When we heard about Judge Neil Gorsuch being nominated for the United States Supreme Court, our first move was to enter his name in Westlaw along with the term “preemption.” That’s the constitutional doctrine most important to our medical device, generic drug, and (unfortunately to a lesser extent) innovator drug clients. It’s also a doctrine more likely to get less attention in what promises to be the upcoming brouhaha.
By far the most important Gorsuch preemption decision is Caplinger v. Medtronic, Inc., 784 F.3d 1335 (10th Cir. 2015), cert. denied, 136 S. Ct. 796 (2016), the very favorable PMA medical device preemption decision that we discussed previously here and named as our #2 best case of the year for 2015. We’re not reprising those posts here. Rather, we’re examining Caplinger for what it might tell us about Judge Gorsuch’s broader views of FDCA preemption. His dissatisfaction with the Supreme Court’s tortured approach to express preemption in medical device product liability cases is very clear.
Caplinger begins with the strong statement that, in enacting the Medical Device Amendments, Congress “[e]xercis[ed] its authority under the Supremacy Clause” in enacting 21 U.S.C. §360k(a). 784 F.3d at 1336. Before looking at relevant Supreme Court authority, Caplinger comments:
At first glance the answer to this appeal might appear easy enough. Section 360k(a) preempts “any requirement” imposed by states on manufacturers that differs from or adds to those found in the FDCA. Given this expansive language one might be forgiven for thinking all private state law tort suits are foreclosed. After all, a “requirement” usually means a request, need, want, or demand. And an adverse tort judgment seems to involve just that: a demand that a defendant appear to answer for its conduct and pay damages for failing some state law duty.
Id. at 1337 (dictionary citation omitted). A string citation about the lower courts’ “struggles” to make sense of medical device preemption followed. Id. at 1337-38.
The problem was, when the Supreme Court got involved, rather than interpreting Congress’ “expansive” preemptive language according to its terms, the Court “issued a number of opinions that embody ‘divergent views’ about the proper role of the MDA’s preemption provision, a fact that has yielded considerable ‘uncertainty’ among the lower courts.” Id. (citation omitted).
Medtronic, Inc. v. Lohr, 518 U.S. 470 (1996), Caplinger explained, “rejected the notion that state law tort suits are always preempted.” 784 F.3d at 1338. Instead, it created an exception for “parallel” claims for which “[t]he fact that a manufacturer is ‘required’ to defend itself against a damages remedy nowhere provided for in federal law is . . . neither here nor there.” Id. However, what is “parallel” “doesn’t appear in the statute, so its meaning was left entirely to judicial exposition.” Id. On top of that, “five justices” in Lohr “took the view that state and federal law duties ‘parallel’ each other not only when they are identical, but also when state law imposes duties on the defendant that are ‘narrower, not broader’ than those found in the FDCA.” Id. That didn’t make much sense, particularly compared to what §360k actually provided:
Now, you might ask, why isn’t a narrower state law requirement at least ‘different from’ a broader federal requirement − and thus preempted by § 360k(a)’s express terms? The Lohr majority acknowledged that a state duty imposing a ‘narrower requirement’ is indeed ‘different from’ the federal rules in a literal sense.’ And when it comes to interpreting the text of a statute, that’s often the sense that matters most.
784 F.3d at 1338 (Lohr citations omitted). Lohr thus created the current problem. “Lower courts have struggled ever since when it comes to trying to decide whether particular state claims do or don’t ‘parallel’ putative federal counterparts.: Id. (second string cite of complaints about the vague rulings omitted).
As an aside, note Judge Gorsuch’s frequent use of contractions. Quite apart from his result, we identify with his writing style.
And that wasn’t Lohr’s only legacy. The FDCA “provides that state laws are preempted to the extent they conflict with ‘any [federal] requirement applicable under this chapter to the device.’” Caplinger, 784 F.3d at 1339 (quoting §360k(a)). “The chapter in question . . . contains the whole of the FDCA.” Id. Once again, Congress used language indicating broad preemption, and once again Lohr cut that language back:
So, again read literally, it would seem “any” federal requirement imposed by the FDCA is capable of preempting any different or additional state requirement. But again the Lohr majority held otherwise, instructing lower courts that for preemption to take place the FDA must first issue some regulation “specific” to a “particular device.”
Caplinger, 784 F.3d at 1339 (quoting Lohr). There might be some preemptive “general” exceptions, but once again “Lohr told us little.” Id.
