Photo of Bexis

We’re returning to the topic of implied Buckman preemption today to discuss three recent decisions, Hughes v. Boston Scientific Corp., ___ F.3d ___, 2011 WL 184554 (5th Cir. Jan. 21, 2011); LeFaivre v. KV Pharmaceutical Co., ___ F.3d ___, 2011 WL 148730 (8th Cir. Jan. 19, 2011); and Goldsmith v. Allergan, Inc., 2011 WL 147714 (C.D. Cal. Jan. 13, 2011).  But before we get to these cases, we have to reiterate some things that we said not too long ago in our Bashing Bausch post.  These have to do with the relationship between implied preemption under Buckman Co. v. Plaintiffs’ Legal Committee, 531 U.S. 341 (2001), and express preemption under Medtronic, Inc. v. Lohr, 518 U.S. 470 (1996), and Riegel v. Medtronic, Inc., 552 U.S. 312 (2008).  Specifically, defendants need to keep in mind what implied Buckman preemption can and can’t do.
Specifically, as a matter of express preemption, Lohr created – under the rationale that remedies don’t matter – an exception for largely undefined state-law claims that are “identical” to FDA regulatory standards.  “Nothing in §360k denies [a state] the right to provide a traditional damages remedy for violations of common-law duties when those duties parallel federal requirements.”  518 U.S. at 495.  We have to live with that.  The Supreme Court was unanimous on the point.  Then there’s the Riegel dictum that MDA preemption “does not prevent a State from providing a damages remedy for claims premised on a violation of FDA regulations; the state duties in such a case “parallel,” rather than add to, federal requirements.”  552 U.S. at 330.
Keep that in mind.  Express preemption allows “parallel” state law duties “premised on a violation of FDA regulations.”  If such a claim exists, it may well fail for other reasons, but it won’t be expressly preempted.
Now we turn to implied preemption.  We know now – even if we don’t like it – after Wyeth v. Levine, 129 S. Ct. 1187 (2009), and Altria Group, Inc. v. Good, 555 U.S. 70 (2008), that with implied preemption, we have a presumption against preemption to deal with.  That is, where, under Buckman Co. v. Plaintiffs’ Legal Committee, 531 U.S. 341 (2001), there isn’t.
When isn’t there a presumption?
There’s no presumption when we’re concerned with the dealings between the FDA and the manufacturer in question.  “[T]he relationship between a federal agency and the entity it regulates is inherently federal in character because the relationship originates from, is governed by, and terminates according to federal law.”  531 U.S. at 348.
Keep that in mind, too.
Finally, Buckman – as a matter of implied preemption – held that preemption (akin to lack of standing) applies to bar certain types of FDCA violation claims by virtue of 21 C.F.R. §337(a), the statute’s preclusion of private enforcement.
Where is this reservoir of implied preemption found?
Where the federal claims aren’t parallel.
And when aren’t violation claims parallel?
Where the claim doesn’t have a state-law equivalent to be parallel to, but instead is wholly a creature of the FDA regulatory scheme:

[T]he fraud claims exist solely by virtue of the FDCA disclosure requirements.  Thus, although [Lohr] can be read to allow certain state-law causes of actions that parallel federal safety requirements, it does not and cannot stand for the proposition that any violation of the FDCA will support a state-law claim. In sum, were plaintiffs to maintain their fraud-on-the-agency claims here, they would not be relying on traditional state tort law which had predated the federal enactments in questions. On the contrary, the existence of these federal enactments is a critical element in their case.

531 U.S. at 353 (emphasis added).
To recap:
(1) Parallel violation claims aren’t expressly preempted by the statute’s “different from or in addition to” language.
(2) Parallel violation claims must parallel “traditional” state tort law.
(3) Buckman implied preemption is the converse – it preempts violation claims that are not grounded in “traditional” tort law.
(4) With Buckman implied preemption, there’s a presumption against preemption, again, with respect to traditional state law violation claims.
(5) With Buckman implied preemption, there’s not a presumption against preemption for claims grounded in the “relationship” between the FDA and (we’ll presume) the defendant.

