Photo of Lisa Baird

Today’s guest post comes from our Reed Smith colleague Jamie Lanphear on a topic near and dear to the Blog’s heart: The new EU Product Liability Directive. As always, our guest posters deserve 100% of the credit, and any blame, for their posts. But, also as usual, our guest posters deliver the goods, so we expect there will be none of the latter to be had. Take it away, Jamie.

*********

While the Blog previously covered the new EU Product Liability Directive (PLD), formally known as Directive (EU) 2024/2853 of the European Parliament and of the Council of 23 October 2024 on liability for defective products and repealing Council Directive 85/374/EEC (Nov. 18, 2024), the prior posts, which can be found here and here, discussed the PLD’s overhaul of product liability law. This post zeroes in on aspects of the Directive related to software, cybersecurity, and digital products.

Passage of the PLD marks a watershed moment for technology companies, software developers, and any business placing digital products on the European market, including medical device companies whose products include software functionality.

Software as a Product: A Paradigm Shift

One of the most significant changes in the new PLD is the explicit inclusion of software—whether embedded, stand-alone, or delivered as a service—within the definition of a “product”:

Products in the digital age can be tangible or intangible. Software, such as operating systems, firmware, computer programs, applications or AI systems, is increasingly common on the market and plays an increasingly important role for product safety. Software is capable of being placed on the market as a standalone product or can subsequently be integrated into other products as a component, and it is capable of causing damage through its execution. In the interest of legal certainty, it should be clarified in this Directive that software is a product for the purposes of applying no-fault liability, irrespective of the mode of its supply or usage, and therefore irrespective of whether the software is stored on a device, accessed through a communication network or cloud technologies, or supplied through a software-as-a-service model. (Emphasis added)

This means that software, firmware, applications, AI systems, and even digital manufacturing files are now subject to the same strict liability regime as traditional physical goods. The Directive also covers integrated and interconnected digital services, such as “a health monitoring service that relies on a physical product’s sensors to track the user’s physical activity or health metrics.”

This expanded scope is intended to reflect the reality that software is now integral to product safety and performance. For companies, this means that, after the date the PLD goes into effect (December 9, 2026) any defect in software, potentially including vulnerabilities or failures in digital services, may trigger liability under in the EU, if it leads to harm.

Cybersecurity Vulnerabilities as Product Defects

The PLD’s new approach to cybersecurity is closely intertwined with the EU’s broader regulatory framework for digital product security. One important piece of legislation in this area is the EU Cyber Resilience Act (CRA), which, together with the NIS2 Directive and sector-specific rules, sets out mandatory cybersecurity requirements for a wide range of digital products and services.

• Mandatory Security Requirements as a Benchmark Defect: Under the new PLD, non-compliance with “safety-relevant cybersecurity requirements” can form the basis of product defectiveness. For example, the CRA requires manufacturers to implement security-by-design, conduct risk assessments, provide security updates, and ensure secure default configurations for products with digital elements. If a company fails to meet these requirements, and a vulnerability leads to harm, non-compliance with the CRA may, in turn, be used to establish defectiveness under the PLD.

• Failure to Provide Security Updates: Both the PLD and the CRA impose ongoing obligations to provide software security updates throughout a product’s lifecycle. Under the PLD, a product is defective if the manufacturer fails to supply necessary updates or patches to address vulnerabilities, providing such updates are within the manufacturer’s control. The CRA similarly requires manufacturers to monitor for vulnerabilities and issue timely updates. If a cyberattack exploits an unpatched vulnerability and causes injury or property damage, the failure to update may provide the basis for strict liability under the PLD.

• Disclosure and Incident Response: The NIS2 Directive and the CRA require companies to have processes for vulnerability management, coordinated disclosure, and incident reporting. The PLD’s new rules on evidence and presumptions mean that if a company cannot demonstrate compliance with these processes, courts may presume defectiveness or causation in favor of the claimant under the PLD—especially in technically complex cases, as we discussed previously

Burden of Proof and Disclosure: Lowering the Bar for Claimants

The new directive also introduces procedural changes intended to make it easier for claimants to bring and succeed in product liability claims, including those involving software and cybersecurity:

• Rebuttable Presumptions: If a claimant faces “excessive difficulties” in proving defectiveness or causation due to technical or scientific complexity (as is often the case with software or AI), courts can presume defectiveness and/or causation if the claimant can show it is likely that the product was defective or that there is a causal link.

Or, in the words of the PLD: “National courts should presume the defectiveness of a product or the causal link between the damage and the defectiveness, or both, where, notwithstanding the defendant’s disclosure of information, it would be excessively difficulty for the claimant, in particular due to the technical or scientific complexity of the case, to prove the effectiveness of the causal link, or both.”

Notably, the Directive instructs courts, when evaluating technical or scientific complexity, to consider certain factors, including “the complex nature of the product, such as an innovative medical device; the complex nature of the technology used, such as machine learning; the complex nature of the information and data to be analysed by the claimant; and the complex nature of the causal link.”

• Disclosure of Evidence: Courts can require companies to disclose relevant evidence in their possession if the claimant makes a plausible case. Additionally, courts may also require that evidence be presented in an easily accessible and easily understandable manner.” The Directive explicitly calls out digital products as those embodying the sort of complexity envisioned: “Taking into consideration the complexity of certain types of evidence, for example evidence relating to digital products, it should be possible for national courts to require such evidence to be presented in an easily accessible and easily understandable manner, subject to certain conditions.”

• No Liability Waivers: Companies cannot contractually exclude or limit their liability under the directive and disclaimers for software defects or security vulnerabilities are not valid: “Member States shall ensure that the liability of an economic operator pursuant to this Directive is not, in relation to the injured person, limited or excluded by a contractual provision or by national law.”

