement: Greece, Luxembourg, Malta, Portugal, and Spain have made no known public progress toward transposition.

Key Divergences to Watch

This is where it gets interesting. The PLD is a maximum-harmonization directive, meaning that Member States generally may not adopt more or less stringent rules than those set forth in the Directive. But the PLD expressly leaves room for national discretion in certain areas. In others, differences emerging in national implementing legislation raise questions about whether Member States have gone beyond the discretion the Directive allows.

Development Risk Defense: This defense—also known as the “state-of-the-art” defense—protects manufacturers from liability if the scientific and technical knowledge at the time the product was released was such that the defect could not have been discovered. Article 18 expressly permits Member States to depart from this defense. Most drafts retain it, but there are some exceptions: Finland will continue its longstanding exclusion of the defense. Hungary limits the defense for medicinal products used as instructed. Germany retains the defense, subject to its existing carve-out for genetic engineering products.

Non-Material Damages: The PLD expressly leaves aspects of damages to national law. Recital 23 provides that non-material losses—such as pain and suffering—are compensable to the extent permitted under national law. As a result, the availability and calculation of non-material damages will continue to vary across Member States, even under the PLD’s maximum-harmonization framework. That variable will likely shape forum-shopping decisions.

The “Excessive Difficulties” Threshold: This one is particularly concerning. Article 10(4)(a) allows courts to presume defectiveness and causation where a claimant faces “excessive difficulties” in proving those elements due to technical or scientific complexity. Finland and Sweden have removed the word “excessive” in their implementing proposals, arguably lowering the threshold for claimants to trigger the presumptions.

However, unlike the development risk defense and non-material damages, the PLD does not expressly authorize Member States to vary this threshold. Whether these divergences represent permissible implementation choices or conflict with the Directive’s maximum-harmonization requirement may ultimately require judicial resolution. But in the meantime, they create exactly the kind of patchwork that encourages forum shopping—claimants filing in the jurisdiction with the most favorable presumption standard.

Disclosure: As we discussed in a prior post, the PLD’s disclosure requirement is a significant development, particularly for jurisdictions that have historically had little or no pre-trial discovery. Member States are taking different approaches. Some are adopting new, PLD-specific disclosure provisions, while others are relying on existing procedural rules.

Regardless of approach, Article 9 requires courts to have authority to order that disclosed evidence be presented in an easily accessible and understandable manner. Belgium’s government has expressly acknowledged that the PLD disclosure requirement goes beyond existing document production under Article 877 of the Judicial Code. The Directive also contemplates that compliance may require compiling or classifying evidence into a comprehensible presentation—not simply producing documents as they exist. That is an obligation that will be new to many defendants.

Distributor Liability: Denmark proposes to keep its existing fault-based distributor liability rules within the PLD framework. Once the injured party proves defect, damage, and causal link, the burden shifts to the distributor to prove it did not make the product dangerous. Whether those rules fall entirely outside the PLD’s maximum-harmonization requirement may become an important issue as distributors defend claims under the new framework.

What This Means

The PLD’s transposition process is unfolding largely as expected: most Member States are hewing close to the Directive’s text, and late movers will likely copy-paste their way to compliance. But the divergences that are emerging—on defenses, the “excessive difficulties” threshold, and disclosure—are important to note. They will create different litigation environments across the EU, and they will shape where claimants choose to file.

For defense lawyers, the practical takeaway is straightforward: do not treat the PLD as a single, uniform regime. Map the differences. Monitor the late transposers. And prepare for the likelihood that the first wave of litigation under the new rules will be as much about testing the boundaries of national implementation as it is about the underlying claims.

December 9 is coming. The question is no longer whether the PLD will change the landscape. It is how—and where—the changes will land first.

Photo of Michelle Yeary

There is a certain kind of plaintiff’s theory that reads like it wandered in from a wellness blog rather than a products liability treatise. Today’s case is one of them. A plaintiff who eats vegan for medical reasons sues a device manufacturer because nobody warned her, personally, that the implant going into her shoulder had a bovine pedigree. No allegation that the warnings to her surgeon were wrong. No allegation that the device malfunctioned. Just a claim that the manufacturer owed her, the patient, a duty to check her diet. The court took one look at that theory and sent it back to the kitchen.

The case is Petty v. Schwartz, No. 71820/2025, slip op. at 1 (Sup. Ct. Westchester Cnty. Sept. 1, 2026), and it is a tidy little pleading-stage win worth flagging even though the underlying facts are, frankly, a little unusual. Plaintiff underwent arthroscopic shoulder surgery, during which her surgeon implanted a bovine-collagen device. Plaintiff alleges that she suffers from an autoimmune disease that causes her immune system to attack cartilage tissue, and that collagen products can trigger or worsen her symptoms — which is why, she alleges, she maintains a vegan diet and lifestyle. She claims the bovine-collagen implant triggered a permanent, severe, and debilitating reaction. Id. at 1-2.

