Photo of Bexis

JAMES M. BECK is Counsel resident in the Philadelphia office of ReedSmith. He is the author of, among other things, Drug and Medical Device Product Liability Handbook (2004) (with Anthony Vale). He wrote the seminal law review article on off-label use cited by the Supreme Court in Buckman v. Plaintiffs Legal Committee. He has written more amicus briefs for the Product Liability Advisory Council than anyone else in the history of the organization, and in 2011 won PLAC's highest honor, the John P. Raleigh award. He has been a member of the American Law Institute (ALI) since 2005. He is the long-time editor of the newsletter of the ABA's Mass Torts Committee.  He is vice chair of the Class Actions and Multi-Plaintiff Litigation SLG of DRI's Drug and Device Committee.  He can be reached at  His LiinkedIn page is here.

We’ve blogged a number of times about the Dormant Commerce Clause (“DCC”) as an additional basis for bolstering both preemption and Due Process arguments.  Here’s another prescription drug-based example.

The state of New York decided to impose a special tax on opioid manufacturers to finance various responses to the so-called “opioid epidemic.”  The tax came in the form of an “a $600 million ‘stewardship fund.’”  Healthcare Distribution Alliance v. Zucker, ___ F. Supp.3d ___, 2018 WL 6651682 (S.D.N.Y. Dec. 19, 2018).  There was a problem with that, however.  What happens with business taxes?  They get passed along (like tort verdicts do) in the form of higher retail prices based on increased costs of doing business.  So the New York legislature, to paraphrase Dr. Seuss, “got an idea.  An awful idea.  They got a wonderful, awful idea.”  No, they didn’t steal Christmas, but they decided to prohibit the manufacturers subject to the tax from passing it along to consumers:

In the provision defining stewardship payments, the [New York statute] states, “No licensee shall pass the cost of their ratable share amount to a purchaser, including the ultimate user of the opioid, or such licensee shall be subject to penalties pursuant to subdivision ten of this section.”  Later, in the penalties provision, the Act notes that “[w]here the ratable share, or any portion thereof, has been passed on to a purchaser by a licensee, the commissioner may impose a penalty not to exceed one million dollars per incident.”

Id. at *3 (quoting N.Y. Pub. Health Law §§3323(2), 3323(10)(c)).

New York, however, is only one state.  The taxed manufacturers, by contrast, sell their products nationwide, as authorized by those products’ multiple FDA approvals.  New York has no power, and the statute had no mechanism, to enforce the prohibition against passing along the cost of “ratable shares” of tax liability in any place other than New York.

Enter the DCC.  What New York did, whether by intent or default, was to pass a tax, to the benefit of in-state “opioid stewardship” programs that would inevitably be paid for solely by opioid consumers in other states, as to whom the statute’s no-pass-through prohibition did not operate.

That arrangement, the court in Healthcare Distribution held, is a burden on interstate commerce that is unconstitutional under the DCC.  First, neither New York, nor any other state, can enact extraterritorial burdens on interstate commerce:

The absolute constitutional prohibition on state regulation of commerce occurring beyond the state’s borders is clear. . . . A statute that directly controls commerce occurring wholly outside the boundaries of a State exceeds the inherent limits of the enacting State’s authority and is invalid regardless of whether the statute’s extraterritorial reach was intended by the legislature. The Constitution is concerned with the maintenance of a national market for interstate commerce. Therefore, even if a statute may not in explicit terms seek to regulate interstate commerce, it can do so nonetheless by its practical effect and design.

Id. at *16 (citations and quotation marks omitted).  Second, states may not discriminate against interstate commerce – such as by imposing taxes that exempt in-state commerce:

The Dormant Commerce Clause also contains an antidiscrimination principle. . . .  [S]tates are aware of the obvious constitutional problems of tariffs. . . .  Instead, the cases are filled with state laws that aspire to reap some of the benefits of tariffs by other means. . . .  [The DCC] examin[es] whether the challenged action shifts the costs of regulation onto other states, permitting in-state lawmakers to avoid the costs of their political decisions. If a regulation unambiguously discriminates in its effect, it almost always is invalid per se.

Id. (citations and quotation marks omitted)

Imposing burdens solely on interstate commerce is precisely what New York’s tax on opioids – combined with the no-pass-through provision limited to New York – did:

[When the statute’s] provisions are given their clearest meaning, the Dormant Commerce Clause violation is clear.  An opioid manufacturer based in Maine that wished to pass on the surcharge it paid on New York transactions by selling opioids at a markup to a pharmacy in New Mexico could face a million-dollar penalty from New York State.  While the statute may not in explicit terms seek to regulate interstate commerce, that it does so nonetheless by its practical effect and design” is abundantly clear.