In short, the new EU PLD signals the start of a new era in which software quality, cybersecurity, and ongoing product support are not just best practices—they are legal obligations. Companies placing digital products on the EU market may wish to evaluate their compliance, engineering, and risk management strategies with the Directive in mind.

Photo of Michelle Yeary

Almost 15 years ago to the day, we posted the following question: why do third party payers and not patients bring RICO claims against drug and device companies for behavior that supposedly makes products cost too much?  We were reporting on a case that dismissed the RICO claims of patients because they were indirect purchasers of their knee implants.  The indirect purchaser rule is a standing doctrine employed most notably in Illinois Brick Co. v. Illinois, 431 U.S. 720, 744 (1977), to bar Illinois, an indirect purchaser of concrete blocks, from bringing an antitrust claim under the Clayton Act against the concrete block manufacturers. Rejecting the argument that Illinois—two levels down the distribution chain from the manufacturers—should be allowed to recover the fraction of the overcharge “passed on” to them, the Supreme Court noted:

Permitting the use of pass-on theories … essentially would transform treble-damages actions into massive efforts to apportion the recovery among all potential plaintiffs that could have absorbed part of the overcharge from direct purchasers to middlemen to ultimate consumers. However appealing this attempt to allocate the overcharge might seem in theory, it would add whole new dimensions of complexity to treble-damages suits and seriously undermine their effectiveness.

Id. at 737.

But we said this post was about RICO claims.  And so it is.  Fifteen years ago, the District of New Jersey found that the rationale for the indirect purchaser rule in antitrust cases applied equally to bar the claims of patient-plaintiffs in RICO claims (the Third Circuit held the same 29 years ago, see McCarthy v. Recordex Serv. Inc., 80 F.3d 842 (3d Cir. 1996)).  Now, the Middle District of Florida reaches the same conclusion in Humana Inc. v. Teva Pharmaceuticals USA, Inc., 2025 U.S. Dist. LEXIS 258748 (M.D. Fla. Apr. 28, 2025), to bar an insurer’s RICO claims.

Defendants include the manufacturer of a drug to treat multiple sclerosis and two “specialty pharmacies” that dispense that drug to plaintiff’s insureds, among others.  Id. at *2.  Plaintiff alleges that the manufacturer drove up the price of the drug through charitable copay-assistance funding.  There are no allegations that the drug did not do what it was supposed to do, only that plaintiff-insurer had to pay the inflated prices.  Plaintiff brought RICO and conspiracy to violate RICO claims as well as a host of state law fraud and consumer claims.  Id. at *3-4.  Defendants’ motion to dismiss asked the court to “extend the Supreme Court’s indirect purchaser bar from Clayton Act cases to civil RICO cases.”  Id. at *7. 

While neither the Supreme Court nor the Eleventh Circuit have addressed the issue, every circuit court to have considered the question has held that the rule applies to civil RICO cases.  Id. (citing cases from 3rd, 6th, 7th, and 8th Circuits).  As have other district courts within the Eleventh Circuit.  Id. at *7-8.  While not binding, the weight of authority is “exceedingly persuasive.”  Id. at *8. 

 First, the RICO statute’s civil remedy provision is modeled after the civil-action provision of the Clayton Act.  Applying the same meaning to similar statutory language is a “well-known canon of statutory interpretation.”  Id.  This is especially true when the comparable provisions share the same purpose—as they do here.  Plaintiff offered and the court found no compelling reason to deviate from the Supreme Court’s interpretation of the same language in the Clayton Act. 

Second, the rationale of Illinois Brick applies to RICO cases.  RICO cases have the same “complicated web of damages at multiple levels,” as antitrust cases.  Id. at *9-10.  Further, Illinois Brick, directly advised lower courts in individual cases not to engage in “an unwarranted and counterproductive exercise to litigate a series of exceptions.”  Id. at *10.  Direction that the Middle District of Florida took to heart.

Third, intentionally mis-quoting an Eleventh Circuit decision, plaintiff argued that a civil RICO plaintiff has standing if his injuries were proximately caused by a RICO violation.  The quote actually reads:  “[A] plaintiff has RICO standing only if his injuries were proximately caused by the RICO violation.”  Id. at *11. The court viewed the omission of the word “only” as an attempt to change the meaning of the legal authority.  Standing and proximate cause are overlapping, but separate concepts.  The Eleventh Circuit case plaintiff attempted to use held that proximate cause is required to plead a civil RICO claim.  But it did not address standing or the indirect purchaser rule.  Id. at *11-12, & n.8.  Therefore, it is easily reconciled with Illinois Brick.  A civil RICO plaintiff “must be a direct purchaser and must demonstrate proximate cause to state a viable civil RICO claim.”  Id. at *11. 

Finally, plaintiff argued that it was a direct purchaser because it made payments to the pharmacies on behalf of its insureds who were prescribed the drug.  But insurance companies are “third-party payors” which by definition means they are not “end-payors” and therefore not direct purchasers.  Id. at *13. 

Plaintiff’s RICO and RICO conspiracy claims were dismissed with prejudice for lack of standing.  As those were the only federal question claims in the case, the court declined to exercise supplemental jurisdiction over the remaining state court claims and those claims were dismissed for lack of subject matter jurisdiction.              

Photo of Bexis

Recently, when putting together our “Staple Suit Cropped” blogpost about Kane v. Covidien LP, 2025 U.S. Dist. Lexis 25718 (E.D.N.Y. Feb. 12, 2025), we realized that, while we had a comprehensive 50-state survey on the questionable status of failure-to-report claims under state law, we did not have a similarly complete reference for preemption of the same reporting-based claims.