The Bad – Hughes v. Boston Scientific

With those propositions in mind, we turn to the most recent case, Hughes, 2011 WL 184554.  Hughes “focus[ed] primarily on [plaintiff’s] claim that [the defendant] failed to provide adequate warnings.”  So we’re not dealing with a manufacturing defect claim.  Is this claim parallel to “traditional” tort claims? That’s the key question.  The answer, evident from the face of the complaint is (to us) no:

[Plaintiff] proceeded on the theory that [defendant] failed to comply with the FDA’s MDR regulations requiring a manufacturer of a Class III device to report incidents.

2011 WL 184554, at *3.  Specifically, the plaintiff claimed that the defendant’s reporting “algorithm” caused a failure to report certain injuries (burns similar to hers) that the FDA required be reported.  Id.  That claim is supported by your usual FDA expert spouting legal conclusions – that the defendant broke the law.
Think about that claim for a minute.  Where’s the parallel tort duty?  What state common-law doctrine (as opposed to statutory-based requirement) requires anybody to report anything to a governmental agency?   Whether it’s child abuse, adverse events, financial improprieties, whatever, the source of all obligations to report to the government – as opposed to duties to warn individuals – is statutory or regulatory.
There’s no state common-law tort duty at all to report anything to the FDA.
Thus, as we see it, Hughes is the same as Buckman, in that the the obligation to report arises from the “relationship” between the FDA and the manufacturer, and nothing else. So that claim should have been preempted.
We call Hughes “bad” because that didn’t happen.
It’s not all bad.  The Hughes court first affirmed the throwing out all of the plaintiff’s standard product liability claims (manufacturing, design, and warnings).  2011 WL 184554, at *5. So far, so good. That illustrates another point we’ve made.  Riegel preemption remains extremely broad.  The easy to prove product liability claims, even under the worst decisions, are gone.  All the plaintiffs have left, no matter how much slack they get cut, is a rather technical regulatory based claim.
The first statement in Hughes that we really don’t like looks innocuous.  The court “[a]ssum[es] that a failure to warn claim may be pursued under Mississippi law as [plaintiff] argues.”  2011 WL 184554, at *6.
Assuming?
Why?
Where’s the legal argument that it doesn’t?
Apparently there wasn’t any.  That doesn’t look like a screw up by the court.  Instead, the defendant – for who knows what reason (money?) – chose to litigate with one hand tied behind its back.
Any defendant that does that – does not give the court state-law bases why these violation claims fail – is more likely to lose.  Period.  A good state-law ground:  (1) may induce the court not to reach preemption at all (that is preferentially to choose the non-constitutional ground), or (2) improves the atmosphere by demonstrating that the claim in question is pretty darn poor in any event and isn’t worth salvaging.
Folks, there are lots of state-law obstacles to tort claims asserting FDCA violations.  We’ve dealt with these extensively in our negligence per se-related posts – particularly here and here.  They need to be utilized.
Defense counsel who litigate only half a case – particularly when the half they do pursue is preemption – aren’t doing their clients any favors by saving a little money.  It’s a case of penny-wise and pound foolish.  All they’re doing is increasing the likelihood of losing.
We don’t like defendants losing, even when it’s not our case.  It’s the reverse of the rising tide floating all boats.  It’s a major reason why we write this blog.
Turning back to Hughes’ treatment of the law, we have the statement “a failure to warn claim limited to an assertion that the defendant violated a relevant federal statute or regulation is ‘parallel’ to federal requirements as defined in Riegel.”  2011 WL 184554, at *6.  That’s true, as far as it goes, which is limited to Riegel and express preemption. The minimal discussion of such claims in Riegel is limited to reiterating Lohr’s conclusion that a violation claim isn’t “different from or in addition to” under the MDA preemption clause.
Hughes analogizes to an earlier case called Gomez. However, Gomez involved a very different kind of violation claim – a “negligence claim alleging that the defendant had defectively manufactured the device.” Hughes, 2011 WL 184554, at *7 (describing Gomez).
But a manufacturing-related claim is about as “traditional” a tort claim as anyone could ask for.  Such a claim is a world away from a claim that the defendant didn’t report stuff to the FDA.  But Hughes missed that distinction entirely, lumping blindly lumping together manufacturing defect claims in other cases with the peculiar warning claim before it, based entirely on failure to report, as unpreempted violation claims.  Id. at *8.
Where’s Buckman?
Where’s any discussion of violation claims that aren’t “traditional,” but rather are grounded in the “relationship” between the FDA and its regulated manufacturers?
Not yet.  That holding was limited to express preemption.
Then Hughes turns to “negligence per se” and dodges most issues – limiting itself to preemption.