The only claim against the device manufacturer was styled as a cause of action for lack of informed consent — a theory that plaintiff more or less conceded, in her own opposition brief, does not work against a manufacturer. Plaintiff and defendant relied on the same New York authority, which holds that a device manufacturer is not a medical practitioner and does not have a physician-patient relationship with the plaintiff. Id. at 3. Not to mention, the manufacturer’s duty to warn runs to the treating physician – the learned intermediary – not the patient directly. Id. at 4.

So instead of defending the pleaded claim, plaintiff tried to argue her allegations, properly construed, stated claims for failure to warn the physician and the wider medical community, negligent undertaking, and negligent marketing and promotion. The court was not interested in construing anything that generously. Under New York’s pleading standard, a court accepts the facts alleged as true and gives plaintiff the benefit of every favorable inference, but bare legal conclusions do not get that benefit. While the standard may forgive “imperfect labeling” of a claim, it does not supply missing allegations. Id.at 3 & 4.

On failure to warn, plaintiff never alleged that the implant’s labeling was inadequate, that contraindications were omitted, that screening information was deficient, or — critically — that any warning deficiency proximately caused her injury. A failure-to-warn claim needs a warning that was actually deficient and a causal link to the injury. Plaintiff supplied neither, just the conclusory assertion that the manufacturer should have told somebody, somewhere, about the bovine collagen. Id. at 4.

On negligent undertaking, the complaint did not allege that the manufacturer undertook to render any service, or that any act by its representatives — two employees present in the operating room — launched or increased plaintiff’s risk of harm. There were no allegations that those representatives gave advice, made recommendations, selected the device, participated in the surgery, or communicated with the surgeon about whether the implant was appropriate for this particular patient. Id. Simply having your people in the room when your product gets used is not, by itself, an undertaking that creates a duty of care to the patient.

On negligent marketing and promotion, plaintiff alleged that the manufacturer provided incentives or otherwise induced orthopedic surgeons generally, and this surgeon specifically, to use the implant. The court called this exactly what it is — conclusory and speculative. There was no allegation tying any marketing or promotional conduct to negligence, and no allegation linking that conduct to plaintiff’s injury. Id. Marketing a product, even enthusiastically, is not a tort, and “the manufacturer wanted its device used” is not a theory of liability.

With all three unpled theories rejected, and the only pleaded claim conceded to be non-viable, the claims against the manufacturer were dismissed in their entirety. Plaintiff also asked, in her opposition brief, for leave to amend if the court ruled against her. The court declined even to consider the request. There was no notice of motion or cross-motion seeking leave to amend, and no proposed amended pleading as required by local rules. A request for affirmative relief tucked into an opposition brief is not a motion, and courts are under no obligation to treat it as one. Which is a good procedural point for defense counsel to remember. If plaintiff wants a second bite, plaintiff has to actually ask for one the right way — a real motion, with a real proposed pleading attached, so the court and the defendant can evaluate exactly what is being proposed. A one-line request in a footnote-adjacent aside does not cut it, and this court was not shy about saying so.

Kudos to the court for not letting plaintiff dress up an admittedly non-viable informed consent claim as three different negligence theories she never actually pleaded. The learned intermediary rule exists precisely so that manufacturers can rely on communicating with the physician, not on independently vetting every patient’s dietary practices, religious beliefs, or lifestyle choices before a surgeon selects a device. Extending informed consent, or its negligence-flavored cousins, to manufacturers based on nothing more than a sales rep’s presence in the operating room would turn every device company into a second physician for consent purposes — a result no court should welcome, and this one correctly refused to create.

As for the vegan angle — we get it, dietary and religious restrictions on medical products are a real and serious issue, and patients are entitled to make informed choices about what goes into their bodies. But the obligation to have that conversation, and to obtain that consent, belongs to the treating physician, not the company that manufactured the device sitting on the shelf. The surgeon is the one who examines the patient, discusses the options, and decides what to implant. Try to put that obligation on the manufacturer instead, and you get a lawsuit that is dismissed as a matter of law — no meat required.

Photo of Bexis

The recent law review article by the always interesting and extremely prolific Lars Noah, “‘Market Shift Liability’ for Generic Drugs:  Market Share Theory’s Eccentric New Cousin,” 13 Tex. A&M L. Rev. 1415 (Spring 2026), warrants discussion.  We usually agree with Prof. Noah, but we don’t buy the central premise of the article:  that innovator liability – that the imposition of liability upon the original manufacturer to obtain FDA approval for a drug for injuries later caused by competing generic knock-offs – is similar to the now mostly passé market share liability theory from the 1980s.

The two theories have completely different origins, for one thing.  Market share liability was a causation dodge, created by various pro-plaintiff courts when plaintiffs were unable to meet the basic substantive product liability element of product identification in DES litigation.  DES was, in a sense the country’s first generic drug (no patent protection), with the result being the anybody with the necessary facilities could make it.  But back then (from the 1940s until 1971), people didn’t keep good records, and the alleged injuries skipped to the next generation and were thus very delayed.  So market share liability cut those plaintiffs a break by removing the but-for-causation requirement that they sue the actual manufacturer of the injurious product.