We’re rectifying that here.

Failure-to-report claims have been asserted against every product that has a preemption defense – branded drugs, generic drugs, and PMA medical devices.  Thus, there are different ways that failure-to-report claims end up preempted.

  • First, reporting-based claims against drugs or medical devices are impliedly preempted under Buckman Co. v. Plaintiffs Legal Committee, 531 U.S. 341 (2001), and 21 U.S.C. §337(a), because they would not exist without the FDA reporting obligations that they claim were violated.  Therefore, “the existence of these federal enactments is a critical element” of the cause of action, and implied preemption applies.  531 U.S. at 353.
  • Second, and relatedly, in the majority of states where no state-law claim exists for failure to make mandatory reports to a governmental agency (see the 50-state survey), Buckman further precludes such claims as purely private attempts to enforce the FDCA/FDA regulations concerning adverse event reporting.
  • Third, in cases involving pre-market approved medical devices, the same absence of any state-law reporting-based claims leads to express, as well as implied, preemption because there is no recognized “parallel” state-law theory of liability that could support a “parallel claim” exception to express preemption under 21 U.S.C. §360k(a).
  • Fourth, generic drugs enjoy their own implied preemption defenses under PLIVA, Inc. v. Mensing, 564 U.S. 604 (2011), and can take similar advantage of Buckman-based preemption precedent.

Obviously, there can be overlap between these three categories, and not all courts keep them separate.

Since the issue is preemption, a federal issue, our primary division of cases is by federal circuit rather than by state.  Of the circuits, the Second, Third, Sixth, Eighth, Tenth, and Eleventh all have precedential decisions holding failure-to-report claims preempted, although the Second has only dealt with express preemption.  The Second, Fifth and Ninth allow “parallel” failure to warn claims to escape preemption if state common law allows them, with the Second being stricter than the others.  The Seventh Circuit has been hostile generally to FDCA-based preemption, but hasn’t decided a reporting-based case.  The First, Fourth, and District of Columbia circuits have yet to decide the question.  We note that no precedential decision from any federal court of appeals has flatly denied preemption in a failure-to-report case since 2013, the 2013 decision was repudiated by the highest court of the state in question (see Ninth Circuit, below), and the United States Supreme Court abolished any “presumption against preemption” in express preemption cases in 2016.  See Commonwealth of Puerto Rico v. Franklin California Tax-free Trust, 579 U.S. 115, 125 (2016).  Thus, defendants have good grounds to seek reconsideration of what adverse appellate authority exists.

Finally, we don’t do the other side’s research for them, so be advised, that while we try to be comprehensive in collecting favorable cases, we aren’t including all adverse decisions. 

Continue Reading Preemption Round Up – Failure to Report
Photo of Steven Boranian

The Ninth Circuit filed its anticipated en banc opinion on personal jurisdiction last week, and the result is the broadening of Internet-based personal jurisdiction in an age of ubiquitous online commerce.  The district court in Briskin v. Shopify, Inc., No. 22-15815, 2025 U.S. App. LEXIS 9410 (9th Cir. Apr. 21, 2025), had ruled that there was no personal jurisdiction in California over a Canada-based Internet payment service provider merely because a consumer in California made a purchase, and a three-judge panel affirmed.  An en banc panel, however, disagreed and published an opinion that holds the defendant to answer purportedly because it “expressly aimed” its services at California, allegedly through its use of “cookies.’ 

The defendant in Briskin is an e-commerce platform that facilitates online sales.  Merchants use the defendant’s software and infrastructure to set up and manage their online stores, and the defendant processes payments and, in some cases, ships products to purchasers.  Throughout the transactions, the defendant’s participation is invisible to consumers, who allegedly see only the online seller when making their purchases.  Id. at *14-*15.

A key fact for the Ninth Circuit majority was that, during the plaintiff’s transactions, the defendant installed tracking cookies on the plaintiff’s phone—software files that allegedly allowed the defendant to track the plaintiff’s behavior, including geolocation, payment information, IP address, etc.  In his class action complaint filed in California, the plaintiff alleged that the defendant gathered and shared private information in violation of California law.  Id. at *15-*18. 

The defendant was actually a group of defendants—a Canadian corporation based in Ottawa and two subsidiaries incorporated in Delaware with principal places of business in New York and Delaware.  So they moved to dismiss for lack of personal jurisdiction, arguing that their platform was agnostic to California (and every other state) and that the mere happenstance that a California consumer made a purchase was insufficient to support personal jurisdiction.  In other words, if they were subject to jurisdiction in California, they would be subject to jurisdiction everywhere.

An unsympathetic Ninth Circuit started with International Shoe and followed a “purposeful direction analysis,” which “focuses on the forum in which the defendant’s actions were felt, whether or not the actions themselves occurred within the forum.”  Id. at *20-*23.  That analysis, in turn, came down to whether the defendants “expressly aimed” an intentional act that causes harm that the defendant knows will be suffered in the forum state.  Think the old law school hypothetical of an archer who fires an arrow across state lines.  That is an intentional harmful act expressly aimed at another forum. 

So what do we do with that in a time when packets of data are traveling across borders (usually without any particular aim) and not arrows?  It is by now established that mere passive nationwide accessibility to cyberspace does not demonstrate “express aiming” at everyone everywhere.  There has to be “something more.”  Id. at *26.  It is not entirely clear to us what the “something more” has to be, other than it must be more than just a foreseeable effect in the forum state.

