Photo of Michelle Yeary

Yesterday the Supreme Court granted plaintiff’s petition for a writ of certiorari in Mallory v. Norfolk Southern Railway Co. – a Pennsylvania Supreme Court case holding that registering to do business in a state does not by itself subject a corporation to general jurisdiction in that state.  266 A.3d 541 (Pa. 2021), discussed here.

As argued by the defendants in opposition to certiorari, and often by us on the DDL Blog, the issue of where a corporation is subject to general jurisdiction was decided in Goodyear Dunlop Tires Operations, S.A. v. Brown, 564 U.S. 915 (2011), and Daimler AG v. Bauman, 571 U.S. 117 (2014).  That’s at most two places, where its incorporated and where it has its principal place of business.  And courts to have considered the obsolete concept of general jurisdiction by “registration” or by consent since Daimler have overwhelmingly rejected the argument.  We detailed the state-by-state tally in our 50-state survey.   Anything else would swallow the rule.  Given that most nationally active corporations, including drug and device manufacturers, are registered to do business in every state, they would be subject to general jurisdiction in every state if jurisdiction by consent were allowed.  You can find more on jurisdiction by consent in our cheat sheet too.

The only notable outlier post-Daimler is Cooper Tire & Rubber Co. v. McCall, which upheld Georgia’s consent-by-registration scheme against a due process challenge. 863 S.E.2d 81 (Ga. 2021).  A petition for certiorari is also pending in that case which turns on a quirk in Georgia law that even the Georgia court recognized was inconsistent with Supreme Court precedent.  It’s quite possible that Cooper Tire will now be held in limbo pending the Court’s Mallory decision.

So, with all of that we will watch for this decision to hopefully put to rest this issue once and for all.

 

 

Photo of Bexis

Federal Rule of Evidence 407, dealing with the inadmissibility of subsequent remedial measures, isn’t all that complicated – as Federal Rules go.  It’s only 73 words long:

When measures are taken that would have made an earlier injury or harm less likely to occur, evidence of the subsequent measures is not admissible to prove:

  • negligence;
  • culpable conduct;
  • a defect in a product or its design; or
  • a need for a warning or instruction.

But the court may admit this evidence for another purpose, such as impeachment or − if disputed − proving ownership, control, or the feasibility of precautionary measures.

It would have been even shorter, if certain courts hadn’t insisted on reading a strict liability exception into Rule 407 that simply wasn’t there.  Thus Rule 407 was amended in 1997:

to provide that evidence of subsequent remedial measures may not be used to prove “a defect in a product or its design, or that a warning or instruction should have accompanied a product.”  This amendment adopts the view of a majority of the circuits that have interpreted Rule 407 to apply to products liability actions.

F.R. Evid. 407, Committee Notes to 1997 amendments.

Trouble is, courts still can’t keep their hands off of Rule 407’s very simple language – if the claimed injury was “earlier” than the “measure” the defendant had “taken,” and there is no other “purpose” for the evidence at issue, it’s excluded.

Back in 2010, we examined one restrictive judicial gloss on Rule 407 – a “policy” based exception that the rule somehow doesn’t apply to government-mandated measures.  That caveat was made up out of whole cloth, since the language of Rule 407 draws no such distinction, nor is any caveat mentioned in any committee note.  Fortunately, that 2010 decision has been a pro-plaintiff anomaly, and FDA required label changes, recalls, etc. continue to be subject to Rule 407-based exclusion.  See, e.g., Yates v. Ortho-McNeil-Janssen Pharmaceuticals, Inc., 808 F.3d 281, 292 (6th Cir. 2015) (“we will not consider evidence of a subsequent improvement to a drug label as evidence of prior failure to warn, even in cases such as this in which the FDA mandated the change in labeling”) (citing Rule 407).  Bexis collects those cases in §11.01[2][a] nn.101-106 of his book.

Having two strikes against them hasn’t stopped the other side from trying to conjure up other Rule 407 exceptions, however.  Our present beef is with a couple of decisions purporting to engraft a causation/intent requirement onto Rule 407 – even though the current rule nowhere discusses causation.  Thus in Godelia v. Zoll Services, LLC, 2019 WL 3883682, at *1-2 (S.D. Fla. Aug. 16, 2019), the court inexplicably held that that a post-injury corporate audit report wasn’t excludable under Rule 407 because it was not prompted by the incident that injured the plaintiff:

The record does not reflect that Defendant retained [the outside auditor] to conduct the audit to make the harm suffered by [plaintiff] less likely to occur or to remedy any issues Defendant had with corrective and preventative action for [the claimed defect]. . . .  Defendant believed that there was a communication issue with the FDA.  In response to this belief, Defendant retained [the auditor] − not to remedy [plaintiff’s claimed] defects or regulatory failures. . . .  Accordingly, this analysis and Audit Report are not subsequent remedial measures and are admissible.

Id. at *2 (footnote omitted).  See In re Davol, Inc./C.R. Bard, Inc., Polypropylene Hernia Mesh Products Liability Litigation, 518 F. Supp.3d 1028, 1036-37 (S.D. Ohio 2021) (relying on “policy” of Rule 407 against “deterrence” to avoid exclusion of a subsequent remedial measure “not trigger[ed]” by “Plaintiff’s injury”).

The interpretation of Rule 407 in those cases goes beyond mere judicial gloss.  Rather, it flies in the face of the most recent (2011) amendments to Rule 407.  Before 2011, Rule 407 used to at least mention causation − the first clause provided that “[w]hen, after an injury or harm allegedly caused by an event, measures are taken. . . .” (Emphasis added).  The 2011 amendments eliminated even that oblique reference, substituting the current language focusing solely on result:  “When measures are taken that would have made an earlier injury or harm less likely to occur. . . .”  Further, since the 2011 amendments omitting causation altogether were not intended to have substantive effect, see Committee Notes to 2011 Amendments (“stylistic only”), the Rule’s current lack of any reference to causation is further indication that Rule 407 was never intended to require any cause-and-effect relationship between a particular plaintiff’s injury and a defendant’s subsequent remedial measure.

Fortunately, most courts have not narrowed Rule 407 on this extratextual basis.  Rather, they take a broad view, excluding evidence of any subsequent remedial measure that, if taken previously, would have made the injury or harm generally less likely to occur, without reference to any particular plaintiff.

[Plaintiffs] seek to sidestep Federal Rule of Evidence 407 by insisting that the change was not a subsequent “remedial” measure because, according to the affidavit of a [defense witness], the change was not prompted by safety concerns.  But [defendant’s] motive for making the change is irrelevant.  All the rule requires is that the measure “would have made the injury or harm less likely to occur.”

Chlopek v. Federal Insurance Co., 499 F.3d 692, 700 (7th Cir. 2007) (quoting text of Rule 407).  Earlier, in Mills v. Beech Aircraft Corp., 886 F.2d 758, 763 (5th Cir. 1989), the court refused to turn Rule 407 exclusion into a battle over the defendant’s subjective intent, holding instead:

In the present case, as in the vast majority of the cases, it is not known why the changes were made.  Instead of attempting to prove or disprove the reasons for subsequent product changes, we should consider the probative value of such evidence on the point at issue . . . whether the product or design was defective at the time the product was sold. . . .  The introduction of evidence about subsequent changes in the product or its design threatens to confuse the jury by diverting its attention from whether the product was defective at the relevant time to what was done later.