Two subsequent Supreme Court decisions “cut back on the scope of [Lohr’s] initial decision. Id. “The first blow” came in Buckman Co. v. Plaintiffs Legal Committee, 531 U.S. 341 (2001). Id. Buckman “addressed” a different FDCA provision (21 U.S.C. §337(a)) “authorizing the federal government to enforce” it. Id. The FDCA’s “language and structure” was “clear evidence that Congress intended that the MDA be enforced exclusively by the Federal Government.” Id. (quoting Buckman, 531 U.S. at 352). But Buckman “left undisturbed the portion of Lohr allowing” parallel claims. That was also perplexing to lower court judges, such as Judge Gorsuch:
So it is that lower courts must now accept both the notion that §337(a) shows Congress intended the federal government to enjoy exclusive enforcement authority over the MDA and the notion that §360k(a) permits private tort suits that do no more than parallel the MDA.
Caplinger, 784 F.3d at 1339 (citation to law review article concerning this “tension” omitted).
Then “[t]he Court retreated further from Lohr in Riegel v. Medtronic, Inc., 552 U.S. 312 (2008).” Id. at 1339. According to Caplinger, Riegel held that the “any state requirement” clause of §360k “should be read literally: any state requirement, whether device specific or generally applicable, is preempted when it differs from or adds to federal requirements. 784 F.3d at 1339. That produced more Caplinger commentary:
While perhaps unremarkable on its own terms, this invited a new tension with Lohr and its suggestion that (at least usually) only device-specific federal requirements bear preemptive power even though the statute’s literal language suggests “any” federal requirement hold that power. True, Lohr and Riegel formally addressed different clauses in §360k(a) − Lohr interpreted the “any” federal requirements clause and Riegel discussed the “any” state requirements clause. But it’s no small mystery why the same word − “any” − should bear such different meanings in two such similar clauses that lie cheek by jowl in the same statutory subsection.
Caplinger, 784 F.3d at 1339-40 (emphasis added).
It was “no easy task” to “apply all these competing instructions. Id. at 1340. Judge Gorsuch synthesized the Supreme Court’s clear-as-mud instructions thusly:
[I]t seems we aren’t supposed to ask whether [plaintiff] wishes to use state tort law to impose on [defendant] a safety requirement that is “different from, or in addition to” a federal requirement so much as whether she seeks to vindicate a state duty that is “narrower” or “broader” than a federal duty. To the extent the state law duty is narrower than or equal to the federal duty it survives, through what seems a sort of Venn diagram approach to preemption. Still, even if the state claim fails that test because it would impose a “broader” duty than can be found in federal law, it appears we may not find the claim preempted just because it conflicts with “any” federal requirement. Instead, we may find the state law claim preempted only if there exists a device-specific federal requirement, though this test admittedly finds no analogue when it comes to the state requirement clause interpreted in Riegel. Finally, should the state claim survive this far, we must ask whether it exists “solely by virtue” of the federal statutory scheme (unacceptable) or “predates” the scheme (acceptable). It’s no wonder that the difficulty of crafting a complaint sufficient to satisfy all these demands has been compared to the task of navigating between Scylla and Charybdis.
Id. (citation omitted). Judge Gorsuch adds, “[o]ne can’t help but wonder if perhaps some of those rules warrant revisiting and reconciliation.” Id.
Great idea.
Then he decides the Caplinger case and finds preemption essentially across the board. All the usual tort claims are preempted, leaving only the purported “parallel” claim to dealt with. Id. Finding rather poor advocacy on the plaintiff side, Caplinger points out that plaintiff did not “attempt[] . . . to identify a single parallel federal statute or regulation.” Id. at 1341. The “directions for use” provisions don’t even apply to prescription drugs – which sort of puts a dent in any argument that (as in Caplinger) uses that regulation to attack off-label use/promotion. Id. Moreover, the federal regulations address “labeling,” whereas plaintiff’s “state law claims go well beyond that, attacking . . . advertising and oral and written representations to her, her doctor, and others.” Id. That’s a “broader” claim, and therefore preempted. Id. Interestingly, even though PLIVA, Inc. v. Mensing, 564 U.S. 604 (2011), and Mutual Pharmaceutical Co., v. Bartlett, 133 S.Ct. 246 (2013), are nowhere mentioned – Caplinger being an express preemption case – we (or to use Caplinger’s phrasing, “one”) find(s) a hint of impossibility preemption: “More than that, once the FDA approves a device’s label as part of the premarket approval process (as it has here), the manufacturer usually may not alter the label’s warnings without prior agency approval.” 784 F.3d at 1341 (emphasis added).
Caplinger then returns to the plaintiff’s poor advocacy (suggesting that Judge Gorsuch isn’t likely to cut lawyers any slack):
So in the end [plaintiff] relies exclusively on a legally inapplicable provision in her effort to establish a parallel claim, fails anywhere to discuss the apparently applicable one, and offers no answer to the conundrum how she might impose a state tort duty on [defendant] to revise a label that federal regulation precludes it from revising.
