Photo of Eric Hudson

We were saddened this week to hear about the passing of Dolly Parton. A prolific and wonderful songwriter, she was also a relentless force for good.  A shining example of her goal of improving the world is her Imagination Library, which she started to make sure as many kids as possible had a chance to read. Since its founding in 1995, the library has donated over three hundred million books to children across the world. As Dolly put it:

When I was growing up in the hills of East Tennessee, I knew my dreams would come true. I know there are children in your community with their own dreams. They dream of becoming a doctor or an inventor or a minister. Who knows, maybe there is a little girl whose dream is to be a writer and a singer.

The seeds of these dreams are often found in books and the seeds you plant in your community can grow across the world.

Dolly planted a remarkable number of those seeds, and in many ways she represented the best of us. Things seem a little darker knowing that such a positive driver of goodness is gone. We will miss her voice, her smile, and her charity, but we trust that her example of doing good will live on.  It’s hard to relate that to our blogging, but ultimately we all have to turn back to the nine to five (OK, OK, we couldn’t help it).

Today’s case is another “LifeVest” wearable defibrillator case (we’ve blogged about prior cases involving this device here and here).  Pellegrin v. Zoll Medical Corp., 2026 WL 2409216 (E.D. La. Aug. 1, 2026). Plaintiff was prescribed a LifeVest for her cardiac care after suffering heart failure. She claimed that she experienced two unnecessary discharges from the device when she was not undergoing a cardiac event. She alleged chronic neck pain following the alleged shocks from the device—although the court noted she had a history of cervical spine problems. The defendant moved to dismiss based on preemption.

Continue Reading More LifeVest PMA Preemption
Photo of Bexis

Shortly after the Supreme Court decided Loper Bright Enterprises v. Raimondo, 603 U.S. 369 (2024), we predicted that that the other side would attempt to use that decision “for nefarious purposes.”  We chronicled their first, “meat-axe” attempt here, where MDL plaintiffs argued that Loper Bright somehow abolished implied preemption in tort cases altogether.  Not surprisingly, that argument got the back of the judicial hand.  In re Suboxone (Buprenorphine/Naloxone) Film Products Liability Litigation, 761 F. Supp.3d 1069, 1086 (N.D. Ohio 2024) (Loper Bright “says little, if anything, about preemption doctrine” since preemption “involves determining the intent of Congress, not an agency”).  A second broad-brush attempt – this time attacking PMA preemption − failed equally miserably in Ehlers v. Abiomed, Inc., 792 F. Supp.3d 941 (E.D. Mo. 2025). “[B]ecause the Riegel Court did not defer to the FDA’s interpretation of the MDA’s preemption clause,” so Loper Bright did not apply at all.  Id. at 952 n.1 (emphasis original).  We found that argument particularly ironic, since the Supreme Court opinion that actually did rely on FDA statutory interpretation, explicitly employing Chevron deference, was the pro-plaintiff decision in Medtronic, Inc. v. Lohr, 518 U.S. 470 (1996).

But in the past couple of months, plaintiffs in much higher profile product liability cases have had their Loper Bright-based anti-preemption arguments roundly rejected. 

In Monsanto Co. v. Durnell, 146 S. Ct. 2001 (2026), which we’ve already discussed more generally here and here, the Supreme Court rejected the plaintiff’s claim that the EPA’s “procedures for registering pesticides and approving pesticide labels” were ultra vires under Loper Bright, and therefore (plaintiff argued) could not have preemptive effect.  146 S. Ct. at 2013.  That was flatly “incorrect.”  Id.  Given the statute’s express delegation of power to EPA to enact regulations and to “determine” labeling compliance, Loper Bright, had no effect at all on preemption.

[The statute] empowers EPA to “prescribe regulations. . . .”  And [it] expressly directs EPA to register pesticides and determine that [their] labeling complies with [the statute’s] many specific requirements.  During that extensive registration process, EPA critically evaluates the pesticide label to ensure that the label contains all warnings necessary to protect human health.  After EPA makes a determination about the appropriate warnings for a pesticide’s label, a manufacturer is legally required to use that label unless and until EPA subsequently approves or requires a new label.

Id. (citations omitted).  Given the EPA’s regulatory power, Durnell considered it just like the FDA – indeed, Riegel v. Medtronic, Inc., 552 U.S. 312 (2008), the FDA PMA preemption decision, was “dispositive here.  If FDA’s premarket approval of medical devices preempted additional state-law requirements, so too must EPA’s registration of pesticides and approval of pesticide labels.”  146 S. Ct. at 2012.

Even more recently, in Yousefzadeh v. Johnson & Johnson Consumer, Inc., ___ F.4th ___, 2026 WL 2192415 (2d Cir. July 30, 2026), another appellate court rejected a similar Loper Bright-related argument in the OTC drug preemption context.  “[T]he FDCA authorizes the FDA to promulgate regulations, including labeling requirements, that govern OTC drugs; these regulations therefore fall within the scope of the FDA’s authority and have preemptive effect.“  Id. at *7 (citations and footnote omitted).  The footnote directly addressed Loper Bright:

To the extent Consumer Plaintiffs contend that Loper Bright [citation omitted], upends this rule by overturning Chevron [citation omitted], and ending the era of Chevron deference, that argument fails.  Following Loper Bright, . . . [a] statute may nevertheless still authorize an agency to “give meaning to a particular statutory term,” and to “prescribe rules to ‘fill up the details’ of a statutory scheme” through regulation.  Thus, Loper Bright does not subvert agencies’ authority to promulgate preemptive regulations pursuant to the authority vested in them by statute − authority the FDCA vests in the FDA here.  Indeed, the Supreme Court has given preemptive effect to federal regulations since before Chevron.

Id. at *7 n.9 (Loper Bright citations and quotation marks omitted).

So far every court that has been presented with a Loper Bright argument against preemption in a product liability case has rejected it – uniformly recognizing that Loper Bright: (1) only deals with agency interpretation of statutes, (2) does not reduce in the least the power of the FDA (and similarly situated agencies) to promulgate regulations, and to approve products, pursuant to congressional delegations of power, and (3) that such discretionary agency actions have full preemptive force.

Now, if only a defendant would only use Loper Bright equally aggressively to contest the continued viability of Lohr.

Photo of Stephen McConnell

Last week we reviewed our 2020-25 Best and Worst lists to see if any patterns emerged in terms of the federal circuits.  Our tentative conclusions were that the Fifth Circuit is pretty defense-friendly and that the Third Circuit is pretty active, both good and bad.  

This week we surveyed state court cases from the same lists. We ended up with an even smaller sample, so any conclusions are even less valid.  But let’s pretend we are an expert witness in Cook or Alameda County, and offer broad opinions premised on bare-bones data and shoddy methodology.  

Here is what we found:

2020 best: Rios (Ill.), Boone (Ct.), Cavanaugh (Fla.), Stiens (Ky.), 

2020 worst: Hammons (Pa.), Russell (Ky.), Mize (Cal.), Risperdal (Cal.), 

2021 best: State ex rel. Hunter v. J&J (Okla.), Albert (Pa.), Hyrmoc (N.J.)

2021 worst: No state cases

2022 best: Dearinger (Wash.), Amiodarone (Cal.), Kline (Cal.)

2022 worst: Glover (Ct.), People v. J&J (Cal.), Blackburn (Ala.), Center for Inquiry v. Walmart, Inc. (D.C.)

2023 best: Center for Envtl. Health (Cal.), Onglyza (Cal.), Hyrmoc (N.J.), Alcazar (Ind.), M.T. v. Walmart (Kansas)

2023 worst: State ex rel. Shikada (Haw.), Estate of Cronin (Nevada), 

2024 best: Shears (W. Va.), In re Natl. Prescript. Opiate (Ohio), Himes (Cal.)

2024 worst: Gilead (Cal.), Himes (Cal.), Providence Health Syst.-Oregon (Oregon), Zantac (Del.), Herzog (Cal.), 

2025 best: Zantac (Del.), Eastern Maine Med. Cntr. (Maine), Silverstein (N.Y.) 

2025 worst: Hall (Wash.), In re Ranitidine (Cal.)

