If asbestos litigation reminds of the 100 Years War, and the Bone Screw litigation recalls (for us) the winning side of the Franco-Prussian War, or Napoleon’s Italian Campaign − pelvic mesh litigation seems like something out of a different war. Try World War I. It’s not quite the Somme (at least not yet) but more like Verdun, and our side has yet to find its “They shall not pass” moment.
Our latest example is Kaiser v. Johnson & Johnson, ___ F.3d ___, 2020 WL 205651 (7th Cir. Jan. 14, 2020). Kaiser is the latest rotten precedent out of the pelvic mesh litigation. Rejecting what it described as the defendant’s “broad-spectrum attack on the judgment,” id. at *1, the appellate court affirmed a $20 million verdict that included a (reduced) $10 million in punitive damages. Id.
When a court takes the position that the FDA does not exist (except when plaintiffs say it can), then perhaps this type of result is logical. Logical, but not rational. As with most pelvic mesh cases, design defect is front and center. In Kaiser, Indiana law governed, which meant that the unfortunate Indiana Supreme Court decision in TRW Vehicle Safety Systems, Inc. v. Moore, 936 N.E.2d 201 (Ind. 2010), controlled. With surprising little analysis – mostly a footnote, Moore rejected an alternative design requirement under Indiana’s negligence-based product liability statute:
For actions based on an alleged product design defect, however, the Act . . . specifies a different standard of proof: “[T]he party making the claim must establish that the manufacturer or seller failed to exercise reasonable care under the circumstances in designing the product.” Ind. Code §34-20-2-2. Thus the statute itself prescribes the applicable standard of care. We decline to require proof of any additional or more particular standard of care in product liability actions alleging a design defect.2
2The American Law Institute recommends a different approach, prescribing . . . adoption of a reasonable alternative design . . ., and [that] the omission of the alternative design renders the product not reasonably safe.” Restatement (Third) of Torts: Products Liability § 2(b) (1997). Our legislature did not adopt this analytical framework but instead enacted in 1998 a negligence standard for product liability claims based on defective design.
936 N.E.2d at 209 & n.2 (second citation to statute omitted). That’s an unusual ruling, because negligent design – which Moore repeatedly recognized it was applying, id. at 209-10 (four references to negligence) – at common law almost always has included an alternative design element, as we discussed here. But the Indiana high court is the final arbiter of Indiana law, and if it chooses to interpret a statutory carve-out from strict liability as an invitation to expand liability, that’s for the legislature to fix. So we don’t like it, but that’s where Indiana law apparently stands at the moment.
Thus, Kaiser recognized the abrogation of prior Seventh Circuit precedent that had enforced an alternative design requirement in Indiana product liability cases. 2020 WL 205651, at *11 (“[o]ur circuit caselaw cannot be reconciled with [Moore]”). OK, we’re big fans of Erie restraint, so we have to take the good with the bad.
But if, as Moore held, the Indiana product liability imposes an negligence “reasonable” person standard in design defect cases, why the heck is the next section of Kaiser entitled “Unreasonably Dangerous”? We frankly don’t know, and we think Kaiser got this completely wrong. The “unreasonably dangerous” element of Indiana statutory product liability comes from Ind. Code §34-20-2-1 (creating general “defective condition unreasonably dangerous” standard). But §34-20-2-2, construed in the just-discussed Moore case, provides
The rule stated in section 1 of this chapter applies. . . .
However, in an action based on an alleged design defect in the product or based on an alleged failure to provide adequate warnings or instructions regarding the use of the product, the party making the claim must establish that the manufacturer or seller failed to exercise reasonable care under the circumstances in designing the product or in providing the warnings or instructions.
(Emphasis added). So the “unreasonably dangerous” test – and the consumer expectation gloss that Ind. Code §34-6-2-146 gives to that test − doesn’t even apply to warning or design claims, which are governed by negligence principles according to §34-20-2-2.
Kaiser is an example of “heads plaintiffs win; tails defendants lose.” Consumer expectation is the statutory definition of “unreasonably dangerous,” but the very same statutory carve-out that Kaiser relied upon to avoid defendant’s alternative design, expressly excludes design and warning claims from said unreasonably dangerous requirement, the one that Kaiser invoked to conclude that “a reasonable jury could conclude that [the product] was unreasonably dangerous” despite the implanting surgeon “testif[ying] that he was aware of many of [product’s] risks.” 2020 WL 205651, at *12. Thus, it appears to us that, when negligence principles benefited plaintiff’s design defect claim in Kaiser, negligence applied, but when strict liability principles could defeat a defense argument, then the same claim was treated as strict liability.
And then there’s the FDA, which cleared this device, and which would have to clear a supplemental application before any design change materially affecting device safety and effectiveness could occur.