Id. at 763 (citation and quotation marks omitted) (emphasis added).  See Bush v. Michelin Tire Corp., 963 F. Supp. 1436, 1449 (W.D. Ky. 1996) (“The rule’s language does not go to Defendant’s intent in adopting the later measures.  It simply asks whether the later measures could have prevented the earlier accident.”); see also Maiorano v. Home Depot U.S.A., Inc., 2018 WL 2128609, at *3 n.3 (S.D. Cal. May 9, 2018) (quoting and following Chlopek).  Cf. Hassebrock v. Air & Liquid Systems Corp., 2016 WL 4496917, at *6 (W.D. Wash. April 11, 2016) (Rule 407 barred admission of the defendant’s subsequent remedial measures after the date of plaintiff’s last exposure to asbestos, even though that exposure was not from the defendant’s products).

The issue has also come up in a number of state cases applying Rule 407 analogs.  We caution that, in all such cases, the language of the relevant rule might not be identical to Rule 407, but here are a few state high court examples:  Johnson v. State, Dept. of Transportation, 233 P.3d 1133, 1137 (Ariz. 2010) (“Rule 407 requires the exclusion of evidence of subsequent measures to prove a party’s negligence or culpable conduct, even when such measures are taken without specific knowledge of the accident in question”); Martin v. Norfolk Southern Railway Co., 271 S.W.3d 76, 88 (Tenn. 2008) (Rule 407 applied even though action was “carried out pursuant to corporate policy” rather than in response to an accident because “it corrected an allegedly dangerous condition and made the [area] safer”); Webb v. CSX Transportation, Inc., 615 S.E.2d 440, 448 (S.C. 2005) (“narrow interpretation” of Rule 407 “that only measures taken in direct response to the accident qualify for exclusion” rejected because it “ignores the literal language of the rule”); Doe v. Johnston, 476 N.W.2d 28, 34 (Iowa 1991) (Rule 407’s “policy would not be served if evidence of defendants’ changed behavior could be used to prove liability just because defendant was unaware that any injury or accident had occurred”).

Decisions applying Rule 407 in the specific context of prescription medical product liability litigation also apply the rule without detours into whether the plaintiff’s injury somehow caused the defendant to take the remedial measure at issue.  “[A] plaintiff could not avoid the effect of Rule 407 simply by arguing a label change was not prompted by safety concerns.”  Hill v. Novartis Pharmaceuticals Corp., 944 F. Supp.2d 943, 961 (E.D. Cal. 2013).

[T]he court did not hold that the absence of a remedial motive prevented exclusion under Rule 407 . . .[, and] other courts specifically addressing this precise issue have held otherwise.  The Court reaffirms its conclusion at the pretrial conference that subjective intent or motive in taking a remedial measure is not a dispositive prerequisite for exclusion under Rule 407.

In re Mentor Corp. ObTape Transobturator Sling Products Liability Litigation, 2010 WL 2015146, at *1 (M.D. Ga. May 20, 2010).  See Herrera-Nevarez v. Ethicon, Inc., 2017 WL 3381718, at *3 (N.D. Ill. Aug. 6, 2017) (rejecting argument that Rule 407 did not apply because “the change was not made for safety-related or other remedial reasons”); In re Depakote, 87 F. Supp. 3d 916, 925 (S.D. Ill. 2015) (“motives are irrelevant to the Court’s Rule 407 analysis”) (both Chlopek-bound courts).

As this discussion indicates, there hasn’t been a lot of litigation over what caused a defendant to undertake a particular remedial measure in prescription medical product liability litigation.  Like the would-be strict liability and government mandate exceptions before it, this purported exception has no basis in the text of Rule 407, and that should be the end of the inquiry.  We think things should stay that way.  “When the express terms of a statute [here a rule, but the proposition is the same] give us one answer and extratextual considerations suggest another, it’s no contest.  Only the written word is the law, and all persons are entitled to its benefit.”  Bostock v. Clayton County, Georgia, 140 S. Ct. 1731, 1737 (2020).

Photo of Steven Boranian

The Ninth Circuit has certified a question to the California Supreme Court on the learned intermediary doctrine that immediately caught our attention:  In a failure-to-warn claim against a prescription medical product manufacturer, is the plaintiff required to show that a stronger warning would have altered the physician’s decision to prescribe the product?  Or can the plaintiff establish causation by showing that the physician would have informed the plaintiff of the stronger warning and that a prudent person in the patient’s position would have declined treatment?

It’s an important question.  Time and again, we have seen cases where there is no evidence that stronger warnings would have had any impact on the physician’s prescribing physician, which should be, and often is, the basis for summary judgment under the learned intermediary doctrine.  The essence of the doctrine is that a prescription medical product manufacturer’s duty to warn runs to the physician—the learned intermediary—not the patient.  Thus, if there is no evidence that stronger warnings would have had an impact on the physician, the plaintiff cannot prove that an alleged inadequacy in the warnings caused his or her injury.  We have written on this more times than we can count, including for example here, here, and here.

But what about cases where the plaintiff says, sure a stronger warning would not have influenced my doctor, but what about me?  If my doctor had informed me, I would not have consented to the treatment.

The Ninth Circuit recently confronted a version of this scenario in Himes v. Somatics, LLC, No. 21-55517, 2022 WL 983176 (9th Cir. Apr. 1, 2022) (to be published in F.4th), and Himes v. Somatics, LLC, No. 21-55517, 2022 WL 989469 (9th Cir. Apr. 1, 2022) (unpublished).  In Himes, two plaintiffs sued a manufacturer of electroconvulsive therapy devices, alleging that the defendants failed adequately to warn against risks like memory loss.  The district court granted summary judgment against both plaintiffs and ruled (1) that the learned intermediary rule applied and (2) absent evidence that stronger warnings would have affected the physicians’ decisions to prescribe the electroconvulsive therapy, the plaintiffs’ claims failed as a matter of law.  Himes v. Somatics, LLC, No. 21-55517, 2022 WL 989469, at *1.  These are tried-and-true applications of California’s learned intermediary rule.

On appeal, the Ninth Circuit had a slightly different take.  The court quickly rejected the plaintiffs’ contention that the learned intermediary doctrine does not apply when the manufacturer has not provided sufficient warnings to physicians.  Plaintiffs in failure-to-warn cases always allege that the manufacturer’s warnings were not sufficient, and the learned intermediary rule is designed to address exactly that situation.  The Ninth Circuit agreed with this and observed that the plaintiffs were essentially asking the court to write the learned intermediary rule out of California law.  Id. (“Because the adequacy of warnings is always challenged in failure-to-warn claims, ‘[i]f the learned intermediary rule became inapplicable when a plaintiff alleged that warnings were inadequate, the doctrine would never operate in California.’”) (quoting Sanchez v. Bos. Sci. Corp., 38 F. Supp. 3d 727, 734 (S.D. W. Va. 2014) (applying California law)).