For the defendants in Briskin, the Ninth Circuit’s majority opinion found express aiming at California because the plaintiff alleged that the defendants “targeted” California consumers to collect and exploit payment information and other personal identifying information “that it extracts from the software it permanently installs on their devices.”  Id. at *34-*35.  According to the majority, it was not mere “happenstance” that California consumers chose to do business with online merchants who used the defendants’ platform. 

Instead, “it is clear that [Defendant] expressly aimed its conduct at California through its extraction, maintenance, and commercial distribution of the California consumers’ personal data in violation of California laws.”  Id. at *36.  We are not sure how clear that is, but for the Ninth Circuit majority, the analogy was to someone who physically entered a Californian’s home to take personal information for its own commercial gain.  In that case, there would be “no doubt” over specific personal jurisdiction in California.  Id. at *37. 

The obvious problem with the Ninth Circuit’s conclusion is that the defendants operate nationwide without differentiation and thus did not “target” California any more than they “targeted” anywhere else.  The Ninth Circuit was again unsympathetic and rejected the notion that a nationwide company can do business everywhere, but be jurisdictionally nowhere except its principal place of business and state of incorporation.  Id. at *37-*38.  The court also expressly overruled cases requiring some sort of differential treatment of the forum state before finding “express aiming” sufficient to support specific personal jurisdiction.”  Id. at *40-*41.  “Express aiming” thus seems not to require any aiming at all. 

The defendants justifiably protested that the Ninth Circuit’s ruling could lead to specific jurisdiction in all 50 states.  The majority’s response was blunt:  “That may be true, but not unfair.” 

There were two concurring opinions.  For one concurring judge, the majority did not go far enough.  When the alleged conduct is purely automated, the operation of the automated system is the relevant tortious conduct, which occurred in California.  That forum conduct is sufficient to support specific personal jurisdiction.  There is no need for “something more.”  Id. at *56-*64.  Another concurring judge looked to the Constitution and due process and searched for analogies to physical presence.  Through that lens, it does not matter whether a defendant targeted the forum state over others, so long as the defendant is sufficiently present in the forum state through its alleged business operations.  Id. at *66-*77.

Finally, there was one dissent, who condemned the majority’s reliance on the defendants’ knowledge of the plaintiff’s location when they installed cookies on his phone.  Personal jurisdiction turns on the defendant’s contacts with the forum state—and not the people who reside there.  The dissenter also roundly mocked the majority for pegging jurisdiction on something as transient as software files on a mobile device. What if the plaintiff made his purchase in California, then traveled to Nevada or Oregon, toting his phone and his cookies all along the way?  Is there personal jurisdiction now in those states, too (“traveling cookie” jurisdiction)?  If the defendants’ inroads in California are so strong, then why not general jurisdiction—a position that not even the plaintiff has advanced?  Id. at *77-*84.

It is easy to view the Briskin opinion as California’s latest attempt to make every controversy justiciable in California, and maybe that is what it is.  Regardless, “something more” is still required to establish specific personal jurisdiction based on e-commerce, and Briskin will not be the last word. 

Photo of Lisa Baird

Five years ago, the COVID-19 pandemic had the world twirling off its bearings, but “Zoom” became a verb and that helped some.  Video conferencing not only became a way to stay connected to friends and family, it became a lifeline that allowed lawyers to continue to take depositions, appear for oral arguments, and even conduct trials.

Remote appearances are so commonplace now, it is easy to assume that everyone knows the rules about what to do and what not to do.  That the days of a lawyer having to declare that “I am not a cat” are long behind us.

Yet, people are people, and mishaps still occur.  Today’s case is a little old (January 2024), but cautionary tales remain evergreen.  Hernandez v. La Fortaleza, Inc., No. A-0367-22, 2024 WL 65217, 2024 N.J. Super. Unpub. LEXIS 22 (N.J. Super. Ct. App. Div. Jan. 5, 2024), involved testimony during the virtual trial of a slip-and-fall matter, but the mistakes just as easily could have occurred during a remote deposition (we hope, never a virtual trial) in a pharmaceutical or medical device lawsuit.

The plaintiff was called as the first witness, and on direct was asked to show on a photograph the location where she supposedly had fallen. 

She provided a vague verbal answer and fumbled a bit with her computer’s cursor, and that seems to have resulted in some off-camera prompting by her husband. The court admonished the husband to remain quiet and visible to the camera at all times. 

The trial then broke for lunch.  The court instructed plaintiff’s counsel that he could take the time to straighten out the exhibit and how to use the cursor, but warned the husband not to say anything.  Plaintiff’s counsel had agreed that he would not talk to plaintiff “about her testimony at any time during the testimony, even if” the court broke for lunch, and defense counsel also reminded everyone there should be “zero coaching” during the break.

You know what occurred next:  The plaintiff’s attorney proceeded to coach his witness − pretty blatantly.   

As overheard by the court’s clerk and captured by the court’s recording system:

[Attorney]: This is the important part of the case. You show this picture, okay? Okay? and I’m going to say, I don’t know if we got this far already but do you recognize this picture? Yes. This side of the restaurant? Does it show – does it fairly and accurately show the way the restaurant looked on the day that you fell? Yes. You must say that or the picture cannot be good. Okay? So, I want you – and the answer has to be, yes, because if you say, no, we can’t do it. But you will say the same thing, I’m going to ask you the same question later.

Anyway, okay. Do you see – do you see the bench that you were going to at the time you fell? You’re going to answer, yes, right. And I will say to you, I’m going to move the cursor – and I will say I’m going to move the cursor. You tell me where is the bench? So, when I get up here, just, right there. Okay? The bench, right next to the lady. Okay? Okay.