It is no surprise that California produced the most bad cases – six. But it also produced the most good cases – five. And one of those bad cases was Gilead, with its wretched duty to innovate, which was recently reversed by the California Supreme Court in a decision that will make our 2026 Best list.  So while we still think California’s view on admissibility of expert testimony is feeble (we wish Sargon was so gone), the real takeaway is that California is the biggest state with the most litigation, so it hogs the limelight for both good and ill. (We will resist saying anything too snarky about the west coast, but it is hard to miss the fact that, besides California, there are very bad cases from Nevada, Oregon, Washington, and, most western of all, Hawaii.) New Jersey is the only other state with more than one case from the Best list.  Pennsylvania and Kentucky (we would not have guessed that last one) each planted one case among the Best and one among the Worst.  So did Delaware, but it is the same case: the Delaware Supreme Court reversed the lower court’s bad expert rulings in the Zantac litigation. That reversal was not merely on the 2025 Best list, it topped the list.

As with the federal cases, our analysis is probably – check that, certainly – plagued by selection bias.  The truth is that good and bad cases can show up anywhere. Count on California and New Jersey to keep us busy, and that is no cause for despair.  For every bad case emerging from those benighted jurisdictions, there is usually a good one, as well.  

Photo of Michelle Yeary

We have all had the experience of ordering something online, watching the delivery truck icon crawl across the map for three days, and then opening the box to find everything is exactly fine. No harm, no foul, just a mildly stressful shipping odyssey. Now imagine a plaintiff’s lawyer showing up at your door insisting that because the box could have arrived damaged, you should treat it as if it did arrive damaged and sue somebody for the full purchase price. That is, more or less, the theory a relator tried to sell the court in Ellis v. CVS Health Corp., 2026 U.S. Dist. LEXIS 137176 (E.D. Pa. Jun. 22, 2026), a False Claims Act (“FCA”) case decided earlier this summer, and we are pleased to report the court was not buying what he was selling.

The case is a qui tam action concerning biologics, complex protein-based medications that need to be shipped within a specific temperature range because their chemical structure. Their medical value can degrade if they get too hot or too cold for too long. That much is undisputed. It is also undisputed that shipping biologics to patients’ homes, so-called “cold-chain” shipping, carries an inherent risk of what the industry calls “excursions,” meaning brief exposures outside the labeled storage temperature. Id. at *31.

By the time the case reached summary judgment, what remained was a single theory — that three specific temperature-sensitive drugs were “worthless services” under the FCA because of the shipping temperature issues, meaning the government had effectively paid for nothing when it reimbursed pharmacies for these medications. The missing ingredient was any actual proof that any drug was actually worthless.

Worthless services claims under the FCA require the plaintiff to show that what the government paid for had no medical value, not just diminished value — “[s]ervices that are ‘worth less’ are not ‘worthless.’” Id. at *23. After years of litigation and nearly six years of a government investigation that ultimately declined to intervene, plaintiff still could not identify a single dose of any of the drugs that was actually damaged by a shipping excursion. The manufacturers’ own stability studies, which measure how much temperature stress a given drug can withstand before losing potency, showed that the drugs could tolerate colder temperatures for longer than the excursions plaintiff alleged. A fact plaintiff conceded. Id. at *14.

So, plaintiff tried to build a workaround. Lacking proof that any given biologic was actually compromised, plaintiff argued that the shipping process was so “systemically vulnerable,” and defendant’s methods so “utterly deficient,” that the medications could simply be presumed compromised, or at least presumed unmarketable, without any need to show actual impairment. The court was having none of it. In the court’s words, this theory “would take the concept of worthless services to an untenable extreme, finding medical worthlessness with no evidence that any drug’s efficacy was impaired in any way.” Id. at *4.

That is about as clean a rejection as a defendant could hope for, and it is worth sitting with for a second. The court did not say plaintiff’s proof was thin. The court said the entire theory, taken on its own terms, would prove too much. Under plaintiff’s logic, any risk of temperature deviation, no matter how brief, no matter whether it ever actually affected the drug, would be enough to render an entire product line “worthless” and trigger fraud liability under a statute whose whole premise is proof of a false claim.

Plaintiff threw a fair amount of expert firepower at the problem, including a pharmacy practice expert, a packaging design engineer, and a chemist. The packaging and pharmacy experts offered opinions that defendant’s shipping practices were inadequate. But as the court noted, criticizing a company’s process is not the same as proving damage to the product, and “[f]laws in [defendant’s] procedures do not inexorably prove flaws in the drugs.” Id. at *17. The chemistry expert came closest to the mark, theorizing that different rates of freezing and thawing could cause different degrees of protein damage through a phenomenon called cryo-concentration. Interesting theory. Untested theory. On cross-examination, the expert admitted that confirming which effect actually occurred would require running an actual test replicating the shipping conditions, and he never ran one. He just “extracted” data from other people’s studies and reached his own conclusion. The court was unimpressed.  Hypothesis is not a substitute for evidence, particularly when “substantial data . . . flatly contradicts an expert’s theory.” Id. at *19. An opinion that a bad process could theoretically cause harm is not the same as evidence that harm occurred, and a court is not required to let a jury fill that gap with speculation.

When the science did not cooperate, plaintiff pivoted to an economics-flavored argument that the drugs were “economically worthless” because they were technically eligible for return under defendant’s customer-friendly replacement policy, regardless of whether their potency was ever actually affected. Plaintiff’s expert economist testified that the drugs would be economically worthless “even if the stability, efficacy, and safety of the medicine were not affected” by the shipping process. Id. at *21. The court shut this down explaining that this theory rests on a “presumed market value,” which is “a construct not found in the False Claims Act, or any decision applying it.” Id. And logically, the court noted, “if the medication had therapeutic value, securing payment for it cannot constitute fraud.” Id. at *21-22. We appreciate a court that is willing to call a made-up legal concept exactly what it is. Whether a drug has therapeutic value is, as this court correctly framed it, a binary question. It either does or it does not. Maybe doesn’t get the job done.

Plaintiff’s argument that the court spent the most time on was that the FDA-approved label and the United States Pharmacopeia amounted to absolute standards, such that any excursion outside the ranges they identify was a per se violation supporting a finding of worthlessness. The court walked through how worthless services claims can, in appropriate cases, borrow standards from regulations or industry practice to define what counts as worthless, particularly where the product either works or does not work on a “black-and-white” scale, and where noncompliance is so extreme it amounts to gross negligence. But the court found that neither the FDA label nor the USP came close to establishing the strict, zero-tolerance rule plaintiff needed.

On the FDA label, the court noted that the label simply specifies one of a few general storage categories and “do[es] not purport to address the effects of excursions on the medications or . . . how the drugs should be transported.” Id. at *29. The USP fared no better for plaintiff. The court pointed to language explicitly acknowledging that drugs “may be held at temperatures outside their labeled storage requirements” and that excursions “may be acceptable provided stability data and scientific/technical justification exists demonstrating that product quality is not affected.” Id. at *33-34. In other words, the USP itself assumes excursions happen and builds in a process for evaluating them rather than banning them outright.

The court also found defendant’s conduct cut against any finding of gross negligence. Defendant spent time and money testing and re-testing the packaging and maintained a policy of promptly replacing any drug reported to have arrived frozen, which the court noted was one of the two compliance options the USP itself offers pharmacies when a deviation is reported. Id. at *34. An organization that is trying, even imperfectly, is not an organization acting “without even slight care.” Id. at *27.

We flag this decision not just for our FCA enthusiasts, but for anyone defending economic loss class actions built on a similar architecture — plaintiff cannot prove that any specific unit of a product actually failed, so plaintiff instead argues that the entire product line, or an entire class of transactions, should be treated as categorically worthless or defective because of a generalized, systemic risk that harm could have occurred somewhere, to someone, at some point. Strip away the FCA-specific vocabulary and this court’s reasoning offers a useful playbook for knocking down economic loss claims regardless of the statute involved. 

The relator here had six years of government investigation and years of party discovery to find one damaged dose and could not do it. In the end all he could do was advance multiple theories asking the court to treat speculation as proof. Instead, the court granted summary judgment. The box arriving a little cold did not mean everything inside it was ruined. Some shipments really do arrive exactly fine. This one, for defendant, arrived on time and intact.