Innovator liability has nothing to do with causation, but-for, or proximate.  It’s a preemption dodge, pure and simple.  It didn’t gain any significant traction until after the Supreme Court, in PLIVA, Inc. v. Mensing, 564 U.S. 604 (2011), and Mutual Pharmaceutical Co. v. Bartlett, 570 U.S. 472 (2013), preempted virtually all product liability claims against generic drug manufacturers.  So the theory wasn’t “a thing“ until after preemption required plaintiffs to look elsewhere for somebody to sue.  Indeed, innovator liability creates, rather than solves, causation problems as warning causation depends on prescribing doctors supposedly relying on the inadequate labels of drugs that they never actually prescribed, and that may have been substituted after-the-fact by the actions of independent pharmacies.  As we’ve discussed elsewhere, warning liability does not exist when the actor never read the allegedly defective warnings.

But while we find that aspect of “Market Shift Liability” a bit of a stretch, we wholeheartedly agree with Prof. Noah’s comprehensive critique of what we will continue to call “innovator liability” – since we are unapologetic defense hacks.  See 13 Tex. A&M L. Rev. at 1416 n.2 (quoting pro-plaintiff article that called “innovator liability” a “catchy name” created by “lawyers who represent drug companies”).  Cf. Id. at 1452 n.176 (we being the unnamed “bloggers” referred to in this footnote).

So, we’ll skip the article’s extended discussion (pp. 1421-52) of the development of both market share liability and innovator liability and go directly to that critique.  Note:  that’s not to say these origin stories aren’t interesting in their own right – Prof. Noah’s thorough and often wry exposition is quite readable – but it’s not what prompted us to find the article blogworthy.

So what does the esteemed Prof. Noah find objectionable about innovator liability?  Like us, just about everything.

  • “[V]iew[ing] the generic drug as a product of the brand-name manufacturer’s choices . . . seems only slightly less plausible than viewing NDA holders as pushers of prescription drugs sold by their competitors.”  13 Tex. A&M L. Rev. at 1453.
  • Pro-innovator liability articles “mak[e] a series of indefensible assumptions that create[] a caricature far removed from the messy reality surrounding these tort claims,” and “entirely ignor[e] the far more important role of learned intermediaries.”  Id. at 1454 (footnote omitted) & 1456 (footnote omitted).
  • Innovator liability will “accelerate” abandonment of the market by “brand-name manufacturers.”  Id. at 1458.  “If exit occurs prior to the approval of any ANDAs, the first to apply may have to file a petition to confirm that the NDA’s withdrawal did not occur because of concerns about safety or effectiveness.  Id. (footnote omitted).  This would create a “regulatory void . . . , leaving the FDA alone to drive future revisions to the labeling.” Id
  • To avoid potentially catastrophic “long tail” innovator liability, brand manufacturers could well seek to “persuade the FDA to withdraw the license in a way that would altogether prevent subsequent generic entry,” by engaging in liability reducing “product hopping.”  Id. at 1461.  What some view as “anticompetitive” behavior would gain the “legitimate business justification of “minimiz[ing] exposure to tort” liability.  Id. at 1462-63.
  • Treating drug labeling “as completely disembodied from the drug itself” is “visiting liability on purveyors of ideas and information that foreseeably might lead to physical injury,” which is recognized as having serious First Amendment implications.  Id. at  1465.
  • First Amendment implications “become[] more pointed . . . as misrepresentation claims threaten to penalize speech further removed from efforts to sell an allegedly injurious product.”  Id. at 1467 (footnote omitted).  There are “alarming free speech consequences of starting down this freewheeling path.”  Id. (footnote omitted).
  • Treating “the NDA sponsor as the ghost author for ANDA sponsors” might even (were it not for preemption) “allow plaintiffs injured by generic prescription drugs to assert design defect claims against brand-name manufacturers.”  Id. at 1470-71 (footnote omitted).
  • Innovator liability “seemingly would impose liability on the sponsor of an approved NDA that never came to market . . . because it could foresee that ANDA sponsors would mimic its never-used labeling.”  Id. at 1472-73 (footnote omitted).
  • “[P]redicating [innovator] liability on the supposed control exercised by the brand-name drug manufacturer hardly makes sense,” as “basing such claims instead on alleged misrepresentations arguably obviates any requirement of identity between the advertised product and the substitute product that caused an injury.”  Id. at  1474.  That could lead to liability based on warnings concerning “substitute product[s] posing somewhat different risks” such as “pharmacy-compounded versions.”  Id.

Thus, Prof. Noah concluded that innovator liability was “counterintuitive” and the rationales offered by courts and commentators to justify it were “muddle-head[ed].”

Even with its limited uptake to date, however, market shift liability threatens to cause far greater disruption of the pharmaceutical sector (as well as tort doctrine) than market share liability ever did.  For these and other reasons, innovative activity should not get singled out while imitative activity largely escapes any obligation to compensate patients injured because of alleged defects in prescription drugs. . . .  [C]ourts in other jurisdictions need to resist the temptation to produce lazy knock-offs of [California innovator liability].

13 Tex. A&M L. Rev. at 1476.

Defense counsel (and their amici curiae) confronted with plaintiffs advocating innovator liability should consult this article and consider relying upon not only the arguments it makes, but also on the copious material that the footnotes we’ve omitted provide.  In addition, feel free to utilize the Blog’s own material on innovator liability – our score card collecting all the cases that have addressed it in chronological order, and our 50-state survey.  We keep both of these current.