Having ruled that the learned intermediary doctrine applied, the Ninth Circuit confronted the proper standard for warnings causation under California law, and on that point, the evidence for the two plaintiffs diverged.  For one plaintiff, the physician never reviewed the manufacturer’s warnings.  As a result, it was clear that the plaintiff could not prove that different or additional information would have affected the physician’s prescribing decision because “a reasonable jury could not find that [the] treating physician . . . would have known about any stronger warnings issued by [the defendant].”  Id. at *2.

For the second plaintiff, the physician testified that he paid attention to communications from manufacturers regarding safety risks and that he would include information about those risks in his patient consent process.  Id. at *2-*3.  Thus, even though there was no evidence that a stronger warning would have altered the physician’s prescribing decision, it was conceivable that a stronger warning might have influenced a patient’s decision to grant informed consent.  Is this enough to prove warnings causation?

The Ninth Circuit didn’t know:

The resolution of this appeal turns on the proper causation standard applied to [the plaintiff’s] claim.  If the district court and [the defendants] are correct that in failure-to-warn claims, a plaintiff must show that stronger manufacturer warnings would have altered the physician’s prescribing conduct, [the plaintiff’s] claims fail.  If, on the other hand, a plaintiff can establish causation by showing that a physician would have communicated the stronger warning to the patient and that a prudent person in the patient’s position would have declined the treatment after receiving the stronger warning, [the plaintiff’s] claims survive summary judgment.

Id. at *3.  The Ninth Circuit certified the issue to the California Supreme Court for determination in a separate, published order.  Himes v. Somatics, LLC, 2022 WL 983176.

We will see how the California Supreme Court rules, but one thing is for sure.  Plaintiffs cannot prove warnings causation with self-serving, after-the-fact declarations stating subjectively that they would not have consented to therapy “had they known” about purported additional risks.  The below quote from the Ninth Circuit is kind of long, but really important (even if buried in a footnote):

We disagree with the appellants’ [plaintiffs’] contention that in establishing causation through warnings, the effect of a stronger warning on a patient could be determined through the patient’s subjective post-hoc declaration.  As the Supreme Court of California has explained in the physician failure-to-inform context, “[s]ince at the time of trial the uncommunicated hazard has materialized, it would be surprising if the patient-plaintiff did not claim that had he been informed of the dangers he would have declined treatment.  Subjectively he may believe so, with the 20/20 vision of hindsight, but we doubt that justice will be served by placing the physician in jeopardy of the patient’s bitterness and disillusionment. . .  .”  We thus conclude that if it were possible to establish causation through warnings communicated to the patient by the physician, the effect on the patient must be determined based on what a prudent person in the patient’s position would have done with the benefit of stronger warnings.

2022 WL 989469, at *3 n.3 (emphasis added).  In other words, if the impact of additional warnings on a patient (as opposed to a physician) can form the basis for warnings causation, it could only be under an objective standard.  The key phrase is “prudent person in the patient’s position.”  Anyone who has experienced an alleged injury can convince himself or herself in retrospect that he or she would not have consented “had they known.”  If that statement could break the warnings causation chain, then that too would essentially write the learned intermediary doctrine out of California law.  The Ninth Circuit correctly did not bite, and neither should the California Supreme Court.  We will keep you posted.

Photo of Andrew Tauber

Defendants love and plaintiffs hate Bell Atlantic Corp. v. Twombly, 550 U.S. 544 (2007), and Ashcroft v. Iqbal, 556 U.S. 662 (2009), which together stand for the proposition that to state a claim and avoid dismissal a plaintiff must allege facts that plausibly suggest that the defendant is liable. Today’s case, Poozhikala v. Medtronic Inc., 2022 WL 1076173 (C.D. Cal. 2022), illustrates why defendants and plaintiffs have contrary views of Twombly and Iqbal.

Poozhikala involves a purportedly defective implantable cardiac defibrillator (ICD) whose battery was alleged to have suffered rapid battery depletion. The plaintiff asserted all the usual claims and then one: manufacturing defect; failure to warn; breach of warranty; and fraud.

Citing Twombly and Iqbal, the court made quick work of the plaintiffs’ (second amended) complaint. The court concluded that each of the plaintiff’s claims were expressly preempted by 21 U.S.C. § 360k(a) because the plaintiffs had failed to adequately allege facts plausibly suggesting that the defendant had violated any federal requirement. Sufficient allegations of a federal violation are necessary to avoid preemption under § 360k(a) because it bars any state-law requirement that is “different from, or in addition to,” the federal requirements imposed on a medical device (at least if it is a Class III device that has received premarket approval or was used under either the Investigational Device Exemption or the Humanitarian Device Exemption).

Although Poozhikala does not break new ground, is provides a nice compilation of points that bear repeating.

First and foremost, the court reiterated that for a complaint to survive a motion to dismiss, “the ‘[f]actual allegations must be enough to raise a right to relief above the speculative level.’” 2022 WL 1076173, at *2 (quoting Twombly, 550 U.S. at 555). Of course, most 12(b)(6) rulings recite this standard. All too frequently, however, courts ignore the standard when actually evaluating the sufficiency of the particular complaint before them. Not so in Poozhikala. The court did not pay mere lip service to Twombly and Iqbal; it actually applied them.

The court began with the plaintiff’s allegation that they were “informed and believe[]” that the defendant “violated applicable federal statutes and regulations relating to medical devices.” That did not suffice, said the court, because “Plaintiffs must allege facts plausibly showing Defendant violated federal standards” rather than make “mere[] conclusory assertions.” 2022 WL 1076173, at *4.

Nor was plaintiffs’ laundry “list of possible violations” sufficient because they “provide[d] no factual support” plausibly suggesting that the defendant had actually violated the listed provisions. 2022 WL 1076173, at *4.

Then the court turned to the plaintiffs’ assertion that their complaint “specified factually how the Defendant[] violated the federal regulations.” The court acknowledged that the complaint “does include more targeted allegations of regulatory violations,” but “closer examination,” said the court, “reveals that these too lack factual support.” 2022 WL 1076173, at *4. A closer look? How refreshing. If only every court took its obligations under Twombly and Iqbal as seriously as the Poozhikala court.

One allegation we frequently encounter, particularly in support of manufacturing-defect claims, is that the defendant violated one or more of the Current Good Manufacturing Practices (CGMPs) codified at 21 C.F.R. Part 820. As readers of this blog are aware, case law is divided over whether alleged CGMP violations can ever support a non-preempted claim (with the more persuasive cases holding that they cannot). The Poozhikala court made no grand pronouncements on the issue; indeed, it did not discuss the competing decisions. Rather, it disposed of the general claim as inadequately pleaded. Echoing other cases and quoting Weber v. Allergan, Inc., 940 F.3d 1106, 1114 (9th Cir. 2019), the court held that “[m]ere evidence suggesting that [the plaintiff’s device] was defective does not show that [the manufacturer] failed to comply with the FDA’s Current Good Manufacturing Practices,” and that “evidence that some other [devices] produced by [defendant] were defective” likewise “does not demonstrate noncompliance.” 2022 WL 1076173, at *6.

The court worked its way through a bevy of alleged violated regulations, but found each wanting. Most because the plaintiffs alleged the “violations in wholly conclusory fashion that regurgitate near verbatim the regulations themselves.” 2022 WL 1076173, at *5. “Such regurgitations, without factual support, are,” said the court, “insufficient.” Id.