Photo of Bexis

We published a 50-state survey on the widespread rejection of so-called market share liability back 2010, which we recently referenced in our post celebrating the demise of an “ultra processed” food strike suit in Pennsylvania.  As that post discussed, one of the liability theories that failed was market share liability.  That got us thinking that we should update our 50-state market share liability survey.  So that’s what we’re doing today.  This is an add on post to our prior survey, so these two posts should be consulted together.  We’re not repeating things here.

Federal Law

Nothing beyond previous rejection of market share liability.

Alabama

Alabama has passed a statute that, while aimed at innovator liability, has the added advantage of also precluding market share liability. 

In any civil action for personal injury, death, or property damage caused by a product, . . . the plaintiff must prove . . . that the defendant designed, manufactured, sold, or leased the particular product the use of which is alleged to have caused the injury on which the claim is based, and not a similar or equivalent product.  Designers, manufacturers, sellers, or lessors of products not identified as having been used, ingested, or encountered by an allegedly injured party may not be held liable for any alleged injury. . . .

Ala. C. 1975, §6-5-530 (emphasis added).

Alaska

Still nothing.

Arizona

Nothing new to add to this state’s previous rejections of market share liability.

Arkansas

In asbestos litigation, “Arkansas has not adopted alternative or market share liability, but has retained the traditional requirement of proximate cause in all tort cases.”  Thomas v. Borg-Warner Morse TEC LLC, 362 F. Supp.3d 610, 616 (E.D. Ark. 2018) (quoting Jackson); accord Tucker v. Southwest Energy Co., 2012 WL 528253, at *2 (E.D. Ark. Feb. 17, 2012) (“Arkansas has not recognized market-share liability”) (fracking chemicals).

California

While California has not repudiated Sindell, recent decisions have refused to expand the scope of market share liability.  See Wiersema v. Target Corp., 2025 WL 1594275, at *3 n.1 (C.D. Cal. March 10, 2025) (market share liability cannot apply to corner guards for shelving); Farris v. 3M Co., 2019 WL 1904206, at *4 (N.D. Cal. April 29, 2019) (yet another rejection of market share liability in asbestos litigation); Pooshs v. Philip Morris USA, Inc., 904 F. Supp. 2d 1009, 1032 (N.D. Cal. 2012) (market share liability cannot apply to cigarettes).

Applying California law, In re Yasmin & Yaz (Drospirenone) Marketing, Sales Practices. & Relevant Products Liability Litigation, 2010 WL 3937414, at *8 (S.D. Ill. Oct. 4, 2010), held that market share liability could not lie against a distributor of prescription drugs absent proof that the distributor had in fact supplied at least some of the drugs at issue).

Colorado

Still nothing.

Connecticut

Nothing new to add to this state’s previous rejection of market share liability.

Delaware

Market share liability “has not been applied in Delaware in any setting.”  Sherman v. Ellis, 2020 WL 30393, at *12 & n.74 (Del. Super. Jan. 2, 2020), rev’d, 246 A.3d 1126 (Del. 2021) (going further and requiring strict “but for” causation in legal malpractice cases).  Accord Robinson v. Union Carbide Corp., 2019 WL 3822531, at *10 (Del. Super. Aug. 15, 2019) (reiterating rejection of market share liability in asbestos litigation); State Farm Fire & Casualty Co. v. Middleby Corp., 2011 WL 683883, at *3 (Del. Super. Ct. Feb. 8, 2011) (Delaware’s rejection of market share liability precludes burden shifting in fire case).

District of Columbia

Nothing new to add to the District’s previous rejection of market share liability.

Florida

An attempt to apply market share liability outside of product liability was “summarily rejected” under Florida law in In re: Chiquita Brands International, Inc., 2019 WL 11497632, at *32 (S.D. Fla. Sept. 5, 2019), and that rejection of percentage share causation was affirmed Carrizosa v. Chiquita Brands International, Inc., 47 F.4th 1278, 1335-36 (11th Cir. 2022).

Georgia

Georgia passed a statute expressly prohibiting market share liability.  O.C.G.A. §51-1-11(d) (“Irrespective of privity, a manufacturer shall not be held liable for the manufacture of a product alleged to be defective based on theories of market share or enterprise, or other theories of industry-wide liability.”).  Williamson v. Walmart Stores, Inc., 2015 WL 1565474 (M.D. Ga. April 8, 2015), held that plaintiff could not get around that prohibition by suing only intermediate sellers.  Id. at *8 (plastic container case).

Hawai’i

Nothing new to add.

Idaho

Nothing new to add to this state’s previous rejection of market share liability.

Illinois

The Illinois Supreme Court’s rejection of market share liability, even in DES cases, was cited in Anderson v. Anderson, 959 N.E.2d 1167, 1173 (Ill. App. 2011) (car crash case).

Indiana

Nothing new to add to this state’s previous rejection of market share liability.

Iowa

A Delaware court, in an asbestos case, recognized that “Iowa has expressly rejected market share liability.”  In re Asbestos Litigation, 2014 WL 4180186, at *4 (Del. Super. Aug. 21, 2014) (footnote omitted), aff’d, 163 A.3d 709 (Del. 2017).

Kansas

Still nothing.

Kentucky

Kentucky law does not recognize market share liability in toxic chemical cases.  Brown v. Arch Wood Protection, Inc., 265 F. Supp.3d 700, 707 n.10 (E.D. Ky. 2017).

Louisiana

“Plaintiff cannot rely on alternative theories of liability, such as market share liability, because they are not recognized under Louisiana substantive law.”  Laurent v. New Orleans City, 2015 WL 846567, at *1 (E.D. La. Feb. 25, 2015) (asbestos case).  “Louisiana courts do not apply a market share liability theory of causation.”  Moore v. BASF Corp., 2011 WL 5869597, at *5 (E.D. La. Nov. 21, 2011) (benzene exposure).

Maine

Nothing new to add to this state’s previous rejection of market share liability.

Maryland

Maryland’s highest court has rejected market share liability in the asbestos context.  Wallace & Gale Asbestos Settlement Trust v. Busch, 211 A.3d 1166, 1176 (Md. 2019) (“Maryland rejects such an approach [market share liability] in asbestos cases.”) (following Reiter); Reiter v. Pneumo Abex, LLC, 8 A.3d 725, 730 (Md. 2010) (“market share liability . . . is not recognized under Maryland law”); accord Sherin v. Crane-Houdaille, Inc., 47 F. Supp.3d 280, 292 n.30 (D. Md. 2014).  Notwithstanding this precedent, a federal court predicted that Maryland would allow “comingled” market share liability in Maryland v. Exxon Mobil Corp., 406 F. Supp. 3d 420, 457-58 (D. Md. 2019).

Massachusetts

Nothing new to add to this state’s previous rejections of market share liability.

Michigan

Nothing new to add.

Minnesota

Nothing new to add to this state’s previous rejections of market share liability.

Mississippi

In Dickens v. A-1 Auto Parts & Repair, Inc., 2019 WL 508074, at *1 (S.D. Miss. Feb. 8, 2019), an asbestos case, the court “agree[d]” that market share liability has “never been recognized in Mississippi.”

Missouri

Nothing new to add to this state’s previous rejections of market share liability.

Montana

Still nothing.

Nebraska

Nothing new to add to this state’s previous rejection of market share liability.

Nevada

Still nothing.

New Hampshire

In the only adverse development since our original survey, the New Hampshire Supreme Court allowed “comingled” market share liability in an MTBE/gasoline case.  State v. Exxon Mobil Corp., 126 A.3d 266, 298 (N.H. 2015) (“because the State could not identify the tortfeasor responsible for its injury, under market share liability the burden of identification shifted to [defendant]”).

New Jersey

A New Jersey appellate court rejected market share liability in a pelvic mesh case.  Kline v. Johnson & Johnson, 2020 WL 1082430, at *2 (N.J. Super. App. Div. March 6, 2020).  The same court also rejected market share liability in a vaccine case.  Moreno v. American Home Products, Inc., 2010 WL 4028605, at *5 (N.J. Super. App. Div. July 12, 2010).