Now in this picture, do you see where you fell and I’m going to put it right where the – where the bench – from where the bench it, where did you fall?

I will move the map. It’s out of the (indiscernible). So, after we identify the bench, I’m going to put the [cursor] right here now and say, do you see the area where you fell? Yes. Okay. Now, how do I have to move the – you – because it has to be her voice. How do I move the cursor to find the place where you fell?

So you should —

[Husband]: Back.

[Attorney]: Okay. So, down, right, like that?

[Attorney]: So, you say, move down. So, I’m going to move it, move it, move it. You have to tell me when to stop. Right there, right? Okay. How about this distance from the curb? Would it be right this way or further here? It would be in the middle?

[Husband]: Yeah.

[Attorney]: Okay. So, we’re going to – first we find the bench. That’s the easy part. Then I’m going to say, okay, I’m going to put the cursor right in front of the bench. Now, tell me how to move the cursor to show how you fell and what you would have to say, you have to move – you could either say down or to the camera, okay, which you prefer.

[Plaintiff and husband speaking in Spanish].

[Attorney]: Okay. So, you’re saying down? So, I’ll say, okay, I’m going to start to move it down and tell me when to stop. I’m going to go down, down, down, down, down.

[Plaintiff]: Stop.

Not surprisingly, the judge concluded that a mistrial was required, and later dismissed the complaint with prejudice and ordered plaintiff’s counsel to pay the fees and costs of the trial and the motion to dismiss.

Because of the coaching, “[t]he well of information that could be presented to the jury has been forever poisoned. There is no way to sufficiently determine that [p]laintiff can testify, especially considering that [p]laintiff had no recollection of her own fall and her attorney invited her to perpetuate a falsity at his direction to create an issue of liability.”  Even worse, the coaching came immediately after admonitions to plaintiff’s counsel and her husband.

Not content to let a bad situation be, plaintiff and her counsel decided to make it worse:  They filed an appeal arguing that there had been “no fraud on the court” and that the sanction of dismissal was “too severe.”

Was there a fraud on the court?  Check. 

A fraud on the court occurs

“where it can be demonstrated, clearly and convincingly, that a party has sentiently set in motion some unconscionable scheme calculated to interfere with the judicial system’s ability impartially to adjudicate a matter by improperly influencing the trier or unfairly hampering the presentation of the opposing party’s claim or defense.”

2024 WL 65217 at *5. 

In this instance, both the plaintiff’s attorney and the husband “sentiently set in motion” a scheme by directing plaintiff to say where the accident had occurred when she had no independent recollection.

Was the dismissal sanction too severe?  Nope.

Although dismissals with prejudice are a drastic remedy to be used “only sparingly” the wrongfulness of the coaching was clear, the plaintiff and her husband were “willing participants” in the wrongful conduct, the coaching “poisoned” and “tainted” the judicial process, and no lesser sanction could undo the harm caused.

The ethical violations came to light in Hernandez due to a technology mishap, which makes us wonder how much other coaching goes on without notice.  To help guard against bad actors on the other side, our Remote Depositions in MDLs 2.0 post can help point you to some best practices.  For best practices concerning conversations with witnesses during breaks, see our Depositions – When Can You Talk To Your Own Witness? post.

Photo of Stephen McConnell

Late last year we said that almost every legal conference these days has a session on artificial intelligence. It is de rigeur. That is also true with respect to litigation funding. It is a hot issue. Our Inn of Court (University of Pennsylvania) did a presentation on litigation funding that, despite the fact that it is such a well-trodden topic, managed to show us a few things we had not seen before. For example, in some cases it turned out that the lawsuit was being funded by a competitor of the defendant.  When that competitor litigation funding comes from overseas, it seems even more menacing.  And, of course, most of you remember how a tech billionaire with a grudge against a media outlet funded a celebrity’s case against said media outlet. Hello, lawsuit. And, after a ruinous verdict, goodbye, defendant. 

But the issues surrounding litigation funding are usually of a more quotidian sort.  A recent example is Gilead Sciences, Inc. v. Khaim, 2025 U.S. Dist. LEXIS 75105, 2025 WL 1151412 (E.D.N.Y. April 21, 2025). In resisting discovery of third-party litigation funding, parties often make claims that the information is privileged.  This case holds that information about who is paying for a party’s lawyer and how is not necessarily privileged.  Quotidian or not, the case is still interesting.

Actually, maybe quotidian is the wrong word. Gilead is an unusual case, with a drug company as the plaintiff in civil litigation against defendants alleged to be involved in counterfeiting the plaintiff’s FDA-approved drugs.  The drugs were for the treatment or prevention of HIV. Following the receipt of multiple complaints from patients and pharmacies who had purchased counterfeit drugs bearing the plaintiff’s brand name, the plaintiff manufacturer filed a complaint against various defendants for trademark infringement, false descriptions, false advertising, and trademark dilution, all in violation of the Lanham Act, as well as deceptive business practices in violation of New York General Business law section 349, and common law unjust enrichment and unfair competition. If the plaintiff succeeded in its action against the defendants, one form of available damages would have been to recover the defendants’ “ill-gotten profits.” 

The plaintiff sought to test a representation by one of the defendants to the effect that she “earned almost no money from the multi-million dollar fraud for which she helped launder and conceal the profits, and that her legal fees are not paid out of those concealed profits but rather from generous benefactors.” (Any fan of American theater cannot help but think of Blanche Dubois’s last, sad statement in “A Streetcar Named Desire” that she had “always depended on the kindness of strangers.”) To determine that defendant’s ability to satisfy a judgment, the plaintiff sought discovery (Rule 45 requests for production) from the defendant’s counsel about who was paying the defendant’s fees and how.  The defendant claimed privilege and the court rejected the claim.  