784 F.3d at 1341 (citations omitted). Maybe somebody else could do a better job, but “[n]either are the courts under an obligation to perform that work for [this plaintiff], searching out theories and authorities she has not presented for herself.” Id. at 1342.
Finally, Caplinger addressed plaintiff’s claim (based on an argument allowed by one federal district judge in Ramirez v. Medtronic Inc., 961 F. Supp.2d 977 (D. Ariz. Aug. 21, 2013)), that the three prior Supreme Court decisions “don’t much matter” because “the fact that her suit concerns an off-label use is enough all by itself to insulate all her claims from preemption.” 784 F.3d at 1343. Judge Gorsuch had no problem calling out this end run for what it was:
In not a single one of its many and involved encounters with the MDA has the Supreme Court so much as hinted at this alternative path around preemption. But [plaintiff] says its past obscurity shouldn’t stop us from recognizing it now. This we decline to do.
Id. at 1343-44 (footnote omitted).
Importantly, “[t]extually, §360k(a) simply does not contain the distinction [plaintiff] would have us draw between suits addressing on-and off-label uses.” Id. at 1344. The statute refers to “devices”; “[n]othing depends” on “uses” – on- or off-label. Id. “[B]y its terms, the statute preempts any effort to use state law to impose a new requirement on a federally approved medical device.” Id. “Knowing about (even encouraging) off-label uses . . ., Congress proceeded . . . to preempt any state tort suit challenging the safety of a federally approved device without qualification about the manner of its use.” Id. (discussing 21 U.S.C. §396). Congress’ unqualified language thus “becomes all the harder to ignore, a sort of dog that didn’t bark.” Id.
Ultimately all the plaintiff’s arguments in Caplinger meet “a similar textual dead end,” because “any” means “any.” As for plaintiffs’ extra-textual preemption “precondition[s],” “[w]e discern nothing in the judicial power that might permit us to undertake such a revamping of Congress’s handiwork.” 784 F.3d at 1345.
That category specifies its preemptive reach plainly and broadly: any state requirement that adds to federal requirements and that relates to the safety or effectiveness of the device is preempted. No other qualification exists.
Id. Judge Gorsuch went on to discuss other PMA preemption cases, and also congressional intent, but only after, and subsidiary to, his textual analyses. Id. at 1345-47. The Caplinger decision concluded:
Not everyone may agree with how Congress balanced the competing interests it faced in this sensitive and difficult area. We can surely imagine a different statute embodying a different judgment. But strike a balance Congress had to and did, and it is not for this court to revise it by beating a new path around preemption nowhere authorized in the text of the statute and nowhere recognized in any of the Supreme Court’s many forays into this field.
Id. at 1347 (emphasis added).
That’s one – and thankfully for the patience of our readers, the most important one – of Judge Gorsuch’s preemption decisions. His second most interesting (but only in the America’s Cup sense of “second”) opinion is Cook v. Rockwell International Corp., 790 F.3d 1088 (10th Cir. 2015), cert. dismissed, 136 S. Ct. 2055 (2016), which involved a state tort (nuisance) action involving the handling of nuclear materials. Express preemption (as in Caplinger) was not at issue, since the first holding in Cook was that express preemption had been waived. 790 F.3d at 1093. As for implied preemption, conflict preemption had been “disclaimed [by defendants] in this appeal.” Id. at 1098. The dispute in Cook was over field preemption, and “we see no field preemption in the Act of the sort the defendants describe.” Id. at 1094. Cook involved two “competing” preemption “narratives” – one (plaintiffs’ version) “entirely familiar” and the other (defendant’s version) not “metaphysically impossible but stating it in concise terms does make you wonder.” Id. As to the defendant’s field preemption argument, Cook applied a presumption against preemption:
[T]he Supreme Court has instructed us to “start with the assumption that the historic police powers of the States were not to be superseded . . . unless that was the clear and manifest purpose of Congress.” So to the extent Congress’s statutory direction is susceptible to more than one reading, we have the “duty to accept the reading that disfavors pre-emption.” A duty that is only “heightened” where (as here) the area of law in question is one of traditional state regulation like public health and safety. . . . [A]pplying the traditional tools of statutory interpretation to the Act before us, it quickly becomes clear that nothing in its language, structure, or history favors the defendants’ curious statutory construction over the plaintiffs’ prosaic one − let alone favors it so clearly that we might overcome the presumption against preemption.
790 F.3d at 1094 (citations to the usual presumption against preemption cases omitted).
Where does Cook start its implied field preemption analysis?
Where do you think?
“Start with the text.” Id. A fairly lengthy discussion of the Price Anderson Act’s (“PAA”) provision ensued, after which Judge Gorsuch concluded:
Where does any of this language – expressly − preempt and preclude all state law tort recoveries for plaintiffs who plead but do not prove nuclear incidents? We just don’t see it. Congress knows well how to preempt a field expressly when it wishes. . . . There’s just nothing like that in the statutory text before us.