Plaintiffs’ allegations weren’t entirely conclusory. They alleged some facts with respect to two purported violations. But those too were insufficient to sustain the plaintiffs’ claims.

The plaintiffs alleged that the defendant had violated 21 C.F.R. § 803.50 by failing to properly report and investigate an adverse event. In support of that allegation, plaintiffs pointed to a particular adverse event that was noted in a recall notice. Plaintiffs alleged that manufactured had failed to investigate and report the event to the FDA as purportedly required.

There were, the court found, two problems with the failure-to-report allegation.

To start, the plaintiffs did “not allege any relevant connection between the source of their injury … and the product recall notices, which do not discuss” the allegedly defective component at issue. 2022 WL 1076173, at *5. That is significant, because to avoid express preemption under § 360k(a), a plaintiff must plead not only a federal violation but a causal connection between that violation and the plaintiff’s alleged injury.

But, the court continued, even if the plaintiffs had sufficiently alleged a causal nexus between their alleged injury and the injury identified in the recall notice, they still hadn’t adequately pleaded a failure to report in violation of federal regulations. The court found that the plaintiffs had “fail[ed] to plausibly allege Defendant did not comply with the federal reporting and investigating requirements” because they only “speculate that Defendant did not report or meaningfully investigate” the adverse event. 2022 WL 1076173, at *5.

The court found the plaintiffs’ allegation that the defendant had violated 21 C.F.R. § 814.39(a) by failing to submit a PMA Supplement to be neither legally or factually supported. The court began with the plaintiffs assertion that the defendant “violated 21 C.F.R. § 814.39(a) after Defendant became aware of issues related to [the ICD] because Defendant was ‘required to submit a PMA supplement for review and approval … before making a change affecting the safety or effectiveness of the device.’” According to the court, “Plaintiffs are correct that a PMA supplement is required after Defendant makes a change affecting safety or effectiveness, but nowhere does the [complaint] allege Defendant made a change.” 2022 WL 1076173, at *4. The court misspoke as to the timing of PMA Supplements. Unless submitted under the changes-being-effected provision of 21 C.F.R. § 814.39(d), a PMA Supplement must be submitted before any changes are made. But the fundamental point stands: allegations of a regulatory violation cannot overcome preemption when the complaint fails to allege facts sufficient to plausibly suggest that the regulation in question was violated.

Furthermore, the court observed, “[n]othing within the regulation requires Defendant to submit a PMA supplement based on their awareness of potential issues with [the ICD] as Plaintiffs contend.” 2022 WL 1076173, at *4. At one level, this observation goes to the plaintiffs’ failure to adequately plead a regulatory violation. At another level, it points to an important point that we have noted before: § 360k(a)’s prohibition on state-law requirements different from or in addition to the federal requirements imposed on a medical device means that states cannot compel device manufacturers to do what federal law permits but does not require. As the Seventh Circuit has put it:

Where a federal requirement permits a course of conduct and the state makes it obligatory, the state’s requirement is in addition to the federal requirement and thus is preempted.  Where a federal requirement permits a course of conduct and the state makes it obligatory, the state’s requirement is in addition to the federal requirement and thus is preempted.

McMullen v. Medtronic, Inc., 421 F.3d 482, 489 (7th Cir. 2005). Or as the Supreme Court held when interpreting the Federal Meat Inspection Act’s preemption clause, which is worded similarly to § 360k(a), a prohibition on different or additional state-law requirements means that state law cannot covert something that an entity “may” do under federal law into something that the entity “must” do under state law. National Meat Association v. Harris, 565 U.S. 452, 460 (2012).

As good as the foregoing is, the Poozhikala court delivered yet other pronouncements that do not change the law but certainly reinforce what it is in helpful ways.

For example, the court took judicial notice of an FDA recall notice. 2022 WL 1076173, at *1 & n.2. There is nothing earth-shattering about that, but it is a good reminder that motions to dismiss do not have to limit themselves to the four corners of the complaint but can instead rely on judicially noticeable materials beyond the complaint. It is often helpful to cite FDA materials, either to support certain assertion (such as the fact that a particular device has received premarket approval) or to rebut plaintiffs’ assertions (such as mischaracterizations of inspection findings in a Form 483 or of the reasons for an FDA recall notice).

The court also shot down, yet again, the now frivolous assertion that “there is a strong presumption against preemption.” The court rightly held that the “presumption does not apply here.” 2022 WL 1076173, at *4 n.3. Having already held as much in a previous ruling, the court did not spell out the reasons why that is so. But, as we have explained, the Supreme Court’s decision in Puerto Rico v. Franklin California Tax-Free Trust, 579 U.S. 115 (2016), clearly holds that when a “statute contains an express pre-emption clause, [courts] do not invoke any presumption against pre-emption.” Id. at 125 (internal quotation marks omitted).

A court that takes Twombly and Iqbal, and the Supreme Court’s other pronouncements, seriously. We can live with that. Quite happily.

Photo of Michelle Yeary

This post is from the non-Reed Smith side of the blog.

Or maybe we should say the court cooked up a particularly nasty version of Cincinnati Chili.

The mesh case of the week, Perry v. Ethicon, Inc., 2022 U.S. Dist. LEXIS 56268 (S.D. Ohio March 29, 2022), is the worst sort of judge-made law. It is not merely a judge making things up that the legislature never intended; in fact, the court’s opinion actively undermines legislation. Ohio expressly abolished all common-law product liability actions when it enacted the Ohio Products Liability Act (OPLA), as the court acknowledged:  “All common law products liability claims are explicitly abrogated by the OPLA.”  Id. at *7.  The OPLA was part of a trend of bringing statutory order, clarity, and efficiency to an otherwise messy, redundant, incoherent product liability landscape. In particular, the Ohio legislature enacted the OPLA after Ohio courts persisted in finding ways around more limited legislative efforts to stave off some of the silliness of the common-law.

It’s like a tug of war between legislators and judges. Who gets the last word?

In Perry, a federal court once again sticks a thumb in the legislative eye, holding that, in addition to asserting OPLA statutory claims, any plaintiff in a product liability case can also assert a common law claim for “economic loss” – which the decision defines broadly as “that the value she paid for the device was more than the value of the product she received.” Id. at *11.  Any plaintiff alleging personal injury from a defective device could make this claim. Nobody pays for a product expecting it to fail or be defective. Thus, the Perry ruling effectively overturns the Ohio legislature’s express abrogation of common-law claims in OPLA cases. For the remarkable proposition that a personal injury claim of product defect also smuggles in a separate claim for economic loss that evades the OPLA, the only citation is to a pre-OPLA case about economic loss. With no Erie analysis at all, the Perry decision vastly expands state law in a way that is not only novel but flies in the face of state statutory law.

There follows the usual TwIqbal analysis of the common-law claims, but that hardly matters in comparison to the “economic loss” ruling.