Id. at *17 (citation and quotation marks omitted).  That’s the regulatory part.  If, however, the statute were construed not to apply to interstate commerce so as to avoid the Scylla of extraterritoriality, it falls directly into the Charybdis of discrimination:

If the [New York statutory] pass-through prohibition applies only to in-state purchasers, New York would clearly reap some of the benefits of tariffs by other means.  New York opioid customers would be protected from any price increases in their purchases, and New York would receive a source of funding subsidized by the out-of-state purchasers of opioids. . . .  [O]ut-of-state drug purchasers, with no representation in New York’s legislature or executive, would bear the cost of New York’s policy program.  This shifting of burdens and benefits is antithetical to the idea of intra-national free trade and demonstrates why the Dormant Commerce Cause exists, i.e., to prohibit discrimination as to “any part of the stream of commerce − from wholesaler to retailer to consumer.

Id. (citations and quotation marks omitted).

There were a lot of other issues that the court in Healthcare Distribution had to plow through between page *3 and *16, but they were all ultimately invalid procedural roadblocks thrown up by New York in order to protect the unconstitutional windfall it was attempting to confer upon itself (and its citizens) at the expense of the rest of the country – justiciability, the Tax Injunction Act, tax comity, abstention, ripeness, and standing.  If any of those interest you, be our guest.  We’re satisfied with the unconstitutionality of state attempts to tax interstate commerce in prescription drugs.

Over the past few weeks, our loyal readers have descended into “The Lows” and then climbed to “The Highs” with us as we reviewed the 10 worst and 10 best cases of 2018.

If you found yourself wanting more information on these cases and their impact – perhaps with a side of CLE credit – we’re pleased to announce that five of your bloggers (Bexis, Eric Alexander, Steven Boranian, Steve McConnell, and Rachel Weil) will be presenting a free 90-minute webinar on “The good, the bad and the ugly: The best and worst drug/medical device decisions of 2018” on Wednesday, January 16 at 12 p.m. EST.

This webinar is presumptively approved for 1.5 general CLE credit in California, Illinois, New Jersey, Pennsylvania, Texas and West Virginia. For lawyers licensed in New York, this course is eligible for 1.5 credit under New York’s Approved Jurisdiction Policy.

The program is free and open to anyone interested in tuning in, but you do have to sign up, which you can do here.

We last reviewed the case law on predictive coding (also called “technology assisted review” (“TAR”)), about 2 ½ years ago.  Back then, we concluded:

The case law has exploded.  Where only a handful of cases existed back then [2012], now we find dozens.  Substantively, we’re happy to report that courts don’t seem to have anything bad to say about using computers to undertake relevance review for documents subject to production in litigation.

How are things now?

“[I]t is ‘black letter law’ that courts will permit a producing party to utilize TAR” when “there was never an agreement to utilize a different search methodology.”  Entrata, Inc. v. Yardi Systems, Inc., 2018 WL 5470454, at *7 (D. Utah Oct. 29, 2018).  Predictive coding continues to enjoy judicial approval:

The predictive coding process required a significant amount of attorney time at the outset to devise and implement, and required the creation of an agreed to strict set of procedural and process safeguards, due to its relative novelty.  Despite the initial “front loaded” investment of time required, the predictive coding system provided a unique way to, in part, realistically manage the immense amount of information needed to be produced and reviewed in this MDL.  The predictive coding system, although not perfect or fully realized, nonetheless, provided an innovative efficiency to the discovery process when compared to the existing, prevailing methods of review.

In re Actos (Pioglitazone) Products Liability Litigation, 274 F. Supp.3d 485, 499 (W.D. La. 2017).  See Story v. Fiat Chrysler Automotive, 2018 WL 5307230, at *3 (Mag. N.D. Ind. Oct. 26, 2018) (“The Court encourages counsel . . . to consider that key word searches or technology assisted review are appropriate and useful ways to narrow the volume of an otherwise overly-broad request”); Duffy v. Lawrence Memorial Hospital, 2017 WL 1277808, at *3 (Mag. D. Kan. March 31, 2017) (“Technology-assisted review can (and does) yield more accurate results than exhaustive manual review, with much lower effort.”) (citations and quotation marks omitted); FCA US LLC v. Cummins, Inc., 2017 WL 2806896, at *1 (E.D. Mich. March 28, 2017) (“Applying TAR to the universe of electronic material before any keyword search reduces the universe of electronic material is the preferred method.”); Davine v. Golub Corp., 2017 WL 549151, at *1 (Mag. D. Mass. Feb. 8, 2017) (“Defendants are entitled to rely on their predictive coding model for purposes of identifying relevant responsive documents”); In re Bair Hugger Forced Air Warming Products Liability Litigation, 2016 WL 3702959, at *1 (D. Minn. July 8, 2016) (MDL order providing for predictive coding). Cf. Dremak v. Urban Outfitters, Inc., 2018 WL 1441834, at *8 (Cal. App. March 23, 2018) (unreported) (finding predictive coding “reasonable and necessary to the litigation” and taxing the cost to unsuccessful plaintiffs), review denied (Cal. July 11, 2018).

