Photo of Bexis

We’re not surprised by the preemption ruling in In re Smith & Nephew Birmingham Hip Resurfacing (BHR) Hip Implant Products Liability Litigation, 2021 U.S. Dist. Lexis 100041 (D. Md. May 17, 2021), but that doesn’t makes us any less appalled.

First, here’s what happened.

This latest Birmingham Hip ruling was in preparation for an MDL “bellwether trial” – that peculiar MDL process by which theoretically “representative” cases are selected for trial, and then rendered as unrepresentative as possible by both sides throwing far more resources into the trial than the case could possibly be worth if it were a traditional one-off case.  Well, through the luck of the draw, this particular bellwether plaintiff happened to be from North Carolina.  2021 U.S. Dist. Lexis 100041, at *33.

The next piece of essential information is that the medical device at issue in the Birmingham Hip MDL was premarket approved (“PMA”) by the FDA via the agency’s most rigorous, and thus most preemptive, regulatory process.  2021 U.S. Dist. Lexis 100041, at *44.  Under the broad tort preemption applicable to PMA medical devices pursuant to Riegel v. Medtronic, Inc., 552 U.S. 312 (2008), most of the ordinary product liability claims – warning and design defect in particular – are unavailable, so plaintiffs have to find some sort of “parallel” claim that also asserts some sort of violation of FDA requirements as well as a supposedly traditional common-law claim.

One such “parallel” claim that PMA plaintiffs like to assert is that the defendant supposedly didn’t comply with ever jot and tittle of the FDA’s requirement to report adverse events (and certain other information).  And of course that was the case in Birmingham Hip – plaintiffs claimed that the defendant “did not include the ad hoc reports discussed above or the data underlying them in the PMA reports.”  2021 U.S. Dist. Lexis 100041, at *46.

But as we recently discussed, North Carolina is a no-go for FDCA-based (or any other kind of) failure-to-report claims:

TBI correctly classified North Carolina as a state that “does not recognize a parallel duty on manufacturers to report to the FDA.”  2021 WL 1050910, at *30 (quoting McNeil-Williams v. DePuy Orthopaedics, Inc., 384 F. Supp.3d 570, 575 (E.D.N.C. 2019)).  McNeil-Williams held that to allow FDCA-based failure-to-report claims “expands [the] law in a manner not consistent with North Carolina law.”  Id. at 576.

Plaintiff’s primary asserted theory of negligence liability fails, however, because North Carolina law does not recognize a parallel duty on manufacturers to report to the FDA as plaintiff asserts.  Rather, North Carolina law recognizes a duty to warn only users or medical practitioners in certain circumstances. . . .  Plaintiff cites no case, and the court has found none, where North Carolina courts have recognized a duty under North Carolina law to inform the FDA of adverse reactions, defects, and other injury information.

Id. (citations omitted).  McNeil-Williams refuted plaintiff’s reliance on Williams v. Smith & Nephew, 123 F. Supp. 3d 733, pointing out (as we did, see Maryland) that Williams “did not cite to any Maryland case law for such proposition.”  384 F. Supp.3d at 576.  Similarly, McNeil-Williams nixed reliance on Stengel, observing (as we did, see Arizona) that “the Arizona Supreme Court expressly disavowed the reasoning of Stengel on the very proposition that is at issue in this case.”  Id.

That’s not all. Burrell v. Bayer Corp., 260 F. Supp.3d 485 (W.D.N.C. 2017), held failure-to-report claims preempted because of their dependence on federal, rather than state, law.  “A requirement to report adverse events exists under the FDCA, and plaintiff’s cause of action is being brought because . . . defendants allegedly failed to meet these reporting requirements.  Accordingly, the plaintiff’s failure-to-warn claim is preempted.”  Id. at 492 (citations omitted).  Secondarily, Burrell pointed out that any failure-to-report claim could not be causal because the FDA did nothing once it learned of the allegedly unreported adverse events.  “[T]he claims must also survive plausibility challenges. . . .  [T]he adverse event reports were provided to the FDA by the time the plaintiff had her surgery.”  Id. at 495.

A similar no-duty conclusion was reached under North Carolina law in Wilkerson v. Christian, 2008 WL 483445 (M.D.N.C. Feb. 19, 2008):

[Defendant’s] alleged failure to file a . . . report with the FDA is not grounds for the application of fraudulent concealment.  Even assuming, as Plaintiff alleged, that the federal regulations required [defendant manufacturer] to submit certain information to the FDA, the duty [defendant] owed was to the FDA.  Accordingly, [defendant] violated no preexisting duty to Plaintiff to disclose this information when it failed to file a . . . report.