The Perry complaint was an extraordinary mish-mosh. If only for reasons of aesthetics, it should have been dismissed with instructions to do better. The complaint asserted numerous claims under the OPLA, and also claims for breach of implied warranty, fraudulent concealment, constructive fraud, common law fraud, negligent pharmacovigilance, and unjust enrichment. Other courts reviewing a similar potpourri of mesh claims (we wrote about one last week), had little difficulty separating the wheat from the chaff, and there was a whole lot more chaff than wheat. Most of these claims are duplicative make-weight. But the court labored mightily and brought forth a gnat. The court dismissed the UCC implied warranty claim for lack of privity, id. at *12-16, and the non-existent claim for negligent pharmacovigilance. Id. at *23.  So we’ve got that, which is nice. The court also dismissed plaintiff’s fraud claims for want of particularity, though the plaintiff was given leave to amend.  Id. at *19-22.

The court also dismissed the strict liability manufacturing defect claim, which is clearly the right outcome because the complaint did not hint as to how the mesh device deviated from specifications or standards. Id. at *31-33.

The entire Perry complaint is about personal injuries allegedly caused by an allegedly defective product. If that claim isn’t squarely within the metes and bounds of the OPLA then we are left wondering what is.

Photo of Bexis

Not too long ago we researched precedent that forbade persons claiming to be “FDA experts” from opining that products are “adulterated” or “misbranded.”  In that post, we mentioned that this research is a subset of a “general” precedent “precluding expert opinions on questions of law,” which we didn’t get into because Bexis’ book addressed it.  But since we recently ran across an unusually favorable new case on this subject, we thought we’d address it again.  It’s not a tort case – it’s not even a civil case – so it demonstrates how widespread the problem legalistic “expert” witnesses extends, and also how wide-ranging relevant caselaw can be.

The defendant in United States v. Xue, 2022 WL 1027634 (E.D. Pa. April 6, 2022), was being prosecuted for “conspiracy to steal trade secrets.”  It doesn’t matter what the secrets were or who was involved.  The prosecution offered up perfectly well-qualified expert witnesses who proposed to testify about a variety of things, but the one that matters for present purposes is that they intended to characterize the material at issue as “trade secrets.”  The defendant opposed admission of that testimony on the ground that “trade secret” – like “adulterated”/”misbranded” in our prior post – was a legal term of art, and thus improper as a basis for an expert opinion.  Id. at *9.

The court agreed with the defendant, and wrote an unusually thorough explanation of why this testimony was being excluded.  “One of the ‘ultimate issues’ in this case [was] whether the alleged misappropriated information falls within the statutory definition of ‘trade secret.’”  Id. at *9-10 (citing 18 U.S.C. §1839 (3)(A-B), which defines “trade secret”).  While F.R. Evid. 704(a) allows an expert witness to opine on an “ultimate issue,” nothing in the Federal Rules allows experts to opine on the law.  Id.  Rather, “‘an expert witness is prohibited from rendering a legal opinion.’”  Id. (quoting Berckeley Investment Group, Ltd. v. Colkitt, 455 F.3d 195, 217 (3d Cir. 2006)).

Thus, “courts commonly exclude ‘legal terms of art’ from expert testimony.”  Id. at *10 (citation omitted).  In the Third Circuit, Flickinger v. Toys R Us-Delaware, Inc., 492 F. Appx. 217 (3d Cir. 2012), prohibited an expert from opining on what constituted “exclusive control,” “dangerous condition,” “substantial cause,” and “negligence.”  Id. at 224.  Numerous district courts in the Third Circuit have reached similar results.  E.g., Dalgic v. Misericordia University, 2019 WL 2867236, at *13 (M.D. Pa. July 3, 2019) (expert may not opine about “proximate cause”); Perez v. Townsend Engineering Co., 562 F. Supp.2d 647, 652 (M.D. Pa. 2007) (precluding expert witness from testifying that a product was “defective,” “unreasonably dangerous,” or was the “proximate cause” of injury); Gallatin Fuels, Inc. v. Westchester Fire Insurance Co., 410 F. Supp.2d 417, 422 (W.D. Pa. 2006) (expert could not opine that an insurer acted in “bad faith”); McCrink v. Peoples Benefit Life Insurance Co., 2005 WL 730688, at *4 (E.D. Pa. March 29, 2005) (same).  See Xue, 2022 WL 1027634, at *11.

Turning to other circuits, Xue began, id., with Bexis’ favorite precedent for precluding expert testimony on conclusions of law:  Burkhart v. Washington Metropolitan Area Transit Authority, 112 F.3d 1207, 1212 (D.C. Cir. 1997), which determined that the term “as effective” was not subject to an expert opinion because it was a legal “term of art with a meaning ‘separate’ and ‘distinct’ from the vernacular.”  Id. at 1213.

[The expert’s] testimony as to the applicable legal standard was plainly erroneous, thus demonstrating the danger in allowing experts to testify as to their understanding of the law.  Each courtroom comes equipped with a “legal expert,” called a judge, and it is his or her province alone to instruct the jury on the relevant legal standards.

Id. (citations omitted).  Other appellate decisions cited in Xue are:  United States v. Barile, 286 F.3d 749, 759 (4th Cir. 2002) (prohibiting expert testimony on FDCA term “materially misleading”); Torres v. County of Oakland, 758 F.2d 147, 151 (6th Cir. 1985) (prohibiting expert testimony on statutory term “as effective”).

Because “the term ‘trade secret’ is a term of art with specialized legal meaning,” it was properly the subject of a jury instruction from the court, not opinions from the parties’ expert witnesses.  Xue, 2022 WL 1027634, at *12.

Ultimately, the jury must decide whether the Government has proved beyond a reasonable doubt that the documents in issue contain “trade secrets,” after carefully applying the facts they find to the law.  Thus, the expert witnesses may not testify using the term “trade secret” because doing so would usurp the District Court’s pivotal role in explaining the law to the jury.

Id. (citation and quotation marks omitted).  Therefore:

As an alternative to “trade secret” or “secret,” the witnesses may testify that the information was “confidential” or “proprietary,” about any steps taken to protect the confidentiality of the information, and about industry customs and practices.  The witnesses may also use synonyms for the words “trade” and “secret.”

Id.

Finally, when we first came across Xue, we compared the list of cases Xue cited to those cited in Bexis’ Book, §10.06, fn 1.  One reason we wrote this post is that we were surprised by the lack of overlap.  Only Burkhart and Perez appeared in both places.  To some extent, that’s because Bexis’ footnote included extensive citation to state-court cases, but it’s also due to the section needing updating (the most recent case is ten years old).  Hence, Xue was worth writing up simply because it adds more recent cases that exclude expert opinions on issues of law.

Photo of Bexis

A few years ago we did a couple of posts about the learned intermediary decision that arose from the Zimmer Nexgen Knee Implant MDL, and was subsequently affirmed by the Seventh Circuit Court of appeals.  We designated the opinion In re Zimmer Nexgen Knee Implant Products Liability Litigation, 218 F. Supp. 3d 700 (N.D. Ill. 2016), “Probably the Best Wisconsin Law Decision We’ve Ever Seen.”  Then we described In re Zimmer, NexGen Knee Implant Products Liability Litigation, 884 F.3d 746 (7th Cir. 2018), as “The Best Wisconsin Law Decision We’ve Ever Seen, Affirmed.”  We did so because Wisconsin was, at the time, one of the few states – and by far the largest – with no appellate precedent adopting the learned intermediary rule.  The Nexgen litigation changed all that.