Photo of Steven Boranian

Federal officer jurisdiction is having a moment.  In November 2024, we called federal officer jurisdiction the quiet middle child of the federal jurisdiction family, but then earlier this year the Supreme Court decided Chevron USA Inc. v. Plaquemines Parish.  That opinion held that federal officer jurisdiction was broader than many previously thought, which we commented on here. 

The Supreme Court’s opinion got Bexis to start beating the federal officer jurisdiction drum (actually, in true Bexis form, he was restarting a riff that he took up way back in 2020).  We more recently posted on a case called Griffin v. Optum, Inc., where the Eighth Circuit approved federal officer jurisdiction over claims against pharmacy benefit managers (“PBMs”) in the long-running opioid litigation.  Because the defendant PBMs performed their functions under contracts with a federal agency, they successfully removed a lawsuit filed by a state attorney general to federal court, despite the AG’s efforts to “disclaim” recovery based on any federal-related activity. 

The Second Circuit has now joined the chorus, and it too has approved federal officer jurisdiction over opioid-related claims against PBMs.  In County of Westchester v. Express Scripts, Inc., No. 24-1639, 2026 WL 2589574 (8th Cir. Sept. 2, 2026), multiple New York counties sued PBMs over harms allegedly caused by the use and abuse of opioid pain medications.  We have seen this play before.  The counties alleged that PBMs contributed to the alleged harms through their benefit management function, such as managing drug formularies and rebates. 

The PBMs, however, performed formulary and rebate work in part under contracts with the Defense Department, the federal Office of Personnel Management, and the VA.  So they removed the cases to federal court under federal officer jurisdiction. 

This is where it gets interesting.  The counties cynically amended their complaints to add broad disclaimers purporting to abandon “any claims or recovery arising out of the Removing Defendants’ work on behalf of any federal entity,” and they moved to remand.  We have seen this play before too, and in the Eighth Circuit’s Griffin case, the “disclaimer” did not work.

It did not work in the Second Circuit either.  The central problem with the disclaimers is one of indivisibility.  As the court explained, the PBMs did not maintain separate federal and non-federal books of business.  Instead, they conducted singular negotiations with pharmaceutical manufacturers on behalf of all clients—federal plans and private plans alike.  In addition, the alleged community-wide harms were similarly indivisible, flowing from conduct that could not be neatly separated into federal and non-federal components.

The “disclaimers” therefore were ineffective.  In so holding, the Second Circuit followed opinions from multiple other circuits.  Relying for example on the First Circuit’s reasoning in Puerto Rico v. Express Scripts, Inc., the Second Circuit explained that a valid disclaimer must “clearly carve out certain factual bases, whether by time span or location,” so that no state court would ever need to evaluate whether the defendant acted under a federal officer.  By contrast, a disclaimer that requires the state court “to determine the nexus between the charged conduct and federal authority” is not valid because it is circular.  If permitted, such a disclaimer would require federal contractors to prove in state court that they were acting under federal contracts, which would negate their right to have a federal defense judged in federal court in the first place. 

The Second Circuit found that the plaintiffs’ disclaimers fell squarely into the circular category.  Whether in state or federal court, the plaintiffs were targeting rebate negotiations that were the same for federal and non-federal programs.  Federal defenses would likewise cover the whole of the alleged conduct.  In other words,

As a result, “Whether in state or federal court, Defendants will continue to press that the alleged conduct and resulting harm necessarily implicate their federal work and are subject to federal defenses.  More fundamentally, in deciding whether to enforce the disclaimers, either court will have to ascertain whether the harms from opioids distributed via federal and non-federal programs can be disaggregated at all.”  County of Westchester, at *8.  That determination belongs in federal court.

Having determined that the counties’ disclaimers were ineffective, the Second Circuit held that federal officer jurisdiction under 28 U.S.C. § 1442(a)(1) existed.  More specifically, the plaintiffs’ claims met the three-part test for federal officer removal under the Supreme Court’s Plaquemines Parish opinion. 

First, the PBMs were acting under a federal officer.  They administered formularies, processed claims, negotiated rebates, etc.—tasks that federal agencies would otherwise have had to perform themselves.  One PBM defendant operated under a detailed Statement of Work subject to weekly agency briefings and formal reviews.  Another had a similar deal with the VA, showing an “unusually close relationship involving detailed regulation, monitoring, or supervision” that satisfies the “acting under” prong. 

Second, the PBMs were sued for conduct “relating to any act under color of [federal] office.”  Under Plaquemines Parish, this requirement “sweeps broadly.”  The court found a clear relation between the plaintiffs’ allegations and the PBMs’ federal work because the rebate negotiations implicated federal clients, and they could not be bifurcated between federal and non-federal plans. Even the diffuse public-nuisance claims—targeting community-wide costs—necessarily encompass harms flowing from federally-administered prescription drug programs.