New Mexico

Still nothing

New York

New York continues to limit market share liability to the now defunct DES fact pattern.  See S.F. v. Archer Daniels Midland Co., 594 F. Appx. 11, 13 (2d Cir. 2014), affirming 2014 WL 1600414 (W.D.N.Y. April 21, 2014) (market share liability does not apply to high-fructose corn syrup); Zeitlin v. Palumbo, 532 F. Supp.3d 64, 71 (E.D.N.Y. 2021) (market share liability does not apply to robocalls); Girau v. Europower, Inc., 2017 WL 4124340, at *5 (S.D.N.Y. Sept. 14, 2017) (market share liability does not apply to hydraulic hose); Pace v. Air & Liquid Systems Corp., 171 F. Supp.3d 254, 263 (S.D.N.Y. 2016) (market share liability does not apply in asbestos litigation); Moreno v. American Home Products, Inc., 2010 WL 4028605, at *6-7 (N.J. Super. App. Div. July 12, 2010) (New York would not apply market share liability in a vaccine case).

North Carolina

Market share liability continues to be rejected in North Carolina asbestos litigation.  Williamson v. Bridgestone Americas, Inc., 625 F. Supp.3d 466, 469-70 (E.D.N.C. 2022) (“North Carolina law does not recognize alternative causation theories, such as market-share liability.”); Cox v. AGCO Corp., 2020 WL 3473636, at *4 (E.D.N.C. June 25, 2020) (market share liability is “not recognized in North Carolina”).

North Dakota

Nothing new to add to this state’s previous rejection of market share liability.

Ohio

Rejection of market share liability in Ohio is so settled that even the Opioid MDL wouldn’t allow it.  In re National Prescription Opiate Litigation, 2019 WL 4178617, at *2 n.6 (N.D. Ohio Sept. 3, 2019).

Oklahoma

In declaring a statute imposing collective tort liability unconstitutional, Steed v. Bain-Holloway, 356 P.3d 62, 67 (Okla. App. 2015), relied upon Oklahoma’s rejection of market share liability.  A bankruptcy court in Delaware confirmed that “[m]arket share liability is not allowed under Oklahoma law.”  In re Chaparral Energy, Inc., 2018 WL 818309, at *6 n. 49 (Bankr. D. Del. Feb. 9, 2018).

Oregon

Nothing new to add to this state’s previous rejection of market share liability.

Pennsylvania

As recently discussed, Martinez v. Kraft Heinz Co., 2026 WL 1878602 (E.D. Pa. June 30, 2026), market share liability cannot be applied in “ultra processed” food litigation.  The products aren’t fungible and it is essentially impossible to join most of the “market” where so many different products are involved.  Id. at *8.  Market share liability is also inapplicable to public nuisance claims.  Atlantic Richfield Co. v. County of Montgomery, 294 A.3d 1274, 1286 (Pa. Commw. 2023) (lead paint claims).  Pennsylvania’s rejection of market share liability was not limited to “product liability cases.”  Id.  In re Methyl Tertiary Butyl Ether (“MTBE”) Products Liability Litigation, 2025 WL 1795126, at *10 (S.D.N.Y. June 30, 2025), refused to apply “comingled” market share liability under Pennsylvania law.  Market share liability was also rejected as a theory in a case involving medicated cream.  McKnight v. Amazon.com Inc., 2024 WL 2156223, at *3 (E.D. Pa. May 14, 2024) (“Pennsylvania courts are reluctant to apply market share liability.”).

Rhode Island

Nothing new to add to this state’s previous rejection of market share liability.

South Carolina

Nothing new to add to this state’s previous rejections of market share liability.

South Dakota

Nothing new to add.

Tennessee

Nothing new to add to this state’s previous rejections of market share liability.

Texas

Texas unequivocably rejects market share liability. 

[W]e have never embraced the concept of industry-wide liability on grounds that proof of causation might be difficult.  Instead, we have rejected such thinking and held firm to the principle that liability in tort must be based on proof of causation by a preponderance of the evidence.  In a mesothelioma case, we rejected theories of collective liability − alternative liability, concert of action, enterprise liability, and market share liability − and held instead: A fundamental principle of traditional products liability law is that the plaintiff must prove that the defendants supplied the product that caused the injury.

Bostic v. Georgia-Pacific Corp., 439 S.W.3d 332, 340 (Tex. 2014) (footnotes and quotation marks omitted).  Accord Alarcon v. Alcolac, Inc., 488 S.W.3d 813, 828 (Tex. App. 2016) (toxic chemical case); Bezuidenhout v. Abbott Laboratories & Co., 918 F. Supp.2d 144, 147-48 (E.D.N.Y. 2013) (Texas would not allow market share liability even in a DES case).

Utah

Still nothing.

Vermont

Still nothing.

Virginia

Still nothing.

Washington

Nothing new to add.

West Virginia

Still nothing.

Wisconsin

Of the states that jumped on the market share liability bandwagon in DES litigation, Wisconsin went for the worst and most expansive version, as discussed in our prior post.  The Wisconsin legislature thought so, too, and enacted a statute that specifically overruled Thomas v. Mallett, 701 N.W.2d 523 (Wis. 2005), by name.  Wis. Stat. §895.046.  Section 895.046(4) allowed a last-resort (no other viable defendants) market share liability claim solely against manufacturers of “chemically and physically identical” “complete integrated products” that were marketed without any “distinctive” characteristic or labeling.  Id. at subsections (4)(a)(1) & (3).  The product must have been marketed in plaintiff’s “geographic market,” and the plaintiff must join manufacturers of an 80%+ market share.  Nonetheless, in a decision we criticized, Gibson v. American Cyanamid Co., 760 F.3d 600 (7th Cir. 2014), refused to give the statute its intended retroactive application.

Wyoming

Wyoming law remains silent as to market share liability.

Photo of Steven Boranian

The plaintiff in Kha v. Tilray Brands, Inc. tried to use California’s Sherman Law to do an end run around the Food, Drug, and Cosmetic Act in a case involving hemp products, but it did not work.  The district court correctly saw the case as an attempt to impose additional, state-law requirements on food labeling, which the FDCA expressly prohibits.  See Kha v. Tilray Brands, Inc., No. 2:25-cv-10630, 2026 U.S. Dist. LEXIS 160579 (C.D. Cal. July 17, 2026).

The plaintiff in Kha alleged that the third-party website where she purchased her products gave product descriptions that disclosed protein amounts in a way that did not comply with the FDCA and California’s copycat Sherman Law.  You are familiar with the Sherman Law.  It is the California statute that purports to incorporate the FDCA by reference, supposedly making violations of that federal law also violations of state law.  As we have explained before, it matters because there is no private right of under the FDCA, and 21 U.S.C. § 337 expressly reserves the right to enforce the FDCA to the United States—not private parties.  Despite this, some courts have allowed private parties to evade section 337 by permitting private actions under California’s Sherman Law, even though the allegations amount to violations of the FDCA. 

That is what the plaintiff attempted in Kha with her class action complaint, which alleged that the product “labeling” was noncompliant and thus misleading, and that she and other consumers should get some or all of their money back. 

The order dismissing the complaint is interesting for a few reasons.  First, the defendant argued that the third-party webpage was not “labeling” regulated by the FDCA.  The district court, however, disagreed and ruled that written and graphic matter on a website that “supplements and explains” a product can constitute “labeling,” even if not physically attached.  The “textual relationship is what is important.”  Kha, at *7.  Under that standard, the court ruled that the plaintiff had plausibly alleged that the web listing could be “labeling” under the purview of the FDCA.  As far as we know, this is the first case where a court has ruled directly that product descriptions on a third-party online marketplace can constitute FDCA-regulated labeling. 

Second, having just ruled that the FDCA regulated the alleged labeling, the district court ruled that the plaintiff still could pursue a private enforcement action through “parallel state law.”  This is the Sherman Law end run that we have grimaced over before.  Notably, the district court relied on the Ninth Circuit’s opinion in Davidson v. Sprout Foods, 106 4th 842 (9th Cir. 2024) (which we discussed here), but it did not discuss or even cite the earlier Ninth Circuit opinion going the other way, Nexus Pharmaceuticals, Inc. v. Central Admixture Pharmacy Services, Inc., 48 F.4th 1040 (9th Cir. 2022) (discussed here).  In our view, courts should treat private lawsuits framed as enforcing the FDCA as private lawsuits enforcing the FDCA, which federal law prohibits. 