[Defendant’s] preemption defense only requires us to decide which of [plaintiff’s] state law causes of action are foreclosed under § 360k. . . .  We need not decide, therefore, whether Hughes will be able to invoke the doctrine of negligence per se as a matter of Mississippi law.

Id. at *8.
There’s that damn waiver again – the defendant throwing away half its defense.
What do we mean?  Well, how about Sumrall v. Mississippi Power Co., 693 So.2d 359 (Miss. 1997)?  In that case, the Mississippi Supreme Court rejected negligence per se based upon OSHA violations because, similarly to the FDCA, Congress expressed an intent that OSHA violations not become mixed up in private civil litigation:

In Otto v. Specialties, Inc., 386 F.Supp. 1240, 1244-45 (N.D. Miss. 1974), the District Court for the Northern District of Mississippi was forced to make an Erie determination of whether OSHA regulations were admissible as evidence of negligence under Mississippi negligence law. In finding the regulations to be not admissible, the district court stated:

We believe the Supreme Court of Mississippi, if faced with this question, would recognize, as we do, that what is at stake here is a question of judicial buttressing of legislative goals.  We believe that with this recognition would come a realization that, before the judiciary undertakes to supplement legislatively designed sanctions it should first inquire whether any supplementation was foreseen or is needed.  Such an inquiry into OSHA has been made by the federal courts, which have concluded that no private civil remedy is needed to fulfill the goals established by Congress in its adoption of the statute.  That this determination was made in the context of a federal civil remedy and not within the framework of the negligence per se doctrine is to us irrelevant, since both concepts share a common raison d’etre – a judicial addition to statutory penalties thought to be inadequate to the purposes the legislative branch sought to promote.
We concede that the Mississippi courts need not be bound in this matter by the federal determination of OSHA’s purpose and effect.  We believe, however, that the Supreme Court of Mississippi would be persuaded by the logic of those opinions to refuse to permit the utilization of OSHA safety standards in this case, either as conclusive proof or evidence of negligence by [the defendant].

Otto, 386 F.Supp. at 1245.
We are persuaded by the district court’s reasoning and hold that, in light both of it and of this Court’s clearly stated rule that governmental codes and regulations are not admissible unless given compulsory force by the state legislature, evidence of OSHA regulations is not admissible to show negligence.

693 So.2d at 366-67 (emphasis added).  Cf. Chisolm v. Mississippi Dept. of Transportation, 942 So.2d 136, 143 (Miss. 2006) (negligence per se is “a tool for assessing a breach of duty only after a legal duty has already been established.  It cannot be used to create a legal obligation under Mississippi law”).
Why give away the argument that Mississippi negligence per se principles would not, as a matter of state law, recognize a cause of action predicated on the violation of a statute expressly intended by its legislative drafted to be enforced only by governmental entities?
Hughes in this respect is no different than Bausch – where the defendant also gave away this argument, likewise soundly grounded in state supreme court precedent.  Perhaps predictably, Hughes blindly follows Bausch, holding that, simply because state law generally allows “negligence per se,” there’s an unpreempted FDCA-based cause of action.  2011 WL 184554, at *8 (plaintiff “is not foreclosed by §360k from arguing at trial that the doctrine of negligence per se is available”) (citing Bausch).
But again, under express preemption, if the claim is assumed (because nobody bothers to argue otherwise) to be a “traditional” state-law tort claim, then, yes there’s no preemption because of the way Lohr read the “different from or in addition to” language.
So while Hughes so far is bad, it’s not necessarily wrong.
What was left in Hughes is fighting in the trenches, against paid FDA experts opining on what the law supposedly is.  2011 WL 184554, at *9-10.  Apparently, there are some bad facts:

[A]ny danger that the jury in this case may apply the plain terms of the MDR regulations in a different or more stringent manner than the FDA intended is considerably mitigated by the summary judgment evidence indicating that the FDA disapproved of [defendant’s] reporting practices. . . .  [T]he FDA told [defendant] that when information regarding a burn is “ambiguous” as to whether the burn requires medical treatment or intervention, the burn must be reported.