There were extensive negotiations between the plaintiff and the defendant regarding the discovery. The defendant was facing a parallel criminal proceeding and wanted to postpone the discovery until the criminal proceeding terminated.  Her lawyers were good ones and, presumably, expensive. The defendant eventually entered a change of plea – changing from the initial not guilty plea to a guilty plea.  That guilty plea should have helped bring the discovery negotiations to some sort of compromise, but then the defendant wanted to postpone discovery until her sentencing – which still has not taken place.  The plaintiff manufacturer agreed to limit the time period for the fee payment information, but the parties were still at loggerheads.  Thus, the court needed to decide the discovery issue.

Here is what the Gilead court decided:

First, the court held that the requested discovery was not privileged.  Fee arrangements are not within the attorney-client privilege because “they are not the kinds of disclosures that would not have been made absent the privilege and their disclosure does not prevent the attorney from rendering legal advice.”  Thus, information that fees were paid by third persons may be sought to determine the identity of a benefactor.  If there is a headline to this case, that is it.

Under the relevant case law, there can be “special circumstances” that protect disclosure of such information, but they are limited to situations where disclosure really would reveal something privileged. But here, no special circumstances existed that could result in wrecking some privilege.  Information is not privileged merely because the party may strongly fear the effects of disclosure.  (Any ex-prosecutor cannot help but think of the old joke about how the most common objection by criminal defense attorneys is, “Objection, your Honor: prejudicial – tends to show guilt.”)  The defendant failed to show that complying with the discovery request would “inhibit the ordinary communication necessary for an attorney to act effectively, justly, and expeditiously.”  Moreover, the defendant had not shown that production of the information would “affect her ability to obtain informed legal advice in the instant case and in the Criminal Action.”

Second, the court held that the requested discovery was relevant and proportional.  As we all know, the standard for discovery under Fed. R. Civ. P. 26(b)(1) is broad.  A subpoena to a third party per Rule 45 is subject to the Rule 26(b)(1) standard.  The information “concerning Defendant’s financial status, including her ability to pay her legal fees” bears on the plaintiff’s claim for damages available under the Lanham Act.  The plaintiff’s agreement to narrow the time-frame of the request aided it in persuading the court that the information sought was “proportional to the needs of the case.” 

Accordingly, the Gilead court held that the plaintiff manufacturer was entitled to “all non-privileged documents and information sufficient to show (1) the source of all payments from or on behalf of Defendant to [her law firm] since the commencement of her representation in the Criminal Action; and (2) the dates and amounts of all payments from or on behalf of Defendant received by [her law firm] since commencement of her representation in the Criminal Action.”

The implications of the Gilead ruling are potentially far-ranging. You and your clients might like some and dislike others. At a minimum, the next time you are in a squabble over discovery of third-party litigation funding, the Gilead case might offer useful ammo.

Photo of Michelle Yeary

We know that any federal court analysis that starts with Although the state has not recognized the duty . . . is going to be followed by a “prediction” of state law that instead creates unprecedented liability according to the federal court’s personal predilections.  Which is precisely what the court did in CLF v. Coopersurgical, Inc., 2025 U.S. Dist. LEXIS 61420 (D. Ore. Mar. 31, 2025). 

We’ve posted many times that under the Erie doctrine, in the words of the Supreme Court:

[a] federal court in diversity is not free to engraft onto those state rules exceptions or modifications which may commend themselves to the federal court, but which have not commended themselves to the State in which the federal court sits.”

Day & Zimmerman, Inc. v. Challoner, 423 U.S. 3, 4 (1975).  Which is also the law of the Ninth Circuit, which includes Oregon. See Hemmings v. Tidyman’s Inc., 285 F.3d 1174, 1203 (9th Cir. 2002) (federal courts should not predict “potential changes” to state law).  But the CLF court did not feel so constrained in deciding that Oregon would recognize a claim for failure to recall despite the nationwide precedent rejecting such a theory of liability. 

CLF involves claims for the destruction of embryos allegedly due to a culture medium manufactured by defendant that did not contain sufficient levels of magnesium needed for the embryos to survive.  Defendant moved to dismiss on several grounds, nearly all of which were denied and since we aren’t really in the business of touting defense losses, we are not going to discuss each one.

But, creating a heretofore unrecognized claim for failure to recall is the type of bad decision that we can’t just let pass by.  As discussed in much more detail in this post, failure-to-recall theories are among the most widely debunked purported “torts” ever.  Even the Third Restatment of Torts rejected any purported common-law obligation either to recall a product in the absence of any governmental order, or for the anticipatory removal of products from the market earlier than any governmental recall required.  As have almost every court to rule on the issue. 

Despite acknowledging that Oregon has “not recognized a distinct duty to recall” for any product, nowhere does CLF mention Erie or the standards for predictions of state law.  Had it, and in the absence of lower court decisions, the district court likely could not have turned a blind eye to the Restatement or the vast majority of decisions going the opposite way.  Nor does its short discussion address that the product at issue is an FDA-regulated device—which is important because the FDA supervises recalls.  The lack of thoughtful analysis makes the expansion of state law even more egregious. 

Contrary to Erie’s caution against federal courts expanding state tort law, CLF bases its conclusion on nothing more than an unsupported assertion that failure to recall is “consistent with” “broader” state product liability law because “a product recall is a warning.”  CLF, 2025 U.S. Dist. LEXIS 61420, *19-20.  This too ignores that most courts have found failure to recall claims more akin to stop-selling claims, which is why they reject them.  In fact, in this case the product was recalled. So, plaintiff’s argument really isn’t that it should have been recalled, but that it should have been recalled earlier or essentially—never sold.  