Third, and in the end, it did not really matter because the FDCA expressly preempted all the plaintiff’s claims anyway, although the court did not call it “preemption.”  The FDCA expressly preempts state-law requirements for food labeling that are “not identical” to federal requirements.  Kha, at *9 (citing 21 U.S.C. § 343-1(a)(5)).  Here, the product labeling—including the third-party web listing—complied with federal regulations.  The plaintiff quibbled with the content of the protein disclosures, but that did not matter in the end, either.  The products’ nutritional facts were “readily available and visible on the front-page” of the web listing, and “[r]etailers are not required to expressly direct consumers to nutritional information—whether via a reference mark, dagger, or other instruction—especially when this information is in commonly known and recognizable locations.”  Id. at *10-*11. 

Because the product labeling complied with the FDCA and related regulations, the district court dismissed all the plaintiff’s claims.  Again, the court did not call it express preemption, but we will.  The product labeling complied with federal requirements, so any state-law liability—whether under consumer protection laws, warranty laws, or torts like unjust enrichment—would derive from additional requirements, i.e., requirements that were “not identical” to federal law.  In other words, if the FDA does not require whatever additional protein disclosure the plaintiff wanted, then state law cannot require it either.  That is preemption. 

Photo of Eric Alexander

The Supreme Court decided Buckman in the context of a fairly large product liability litigation about implanted medical devices.  In the years that followed, the plaintiffs in drug and device product liability litigation tried a number of arguments to cabin the impact of Buckman.  For instance, they argued that it 1) did not apply unless the cause of action was actually titled “fraud on the FDA,” 2) did not apply to drug cases, 3) did not apply where they were right about information being withheld from FDA, 4) did not apply if there was also potential express preemption, 5) did not apply to cases involving recalled or withdrawn medical products, 6) did not apply to consumer protection claims brought by AGs, and 7) could not limit evidence offered to show fraud on FDA as a way to rile up the jury as long as there was some relevance to some non-preempted theory.  Courts have not been consistent, but plaintiffs were wrong about all of this.  (There are way too many posts to show we are right about the plaintiffs being wrong, so here is a general link to prior Buckman posts.)  Because implied preemption is based on the Supremacy Clause, it does not apply to federal law causes of actions, such as RICO.  But what about where a company sues another company in a commercial state law breach of contract action over the supply of an FDA-regulated medical product?

We know that FDA and CMS regulation of regenerative medicine and wound care products have been in flux recently.  We also know that the uses of products derived from human or animal tissue/cells have been expanding rapidly, along with the technology supporting new products and uses.  So, it is not surprising that the case we are discussing today involves such products and their evolving regulatory status.  We do not normally discuss state trial court-level decisions unless they are from a large coordinated proceeding.  We also do not normally discuss contract cases, but Desai v. StimLabs, LLC, No. 25CV014789, 2026 Ga. Super. LEXIS 3809 (Ga. Super. Ct. May 15, 2026), is an exception to both faux rules because of its treatment of Buckman.  Given the regulatory history it discusses, we would not be surprised if there are other cases like Desai being brought against manufacturers of these products.  Desai centers around the assertion that a dermatology practice was defrauded in connection with its contracts to purchase millions of dollars’ worth of four human cellular- and tissue-based products (“HCT/Ps”) from the manufacturer.  Plaintiff used these products on its patients, charged CMS for the care, and was ordered to pay back close to $5 million, most of which it attributed to charges for the use of defendant’s four HCT/Ps.  It sued the manufacturer under various theories, and the manufacturer moved to dismiss or for a more definite statement on the contract and fraud claims.

Because Desai was decided on a motion to dismiss, the facts are as alleged by plaintiff except for the few areas where the court took judicial notice.  Whether HCT/Ps need pre-market approval depends on whether the intended use is limited to “homologous use only.”  Id. at *3.  Those that are—the products intended to “perform[] the same basic function or functions in the recipient as in the donor”—are called “361 HCT/Ps” and do not need pre-market approval.  Id.  In November 2017, FDA issued a final guidance on how to distinguish between the two basic regulatory categories, which announced limited enforcement—“enforcement discretion”—until May 2021.  Id. at *4.  (We know from a few minutes on-line that the guidance was issued after three years of comment, was amended a month later, and was superseded in July 2020, with the new guidance extending the enforcement discretion period until May 2021.)  At some point in 2020, plaintiff entered into purchase contracts with defendants based on what it claimed were representations that the products were legally marketed as 361 HCT/Ps (without pre-market approval), they were “reimbursable by payors such as Medicare,” and nothing would change for the worse once the enforcement discretion period ended.  Id. at *4-7.  Those representations continued until at least May 2021.  Plaintiff also alleged that the products were marketed as non-361 HCT/Ps because they were intended for broader uses of “reduc[ing] pain, inflammation, scarring, and wound healing.”  Id. at *5-6.  Plaintiff did not allege its own analysis of these issues at the time or the due diligence it did before buying, using, and submitting claims related to the products at issue.  It did allege that defendant received a May 10, 2021, letter from FDA that raised the possibility that “its marketing of the Products may have been unlawful.”  Id. at *8.  Defendant was part of suits filed in 2022 and 2024 challenging CMS’s many denials of Medicare coverage for purported 361 HCT/Ps, including one of the ones plaintiff was buying from defendant.  When plaintiff was audited by a CMS contractor for Texas in 2022 for submissions between April 2019 and April 2022, the focus was apparently on the non-homologous use of HCT/Ps, which is how it learned that there may not actually be Medicare coverage.  Id. at *9-10.  Even with the alphabet soup and many overlapping dates above, it should be apparent that there were factual pleading holes in the complaint.  We will not really address the motion for a more definite statement, though.

At least implicitly, plaintiff contended that the four products were illegally marketed in violation of the FDCA.  Explicitly, it contended that the “marketing materials” for the products misrepresented that they were suitable for non-homologous use under FDA requirements.  Id. at *13.  Other statements allegedly falsely represented that Medicare would pay for their use.  In analyzing Buckman on the FDCA-related allegations, Desai first rejected plaintiff’s argument that Buckman was inapplicable where a private party also claimed to have been defrauded.  “There the Supreme Court considered the substance of the asserted fraud claim and found pre-emption was appropriate because ‘the existence of these federal enactments [was] a critical element’ of the patients’ case.”  Id. at *14.  Citing cases we discussed here, here, and here, Desai rightfully focused on the question of whether the state law claim could exist without the FDCA.  Those tied to FDA approval and alleged misrepresentations about approval could not, because they “would not exist but for the FDA’s regulations.”  Id. at *16.  So, those claims, portions of multiple counts, were impliedly preempted.  (We note that Georgia is among the majority of states that do not recognize a failure to report to FDA as a state law claim.)

The claims based on alleged misrepresentations about Medicare coverage and what CMS would do in the future were not.  As the cottage industry of relators pursuing False Claims Act cases over Medicare reimbursement shows, the Social Security Act and its brethren are not like the FDCA.  Private rights of action exist, and CMS lacks exclusive enforcement authority.  That means there is no Buckman-style preemption of claims based on misrepresentations about Medicare eligibility.  However, fraud cannot be based on “conjecture about a future event” or a reading backward from future knowledge of regulatory development to a support for the intent to mislead in the past.  As HHS regulation of a range of regulated medical products, not just HCT/Ps, seems to swing wildly, these principles are good to keep in mind for cases where Buckman is not a complete defense.

Photo of Stephen McConnell

Above the Law used to be a forum largely dedicated to law firm associates bemoaning their treatment at their “TTT” (third-tier toilet). But now the Above-the-Law substack does a splendid job of covering law industry topics. David Lat is a knowledgeable and fair-minded commentator. He is smart. He did, after all, work at Wachtell.  He has informed connections. Last week, Above the Law discussed how to rank the federal judicial circuits by importance.  If “importance” hinges on caseloads or population, the Ninth Circuit is comfortably in first place. By contrast, applying those same criteria, the D.C Circuit is a cellar-dweller.  Then again, if “importance” refers to prestige, which is a function of high-profile cases and feeding judges and clerks to the Supreme Court, then the D.C. Circuit vaults to the top.  Four of the nine current SCOTUS Justices were formerly D.C. Circuit judges.  (Roberts, Thomas, Kavanaugh, and Jackson.) Justice Sotomayor is from the Second Circuit, Alito the Third Circuit, Coney Barrett the Seventh Circuit, and Gorsuch the Tenth Circuit. Justice Kagan came straight from academia.  (As a former Ninth Circuit clerk, we are appalled that the Ninth Circuit is unrepresented on the High Court. And as a U. of Chicago grad, we are miserable that nobody from our school has ever made the cut.)