Another thing that appears settled is the right of the party producing electronically stored information (“ESI”) to choose the means by which it conducts electronic discovery.  Several cases have addressed this issue, with the result being that neither side can force the other to use this technology.  In a follow-up to a case mentioned in our prior post, the Tax Court in Dynamo Holdings Ltd. Partnership v. CIR, 2016 WL 4204067 (T.C. July 13, 2016), overruled an IRS objection to a taxpayer’s predictive coding-based production that amounted to the IRS seeking a redo.  In response to the IRS’s demand for a 95% recall rate (returning 95% of all relevant documents) – that had a precision rate of only 3% (97% of the total documents produced would be irrelevant) − the court chastised the IRS for perpetuating two “myths” about discovery:  First, that human review is “perfect,” or at least the “gold standard.”  Id. at*5. It isn’t:

[H]uman review is far from perfect. . . .  [I]f two sets of human reviewers review the same set of documents to identify what is responsive, research shows that those reviewers will disagree with each on more than half of the responsiveness claims.

Id. (citations omitted).  The second “myth” is that a “perfect response” to discovery is necessary.  Id.  It isn’t, nor is it even possible:

[O]ur Rules do not require a perfect response. . . .  Like the Tax Court Rules, the Federal Rule of Civil Procedure 26(g) only requires a party to make a “reasonable inquiry” when making discovery responses.  The fact that a responding party uses predictive coding to respond to a request for production does not change the standard for measuring the completeness of the response.

Id. at *5-6 (citations omitted).  Thus, predictive coding, even if imperfect for all the reasons stated by the IRS, was nonetheless proper discovery.  “Petitioners made a reasonable inquiry in responding to the Commissioner’s discovery demands when they used predictive coding to produce any documents that the algorithm determined was responsive.”  Id. at *6.

The converse was true in In re Viagra (Sildenafil Citrate) Products Liability Litigation, 2016 WL 7336411, at *1 (Mag. N.D. Cal. Oct. 14, 2016).  The MDL plaintiffs attempted to force the defendant to conduct ediscovery by predictive coding rather than the defendant’s chosen search term-based methodology. That was a no go, as the rules did not give the requesting party the right to dictate the producing party’s search methods:

[N]o court has ordered a party to engage in TAR and/or predictive coding over the objection of the party.  The few courts that have considered this issue have all declined to compel predictive coding. . . .  [T]he responding party is the one best situated to decide how to search for and produce ESI responsive to discovery requests.  The responding party can use the search method of its choice.

Id. (citations and quotation marks omitted).  Accord Rockford v. Mallinckrodt ARD, Inc., 326 F.R.D. 489, 493 (N.D. Ill. 2018) (“This Court will not micromanage the litigation and force TAR onto the parties.”); T.D.P. v. City of Oakland, 2017 WL 3026925, at *4-5 (Mag. N.D. Cal. July 17, 2017) (rejecting plaintiff’s effort to force defendant to use predictive coding); Hyles v. New York City, 2016 WL 4077114, at *2-3 (Mag. S.D.N.Y. Aug. 1, 2016) (even though “TAR is cheaper, more efficient and superior to keyword searching” “it is not up to the Court, or the requesting party . . ., to force . . . the responding party to use TAR when it prefers to use” something else).

What can happen down the road, post-production, if the requestor still thinks that the producer’s methods are inadequate?   Requestors must have “specific examples of deficiencies” before seeking discovery into an opponent’s TAR process.  Entrata, Inc. v. Yardi Systems, Inc., 2018 WL 3055755, at *3 (Mag. D. Utah June 20, 2018), objections dismissed, 2018 WL 5470454 (D. Utah Oct. 29, 2018).

One way of getting a handle on the recall rate “is to randomly sample the null set.”  Rockford, 326 F.R.D. at 493.  This so-called “null set” represents the universe of electronic documents reviewed and found non-responsive.

Conducting a random sample of the null set is a part of the TAR process.  The purpose of randomly sampling the null set after a TAR review is to validate the process and provide reasonable assurance that the production is complete.  Validation and quality assurance are fundamental principles to ESI production.  The process provides the reasonable inquiry supporting the certification under Rule 26(g) [of reasonable completeness].