Third, the PBMs were asserting a colorable federal defense.  Namely, a government-contractor defense, a defense under a plan for military personnel, and federal preemption under the Federal Employees Health Benefits Act.  The court noted that these defenses need not be clearly sustainable—only colorable—and the plaintiffs’ sole counterargument was that they had disclaimed claims implicating those defenses, which the court had already rejected.

Federal officer jurisdiction is no longer the quiet middle child, if it ever was.  Following County of Westchester and other cases, including from the Supreme Court, we would say that federal officer jurisdiction is ready to shine. 

Photo of Eric Alexander

Over the life of the Blog, there have been many posts taking issue with the pro-plaintiff leanings of California courts, which have resulted in many questionable decisions on issues ranging from expanding personal jurisdiction to shrinking preemption to re-writing the duties of medical product manufacturers under California law.  We have also thrown more than a little shade at the California federal courts, which, while less unitary, have been worse than the average federal court on the issues that come up in our sort of cases.  For instance, a number of California federal courts have been hostile to removals and even more have often found ways to deny motions to dismiss based on preemption and inadequate pleading.  We cannot say that the tide has turned with the California federal judiciary, but we have seen a number of decisions over the last year or so where we were pleasantly surprised at how these courts came out on the issues that matter to us.  In Mizuno v. Sanofi Vaccines U.S. Inc., No. 2:26-cv-012115-CKD, 2026 WL 2392200 (E.D. Cal. Aug. 17, 2026), a long-time magistrate judge dismissed all the claims against prescription drug manufacturers without leave to amend.  Even though it raised one of our yellow flags, Mizuno is a good decision from a somewhat unexpected source.

As should happen, the plaintiff’s obvious sloppiness and apparent laziness worked against him.  (Some of the info below is from the docket not the decision.)  First, the case was brought in state court against two companies not based in California.  When defendants removed, plaintiff did not move to remand.  Second, plaintiff did not decline to have the case assigned to a magistrate for all purposes.  The decision says the parties consented, id. at *1, but actually neither side objected to the assignment in time.  Third, when the defendants moved to dismiss the original complaint, plaintiff responded by filing an amended complaint, but it did not address the issues raised in that motion.  That failure factored into the Mizuno court’s decision not to give plaintiff a third chance to file a complaint that stated a non-preempted, non-time barred recognized claim.

It may also be the case that plaintiff’s complaint omitted some of the basic facts about the drug at issue and his use of it.  The decision does not say what the drug was approved for, why it was prescribed to plaintiff, who prescribed it, or much of anything about plaintiff’s actual medical course.  The first approved indication would have been apparent from the materials the court took judicial notice of at defendants’ request, hence our yellow flag.  Usually when the court leaves out the serious medical need being addressed though the use of the medical product over which the plaintiff is suing that is a bad sign.  Here, the approved indication makes what the court identified as the shortcomings in plaintiff’s complaint that much worse, assuming plaintiff’s use was within or close to the indication.  Here we definitely go beyond the decision to set out the facts of the case.  Defendants’ prescription medication was approved for “the treatment of adult and pediatric patients 12 years and older with chronic graft-versus-host disease (chronic GVHD) after failure of at least two prior lines of systemic therapy.”  GVHD is a potentially debilitating and life-threatening inflammatory and fibrotic condition that arises in a high percentage of patients after a bone marrow or stem cell transplant.  Its most common manifestation at the time of diagnosis is in the skin.  The transplants that trigger the chronic GVHD are themselves for very serious conditions like leukemia.  So, when the Mizuno plaintiff brought product liability claims for injuries to his skin and the court considered the motion to dismiss the amended complaint, this important context was missing.

The first issue the Mizuno court addressed was whether the second amended complaint was time-barred on its face under California’s two-year statute of limitations.  Plaintiff alleged that within a month of starting his medication, he experienced “a significant increase in rash, itching, and pain”—skin conditions he apparently had before starting the medication—and discontinued it.  Id. at *3.  He then had a videoconference with his prescribing physician to discuss his belief that the new medication was to blame.  Id.  Yet he took two years and ten days to sue.  Id.  Courts often seem to mention the discovery rule without delving deeply into allegations supporting it before letting late plaintiffs slide.  However, because the amended complaint in Mizuno did not identify a date within two years of filing when his claim allegedly accrued, let alone facts supporting the “manner of discovery,” plaintiff could not rely on the discovery rule.  Id.  Indeed, because the first motion to dismiss raised the statute of limitations, the failure of the amended complaint to plead the discovery rule properly “suggests plaintiff cannot fix the defect.”  Id.  Given that plaintiff was apparently taking the medication to treat chronic GVHD, including skin manifestations, his complaint to his physician that he believed the medication increased his skin problems is a pretty strong indication of the date of accrual of his claims.  Still, plaintiff might have gotten a chance to re-plead were it not for the main reason we are discussing Mizuno, implied preemption.