The Second Circuit, with hegemony over financial industry cases, usually slots in second in prestige.  Judge Henry Friendly was very highly regarded.  So was Judge Amalya Kearse and many other Second Circuit judges. Have you ever heard of Learned Hand? Then the Ninth Circuit is in the show position, because size matters.  Before he was elevated to SCOTUS, Anthony Kennedy was a feeder judge.  So was Cliff Wallace. So was Alex Kozinski until, well, things ended badly. Collins, Fletcher, and Friedland are strong feeder judges. Bumatay is thought to be on the SCOTUS short list. Van Dyke is … notorious.

The point of the Above the Law article is that the factors driving circuit “importance” are somewhat dynamic. For a while, the Seventh Circuit enjoyed outsized significance when Posner, Easterbrook, Wood brought Chicago school law and economics heft to their opinions.  But Posner and Wood retired, and the Seventh Circuit might have lost some luster. (And we will never forgive it for Bausch.) Further, the roster of feeder judges has now spread well beyond the Acela corridor. Some of the most prized clerkships are now in, for example, the Fifth Circuit, where  Judges James Ho and Andrew Oldham (both U. of Chicago grads) are on the Republican short list for future SCOTUS appointments.  Judge Amul Thapar is also on that list, and he is in the Sixth Circuit.  (For many years Sixth Circuit Senior Judge Danny Boggs was viewed as a future Justice. One Boggs clerk referred to Boggs as the Brazil of Circuit Judges – always mentioned as the next Big Thing, but never quite getting there. Boggs was famous for administering a brutal quiz to clerkship applicants. Questions included: What is measured by an anemometer? Distinguish Roy M. Cohn from Roy G. Biv. Who had a famous Last Theorem? Last Tape? Last Stand?) On the Eleventh Circuit, Judges Pryor, Brasher, and Newsom (Chicago!) place many clerks with SCOTUS. Naturally, we have a soft spot for the Third Circuit. We can see its courthouse from our window.  The quality of writing is quite high in the Third Circuit.  Judges Bibas, Krause, and Hardiman, for example, wield sharp pens.

We encourage our Penn Law students to consider applying for clerkships.  Save for those few students for whom a SCOTUS clerkship is a realistic possibility, our advice is that location does not matter much. To be sure, there is nice c.v. value to having a clerkship in the circuit where one ends up working. But the central truth is that the privilege of working on judicial opinions is instructive and enjoyable. (Mostly. From what we can tell, 90% of judicial clerkships are terrific jobs, while the unpleasant clerkships with ogre judges are a very small minority).  Our judge, William Norris (Ninth Circuit), would sit next to us at the keyboard as we edited opinions line by line. It was a fabulous learning experience.

The Above the Law article got us thinking about how we would rank the circuits with respect to drug and device law opinions. Inasmuch as preemption is our favorite defense, we figured that the D.C. Circuit would be in the mix in a big way. (Spoiler alert: we were wrong about that.)  We hypothesized that the proliferation of pharmaceutical companies along I-95 would result in many consequential Third Circuit opinions, and the locations of device manufacturers would do something similar for the Seventh Circuit.  If importance can also include the bad sort, we posited that the Ninth Circuit’s matador treatment of Rule 702 would be prominent.  

How to test these ideas?

We decided to canvas our blog’s Top Ten Best and Worst lists for 2020-2025 (so that is six years) to see how the circuits lined up.  We took into account only federal cases, of course, and added up numbers looking both at circuit cases, as well as district court cases grouped by circuit.  It is hardly a perfect methodology. The sample size is definitely too small.  There is also a danger of double (or worse) counting. Some litigations produced multiple opinions, sometimes on both the district and circuit court levels.  More on that later.

Here are the cases we counted for each year:

2020 best: Mirena (2d), Taxotere (5th), Dolin (7th), Viagra (N.D. Cal.), Keen (E.D. Pa.), Ridings (W.D. Mo)

2020 worst: Kaiser (7th), Booker/Bard Filters (9th), Gremo (D.N.J.), Gross (E.D. Pa.), Genentech (10th), Valsartan (D.N.J.)

2021 best: Brooks (10th), Zofran (D. Mass.), Salinero (11th), Ignacianos (2d), Taxotere (5th), Incretin (S.D. Cal.), Zantac (S.D. Fla.)

2021 worst: Bair Hugger (8th), Hamer (3d), Blackburn (11th), Valsartan (D.N.J.), Gaetana (D.N.J.), Johnson (E.D. Mo.), Smith&Nephew (D.Md.), Munson (N.D. Miss.), Abrams (9th), Fluoroquinolone (D. Minn.), Allergan (D.N.J.)

2022 best: Nexus Pharms. (9th), Zantac (S.D. Fla.), Fosamax (D.N.J.), Nelson (5th), City of Huntington (S.D. W. Va.), Onglyza (E.D. Ky.), Zostavax (E.D. Pa.)

2022 worst: Nicholson (8th), Thacker (6th), Jacob (11th), DeCostanza (E.D.N.Y.), Bueno (S.D. Cal.), Blackburn (11th)

2023 best: Zofran (1st), Cates (11th), Acetaminophen (S.D.N.Y.), Fresenius (D. Mass.), Kampmann (C.D. Ill.)

2023 worst: Valsartan (D.N.J.), Painters (C.D. Cal.), Sykes (7th), Holley (N.D. Cal.), Wilson (S.D. Ill.), CPAP (W.D. Pa.), Crockett (E.D. Pa.), Keralink (4th)

2024 best: Hickey (5th), Onglyza (6th), Bueno (S.D. Cal.), Dickson (W.D. La.), Zostavax (3d), Oglesby (5th), Oral Phenylephrine (E.D.N.Y.)

2024 worst: Fosamax (3d), Huertas (3d), Valsartan (D.N.J.), Dressen (D. Utah), Ahmed (D. Ala.)

2025 best: Sprafka (8th Cir.), Gardasil (4th), Wilson (S.D. Ill.), Valsartan (D.N.J.), Gardasil (W.D. N.C.), DepoProvera (N.D. Fla.), Peptides (E.D. Pa.)

2025 worst: City of Huntington (4th), Dibble (8th), Avandia (E.D. Pa.), In re Phillips (W.D. Pa.), CLF 007 v. Cooper Surg. (D. Or.), Boncher (E.D. Pa.), Hollestein (D.N.J.), Thacker (E.D. Ky.)

What do these results tell us? First, there are no D.C. Circuit cases. None. Let that be a monument to our prescience. Second, federal cases make up approximately two-thirds of the cases on our Best and Worst lists.  A decent chunk of the action remains in state courts.  That is not much of a surprise.  We all know that most plaintiff lawyers prefer to be in state courts.  Maybe we should do a separate analysis of the state cases.

Another thing you notice is that, wow, rotten opinions from the Valsartan MDL really were as numerous as cockroaches in a greasy spoon joint. Thankfully, there is a new sheriff in town for that case. That is why, after years of crowding our Worst lists, a Valsartan opinion made it onto our last Best list. There are, of course several repeat players besides Valsartan: Blackburn, Bueno, Thacker, Fosamax, Taxotere, Huntington, etc. Some cancel each other out (reversals), and some compound things (affirmances). Look, we are not doing a regression analysis here. 

For the good cases, if we look strictly at circuit court rulings, the Fifth Circuit has four cases, which is double that of its nearest competitors, the Second and Eleventh Circuits. If we count the lower court decisions, the Third Circuit leaps up to tie the Fifth Circuit, with six cases. The Second and Ninth Circuits each produce four decisions if you count the lower courts.

For the bad cases, if we look strictly at circuit court rulings, the Third, Eighth, and Eleventh Circuits lead the pack with three decisions each. If we count the lower court decisions, things get kind of crazy.  The Third Circuit now has 17 decisions, far more than the seven from the Ninth Circuit or six from the Eighth Circuit. What does that off-the-chart number tell us?  Does it confirm that the Pennsylvania and New Jersey courts hog the pharma caseload? And in a really bad way? Or does it simply mean that the location of most of the DDL bloggers in Princeton and Philly makes us a bit short-sighted? Maybe there should be a DDL version of that famous New Yorker cartoon; Philly and Jersey occupy most of the landscape,  with far-flung jurisdictions barely in view at all. Perhaps what really matters is the friends we have made along the way, and most of our friends are hunkered down along the Delaware River.