Id. at *12.

Outside the FDCA context, failure-to-report claims have been rejected under North Carolina law.  These include reporting of child abuse.  See Ostwalt v. Charlotte-Mecklenburg Board of Educ., 614 F. Supp.2d 603, 608 (W.D.N.C. 2008) (finding no common-law duty to report abuse).  Similarly, no failure-to-report liability is recognized in North Carolina for failure to comply with reporting requirements imposed by the Bank Secrecy Act.  See Taylor & Co. v. Bank of America Corp., 2014 WL 3557672, at *3 (Mag. W.D.N.C. June 5, 2014) (joining those courts “unwilling to create a common law duty of care” predicated on failure to make mandatory suspicious activity reports), adopted, 2014 WL 3557679 (W.D.N.C. July 18, 2014).

There is simply no basis for teasing a failure-to-report claim out of North Carolina law.

Which leads us to, second, what’s truly appalling about Birmingham Hip.  It simply decided it wouldn’t follow state law, because to do so would have resulted in preemption – which is the proper result under Riegel.  “[R]emov[ing] all means of judicial recourse . . . is exactly what a pre-emption clause for medical devices does by its terms.”  552 U.S. at 326 (citation and quotation marks omitted).

The defendant properly relied on McNeil-Williams, the first and foremost case we also cited in our post, above.  Birmingham Hip instead declared that it didn’t matter whether a state allowed failure-to-report claims or not.  2021 U.S. Dist. Lexis 100041, at *52.  This result, and reasoning, is facially absurd.  Even though there is no dispute that an individual plaintiff in a one-off case would have a failure-to-report claim dismissed by any North Carolina judge, in this MDL – purporting to follow that same state’s law – a plaintiff can pursue the same failure-to-report claim, supposedly as a “parallel” claim, since “parallel” is (again, in this MDL) is now devoid of any significant meaning.  Birmingham Hip thus becomes the apotheosis of an MDL judge ignoring state law in order to force settlement by increasing the defendant’s liability exposure.

For what it’s worth, we object.

The MDL court didn’t care for what it called the “developing line of federal case law that requires plaintiffs alleging failure to warn claims against a medical device manufacturer to point to a specific and traditional state law duty to report information to the FDA (or at least to a regulatory body) in order to establish a parallel state law duty that survives preemption.”  Birmingham Hip, 2021 U.S. Dist. Lexis 100041, at *53 (citations omitted).

So this decision simply “declined to follow” the vast majority of cases requiring actual “parallel” claims – claiming this was “inconsistent” with Medtronic, Inc. v. Lohr, 518 U.S. 470 (1996).  Id.  Specifically, Birmingham Hip cited to Lohr’s holding that:

[T]he MDA was not intended to preempt “State or local requirements of general applicability where the purpose of the requirement relates either to other products in addition to devices or to unfair trade practices in which the requirements are not limited to devices[.]”

2021 U.S. Dist. Lexis 100041, at *53-54 (quoting Lohr, 518 U.S. at 501-02).  Birmingham Hip then twisted the Lohr holding into a supposed rule that “[a] general state common-law duty to warn . . . is not directed only to medical devices and does not “impede the ability of federal regulators to implement and enforce specific federal requirements.’”  2021 U.S. Dist. Lexis 100041, at *54 (also quoting Lohr).  So any state law duty to warn – which every state has – “may be premised on the failure to comply with federal regulations mandating certain reports to the FDA.”  Id. (citation omitted).

This holding is completely and utterly wrong on several levels.

Most obviously, it relies on Lohr, a §510(k) case, rather than Riegel, which like Birmingham Hip, involved a PMA device.  Here’s what Riegel had to say about the same “state or local requirements of general applicability” that Birmingham Hip found precluded preemption:

[Plaintiffs] contend that the duties underlying negligence, strict-liability, and implied-warranty claims are not pre-empted even if they impose “requirements,” because general common-law duties are not requirements maintained “with respect to devices”. . . .  The language of the statute does not bear [plaintiffs] reading.  The MDA provides that no State “may establish or continue in effect with respect to a device . . . any requirement” relating to safety or effectiveness that is different from, or in addition to, federal requirements.