But this is mesh world. In order to ensure that plaintiffs can win – and thus force defendants to settle – the FDA does not exist, except when plaintiffs want it to exist. Kaiser stated, about the “original” 1988 FDA clearance of the predicate device on which the agency’s later clearance of this particular product was based:
When it promulgated that rule, the FDA cautioned that the “surgical mesh has not been implanted in a sufficient number of patients by a sufficient number of medical practitioners to provide adequate evidence on the long-term biocompatibility of these devices.” General and Plastic Surgery Devices, 53 Fed. Reg. 23856, 23862 (June 24, 1988). Due to “insufficient evidence of safety and effectiveness,” the FDA assigned these surgical meshes to Class II. Id.
Kaiser, 2020 WL 205651, at *4.
That didn’t make sense to us. If there weren’t “adequate evidence on the long-term biocompatibility” of what was supposed to be a permanent implant, then the FDA wouldn’t have cleared that product at all, and certainly wouldn’t have put it in an intermediate classification like Class II.
Something wasn’t right, so we looked up the Federal Register reference that Kaiser cited. Sure enough, to justify ignoring the FDA’s safety and effectiveness conclusion, Kaiser mangled it beyond recognition. Here, in context, is what the FDA was actually discussing in the snippet cited in Kaiser:
Two comments said that the device is made from well-established materials and that general controls would provide reasonable assurance of the continuing reliability of the device. The comments requested that the device be classified into class I instead of class II, based on safe and effective clinical use over the years.
FDA disagrees with the comments. FDA is classifying the device into class II to control the risks to health of infection and foreign body reaction which may result in implant rejection. Surgical mesh is intended to be implanted in the human body. Section 513(d)(2)(B) of the act (21 U.S.C. 360c(d)(2)(B)) requires that FDA classify all implants into class III unless the agency determines that, for a particular implant, premarket approval is unnecessary to provide reasonable assurance of its safety and effectiveness. FDA believes that surgical mesh has not been implanted in a sufficient number of patients by a sufficient number of medical practitioners to provide adequate evidence on the long-term biocompatibility of these devices. Consequently, FDA believes that insufficient evidence of safety and effectiveness is available at this time to support classifying surgical mesh into class I.
General & Plastic Surgery Devices; General Provisions and Classifications of 51 Devices, 53 Fed. Reg. 23856, 23681-892 (FDA June 24, 1988) (emphasis added).
The FDA’s decision was almost the exact opposite of what Kaiser cited it for. The FDA rejected only a proposal to classify the mesh in question into its lowest classification, Class I. The FDA acknowledged that, if there was not “reasonable assurance of [a device’s] safety and effectiveness,” a Class III designation would have been “require[d].” It nonetheless put these mesh predicates in Class II. Why? Elsewhere in the same document, the FDA also rejected comments that the mesh predicates be placed in Class III. Class III was unnecessary because:
- “FDA has determined that requirements of premarket approval are unnecessary to control the risks to health presented by the devices, including the risk of bioincompatibility.” Id. at 23861.
- “FDA believes that the biocompatibility of the materials now being used in these devices has been established through their successful use for a number of years.” Id.
- “Clinical experience with the devices listed above has established the persons for whose use the devices are intended and the proper conditions of use. FDA has determined that the probable benefit to health from proper use of these devices outweighs and likelihood of illness or injury resulting from their use.” Id.
- “FDA believes that establishment of performance standards will provide reasonable assurance of the safety and effectiveness of these devices and sufficient evidence is available to establish such standards.” Id.
Kaiser thus mischaracterized the 1988 administrative record to suggest a finding of “insufficient evidence of safety and effectiveness” for any purpose. That can’t be so, because such a finding would have precluded any clearance at all. The full administrative record demonstrates that Kaiser’s implication was false – and, in fact, the FDA actually made a positive finding of “reasonable assurance of the safety and effectiveness of these devices” when they were subject to Class II “performance standards.”
We went through this regulatory history in detail because we think that the judicial gyrations in Kaiser, in and of themselves, demonstrate why the exclusion of FDA evidence in Kaiser, 2020 WL 205651, at *16, as supposedly “minimally” probative was error and should have been reversed. The statement in Kaiser that “[t]he FDA warned that [product’s] original predicates − the surgical meshes in the 1988 rulemaking − did not come with a reasonable assurance of safety,” id. simply ain’t so. Facts matter, and the fact is that in 1988 “FDA believe[d] that establishment of performance standards will provide reasonable assurance of the safety and effectiveness of these devices.” 53 Fed. Reg. at 23861. That Kaiser found it necessary to mischaracterize the regulatory history so thoroughly only underscores how relevant and probative that history really was.
Finally, preemption. We’ll give Kaiser this – is did treat the defendant’s implied preemption argument seriously. The express preemption argument for this product is a difficult one, since its pedigree stretches back to 1988, which was before the Safe Medical Devices Act of 1990 revamped the §510(k) process. To find express preemption, one would have to argue that Medtronic, Inc. v. Lohr, 518 U.S. 470 (1996), is wrong, and not merely based on a superseded statute (for more background see our discussion here). That may be true. Compare:
Despite its relatively innocuous phrasing, the process of establishing this “reasonable assurance,” [of safety and effectiveness] which is known as the “premarket approval,” or “PMA” process, is a rigorous one.