Did you notice that Acetaminophen case from our 2023 Best list? This year the Second Circuit reversed. That opinion is a strong candidate to make our 2026 Worst list. Oh well.

So there it is.  Subject to the tyranny of small numbers, seasoned with a moderate dose of selection bias, what we learn is that good and bad DDL cases are all over the place.  The Third Circuit is important as a vast mixed bag. Consider that hypothesis at least partially validated.  The Fifth Circuit is pretty good for our side of the v.  The Ninth Circuit? Not so much. What else is new?

To be sure, we are not announcing any firm, major conclusions from this review.  We certainly would not suggest that the database and methodology we deployed would pass a Rule 702 inquiry.  

Except maybe in the Ninth Circuit. 

Photo of Michelle Yeary

Every sport has that one venue where the home team gets every close call.  Fenway in the late 2010s. Lambeau Field in a January snowstorm. Or, if you are a Baltimore Orioles fan — Yankee Stadium in Game 1 of the 1996 American League Championship Series, where a twelve-year-old kid with a glove reached over the fence, and an umpire who should have known better decided he had not seen what everyone else on the planet had just watched on the replay (yes, still bitter). And now, apparently, the Complex Litigation Docket in Waterbury, Connecticut, is the place for plaintiffs suing over Filshie Clips.

We have chronicled the Filshie Clip litigation extensively on this blog, and the pattern has been remarkably consistent — plaintiffs allege that the clips migrate more often than the 0.13% rate reported to the FDA, that the defendant should have reported more of those migrations as adverse events, and that better reporting would somehow have led to better warnings that would somehow have changed a physician’s or patient’s decision to use the product. And in courtroom after courtroom, from Texas to Georgia to Illinois to Florida to Kentucky, that theory has gone down swinging on preemption grounds. We have practically run out of ways to say “another win for the defense.”

Then the litigation came to Connecticut, and the string snapped. The case is In re 2024 Connecticut Filshie Clip Litigation, 2026 WL 2322193 (Conn. Super. Aug. 4, 2026). This is the consolidated proceeding for all Filshie Clip cases filed in 2024.  Defendant moved to strike the plaintiffs’ complaints in their entirety as preempted by the FDCA. The court split the baby. It struck the design defect claims, but let the failure-to-warn claims — which are really failure-to-report claims wearing a warnings costume — survive.

On the design defect side, the ruling is exactly what it should be. The Filshie Clip is a Class III device that went through PMA, so Riegel preemption applies with full force. The plaintiffs never alleged that the clip’s design deviated from what the FDA approved, and without that allegation, a design defect claim is nothing more than an attack on the FDA’s own safety and effectiveness determination. No surprises there, and no complaints from us.

The failure-to-warn ruling is where things go sideways, and it goes sideways for a reason we predicted years ago. Back in 2022, we wrote about the Connecticut Supreme Court’s decision in Glover v. Bausch & Lomb, and we did not mince words about it then either. The Second Circuit had certified a question to the Connecticut Supreme Court asking whether the Connecticut Product Liability Act (“CPLA”) recognizes a cause of action based on a manufacturer’s failure to report adverse events to the FDA. The right answer, and the answer nearly every other state has given, was no. Connecticut said yes.

The theory, as we described it at the time, requires treating the FDA as a kind of substitute learned intermediary — as if federal regulators, rather than a treating physician, are “in the best position” to receive warnings and pass them along to patients. That is a strange fit for a state statute built around the traditional learned intermediary framework, and it opens exactly the door that Buckman was supposed to close — private plaintiffs using state tort law to enforce a manufacturer’s reporting obligations to a federal agency. It was a bad decision in 2022 and nothing about watching it play out in the Filshie Clip litigation has changed our minds.

In fairness, the current ruling is almost mechanical given that the Second Circuit had already blessed the outcome twice in the Glover trilogy — Glover I certifying the question, Glover II answering it, and Glover III applying the Connecticut Supreme Court’s answer to hold that the claims survived preemption. Once the In re 2024 Filshie Clip court had that binding roadmap sitting on the shelf, denying the motion to strike as to the failure-to-warn claims was less a judicial decision than a formality. The court dutifully found that the plaintiffs’ allegations fit within the same “narrow gap” Glover carved out — plaintiffs suing for conduct that happens to violate the FDCA, but not suing because it violates the FDCA. It is a distinction that sounds meaningful in the abstract and evaporates the moment you apply it to a complaint that is, from top to bottom, about underreporting adverse events to a federal regulator.

The court also brushed aside the mountain of contrary Filshie Clip authority from other jurisdictions because those states, unlike Connecticut, do not have a duty that “parallels” the federal reporting requirement. That is technically accurate, and it is exactly the problem. As we have noted repeatedly, and as our 50-state survey documents, the overwhelming majority of states do not recognize any tort duty to report adverse events to the FDA, which means failure-to-report theories in those states are preempted twice over. Expressly, because they seek to impose a requirement different from or in addition to federal law, and impliedly, because without a genuine parallel state duty, the claim is just private FDCA enforcement dressed up in state-law language. Connecticut is an outlier that manufactured its own parallel duty out of whole cloth, and now it is reaping exactly the litigation it invited.

There is a reason this result was entirely predictable rather than some bolt from the blue. The plaintiffs’ bar keeps running the identical playbook against this identical product. The Filshie Clip is a contraceptive device — titanium clips with a silicone lining, placed on the fallopian tubes for tubal ligation — and it has become one of the plaintiffs’ bar’s preferred vehicles for a broader campaign against reproductive choice. These cases are based on an inflated migration rate untethered to what the FDA actually requires to be reported. The FDA’s reporting obligations are triggered only by events that are potentially injurious; the overwhelming majority of clip migrations are asymptomatic and cause no injury at all. Plaintiffs’ lawyers know this, which is exactly why their complaints omit it. In other jurisdictions, that omission sometimes is enough to survive a motion to dismiss, but as the summary judgment records have shown, the theory falls apart once actual evidence is required.

What Glover does in Connecticut is worse. It does not just let artful drafting slip past a skeptical judge, it manufactures the underlying cause of action itself, so there is no field-of-dreams uncertainty about whether the claim exists at all once discovery starts. The tragedy is how much discovery expense a defendant has to absorb before any of these claims meet their inevitable end, and Connecticut made that expense a certainty rather than a probability.

In addition, the In re 2024 Filshie Clip opinion leaves two loose threads that matter quite a bit for what happens next. First, the court never tells us how many of the plaintiffs in this consolidated docket are actually Connecticut residents, or had their clips implanted in Connecticut, such that Connecticut law — and Glover’s idiosyncratic reporting duty — would even govern their claims. One of the Filshie Clip defendants happens to be headquartered in Connecticut, which invites the obvious question whether plaintiffs from elsewhere are trying to borrow Connecticut law simply because the defendant can be found there. Second, the opinion does not say how many plaintiffs are part of this litigation at all. Both numbers matter for a choice-of-law fight that could meaningfully shrink the practical reach of this ruling, and both are conspicuously absent from the decision.

So, is Waterbury actually the house that Jeffrey Maier built for the Filshie Clip plaintiffs’ bar, the one park where the umpire calls it a home run no matter what actually happened on the field? Not exactly. Design defect claims still struck out looking, and if the choice-of-law questions get asked (and they should be), the failure-to-warn survivors on this roster may turn out to be a much smaller team than the complaint suggests. Glover is the twelve-year-old fan reaching over the wall — it does not change what the ball actually did, it just changes what the umpire is willing to call. But make no mistake, as long as Glover remains good law, Connecticut will keep ruling that pitch a home run when every other park in the country would call it exactly what it is. We told you back in 2022 that this call was coming. Now we are just watching the kid reach over the fence.