Plaintiff’s claim for negligent failure to warn—he had dropped his strict liability claims from the first complaint—was so conclusory that it probably would have failed TwIqbal had preemption not been raised.  Back when Twombly and Iqbal came out, we welcomed the authorization for courts to look at the plausibility of claims based on facts asserted in the complaint, which was a step up from the standard practice of allowing conclusory allegations of the elements of claims to suffice.  Still, winning preemption of a warnings claim as to a branded prescription drug at the motion to dismiss stage was a long shot back then.  Once Albrecht upgraded the Levine mess, however, it became clear that federal courts do have to decide preemption as a matter of law.  The majority position, expressed most notably in the Second Circuit’s Gibbons decision (discussed here and ranked here), is that plaintiffs who want to avoid preemption of their branded prescription drug warnings claims have to plead that there was newly discovered evidence after approval of the drug’s label that would have allowed the manufacturer to change the relevant warnings unilaterally under the CBE regulations.  (We carved out branded prescription drugs from generic prescription drugs because the manufacturers of the latter can never change the substance of their labels unilaterally.)  Mizuno is one of several California federal courts we have seen apply this requirement without making up some loophole (such as the need for the plaintiff to get lots of discovery before it can plead a non-preempted claim).  One was the Owens case discussed here, which Mizuno cited along with Gibbons and cases we discussed here and here.  Id. at *4.  With the proper standards in mind, it was clear that plaintiff’s generalities did not identify newly acquired evidence that could have led to a labeling change through a CBE submission:

Here, the first amended complaint alleges in general and conclusory fashion that “[f]ollowing Rezurock’s approval, Defendants received and possessed post-marketing adverse event reports through FDA-mandated reporting systems, including FAERS” and “[o]n information and belief, Defendants’ pharmacovigilance systems—including FAERS data review, internal signal detection algorithms, and periodic safety update reports—identified dermatologic events [of increasing severity and chronicity] that differed in severity, persistence, and clinical presentation from those described in the approved labeling.”

Id.  Without facts supporting a basis for the use of a CBE, plaintiff’s warnings claim was preempted.  The same ruling held for the slightly repackaged negligence claim and the strict liability failure to warn claim that plaintiff had dropped.  Id. at *5 & 6.

Negligent design was even easier to ditch, a dynamic we have lauded more than a few times.  Changing the drug’s design to avoid the alleged defect of a risk of skin injuries would have been a major change that a manufacturer cannot make unilaterally.  Id. at *5.  Under Bartlett and some progeny within the Ninth Circuit, that claim is preempted.  Id.

Lastly, plaintiff asserted an implied warranty claim that California does not recognize for prescription drugs.  Id. at *5.  Plaintiff’s proposed pivot to pleading an express warranty claim instead was rejected because “[t]he court finds no indication that plaintiff can plead facts to state a plausible claim,” given his inability to plead facts about specific representations in his two complaints.  Id.  Again, with the added context of the drug’s indication and plaintiff’s actual medical situation, it is hard to imagine an express warranty from the manufacturer directly to the plaintiff that the medication would prevent any worsening of his preexisting skin problems.  The actual physician labeling, which plaintiff surely never read, and the patient brochure clearly made no such promise.  To the extent the alleged warranty came from the labeling, it would have been preempted anyway.  That was the end of Mizuno, dismissal without leave to file a third complaint.  Not waiting for the inevitable third, or even fourth, strike is good practice.

Photo of Bexis

Some of us are on the mailing list of the Rabiej Litigation Law Center.  Recently, we received an email entitled “Rare Mass-Tort MDL Consensus Among Opposing Counsel & Judges.”  Since such a broad consensus is indeed “rare,” we read the email.  Apparently, a recent conference resulted in “a rare consensus:  holding ‘pilot trials’ very early in an MDL provides substantial value to everyone involved.”  Here is how the email described the “pilot trial” concept:

[A]ttendees were confident that pilot trials could successfully substitute for later bellwether trials, so long as the main issues across the cases are properly represented.  These pilot trials would quickly familiarize the court with the central issues, paving the way for early preemption and Daubert rulings while offering both parties clear insight into potential trial outcomes to foster earlier resolutions.

*          *          *          *

In addition to ”picking representative pilot cases, . . . participants highlighted several priority topics . . .”:

  • Judicial Selection of Cases:  Having judges choose pilot cases avoids the typical disagreements between opposing counsel.
  • Standardized Discovery:  Product-liability standard case timelines for document production and expert disclosures should apply to pilot trials.
  • Binding Preemption & Daubert Precedent:  Preemption and Daubert rulings in pilot trials ought to bind all cases in the MDL, provided notice and an opportunity to be heard are given.

The email mentioned only one example:  that a “pilot-trial proposal was introduced in the Depo-Provera MDL” but never came to fruition due to a settlement.  Our understanding is that the Depo-Provera proposal really wasn’t much different than a bellwether trial, only earlier.  Nonetheless, from a defense perspective that proposal had several meritorious features:

  • The chosen “pilot” plaintiffs were reasonably representative; they were not either side’s best or worst cases.
  • Plaintiffs were not allowed to rig the selection process by dismissing their weakest cases to maximize the possibility of a good result.
  • The initial work-up centered on two defense-friendly issues:  general causation and preemption, which ensured focused discovery and motion practice.
  • As already mentioned, the overall handling of cases in the Depo Provera MDL required early plaintiff vetting on proof of use and injury

Still, as to the “pilot trial” concept generally, color us unconvinced for essentially the same reasons that we don’t like the aforementioned “bellwether trial” concept either.  As Bexis wrote about a similar MDL conference back in 2018:

There was considerable discussion of bellwether trials, but we have doubts that rules could solve the problems that were mentioned.  Our view is to interpret the MDL statute as written, and restrict MDLs to “coordinated or consolidated pretrial proceedings.”  28 U.S.C. §1407(a) (emphasis added).  A textualist approach to the MDL statute would, and we think should, mean no trials, “bellwether” or otherwise, in the MDL transferee court.  Simple.  No rules change needed. . . .  [I]f someone wants to preserve it for appeal, this issue could be the next Lexecon.