Lohr, 518 U.S. at 477 (emphasis added), with:
Clinical experience with the devices listed above [including mesh] has established the persons for whose use the devices are intended and the proper conditions of use. FDA has determined that the probable benefit to health from proper use of these devices outweighs and likelihood of illness or injury resulting from their use. . . . FDA believes that establishment of performance standards will provide reasonable assurance of the safety and effectiveness of these devices and sufficient evidence is available to establish such standards.
53 Fed. Reg. at 23861 (emphasis added). This regulatory record seems to us to establish that the FDA in fact used the same “reasonable assurance” standard for this Class II device that Lohr described as “rigorous.” That was in 1988, prior to SMDA, and indicates that the FDA was already holding Class II medical devices to the “reasonable assurance” standard. Lohr’s blanket holding that all non-pre-market approved devices are “marketed without” this “rigorous” FDA review, 518 U.S. at 478, is not borne out by the regulatory history of at least this particular Class II mesh device.
Unlike a lot of prior decisions, which fail to distinguish between types of preemption in Class II device cases and simply chant “Lohr, Lohr, Lohr” regardless of express or implied preemption, Kaiser gave the implied preemption argument full (if questionable) treatment. “[Plaintiff] argues that Lohr forecloses [defendant’s] preemption defense without further inquiry. We disagree. Lohr addressed a question of express preemption; [defendant] argues here for implied preemption.” Kaiser, 2020 WL 205651, at *7. Kaiser also correctly identified the test imposed by the Mensing independence principle. “The question for ‘impossibility’ is whether the private party could independently do under federal law what state law requires of it.” Id. at *8 (quoting PLIVA, Inc. v. Mensing, 564 U.S. 604, 620 (2011)).
What did the FDA have to say about independent action and design defects? In the same 1988 regulatory action, the agency stated, “FDA believes that a change in a material used in a device intended to be implanted is a significant change in a device that could affect its safety and effectiveness.” 53 Fed. Reg. at 23861 (emphasis added). Thus, “FDA will require that new or significantly changed materials be subject to requirements of premarket approval.” Id. For preemption purposes, this means that any change in the composition of this product would require FDA pre-approval, which under Mensing requires preemption.
So how did Kaiser, after setting up the proper test, nonetheless get around preemption? By creating a circuit split. In Mutual Pharmaceutical Co. v. Bartlett, 570 U.S. 472 (2013), the Court held that state law cannot demand, consistently with the Supremacy Clause, that the manufacturer of a product that the FDA allows to be marketed – even via abbreviated review – stop marketing its product. Id. at 488. In Yates v. Ortho-McNeil-Janssen Pharmaceuticals, Inc., 808 F.3d 281, 300 (6th Cir. 2015), the court held that a claim that the product design the FDA initially approved was “defective” was merely a stop selling claim under a different name. We discussed Yates here.
Kaiser disagreed with Yates:
[T]he Court’s generic-drug rulings turn on whether a “generic manufacturer may change [its] label[ ] after initial FDA approval.” [Mensing], 564 U.S. at 613, 131 S.Ct. 2567. The important point for this case, however, is that [defendant] had complete and independent control over [its product’s] design before it sought §510(k) clearance for the device. It was not impossible to simultaneously comply with federal and state law.
Kaiser, 2020 WL 205651, at *7. That’s all true, but this factual distinction simply ignores the holding in Mensing. We repeat, “[t]he question for ‘impossibility’ is whether the private party could independently do under federal law what state law requires of it.” 564 U.S. at 620. The test Mensing poses turns on “whether,” not “why.”
Before the Manufacturers could satisfy state law, the FDA − a federal agency − had to undertake special effort permitting them to do so. To decide these cases, it is enough to hold that when a party cannot satisfy its state duties without the Federal Government’s special permission and assistance, which is dependent on the exercise of judgment by a federal agency, that party cannot independently satisfy those state duties for pre-emption purposes.
Id. at 623-24 (emphasis added). Similarly, in Kaiser is should have been “enough to hold” that the FDA would have required a new application, whether in place of the one that the defendant originally submitted, or at some later date, to implement the design change that state tort law supposedly required here. The “impossibility” that undergirds preemption here is two-fold: First, that simultaneous compliance becomes impossible when independent agency action is required. Check. Second, once the FDA has said “yes” to marketing,” a state cannot say “no.” Check. Still not sure? There’s always the Shuker approach of asking the FDA directly for its views.
To sum up, Kaiser butchered the regulatory record of this product, and used the dismembered remnants both to justify exclusion of FDA evidence and to wriggle around preemption. Further, Kaiser’s treatment of Indiana design defect law is internally inconsistent so that plaintiff wins either way. Thus, Kaiser is an early candidate for our “worst of “ list of 2020.