Fast forward to last year and we invited our readers with focused state-law practices to submit guest posts on how the learned intermediary rule stood in all the states where there was no state court appellate decision adopting the doctrine.  We did that because, while federal precedent is good, the learned intermediary rule is ultimately a proposition of state law so that state court decisions take precedence.  We received responses for Iowa, South Carolina, and Wisconsin (BTW, we’re still looking for guest posts on Idaho, the Dakotas, three of the smaller New England states and Puerto Rico).

The Wisconsin guest post let us know about a pending decision that – we all hoped – would take that state out of the “no state appellate precedent” category altogether:

Wisconsin has recently made it to the red zone, and may soon cross the goal line, with the Wisconsin Court of Appeals’ forthcoming decision in Rennick v. Teleflex Med. Inc., fully briefed in early 2021, which will hopefully − and finally − make clear that Wisconsin adopts the doctrine, as consistently predicted by Wisconsin federal courts since 2016.

That was precisely the kind of on-the-ground information we had hoped for from this kind of guest post from local practitioners familiar with the lay of a particular state’s legal landscape.

Well, Rennick has now been decided and, continuing with the football analogy of our guest posters, it turned out to be a meaningless exhibition game that, in some way advanced the ball, but at one critical point fumbled it away out of bounds.  See Rennick v. Teleflex Medical, Inc., 2022 WL 1016686 (Wis. App. April 5, 2022) (per curiam).

The plaintiff in Rennick claimed injury from alleged migration of the defendant’s surgical clip.  Id. at *1.  The warning/learned intermediary issue arose because the defendant’s implanting surgeon “would not have altered his decision to use the clips had [defendant] provided a warning . . . about a risk of clip migration,” to which plaintiff responded “that the learned intermediary doctrine did not apply.”  Id. at *2.  Plaintiff relied upon a variety of materials concerning the risk of clip migration, and his own affidavit that “he would have altered his behavior to avoid his injuries had he known about the risk.”  Id.  The trial court granted summary judgment based on evidence that the implanting surgeon “would have used the clips in the surgery even if he had received such a warning” and thus plaintiff “could not prove the cause element of his claims.”  Id. at *3.  The surgeon’s testimony supported two grounds for non-causation:  lack of reliance, and that a warning would not have changed how the prescription product was used.

[Defendant] point[ed] to deposition testimony from [the prescribing surgeon] that he learned the [relevant medical] technique, not from anything [defendant] provided, but by reading medical literature and learning the technique from other physicians.  [Defendant] further supports its argument saying that [the prescribing surgeon] indicated that he would still have used the clips if he had received a warning about the risk of migration because of the overall positive experience he has had and the inability to perform this type of surgery without the clips.

Id. at *4.

The Wisconsin guest post thought that the applicability of the learned intermediary rule was squarely before the appellate court in Rennick, and given the above description of the case history, we would have thought so, too.  But the Rennick decision stated that it was not deciding this issue.  “We need not address the adoption of this doctrine in Wisconsin because we conclude that it does not apply to the facts of this case.”  Id. at *4.  To carry off that dodge, Rennick created a limitation on the learned intermediary rule that has been adopted exactly nowhere else – that it only applies in cases of inadequate, as opposed to absent, warnings:

[T]he doctrine applies when the manufacturer has “inform[ed] the prescribing physician” of the risks associated with its product.  See [Zimmer Nexgen, 884 F.3d] at 751.  Here, it is undisputed that [defendant] provided no warning of any sort of the risk. . . .  Having provided no warning about this risk of migration, [defendant] has not met the threshold requirement for the application of the learned intermediary doctrine.

Rennick, 2022 WL 1016686, at *4 (no citations omitted).

We included that unusual citation parenthetical because the critical statement – that “no warning” precludes the learned intermediary rule – is completely unsupported in Rennick, and we would add, unprecedented.  Rennick simply made something up to avoid deciding the main issue being presented.  Apparently, the “no warnings” issue arose because physician in Rennick put the product to an off-label use, id. at *3 – which means, as we’ve discussed before, a manufacturer cannot include warnings about such off-label uses absent specific FDA approval.

Even then Rennick’s rationale didn’t do a very good job.

We’ve seen a lot of distinctions/exceptions concerning the learned intermediary rule – Section 2.03 of Bexis’ book goes through just about every learned intermediary rule wrinkle known to man, but not this one (we searched it thoroughly).  One certainly doesn’t find this distinction at the Zimmer Nexgen language that Rennick cited.  The full sentence there reads:  “The [learned intermediary] doctrine holds that the manufacturer of a prescription drug or medical device fulfills its duty to warn of the product’s risks by informing the prescribing physician of those risks.”  884 F.3d at 751.  Nothing about what does or does not “inform” prescribers.

That would-be distinction between no warning and an inadequate warning is as illogical as it is unprecedented because, as Rennick itself stated, the doctrine is “an exception to this general rule” that manufacturers “ha[ve] a duty to warn consumers directly.”  2022 WL 1016686, at *4.  Whether the learned intermediary rule applies turns on “who” rather than “what.”  The doctrine doesn’t depend on what, if anything, the defendant conveyed to the physician – only that an intermediary physician was present to be told whatever information was at issue.  While adequacy what satisfies the defendant’s duty; it has no bearing on whether the duty exists in the first place.  Cases on this point are legion.  Here’s a recent example:

[E]ven when warnings are assumed to be deficient, in the context of prescription products, the analysis always relies on the impact of a hypothetical stronger warning on the physician.  After all, because the adequacy of warnings is always challenged in failure-to-warn claims, if the learned intermediary doctrine became inapplicable when a plaintiff alleged that warnings were inadequate, the doctrine would never operate. . . .  We thus conclude that the district court correctly relied on the learned intermediary doctrine to analyze the claims.

Himes v. Somatics, LLC, 2022 WL 989469, at *1-2 (9th Cir. April 1, 2022) (applying California law) (emphasis added).

Moreover, the particular distinction Rennick used has no basis in the law.  See Munoz v. American Medical Systems, Inc., 2021 WL 1200038, at *2 (C.D. Cal. March 30, 2021) (“Where the learned intermediary doctrine applies, the plaintiff must prove that . . . no warning was provided or the warning was inadequate”); Ackermann v. Wyeth Pharmaceuticals, 471 F. Supp.2d 739, 747 (E.D. Tex. 2006) (rejecting distinction between “a case of no warning as opposed to an inadequate warning”), aff’d, 526 F.3d 203 (5th Cir. 2008); Herzog v. Arthrocare Corp., 2003 WL 1785795, at *13 (D. Me. March 21, 2003) (learned intermediary rule applied, “[b]ut when an inadequate warning or no warning is provided to the learned intermediary, the manufacturer or seller has not satisfied its duty to the consumers”).  Cf. Crook v. Kaneb Pipe Line Operating Partnership, L.P., 231 F.3d 1098, 1102 (8th Cir. 2000) (“sophisticated user” doctrine may apply “when no warnings or inadequate warnings are given”) (applying Nebraska law); Petty v. United States, 740 F.2d 1428, 1438 (8th Cir. 1984) (heeding presumption case; “distinction between no warning . . . and an inadequate warning is unpersuasive”) (applying Iowa law).