As far being analogous to warnings, CLF rests its holding on failure to recall being similar to a failure-to-update warning claim.  Another contrived claim that emerged in generic drug litigation for the sole purpose of threading the eye of the needle between Mensing and Buckman preemption.  Hardly a solid basis on which to predict the expansion of state law to recognize a new basis for liability.

Sitting in diversity, federal courts must be cautious when making pronouncements about state law. It’s not enough that no court might have had occasion to reject a novel theory like failure to recall nonetheless being a basis for liability.  There must be state-law precedent somewhere affirmatively supporting a claim for a federal court to predict its adoption.  No such Oregon law exists, and CLF oversteps.

Photo of Bexis

Not too long ago we saw a story in the legal press about a newly filed case in Minnesota where the plaintiff claimed that the FDA was going easy on the defendant because it approved “hundreds of premarket supplements” rather than requiring “a new PMA application.”  Supposedly “[b]y utilizing the [premarket approval] supplement process instead of filing a new PMA application, [the defendant] avoided the rigorous scientific review, public comment and clinical trial requirements.”

That’s barnyard excrement.  This plaintiff isn’t just wrong s/he is loud wrong – which, for plaintiffs, is unfortunately rather common.

Continue Reading Supplemental Authority

Today’s opinion, In re SoClean, Inc., Mktg., Sales Pracs., & Prods. Liab. Litig., No. 22-MC-00152-JFC, 2025 WL 974258 (Sp. Mstr. W.D. Pa. Mar. 20, 2025), involves a lot of case-specific discussion with little applicability to the broader readership of the Blog. But it also contains some general observations regarding invading the province of the FDA that are “so fresh and so clean” (if this litigation name takes you back, as it does us, to circa 2000 Outkast).

Continue Reading Same Rule, Different Setting: Litigants Cannot Usurp the FDA’s Authority
Photo of Eric Alexander

For several years now, the Valsartan MDL has been something of a poster child for the problems with modern serial product liability litigation.  It started with questionable data coming out of a questionable lab, leading to publicity and regulatory actions that outpaced reliable evidence of increased risk from an alleged carcinogenic contamination.  It snowballed from there, which we have chronicled here.  Along the way, there has been an obvious contrast with the many good rulings coming out the Zantac MDL, which is also a litigation based on questionable data coming out of a questionable lab, which led to publicity and regulatory actions that outpaced reliable evidence of increased risk from an alleged carcinogenic contamination.  The law in the Eleventh Circuit, where the Zantac MDL’s court is, is not so much more defense friendly than the law in the Third Circuit, where the Valsartan MDL’s court is, to explain the differences in results between the two MDLs formed about a year apart.  As far as we can tell with some gross overgeneralization, the facts and science are not that different either.  We are loathe to talk about specific judges, but we cannot ignore the possibility that the explanation is that the judge assigned to the Valsartan MDL for its first five years and a few months was simply overly inclined to the plaintiffs’ positions.  In May 2024, the original Valsartan MDL judge retired and the MDL was reassigned to the Chief Judge of the United States District Court for the District of New Jersey.  We now have our first major substantive ruling under the new regime, and things seem to be different now.

The signal that there is a new sheriff in town, without overtly dumping on the prior sheriff, is quite clear in In re Valsartan, Losartan, and Irbesartan Prods. Liab. Litig., MDL No. 19-2875 (RMB/SAK), 2025 U.S. Dist. LEXIS 66185 (D.N.J. Apr. 7, 2025) (“Valsartan ‘25”).  It was obvious that plaintiffs pushed the narrative that the prior judge’s rulings precluded the new judge from ruling on the motion before her.  We will not belabor the rejection of those arguments and the skill evident in avoiding saying that any prior decision was wrong.  The ruling in Valsartan ‘25 related to the admissibility under Rule 702 of plaintiffs’ economist expert’s opinions offered in support of the express warranty claims asserted in a third party payor (“TPP”) bellwether trial.  If that seems pretty specific, it is.  But it is also a lesson in how granular the fit requirement can be, derived as it is from Rule 702(a)’s insistence that expert opinion “help the trier of fact to understand the evidence or to determine a fact in issue.”  The claims, defenses, and applicable law determine what facts are truly at issue in a case.  Rule 702 veterans will be familiar with fit being used to attack causation opinions based on, for instance, exposure to a substance at much higher levels and/or for a much longer time than is alleged in the case.  It is easy to understand that an opinion that ingestion of one pill of a drug with alleged contamination of an alleged carcinogen at the level of one part per billion can cause cancer in a human cannot be based solely on data that long-term ingestion of thousands of times as much of the alleged carcinogen was associated with increased rates of cancer in certain laboratory animals.  Of course, we did not pick that example out of thin air.  The premise of the Valsartan and Zantac litigations, at least from our perspective, has been that allegations of contamination—without proof of any particular plaintiff taking pills that had any contamination, let alone risky levels of contamination—were sufficient to create liability.  Contaminated drugs did not meet specifications and were defectively manufactured.  They should not have been sold.  For generics, they failed the duty of sameness.  They were all adulterated and misbranded.  And the drugs were inherently worthless, and nobody should have ever had to pay for them.  So goes the “logic” of these claims going back to the citizen’s petitions filed by that questionable lab with unquestioned ties to the plaintiff lawyers.  As Valsartan ‘25 said of one of plaintiffs’ specific arguments in support of their expert’s opinion, “But this is both too simplistic and too doctrinaire.”  2025 U.S. Dist. LEXIS 66185, *52.