552 U.S. at 327-28 (citation omitted) (emphasis original).  The Lohr holding (and the plaintiffs’ argument in Riegel) relied on an FDA regulation, 21 C.F.R. §808.1(d)(1), “stat[ing] that the MDA’s pre-emption clause does not extend to certain duties, including ‘[s]tate or local requirements of general applicability where the purpose of the requirement relates either to other products in addition to devices.”  552 U.S. at 328 (quoting 808.1(d)(1)).  Another part of the same regulation, however, acknowledged a “general rule pre-empting state law duties,” including those “established” by “court decision.  Id. at 329.  Thus, §808.1(d)(1), was, at best contradictory.  Therefore,

All in all, we think that §808.1(d)(1) can add nothing to our analysis but confusion. . . .  [T]he regulation fails to alter our interpretation of the text insofar as the outcome of this case [involving a PMA device] is concerned.

Riegel, 552 U.S. at 329-30.  Birmingham Hip is thus erroneous, in the first place, because it follows a portion of Lohr, in a PMA preemption case, that was expressly rejected in Riegel as a reason for avoiding preemption.

If ignoring Riegel were not enough, Birmingham Hip is also completely contrary to the well-established Erie v. Tompkins doctrine that federal courts exercising diversity jurisdiction are not authorized by such jurisdiction to make up expansive forms of state-law liability that have not already been accepted by the states themselves.  This rule is, of course, regularly cited and enforced in the Fourth Circuit where the Birmingham Hip MDL is located.  We collected that precedent back in 2015 in another post about another MDL ruling that played fast and loose with state law.  To that compilation of Fourth Circuit precedent enforcing Erie conservatism, we’ll just add:

First and foremost, we are mindful of our role as a federal court sitting in diversity. In [one of the cases cited in that prior post] we explained, “[a] federal court acting under its diversity jurisdiction should respond conservatively when asked to discern governing principles of state law.”  As a result, “in a diversity case, a federal court should not interpret state law in a manner that may appear desirable to the federal court, but has not been approved by the state whose law is at issue.”

McKiver v. Murphy-Brown, LLC, 980 F.3d 937, 963 (4th Cir. 2020) (quoting and following Rhodes v. E.I. du Pont de Nemours & Co., 636 F.3d 88 (4th Cir. 2011)).

Moreover, not even the cases Birmingham Hip cites as supporting its holding that any extant state-law failure can support a supposedly “parallel” failure-to-report claim don’t stand for that proposition.  The only appellate case Birmingham Hip discusses, 2021 U.S. Dist. Lexis 100041, at *54, beyond a string citation is Hughes v. Boston Scientific Corp., 631 F. 3d 762, 769-70 (5th Cir. 2011).  Hughes, however, did not purport to decide anything at all about state law (in that case, Mississippi).  Rather it decided only a preemption issue “[a]ssuming that a failure to warn claim may be pursued under Mississippi law as [plaintiff] argues,” 631 F.3d at 769.  “The only issue presented to us on this appeal is whether the district court correctly determined that [plaintiff’s] suit is preempted.”  631 F.3d at 771.  And Hughes assumed wrongly.  As discussed in more detail in our failure to report 50-state survey, the Mississippi state legislature thereafter acted to abolish the negligence cause of action that Hughes hypothesized might support such a claim.  The other two cases Birmingham Hip cited, Rosen v. St. Jude Medical, Inc., 41 F. Supp.3d 170, 185 (N.D.N.Y. 2014), and Silver v. Medtronic, Inc., 236 F. Supp.3d 889, 900 (M.D. Pa. 2017), both purported to make actual predictions of state law – not to ignore such law altogether as Birmingham Hip’s rationale did.  We take issue with the results in those two cases (for reasons discussed in the aforementioned 50-state survey), but at least they ostensibly applied, rather than avoided, state law.

Third, here’s why, while we’re appalled by this ruling in Birmingham Hip, we’re not particularly surprised.  The same judge has issued two prior rulings that, as here, purported to allow failure-to-report claims to avoid preemption without following any state law.  We discussed those decisions in the Maryland section of our 50-state survey:

Maryland is another example of federal courts hijacking state law for their anti-preemption agendas in disregard of the Supreme Court’s Erie standards for predicting state law.  Maryland appellate decisions outside the FDCA context are as best equivocal, but a federal court purporting to interpret Maryland law plowed ahead with allowing FDCA-based failure-to-report claims to avoid preemption.  Williams v. Smith & Nephew, Inc., 123 F. Supp.3d 733, 742-43 (D. Md. 2015).  Sure, Maryland state courts may well do the same thing (so far they haven’t), but the difference is that’s their right as state courts.

[Discussion of Maryland appellate precedent omitted]

Without citing any of the above Maryland state precedent – or any reporting-related Maryland precedent at all − Williams, 123 F. Supp.3d 733, 742 (D. Md. 2015), decided to recognize FDCA-based failure-to-report under Maryland law.  Its discussion of the novel aspect of that reporting claim, involving information provided to a third-party governmental entity rather than to the “learned intermediary,” consisted of one sentence:  “And reasonable efforts would, in some circumstances, entail a warning to a third party such as the FDA.”  Id.  That statement is starkly unsupported and as just shown, simply ignores Maryland law.