Id. at *10.  When a defendant has bad facts, it can’t afford to give up available defenses.
Finally, we get to Buckman, and here everything heads south.  Instead of evaluating whether the failure-to-report claim implicates the “relationship” with the FDA, as per Buckman, the court simply says this isn’t fraud on the FDA claim.  “The plaintiffs in Buckman were attempting to assert a freestanding federal cause of action based on violation of the FDA’s regulations; the plaintiffs did not assert violation of a state tort duty.”  2011 WL 184554, at *11.  Well, that’s simply wrong as a matter of history.  Buckman was postured as being predicated on a state law duty – the duty not to commit fraud.  Cf. Cipollone v. Liggett Group, Inc., 505 U.S. 504, 530 (1992) (discussing such state law duty).  What resulted in preemption in Buckman was that the purported fraud occurred entirely in the context of the defendant’s dealings with the FDA.
Hughes then states that there’s a “recognized state tort claim,” 2011 WL 184554, at *11, but doesn’t cite any case in the history of Mississippi law that created any common-law (as opposed to statutory) duty to report something to a government agency.  Hughes thus entirely papers over the yawning gap between a failure to warn and a failure to report claim.  Once again Bausch raises its ugly head.  Id.  But Bausch, wrong as it was on a lot of things, at least involved a manufacturing defect claim predicated on violation of FDA good manufacturing practices.  There’s simply no traditional state law claim in Hughes.  Instead, the court did just what the Mississippi Supreme Court rejected in Chisolm, 942 So.2d at 143 – it allowed plaintiffs to use a reporting violation to “to create a legal obligation under Mississippi law.”
Hughes goes from bad to even worse at the end of its preemption discussion.  We’ll quote the language first:

[Defendant’s] interpretation of Buckman barring this otherwise parallel state claim is inconsistent with the Supreme Court’s reasoning in Riegel, decided long after BuckmanRiegel unequivocally held that parallel state claims survive a defendant’s preemption defense under the MDA because states may impose an additional “damages remedy for claims premised on violation of FDA regulations.”  Riegel, 522 U.S. at 330.  Our conclusion in this respect is also supported by our decision in Gomez, decided years after Buckman, in which we permitted a negligence claim for defective manufacturing to proceed.

2011 WL 184554, at *12.
This passage reminds of a certain sculpture on the campus of Harvard University known as the “statue of three lies.”  In all of three sentences, Hughes committed three major legal blunders.  (1) It ignored that Buckman was entirely an implied preemption case, whereas Riegel was entirely an express preemption case, thus there is no inconsistency at all between the two because they dealt with different doctrines.  (2) It overlooked that the entire violation question in Riegel was waived (552 U.S. at 229 (“we decline to address that argument”)) – so Riegel didn’t “unequivocally hold” anything at all about such claims.  (3) It ignored the fundamental difference between the “manufacturing” claim in Gomez, and the failure-to-report claim before it, with the former being a long-recognized tort claim, and the latter being a creation solely of FDCA reporting requirements.
So we think Hughes is really bad.  As to preemption it’s worse than Bausch, which for all its faults, at least involved a something that bore passing resemblance to a traditional state-law claim.  But us kvetching isn’t going to get either decision off the books.  As we already mentioned, Hughes and Bausch have one notable thing in common.  In neither case was the FDCA violation claim proper on state law, but in both cases the defendants never bothered to make state law arguments.
You can’t do that guys – not with the post-Levine legal climate surrounding preemption in the FDCA context.  We’ve said before that 2011 would probably the year that the contours of the parallel violation exception to PMA preemption would be decided.  So far, we’d have to say that the year’s not off to a very good start.