Photo of Bexis

On July 6, 2026 the FDA announced the results of a study it recently completed on tampons (which are medical devices) and their “Biocompatibility and Toxicology.”  The Agency’s conclusion was:

[T]esting conducted by FDA laboratories and a toxicological risk assessment related to metals found in tampons . . . found that, while trace metals are present in tampons, the amount released during use is too small to cause harm.  This study is consistent with findings from an independent systematic literature review previously commissioned by the FDA.

FDA, “Biocompatibility and Toxicology Program:  Research on Medical Devices, Biocompatibility, and Toxicology,” at Update dated July 6, 2026) (emphasis added).  Here’s a link to the study itself, which is Jin, et al., “Do metals in tampons pose a health risk? A toxicological risk assessment study,” 209(6) Toxic’l Scis. ??? (May 2026).  And here are excerpts from the study abstract (available at the link):

Recently concern has been raised over metal content in tampons and the possibility of toxicological risk to women. As a result, researchers at the United States Food and Drug Administration (FDA) investigated metals present in tampons and estimated the maximum exposure and toxicological risk associated with tampon use.  The levels of metals contained in, and released from 11 tampon products, which are legally marketed in the United States, were analyzed. . . .  19 metals were detected including . . . arsenic (As), cadmium (Cd), and lead (Pb) were found at trace levels.  Based on the in-depth [toxic risk assessment], all detected metals had margins of safety (MoS) indicating that there is negligible toxicological concern from the presence of metals in tampon materials.

(Emphasis added).  The study was written by FDA employees, so it is in the public domain.

Given these conclusions, there isn’t likely to be any “newly acquired information” that would allow a plaintiff to claim that state law demands a warning about any or all of these metals being present in tampons.  That’s important because of Medtronic, Inc. v. Lohr, 518 U.S. 470 (1996).  Tampons are considered Class II §510(k)-cleared medical devices, and until Lohr is overruled, as it should be, the only kind of preemption available to §510(k) devices is implied preemption.

Having said that, we should point out that this isn’t necessarily true for tampon warnings.  An FDA regulation, unique to tampons, 21 C.F.R. §801.430, specifies certain aspects of these products’ warnings – those concerning toxic shock syndrome (“TSS”).  Those FDA-specified warnings have been accorded preemptive effect as “device specific,” even under Lohr’s improper, atextual reading of the relevant “different from or in addition to” preemption clause (21 U.S.C. §360k(a)).  See Murphy v. Playtex Family Products Corp., 69 F. Appx. 140, 143 (4th Cir. 2003), aff’g, 176 F. Supp.2d 473, 487-88 (D. Md. 2001); Haddix v. Playtex Family Products Corp., 138 F.3d 681, 686 (7th Cir. 1998); Papike v. Tambrands, Inc., 107 F.3d 737, 741-42 (9th Cir. 1997); Choi v. Kimberly-Clark Worldwide, Inc., 2019 WL 4894120, at *7-8 (C.D. Cal. Aug. 28, 2019); Wichmann v. P&G Manufacturing Co., 2005 WL 2176922, at *2 (E.D. Mo. Sept. 8, 2005);

However, FDA-mandated tampon warnings under §801.430 are limited to TSS, and in Barton v. Kimberly-Clark Corp., 2025 WL 486316 (S.D. Cal. Feb. 13, 2025), the court rejected an express preemption argument.  The plaintiff’s “allegations d[id] not deal with TSS at all.” and “only target[ed] Defendant’s voluntary Representations” about its tampons.  Id. at *4-5.  Thus, Barton held:

The analysis turns on what “different from, or in addition to” actually means.  In 21 C.F.R. §808.1(d), the FDA interprets the preemption provision to mean that State and local requirements are preempted only when the Food and Drug Administration has established specific counterpart regulations or there are other specific requirements applicable to a particular device under the act, thereby making any existing divergent State or local requirements applicable to the device different from, or in addition to, the specific Food and Drug Administration requirements.  In other words, to consider a State or local requirement as being “in addition to” an FDA regulation, the State or local requirements must be, at a minimum, related to the specific regulation.

Id. at *5 (Papike citation omitted) (emphasis added).  The emphasis in the quotation is to underscore that Barton turned on the Chevron deference-driven limitation that the FDA imposed on §360k(a) – precisely what Loper Bright Enterprises v. Raimondo, 603 U.S. 369 (2024), held that administrative agencies can’t do, and that courts are now required to ignore.

Coming full circle, the plaintiff in Barton was suing over precisely what the FDA has now stated are bogus allegations.  The allegations were, first, that the defendant “failed to disclose, and materially omitted, that [tampons] contain lead – supposedly a violation of California consumer protection laws.  2025 WL 486316, at *1.  Second, plaintiff’s liability theory was based on the contention that “[t]here is no level of exposure to lead that is known to be without harmful effects.”  Id.  That’s exactly the opposite of what the FDA has just concluded – based on its own published scientific study.  Too bad that, according to PACER, Barton resolved last year.

However, plaintiff Barton is a serial litigator and has a second suit in which the preemption issue has yet to be decided.  See Barton v. Procter & Gamble Co., 766 F. Supp.3d 1045 (S.D. Cal. 2025) (deciding motion to dismiss that did not involve preemption).  The allegations about lead in tampons, and that there is purportedly no safe exposure level, are the same.  Id. at 1054.  On such allegations, there are now viable arguments on both express and implied preemption.

On express preemption, the argument would be grounded on the proper extent of the applicable preemption language, as stated in Monsanto Co. v. Durnell, 146 S. Ct. 2001 (2026).

The question, then, is whether the [state-law] failure-to-warn claim − which would require a . . . warning on the [product] label − would impose a labeling requirement that is “in addition to or different from” federal labeling requirements. . . .  The answer is yes. . . .  If a manufacturer does not use the [federally]-approved label, it may be subject to civil and criminal penalties. . . .  [A]bsent such an [agency]-approved or [agency]-required label change, the [regulated] manufacturers may − and indeed legally must − use the [product] label approved by [the agency] at registration.  In sum, federal law requires [defendant] to sell [its product] with the label that [the agency] approved . . . − that is, the label without [the plaintiff’s] warning.  [Plaintiff’s] state tort claim, by contrast, would require [defendant] to add a . . . warning to its labels.  That [state]-law requirement is “in addition to” and “different from” [defendant’s] federal-law labeling obligations.

Id. at 2010-11 (citations and footnote omitted).  A Durnell-based approach would, of course, require a long-overdue direct challenge to Lohr, since tampons are §510(k) devices.

Alternatively, an implied preemption argument is now available in light of the FDA’s recent actions.  While implied preemption has not been much utilized in medical device litigation, the Supreme Court has held, specifically in the medical device, context, that the two forms of preemption operate independently.  Buckman Co. v. Plaintiffs Legal Committee, 531 U.S. 341, 352 (2001) (“neither an express pre-emption provision nor a saving clause bars the ordinary working of conflict pre-emption principles”) (citation and quotation marks omitted).

The basis for implied preemption in the FDCA context is well settled – either:  (1) the plaintiff cannot establish the prerequisites of the FDA’s “changes being effected” regulations, in which case preemption exists because the defendant cannot change its labeling independently of FDA pre-approval (what we call the “Mensing independence principle”); or (2) the CBE rule is satisfied, in which case the defendant must come forward with “clear evidence” that the FDA would not agree to the label change that the plaintiff demanded.  See generally Bexis’ 2020 FDLI article on Merck Sharp & Dohme Corp. v. Albrecht, 587 U.S. 299 (2019), here, at 12-15 (“Impact”).

Implied preemption under the first prong turns on whether “newly acquired information” – data that the FDA did not previously have available – exists, so that a unilateral CBE label change was possible.  The second prong must meet Albrecht’s test for “clear evidence.”  As for tampons and purported lead (or other metallic) “contamination,” the FDA’s July 6, 2026, announcement and the accompanying study published by FDA researchers, should satisfy both tests.  However, the “clear evidence” prong requires more in the way of regulatory technicalities.  But as for the first prong’s “newly acquired information” requirement, since the Agency has just stated that “while trace metals are present in tampons, the amount released during use is too small to cause harm,” we see no chance that any plaintiff could possibly meet applicable FDA standards for such information.  Hence, tampon-related warning claims should be preempted, Lohr or no Lohr, under this prong of FDCA-based implied preemption.

So, it looks like time to revisit preemption in at least some tampon product liability litigation.

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