(Emphasis original).  If “we’re all textualists now,” as Justice Kagan famously announced, then the “pretrial” limitation that Congress created in the MDL statute (28 U.S.C. §1407), should mean just that – “pretrial proceedings” only.  Thus, any sort of “trial” goes beyond the power Congress conferred in the MDL statute, and the Supreme Court would likely so hold, if the issue ever came before it.

We are bolstered in that respect by what the Supreme Court recently held about statutory interpretation in Loper Bright Enterprises v. Raimondo, 603 U.S. 369 (2024):

[S]tatutes, no matter how impenetrable, do − in fact, must − have a single, best meaning.  That is the whole point of having written statutes; every statute’s meaning is fixed at the time of enactment.  So instead of declaring a particular party’s reading “permissible” in such a case, courts use every tool at their disposal to determine the best reading of the statute and resolve the ambiguity.

In an agency case as in any other, though, even if some judges might (or might not) consider the statute ambiguous, there is a best reading all the same. . . .  In the business of statutory interpretation, if it is not the best, it is not permissible.

Id. at 400 (citations and quotation marks omitted).

The permissible scope of multidistrict proceedings is set by a statute, 28 U.S.C. §1407.  That statute provides, in pertinent part:

When civil actions involving one or more common questions of fact are pending in different districts, such actions may be transferred to any district for coordinated or consolidated pretrial proceedings. . . .  Each action so transferred shall be remanded . . . at or before the conclusion of such pretrial proceedings to the district from which it was transferred. . . .

(b) Such coordinated or consolidated pretrial proceedings shall be conducted by a judge or judges to whom such actions are assigned. . . .  The judge or judges to whom such actions are assigned . . . may exercise the powers of a district judge in any district for the purpose of conducting pretrial depositions in such coordinated or consolidated pretrial proceedings.

(c) . . . The panel shall give notice to the parties in all actions in which transfers for coordinated or consolidated pretrial proceedings are contemplated. . . .  

Id. §1401(a-c) (emphasis added).  From reading this statutory language, we think, first, that Congress’ limitation of the scope of MDLs to “pretrial proceedings” was intentional, and second, that the “best,” and therefore only permissible, construction of “pretrial” cannot include “trials” of any sort.

So regardless of both sides’ agreement, either in an MDL or in an MDL-related conference, that a “pilot trial” might be a good idea, as §1407 currently stands, holding one would be ultra vires.  There are plenty of non-“trial”-related ways to obtain binding MDL-wide pretrial rulings on issues like preemption and Rule 702 (we don’t say Daubert).  One example is the recent show-cause order in the Talc MDL.  Creating another atextual “trial” variant isn’t one of them under the MDL statute as it currently stands.

Photo of Michelle Yeary

Every law student learns the same tidy hierarchy on day one of Torts. Negligence is harder to prove than strict liability, because strict liability does not require showing the defendant did anything wrong, only that the product itself was defective. And recklessness is harder still than negligence, reserved for conduct that goes well beyond a failure to use reasonable care. It is a clean pyramid, narrower at each level. Today’s case takes that pyramid, turns it upside down, balances it on its point, and calls it a ruling.

The case is Apuzzo v. Karl Storz Endoscopy-America, Inc., 2026 WL 2494546 (D. Conn. Aug. 25, 2026), involving a broken urological device. We filed this one away at first as a routine malfunction-theory dismissal, the kind we can write up in three paragraphs and move on. Then we kept reading, and the ruling kept getting stranger, all the way to a punitive damages claim that survives on the very allegation the court used to save the design defect and negligence claims.

Plaintiff underwent a routine cystoscopy, during which a cystoscope sheath was inserted exactly as intended. During the procedure the sheath fractured and its tip detached inside the bladder, requiring surgical removal. Id. at *1. Plaintiff’s amended complaint pressed three theories of product liability under the Connecticut Product Liability Act (“CPLA”) — strict liability, negligence, and recklessness. The defendant manufacturer moved to dismiss the whole thing because even after one amendment prompted by the court’s own suggestion that it needed more factual specificity, the allegations are thin.

Plaintiff tried to plead strict liability four different ways — malfunction, manufacturing defect, design defect, and failure to warn. Three of the four went nowhere. Start with the malfunction theory, which is usually a plaintiff’s best friend in a broken-device case because it lets circumstantial evidence stand in for a specific defect. The catch in Connecticut is that the theory only kicks in when the product is unavailable for inspection because it was lost, destroyed, or otherwise inaccessible. Id. at *3. Here, the metal fragment that broke off inside the plaintiff was surgically retrieved. Plaintiff never alleged the device was unavailable for inspection, and an argument about unavailability raised for the first time in an opposition brief does not count. Id. at n.1. Malfunction theory dismissed.