The Ugly – LeFaivre v. KV Pharmaceutical

We’re sticking LeFaivre, 2011 WL 148730, with our “ugly” moniker because while we could see some logical way of arriving at the court’s result, we can’t stomach the route that the Eighth Circuit took to get there.  LeFaivre involved improperly manufactured drugs.  That much is not in dispute.  The FDA “filed a complaint” over certain manufacturing violations, and the defendant (actually more than one, but the court treated them as a unit) entered into a “consent decree” in which it “stipulated . . . that it had sold drugs that were ‘adulterated’” under the FDCA.  Id. at *1.
So LeFaivre is another “bad facts” case.  A civil suit, seeking economic damages for the same conduct was filed.  There isn’t any express preemption in prescription drug cases, so the defendant raised an implied preemption defense under Buckman.  It got lucky.  The trial court threw the case out, finding it to be nothing more than an attempt at private enforcement of the FDCA, which was barred under §337(a) as construed in BuckmanLeFaivre, 2011 WL 148730, at *2.
Recall the last three of the five propositions that we listed above with respect to Buckman preemption.  They have to do with the distinction between “traditional” common-law claims and “non-traditional” claims arising as a consequence of what the FDCA requires manufacturers to do.
Well, as we’ve already said in connection with Hughes, a manufacturing defect claim can be a state-law analogue to an allegation of violation of FDA good manufacturing practice regulations (assuming the regulations aren’t too vague – a state-law ground, and that the violation actually pertains to the product at issue).
So it would have been rational for an anti-preemption court to view the manufacturing violation claims in LeFaivre as paralleling a “traditional” state-law tort theory and, under Buckman, escaping preemption.
But the Eighth Circuit didn’t do that.  Instead, it construed the §337(a) argument as some sort of “field preemption” case – which it really isn’t, unless the “field” is so shrunken as to correspond to the sort of non-traditional tort claims that Buckman addressed.  “The Court’s comments in [Levine] regarding drugs and drug labeling strongly imply that field preemption does not apply in the present case.”  LeFaivre, 2011 WL 148730, at *5.  It also found no conflict preemption, since there was no impossibility or obstacle raised by requiring the defendant to manufacture its drugs in compliance with FDA regulations applicable to all drug manufacturers.  Id.  Well, duh.
Turning to BuckmanLeFaivre didn’t couch its analysis in terms of a manufacturing defect claim having a state-law analogue and thus not being preempted because it wasn’t solely a creature of federal law.  Instead, it construed Buckman as a field preemption case:

The Court in Buckman specifically applied field preemption to state-law fraud-on-the-FDA claims because policing fraud against federal agencies “is hardly a field which the States have traditionally occupied.”  Buckman, 531 U.S. at 347

2011 WL 148730, at *8.
That’s just ugly.  Buckman has nothing to do with “field preemption.”  It is purely an implied conflict preemption case.  Use of the word “field” didn’t occur in the court’s preemption analysis, but rather in the context of whether a presumption against preemption applied:

Policing fraud against federal agencies is hardly a field which the States have traditionally occupied, such as to warrant a presumption against finding federal pre-emption of a state-law cause of action.  To the contrary, the relationship between a federal agency and the entity it regulates is inherently federal in character because the relationship originates from, is governed by, and terminates according to federal law.  Here, petitioner’s dealings with the FDA were prompted by the MDA, and the very subject matter of petitioner’s statements were dictated by that statute’s provisions.  Accordingly – and in contrast to situations implicating federalism concerns and the historic primacy of state regulation of matters of health and safety – no presumption against pre-emption obtains in this case.

531 U.S. at 347-48 (all citations and quotation marks omitted) (emphasis added).  If Buckman had been about “field preemption,” it would have had to deal with all the other defendants and all the other myriad causes of action that had been alleged in the Bone Screw litigation.  Further, as we’ve discussed before, “field preemption” is where the unfortunate presumption against preemption was invented.  Buckman would never have found that the presumption didn’t apply to a field preemption claim..
Having twisted Buckman into something unrecognizable, LeFaivre went on to conclude that “the present case is distinguishable from Buckman because [plaintiff’s] state-law claims are not fraud-on-the-FDA claims, as they focus on harm that is allegedly perpetrated against consumers rather than the FDA.”  2011 WL 148730, at *8.  We don’t know all the facts of LeFaivre, but as described in the opinion, both the nature of the claim – improper/defective manufacturing – and the nature of the facts – a violation already administratively adjudicated and thus well beyond a bare allegation – are such that we can’t say that the result violates either the language or policy of Buckman.
So we can’t describe the result in LeFaivre as necessarily bad (even though the defendant lost the preemption issue), but it sure was ugly.