Then there’s In re Smith & Nephew Birmingham Hip Resurfacing (BHR) Hip Implant Products Liability Litigation, 300 F. Supp.3d 732 (D. Md. 2018), which we discuss here only because the decision is not at all clear what state’s law was being applied.  The failure-to-report claims in Birmingham Hip got the MDL treatment.  Citing nothing but Stengel [overruled by that state’s highest court], that court let those claims proceed:

As already stated, [defendant] was required by the FDA to report adverse incidents.  Thus, state failure to warn claims that support holding [defendant] liable for its alleged failure to report specific information to the FDA are not expressly preempted.

Id. at 745 (citations omitted).  That’s it – two sentences devoid of any discussion of any supposedly “parallel” reporting duty imposed under any given state’s law.

Thus we already knew that this particular judge is a recidivist violator of Erie conservatism, both inside and outside of this particular MDL.  That this case produced strike three did not surprise us at all.

Having said all this, the combined disregard of both Riegel and Erie in Birmingham Hip rises to a level of judicial lawlessness that would be mandamus-worthy, except. . . .

Summary judgment was nonetheless granted in Birmingham Hip on the plaintiff’s failure-to-report claim.  The claim failed on causation grounds on the facts of this particular case.  The plaintiffs’ “theory of causation [was] limited to showing that ad hoc data provided to the FDA would necessarily have been made public such that [it] would have been incorporated into the literature [the implanting surgeon] read prior to [the relevant] surgery.”  2021 U.S. Dist. Lexis 100041, at *55.  Causation thus depended on “the ad hoc data at issue [being] published in the MAUDE database.”  Id. at *56.  But none of the plaintiffs’ experts opined that this was likely to happen.  Id. at *57-58.

Thus, summary judgment was entered on this claim, but not for the fundamental reason that it does not exist in the first place under North Carolina law.  Because mandamus has an irreparable harm element, e.g., In re National Prescription Opiate Litigation, 956 F.3d 838, 845 (6th Cir. 2020); In re Depuy Orthopaedics, Inc., 870 F.3d 345, 353 (5th Cir. 2017), and the offending claim has been dismissed, mandamus is not likely to be successful, no matter how wrong the Birmingham Hip preemption ruling is.

This Birmingham Hip opinion also contains a variety of other non-failure-to-report-related rulings, some of which may also be of interest to Blog readers.  Briefly, the decision allowed express warranty and negligent misrepresentation by “omission” having to do with alleged “comparative” statements made in non-FDA (and thus not covered by preemption) marketing materials.  2021 U.S. Dist. Lexis 100041, at *61-62, 77-79.  It’s not at all clear that North Carolina law recognizes negligent misrepresentation based solely on omissions, id. at *61-62, but we’ve already seen how much fidelity to state law matters in this MDL.

For the reason we discussed (causation), the court entered summary judgment against the plaintiffs’ negligence per se claim, since it was also based on failure to report.  Id. at *71.  For the sake of completeness, we point out one more reason why that claim should have failed – the Fourth Circuit has ruled that mere failure to report is neither sufficiently substantive nor patient-directed to support a negligence per se claim.  “Where a statutory provision does not define a standard of care but merely imposes an administrative requirement, such as the requirement . . . to file a report to support a regulatory scheme, violation of such requirement will not support a negligence per se claim.”  Talley v. Danek Medical, Inc., 179 F.3d 154, 159 (4th Cir. 1999) (emphasis added).

In the best ruling (from a defense perspective) in the decision, the plaintiffs’ claim for failure to train was preempted, since “the [FDA’s] PMA condition that required . . . a training program did not include requirements as to what the training must include.”  2021 U.S. Dist. Lexis 100041, at *72.  Therefore, plaintiffs’ attempts to add training requirements not imposed by the FDA violated the statute’s “different from or in addition to” language.  Id.at *73-74.

Also preempted was plaintiffs’ attempt to argue “adulteration” under North Carolina’s “Little FDCA” statute (plaintiffs’ counsel in MDLs are endlessly creative).  Id. at *75 (“to prove that [defendant] violated the NCFDCA’s misbranding provision, . . . plaintiffs would have to show that the FDA-approved labeling that accompanied the product was misleading, a claim which is preempted”).  Finally, various affirmative defenses were either dismissed, or not, on case-specific reasons or else deferred to trial.  Id. at *83-90.