This nonexistent and nonsensical distinction between “none” vs. “inadequate” warnings is where we believe Rennick fumbled the ball out of bounds.  Some advancing of the ball is implied in Rennick in that, at least in cases involving inadequate warnings, it suggests that the learned intermediary rule applies.  Finally, Rennick amounted to a meaningless preseason game because this per curiam decision is unreported, 2022 WL 1016686, at *7 (“This opinion will not be published.”) and thus not only lacks precedential value, but “may not be cited in any court of this state as precedent or authority.”  Wis. Stat. §809.23(3)(a-b).

That’s fine with us.

Photo of Rachel B. Weil

We write on a gorgeous spring day in suburban Philadelphia, with Passover imminent for us and Easter for many we love.  Today’s decision contains more than enough “good” to sustain the positive vibe. Ward v. Zoll Lifevest Holdings LLC, 2021 WL 7907066 (D.D.C. Sept. 20, 2021) is a D.C. District case from last September that didn’t show up on Bexis’s search until just now.  The plaintiff, who had suffered a heart attack, was fitted with the defendant’s wearable defibrillator.  The device is designed to detect heart arrhythmias and to administer an electric shock to restore the wearer’s normal rhythm.  The device is supposed to sound a warning before it administers the shock.  The wearer can then prevent the shock by pressing a button on the device.  The plaintiff alleged that his device failed to sound a warning before it administered an unnecessary shock, which knocked him to the ground and caused physical and emotional injuries.  He filed suit in federal court, asserting claims for design defect, manufacturing defect, negligence, and breach of express warranty.  The defendant moved to dismiss all of the claims.

Design Defect

The defibrillator is a Class III device, which means it underwent the FDA’s full premarket approval (PMA) process before it was approved for marketing.  The defendant argued that the design defect claim was barred because, under Riegel v. Medtronic, Inc., 552 U.S. 312 (2008), state law design defect claims against manufacturers of Class III devices are expressly preempted by the medical device amendments to the FDA.  As the court explained, the only exception relates to claims that are “truly parallel to the federal requirements at issue” and are not based on state law requirements that are different from, or in addition to, the federal requirements.  Ward, 2021 WL 7907066 at *2 (citation omitted).  To assert such a “parallel claim,” a plaintiff must point to specific federal requirements that the manufacturer violated and that were causally connected to the alleged injuries.

The plaintiff argued that he had asserted just such a parallel claim by asserting that, by failing to sound an alarm before delivering a shock, the device did not conform to its FDA-approved design.  But the Complaint did not say this, instead asserting a traditional design defect claim of “precisely the type . . . that courts have held are expressly preempted by the MDA.”  Id.  at *3 (citation omitted).  The court concluded, “. . . [T]he plaintiff cannot escape preemption by recasting his claim through his opposition brief, [and] the court therefore grants Defendants’ motion to dismiss [the design defect claim].”  Id.

Manufacturing Defect

The defendant argued that the manufacturing defect claim was “merely a design defect claim disguised in manufacturing defect clothing” and should be dismissed for the same reasons.  The court disagreed.  As the court explained, “[m]anufacturing defect claims are capable of threating the ‘narrow gap’ between express and implied preemption . . . .” The only such claim that can survive preemption “is one where the defect was caused by a problem in the manufacturing process that deviated from” the approved manufacturing specifications. Id. at *4.  The court held that the plaintiff had identified at least two such potential problems in the manufacturing process, one that resulted in the failure of the alarm to sound before the shock was delivered and on that resulted in too many unnecessary shocks.   Even if those flaws were widespread in the manufactured devices, the court emphasized, they still could be the result of manufacturing defects.

The defendant also argued that the manufacturing defect claim was inadequately pled because it did not identify the ways in which the device deviated from the FDA-approved manufacturing specifications, but the court held that the defendants “demand[ed] too much at the pleadings stage” and allowed the manufacturing defect claim to move forward.

Express Warranty Claim

The plaintiff asserted that the defendant’s sales representative assured him that any shock the device delivered was only a “mild jolt,” while FDA documents suggested that the shocks were so painful that they should occur only when a patient is unconscious.  The court held that, accepting the allegations as true, this express warranty claim passed muster.

The plaintiff also asserted that the defendants assured him that the device would “save his life.’   The court held,

This statement cannot form the basis of a parallel claim for express warranty.  It is not a statement holding Defendants’ product out as meeting a higher standard than that required by the FDA.  Rather, this representation is one where an essential element of the claim will be proof that a device granted a PMA is not safe or effective, which necessarily conflicts with the FDA’s contrary finding.

Id. at *5 (internal punctuation and citations omitted).   The court concluded, “Such a warranty claim is directly preempted by Riegel.”  Id.  

Failure-to-Warn Claims.

As the court explained, the “failure-to-warn claim involve[d] two separate theories.  The first theory involved the plaintiff’s allegation that he was not provided a complete set of warnings.   While a “parallel” warnings claim can survive if it alleges that a manufacturer did not provide the warnings the FDA requires, this was not what the plaintiff alleged.  Instead, he alleged that he should have been provided additional warnings beyond those the FDA approved, and Riegel preempts such claims.

The second theory asserted that, had defendants reported alarm failures and inappropriate shocks to the FDA, patients such as the plaintiff would have been made aware of the incidents and, thus, “warned” of them.  But, the court explained, “suffice it to say that unless both the state-law claim and the applicable federal law contain a requirement that Defendants report adverse events to the FDA, there is no parallel claim.”  Id. at *6.  But the plaintiff “allege[d] only the possibility, under state law, that manufacturer may discharge their duty to warn by reporting to a third party.   This is not enough to make a state-law claim parallel.”  Id. (citations omitted).   Warnings claims dismissed, and Ward has been added to the D.C. section of our failure to report 50-state survey.

Negligence

Finally, the court held that the negligence claim, like the design defect claim, sought to impose state law requirements that were different from or in addition to federal requirements;  thus, it also was preempted under Riegel.

Ward is mostly good, and it is well-reasoned in the limited instances where it allows claims to proceed.  We will keep you posted on further developments.  In the meantime, have a safe weekend, and enjoy celebrating if you are.

Photo of Steven Boranian

The Incretin-Based Therapies MDL has followed a long and winding road, and it all should come to an end with a recent Ninth Circuit opinion affirming the exclusion of the plaintiffs’ only general causation expert.  It all started in 2013 with the MDL transfer of cases involving multiple diabetes drugs to the Southern District of California.  Along the way, the district judge essentially dismissed the entire MDL on implied preemption grounds, only to have the Ninth Circuit reverse and give the plaintiffs a second chance to uncover preemption-defeating “newly discovered evidence.”  We covered that judicial whipsaw here and here.

After a nearly three-year hiatus pending appeal, the district court again essentially dismissed all claims, this time both on implied preemption and on these plaintiffs’ other Achilles heel—the lack of any admissible expert opinion that these drugs can cause pancreatic cancer.  That order was so long, we blogged about it twice (here and here).