The Good – Goldsmith v. Allergan

We saved the best for last, Goldsmith v. Allergan, Inc., 2011 WL 147714 (C.D. Cal. Jan. 13, 2011) – and a tip of the cyber cap to Kurt Karst at Hyman, Phelps for originally sending it along to us .
We consider Goldsmith to be an unimpeachable (if conservative) use of Buckman preemption in that the decision concluded that a purported state-law violation claim unrelated to any “traditional” state tort theory was preempted.  Goldsmith was a consumer fraud claim out of California alleging that the defendant should have to pay for “allegedly market[ing a drug] as a “multi-use” product to encourage administering physicians to use a single vile [sic] of [the drug] for more than one patient.”  2011 WL 147714, at *1.
Goldsmith offered no allegations that there was anything wrong with the design, manufacturing or labeling of the drug.
Thus Goldsmith did not involve any traditional product liability claim, only an allegation of off-label promotion.
As in Buckman, therefore, the claim arose solely because of an FDCA violation.  State common law doesn’t have a concept of off-label promotion.

[T]he Court agrees with Defendants suggestion that Plaintiff’s [consumer protection] claims are based on conduct promoting [the drug] for off-label use. . . .  These, and related allegations . . . constituted an attempt to shoehorn allegations that Defendant had engaged in off-label promotion in violation of the FDCA into state consumer fraud causes of action.

2011 WL 147714, at *3 (quotation marks omitted).  Allegations of this sort are simply attempts to enforce the FDCA privately in violation of §337(a).  “No matter how artfully the Complaint is pleaded in attempting to enforce the FDCA, Plaintiff cannot enforce the FDCA’s off-label advertising provisions simply by calling it a violation of [a consumer fraud statute].”  Id. at *8.
Goldsmith contains a lot of discussion about pleading fraud under Fed. R. Civ. P. 9(b), and concludes that the complaint was insufficient with respect both the misrepresentation and reliance elements.  2011 WL 147714, at *5 (“Plaintiff never alleges that he saw any ads and relied on them”), at *7 (“Plaintiff has not alleged facts showing that Defendant’s alleged representations were false or misleading”).  Thus Goldsmith contains a caveat – that such allegations might escape Buckman preemption if instead they paralleled a state-law fraud claim.  2011 WL 147714, at *3 (“Plaintiff’s [consumer protection] claims are actionable if they include properly pleaded allegations of false or misleading representations that resulted in Plaintiff’s injuries”).  That also fits within Buckman’s rationale, which as discussed above, predicates §337(a) preemption/lack of standing on the state law claim at issue being purely a creature of the FDCA without any common-law analog.
For these reasons – its fidelity to Buckman in both rationale and result – we bestow our “good” designation on Goldsmith.

The Funky – Funk v. Stryker

Finally, while we were writing this post (yesterday) we noticed Funk v. Stryker Corp., No. 10-20022, slip op. (5th Cir. Jan. 25, 2011), had appeared on the Fifth Circuit’s website.
Even though it’s not a Buckman case, we decided to add Funk because it’s an important appellate decision.  It creates a circuit split with Bausch, as both decisions arise from the same defect allegations involving the same device.  Circuit splits are one criteria the United States Supreme Court uses to decide whether to accept a case for appeal, and if any case deserves further review, it’s Bausch.
But Funk – there’s no other way to say it –is rather funky.  Unfortunately (maybe even for us on the defense side), the plaintiff screwed up his appeal and only appealed from the original, pretty pathetically pleaded, complaint.  Slip op. at 5.  That original complaint was so poor that there wasn’t much question that the claims were all preempted.
Thus, the conflict between Funk and Bausch is more with respect to TwIqbal than preemption.  Contrary to Bausch, Funk holds that a parallel violation isn’t adequately pleaded without specifying both how the FDCA was violated and that the violation caused the plaintiff’s injury:

This complaint is impermissibly conclusory and vague; it does not specify the manufacturing defect; nor does it specify a causal connection between the failure of the specific manufacturing process and the specific defect in the process that caused the personal injury. Nor does the complaint tell us how the manufacturing process failed, or how it deviated from the FDA approved manufacturing process.

Funk, slip op. at 7.  Instead, an adequately pleaded complaint “specifies with particularity what went wrong in the manufacturing process and cites the relevant FDA manufacturing standards [the defendant] allegedly violated.”  Id. at 8.
Because Funk found that the only arguably unpreempted (violation) claim was insufficiently pleaded under TwIqbal, it didn’t have to reach any controversial preemption questions under Riegel or (as discussed above) under Buckman.

Funk does, however, hold that it’s proper to take judicial notice of “publically-available documents and transcripts produced by the FDA, which were matters of public record.”  Slip op. at 9. T hat’s a useful – if we think rather self evident – holding.