The opinion at issue was that the drugs of the defendants in this TPP bellwether trial had no value based on the assumption that they were all contaminated, and thus adulterated and illegal to sell, the entire time they were on the market.  The piggyback opinion was that the plaintiffs were entitled to huge damages for having paid for “worthless” drugs, even if the drugs actually worked really well at treating hypertension and related conditions.  Because the TPP plaintiffs claimed only economic injuries from paying for drugs, their theory did not turn on whether patients ever suffered any harm.  To us, if liability is predicated on state law, then we are talking about the stop selling theory held preempted in Bartlett and we do not even need to get to the admissibility of this particular economist to support it.  Well, the prior judge had rejected a range of preemption arguments, so we can set that concern aside for now.  Of more pressing concern was that the prior judge had accepted essentially the same opinion from the same expert, Dr. Conti, in connection with the denial of a motion to decertify a TPP class.  The new judge found enough leeway in the prior ruling to decide the admissibility of Dr. Conti’s worthlessness opinion on its merits.  Id. at 37-41.  To get there, the court first had to identify what Dr. Conti was being offered to support, and then determine if her proposed opinions would be helpful to the jury in resolving the issues implicated by the TPP plaintiffs’ triable claims.  Prior motions practice allowed the TPP plaintiffs enough wiggle room to offer pretty fuzzy claims that were designed to skirt preemption.

The TPP plaintiffs offered an express warranty theory, which can often be based on voluntary representations beyond FDA-required labeling.  However, these plaintiffs were suing over “the labeling-based representations by the Defendants that they were selling FDA approved, Orange Book A/B rated, USP compliant valsartan that was manufactured in a manner that was compliant with cGMPs and was not adulterated.“  Id. at *10.  For us, that is clearly preempted claim and not an express warranty at all.  Perhaps because its hands were somewhat tied by the prior rulings, the Valsartan ‘25 decision did not directly address state law on express warranty beyond how damages are measured.  Id. at *18 n.10.  To obtain the full refund they sought, the TPP plaintiffs had three damages theories available.  Only one, that the products were inherently valueless because of contamination, was available, because plaintiffs did not allege the drugs failed to perform their intended function of lowering blood pressure and because the drugs had been used, negating the contention that they were unusable.  Id. at *18-25.  This theory dictates the scope of proof at trial and the issues that Dr. Conti’s opinions had to fit:

This entire litigation is about the alleged contamination of lifesaving VCDs with cancer-causing nitrosamines. To try this case without evidence of cancer causation is to ignore the elephant in the room. Causation must be front and center. As a result, not only will the Court permit the parties to present evidence regarding the nitrosamines in question and cancer causation, but it will require it.

Id. at *24-25.  This may seem like an obvious requirement, but it was clear that the plaintiffs envisioned not having it.

As might be expected from an economist, Dr. Conti did not evaluate any science related to contamination and its attendant risk.  Instead, she assumed the drugs at issue were always contaminated and that contamination rendered them adulterated.  Of course, adulteration is an FDCA concept and predicating a purported state law claim on a violation of the FDCA not found by FDA should be a big no-no.  It is also clear that this part of Dr. Conti’s proposed testimony was a legal conclusion, not an opinion based on scientific or technical expertise.  A legal conclusion that supported a preempted claim, which Dr. Conti reinforced by offering that the drugs should not have been sold because they were “adulterated,“ one of her foundational assumptions.  Rather than looking at preemption and that Dr. Conti was being offered as a mouthpiece for legal conclusions, Valsartan ‘25 addressed whether Dr. Conti’s opinions would be helpful to the jury and were based on a reliable methodology.

As to the former,

Conti readily opined there was no legitimate supplier for the adulterated VCDs and therefore the VCDs had no economic value from the day the VCDs were first sold. Conti’s conclusion rests solely on the fact that it was “impermissible” to have sold the VCDs. Because it was impermissible – in hindsight – the value of the VCDs is zero and the Plaintiffs should receive a full refund. These are neither the facts of the case nor an adequate statement of the law as to damages in a breach of express warranty case.

Id. at *43-44 (emphasis in original and internal citation omitted).  Put another way, “Economic supply curves say nothing about whether any defect in the VCDs is ‘fundamental’” and proof of a defect “requires an analysis of causation and risks versus benefits.  But Conti’s methodology and testimony do not even account for the nature, impact, or extent of the adulteration.”  Id. at *47-48.  As to the latter,

The Court finds that Conti’s testimony is unreliable in light of its inconsistencies and the stark lack of scientific or economic basis for her methodology. Conti’s opinion is largely argument and advocacy based on her own ipse dixit, rather than a reliable application of economic principles and methods to the facts of the case. Perhaps this has all come about because of Plaintiffs’ failure to appreciate the critical distinction between alleging an injury in fact for standing, sustaining an injury as an element of the claim to establish liability, and then proving the injury to calculate damages.

Id. at *54-55.  By Rule 702 standards—and Valsartan ’25 noted the 2023 revision, see id. at *26 & n.15, *39-41 for an excellent discussion of those amendments—Dr. Conti’s core opinion did not come close to being admissible.

We know that we have not done a very good job of setting aside preemption as we promised a few paragraphs ago.  We are hopeful that the new direction of the Valsartan MDL signaled in Valsartan ‘25 will result in a serious reexamination of the basic question of whether plaintiffs’ state law claims are preempted.  If that is a bridge too far, then other holes in plaintiffs’ case—at least while it is based on the mere assumption of dangerous contamination—should present real obstacles for the Valsartan MDL plaintiffs.

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