Manufacturing defect fared no better. Plaintiff’s entire theory boiled down to alleging that the device fractured, and fractures like that do not happen absent a defect, therefore it was defectively manufactured. Id. at *4. That is circular reasoning dressed up as a factual allegation. The court called it what it was – a conclusory pleading that identifies no specific component and no deviation from any manufacturing standard or specification. Nothing beyond the bare fact of the break itself. Manufacturing defect dismissed. Id.

Failure to warn went down for the same reason it almost always does at the pleading stage when a plaintiff phones it in. The complaint recited that the warnings were inadequate and that adequate warnings would have prevented the injury, without a single factual allegation about what the warnings actually said or how they fell short. Conclusory in, conclusory out. Failure to warn dismissed. Id. at *5.

So far, so unremarkable. Then we get to design defect, and the court finds the allegations “barely” sufficient. Id. But apparently that is enough in Connecticut, a state admittedly generous on design defect pleading. The standard does not require a plaintiff to identify a specific defect so long as there is some evidence of an “unspecified dangerous condition.” Id. at *4-5. But look at what actually got plaintiff over the line here. The device fractured during normal use, and before the plaintiff’s procedure, the defendant had received other complaints of metal fatigue and fracture in the same device. That is the entire factual predicate. No allegation about what feature of the design caused the fatigue. No allegation about an alternative design that would have avoided it. No allegation about a manufacturing or engineering standard the design fell short of. Just — it broke, and apparently it has broken before.

Strip away the label and squint, and that is the malfunction theory wearing a different coat. The court spent an entire section three pages earlier explaining why a plaintiff can’t get past the door on malfunction theory without a threshold showing of unavailability. Then it let functionally the same inference through the design defect door, with the only addition being an allegation of prior unspecified complaints that says nothing about the design at all. If prior complaints of breakage, without any indication of what aspect of the design was responsible, are enough to plead an unspecified dangerous condition, then a plaintiff never has to identify anything about the design in the first place. The device broke, other units apparently broke too, ergo the design is defective. That is malfunction theory with an extra data point.

Here is where the pyramid really tips over. The amended complaint never got around to pleading a standalone negligence count, but the court found enough scattered negligence-flavored allegations buried in the strict liability count to construct one anyway and rule on it. Negligence requires proof of a breached duty of care, conceptually a heavier lift than strict liability’s focus on the product’s condition rather than the manufacturer’s conduct. So, what did the court have to work with? The same allegation that got design defect over the line — that the defendant had received complaints of metal fatigue and fracture in the device before this plaintiff’s procedure. Id. at *5-6. That is the entire factual basis for the claim. How many complaints? What did they say? Which component, which failure mode, which use conditions? What warnings, if any, already accompanied the device regarding fracture risk? In other words, what facts support an inference that the defendant’s conduct could be judged unreasonable? None of that appears anywhere in the complaint, and none of it troubled the court. The allegation of unspecified prior notice was enough, standing alone, to make it plausible that the defendant knew or should have known the product was dangerous and unreasonably failed to do anything about it. Id. A claim that requires more than strict liability, not less, survived on categorically less information than the strict liability claims the court had just dismissed as too conclusory to proceed.

If you thought the ladder from strict liability to negligence collapsing was strange, wait until you see what happens to recklessness. Recklessness, under Connecticut law, is supposed to require more than negligence and more than gross negligence. The standard is “highly unreasonable conduct, involving an extreme departure from ordinary care, in a situation where a high degree of danger is apparent.” Id. at *7.  It is meant to be the hardest tier to plead, reserved for conduct a jury could punish with punitive damages precisely because it goes well beyond a failure to exercise reasonable care.

The court let the punitive damages claim survive on the theory that the defendant was aware of but failed to respond to notice that its devices had fractured during normal use, and that continuing to sell the device after that notice amounted to a conscious disregard of known risk. Read that sentence again, then flip back to the negligence holding, which survived on the theory that the defendant knew or should have known its product was dangerous and sold it anyway. Those are not two different theories supported by two different sets of facts. They are the identical allegation, doing double duty as both the floor of the pyramid and the ceiling. If the same recycled prior-complaints allegation satisfies both negligence and recklessness, then recklessness has been quietly redefined as whatever gets a defendant past a motion to dismiss on negligence, plus nothing.

This opinion gets the easy calls right and the hard ones backwards. Malfunction theory, manufacturing defect, and failure to warn were all correctly dismissed for exactly the kind of conclusory, circular pleading that should not survive TwIqbal in any circuit. But then, the court used a single unadorned allegation, that the defendant had received unspecified prior complaints about fracture, to sustain design defect and negligence claims, and to bootstrap that same negligence claim into a punitive damages theory. One allegation, three separate legal thresholds cleared, each threshold supposedly requiring more than the last. The pyramid was supposed to get narrower as you climb. Here, it is the same width from the ground floor to the penthouse. So the case moves forward on the strength of an allegation that was, by the court’s own word, barely enough.

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