Id. at 1095 (OSHA discussion omitted). Rejecting some PAA-specific defense arguments, Cook found “nothing” in that language suggesting field preemption, and “[s]urrounding textual features confirm[ed] the point.” Id. “[B]eyond textual clues . . . the larger statutory structure does nothing to alter our impression.” Id. at 1096. Nor did legislative history (which Judge Gorsuch is willing to utilize, unlike Justice Scalia), suggest broad implied preemption. Id. at 1096-97.
As for precedent, Supreme Court PAA precedent did not support field preemption. The “defendants overread these decisions,” which recognized that Congress “has done little to forbid states from indirectly regulating nuclear safety through the operation of traditional after-the-fact tort law remedies.” Id. at 1098. Nor was such an arrangement “unusual”:
Often Congress entrusts before-the-fact regulation to a federal agency while leaving at least some room for after-the-fact state law tort suits. It has done so in the field of motor vehicle safety. It has done so in the field of medical devices. [citing Caplinger] And all the statutory evidence before us suggests it has done the same thing here.
Id. (other citations omitted). Field preemption therefore failed in Cook. “[T]here’s nothing inconsistent about a statutory scheme that provides federal jurisdiction over certain claims . . . while permitting claims involving lesser occurrences to proceed to decision under preexisting state law principles.” Id. at 1099.
Judge Gorsuch’s remaining decisions are considerably less detailed on preemption. Russo v. Ballard Medical Products, 550 F.3d 1004 (10th Cir. 2008), involved whether state-law claims concerning the ownership of certain product improvements (in a medical device) were preempted by federal patent law. Russo applied “obstacle” conflict preemption – “when state law ‘stands as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress,’ as expressed in this case in the Patent Act.” Id. at 1011 (citation omitted). Russo was primarily an application of an on-point United States Supreme Court decision (Kewanee Oil Co. v. Bicron Corp., 416 U.S. 470 (1974)), which had refused to preempt state trade secret laws, as they were compatible with the federal patent regime. 550 F.3d at 1011-13. Broad preemption, Russo decided, would produce “incongruous” results:
Were the law otherwise, it would be incongruous indeed. Any defendant could (and would have a significant incentive to) insulate itself from a trade secret misappropriation claim simply by patenting the stolen idea. And Kewanee Oil’s holding that trade secret and patent law can coexist would be significantly impaired because patent law would preempt many (if not most) trade secret claims.
Id. at 1015. The end result in Russo was that claims “aris[ing] from the misappropriation of a trade secret . . . may be tried under state law” whereas claims “seek[ing] to exclude others from employing a publicly disclosed idea . . . must be tried under federal patent law.” Id. at 1016.
In Kiker v. Community Health Systems Professional Service. Corp., 484 F. App’x 215, 217 (10th Cir. 2012), Judge Gorsuch avoided an ERISA preemption question by finding the underlying claim lacked merit. Similar lack of merit mooted an IDEA preemption argument in Roberts v. International Business Machines Corp., 733 F.3d 1306, 1310-11 (10th Cir. 2013). Howard v. Ferrellgas Partners, L.P., 748 F.3d 975 (10th Cir. 2014), briefly mentioned preemption by the Federal Arbitration Act as a background fact, but Howard was a procedural decision about how to determine arbitrability. Kay Electric Co-op. v. City of Newkirk, Okla., 647 F.3d 1039 (10th Cir. 2011), similarly mentioned preemption in passing, but only decided issues concerning the scope of municipal anti-trust immunity.
Putting all this together, we can tell that Judge Gorsuch is a textualist in preemption matters. That comes through loud and clear in both Caplinger and Cook. We are cautiously optimistic that he will be inclined to do away with the “general vs. specific” and “parallel claim” exceptions to express MDA preemption in Lohr, and would be content to interpret that broad preemption language in §360k in accordance with its terms, leaving Congress to amend the statute if it does not like the scope of its own language. We also think that Judge Gorsuch would be a friend of Buckman preemption, since it is also based on a statutory provision of broad applicability.
How Judge Gorsuch would view generic preemption is harder to tell, as he has never adjudicated an impossibility preemption case. It will probably depend on how convincing the evidence of impossibility is. Since unilateral alteration of the design of prescription medical products is pretty emphatically precluded by the FDA scheme, we would again be cautiously optimistic that such claims would be preempted. We’re less sure, however, of whether Judge Gorsuch would trim Wyeth v. Levine, 555 U.S. 555 (2009), insofar as it applies to warning claims. Since there is no express preemption clause for drugs, this is a place where – at least under the current FDA scheme – his textualist views might require our side to go to Congress (or to the FDA) to seek broader preemption.