This time around, the Ninth Circuit has seen enough, and it affirmed the exclusion of the plaintiffs’ general causation expert and the resulting order granting summary judgment.  In re Incretin-Based Therapies Prods. Liab. Litig., No. 21-55342, 2022 WL 898595 (9th Cir. Mar. 28, 2022).  The science has always been weak in these cases, amounting to little more than speculation and conjecture.  And, to be sure, the science case has not improved for the plaintiffs over time.  The plaintiffs served their expert reports initially in 2015, and while they served an updated report from their general causation expert in 2019, the newer report failed to account for studies published in the interim.  Id. at *1.

It is no wonder why, since every study published between 2015 and 2019 found no causal relationship between the drug (liraglutide) and pancreatic cancer.  Id.  In addition, although the plaintiffs’ expert purported to follow a “weight of the evidence” methodology, he did not explain what that means or how anyone could possibly replicate his results.  Id.  In other words, “the district court had no means to ensure that [the expert’s] ‘conclusions were not mere subjective beliefs or unsupported speculation.’”  Id. (quoting Claar v. Burlington N. R.R. Co., 29 F.3d 499, 502 (9th Cir. 1994)).  Finally, the expert was “alone” in finding a causal link “despite years of research . . . conducted by medical, scientific, and regulatory entities.”  Id.

Given the paucity of scientific support, the Ninth Circuit’s opinion is short and to the point.  Be that as it may, we find this opinion noteworthy for multiple reasons.  First, the result.  By all indications, this MDL has now proceeded for nine years (off and on), despite a profound absence of scientific support and the preemption of all of the plaintiffs’ claims.  Federal preemption can be decided, and often is decided, on the pleadings, not after nearly a decade of litigation.  We are not sure what this says about modern multidistrict litigation, but it’s not good.

Second, in affirming the exclusion of expert opinions, the Ninth Circuit distinguished its own wrongly decided opinion in Wendell v. GlaxoSmithKline, 858 F.3d 1227 (9th Cir 2017), which came in at number nine on our list worst cases of 2017.  Even compared against the low Wendell bar, the expert’s opinion in Incretin still fell short.  The Incretin expert conducted no differential diagnosis and offered no opinions on specific causation.  In re Incretin-Based Therapies, 2022 WL 898595 at *1.  But more importantly, “[N]othing in Wendell absolves expert witnesses of the general and longstanding requirement that they explain their methods with enough detail that their results can be replicated.”  Id.  We would have preferred that the Ninth Circuit overrule Wendell outright, but short of that, we will take the Ninth Circuit’s conclusion and run with it.

Third, the Ninth Circuit came to the correct result, but it could have said much more.  The district court excluded other experts’ opinions and it granted summary judgment also on implied preemption, even after the plaintiffs had the opportunity to search for “newly discovered evidence”—which they presumably did not discover.  The Ninth Circuit kept mum on these important issues.  In fact, the court emphasized that it “need not—and do[es] not—opine on any other theory or argument urged by the parties.”  Id. at *2.  That pointed observation leaves the Ninth Circuit’s prior opinion on implied preemption intact, when it could have and should have acknowledged that implied preemption was ultimately a winning argument for the defense.

But, as Bexis would say, a win is a win, and the Ninth Circuit’s opinion on the plaintiff’s expert makes it eminently blogworthy.

Photo of Stephen McConnell

Bostic v. Ethicon, Inc., 2022 WL 952129 (E.D. Pa. March 29, 2022), is a Pennsylvania mesh case raising a host of familiar issues in a motion to dismiss context. The complaint is of the typically overpleaded (14-count) variety. Dickens was not really paid by the word, but plaintiff lawyers seem to think they might be. Or perhaps they hope that if enough counts make it to the jury, compromise or charity might result in throwing a bone the plaintiff’s way. In any event, by the time the Bostic court was through considering the motion to dismiss, considerably fewer counts remained.

Let’s begin with the disappointing part. The court declined to dismiss the complaint as a “shotgun” pleading – an argument that seems to be more popular (if not necessarily more successful since our post last year on this subject). The defendant pointed out that the complaint was virtually identical to dozens of other complaints filed by other plaintiffs around the country. The complaint also contained very few case-specific allegations. All we learn from the complaint is that the plaintiff had been implanted with the mesh device, the mesh was made from polypropylene, and the plaintiff suffered injuries. The Bostic court reasoned that a defective “shotgun” complaint is one that fails to give the defendant adequate notice. According to the court, the Bostic complaint cleared this rather low hurdle. The defendant wanted some dates that might help it ascertain the viability of certain defenses, such as statute of limitations. The court essentially said Too Bad, and that a complaint need not head off affirmative defenses. Oh well

The court did better with the manufacturing defect claim. The demise of that cause of action was perfectly typical. The complaint alleged that the mesh device used non-medical grade material, was made of polypropylene, was cut in a manner resulting in sharp edges, and lacked a sheath that might have reduced abrasion. But these alleged defects were common with other mesh devices and, more important, were not deviations from the intended design. Thus, the manufacturing defect claim was really a disguised design defect claim. The court dismissed the manufacturing defect claim, but gave leave to amend because the plaintiff promised to specify how the device was different from similar products manufactured by the defendant. Count us as skeptical.

The Bostic court also held that Pennsylvania law would apply comment k across the board to all prescription medical products, including medical devices. This Erie guess was based on several Pennsylvania Superior Court decisions, as well as other decisions by E.D. Pa. Application of comment k compelled dismissal of the strict liability claims for design and warnings, and also dismissal of implied warranty claims. Those dismissals were with prejudice.

The Bostic court dismissed the common law fraud claim because it had not been pleaded with particularity. The complaint did “not allege the date, time or place in which those misrepresentations were made or otherwise inject precision into the allegations.” This dismissal was without prejudice, so the plaintiff can seek to “inject precision.” We expect precisely no precision, but you knew that, didn’t you?

Not content with having an imprecise fraud claim, the plaintiff also ladled on claims for constructive fraud, negligent misrepresentation, and fraudulent concealment. If less is more does that mean that more is less? In this case, more was simply duplicative. The Bostic court held that negligence for failure to warn is the sole theory under which a plaintiff can recover against a device manufacturer when the claim is essentially that the manufacturer knew of dangers associated with the product but concealed that information while fraudulently misrepresenting the product’s safety. Consequently, the Bostic court dismissed the potpourri of fraud/misrepresentation/concealment claims. Since we just this week learned that the word potpourri comes from the French for “putrid pot,” our usage seems appropriate.

As is lamentably typical (there’s that word again), the plaintiff included an express warranty claim that did “not contain any allegations regarding the specific sources of these warranties” and did not offer a single clue as to where the plaintiff read, heard, or saw those express warranties or otherwise knew of them. The Bostic court dismissed the express warranty claim. The plaintiff again promised (promises, promises!) to do better, so the court permitted an amendment.

Finally, we have the ultimate useless, gratuitous, makeweight claim in a personal injury case: unjust enrichment. The plaintiff received and used the product, and thus could not allege unjust enrichment. The plaintiff could “not plausibly allege that Defendants refused to provide her with a service or product in exchange for payment.” The Bostic court granted the motion to dismiss the unjust enrichment count with prejudice. Even better, the court ruled that any amendment “would be futile.” The dismissal was with prejudice.

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