Id. at 494 (citations omitted). “[A] random sample of the null set provides validation and quality assurance of the document production.”  Id.  See Winfield v. City of New York, 2017 WL 5664852, at *11 (Mag. S.D.N.Y. Nov. 27, 2017) (“Plaintiffs have presented sufficient evidence to justify their request for sample sets of non-privileged documents” even though “neither this Court nor Plaintiffs have identified anything in the TAR process itself that is inherently defective”).

Finally, for the technophobes among us, we’ll close this post with a soothing quote from one of the more recent cases:

The Court pauses here for a moment to calm down litigators less familiar with ESI. (You know who you are.) In life, there are many things to be scared of, including, but not limited to, spiders, sharks, and clowns – definitely clowns, even Fizbo.  ESI is not something to be scared of. The same is true for all the terms and jargon related to ESI. . . . The underlying principles governing discovery do not change just because ESI is involved.  So don’t freak out.

Rockford, 326 F.R.D. at 492 n.2 (citations and quotation marks omitted).

This isn’t the first time, but the Blog has a problem with its reporting on cases decided in the ongoing “Opioid” litigation.  As lawyers, our first obligations are to our clients, and in this instance some of our clients in the opioid litigation don’t want us talking about their cases.  So when that happens, we don’t.

That doesn’t mean that we aren’t looking for ways to help, but it can’t be through reporting on what went down in individual cases.  We can do concepts, however. That’s why we have been particularly intrigued with one aspect of the recent opioid decision in Floyd v Feygin, 2018 WL 6528728 (N.Y. Sup. Dec. 6, 2018), that dismissed the plaintiff’s “negligence” claim.  The defendant (not a RS client) in Floyd successfully argued the following:

In support of its motion, [defendant] argues that plaintiff’s negligence cause of action against it is, in effect, an attempt to privately enforce the Controlled Substances Act. [Defendant] asserts that plaintiff lacks the authority to regulate its conduct in this regard or to enforce the Controlled Substances Act since Congress has committed enforcement of the Controlled Substances Act exclusively to the Attorney General and the Department of Justice.

Id. at *5. Thus the defendant argued that this “negligence” claim was preempted.  Id.

As we discussed earlier, the Floyd court agreed that the Controlled Substances Act (“CSA”) could not be used in this way, because there isn’t any private right of action:

It has been held that pursuant to its plain terms, the [CSA] is a statute enforceable only by the Attorney General and, by delegation, the Department of Justice.  Based on the regulatory structure of the [CSA] federal courts have uniformly held that the [CSA] does not create a private right of action.  The manufacture, distribution, and dispensing of opioids is comprehensively regulated by the [CSA], and the DEA is the primary federal agency responsible for the enforcement of the Controlled Substances Act.

Id. (citing: Schneller v Crozer Chester Medical Center, 387 F. Appx. 289, 293 (3d Cir. 2010); Smith v Hickenlooper, 164 F. Supp.3d 1286, 1290 (D. Colo. 2016), aff’d, 859 F3d 865 (10th Cir. 2017); McCallister v Purdue Pharma L.P., 164 F. Supp.2d 783, 793 (S.D.W. Va. 2001); DEA, Practitioner’s Manual, at 4 (2006)) (citations and quotation marks omitted).  Further, the New York equivalent of the CSA likewise carried with it no private right of enforcement. Id. at *6 (discussing 10 N.Y.C.R.R. Part 80).

Because the CSA provided no means of private enforcement, its requirements could not create novel common-law duties that did not otherwise exist.

[P]ursuant to both the [CSA] and New York State regulations, [defendant’s] sole duty with respect to its manufacture and distribution of opioids was to collect and record data, and make reports to federal and state agencies.  Although these regulations define circumstances that trigger reports to federal or state agencies, there is nothing within these regulations that requires a manufacturer to stop or restrict its distribution of opioids. . . .  Thus, [defendant] had no duty to control the prescribing, dispensing, and use of its [drug].

Id. at *6 (citations omitted).

The rationale in Floyd – that a federal statute’s lack of a private cause of action preempts a common-law plaintiff from using its requirements to create a “duty” – is what Buckman Co. v. Plaintiffs Legal Committee, 531 U.S. 341 (2001), is all about.  Curiously, Buckman is nowhere cited in Floyd, although Buckman’s implied preemption principles are applicable to all similarly situated federal statutes.

Which raises the question – how much can we “Buckmanize” the CSA, replacing the FDA with the equivalent federal authority of the “Attorney General?  Buckman, of course depended on statutory designation of the FDA as the sole enforcer of the FDCA. So we looked through the CSA looking for similar designations. We found:

21 U.S.C. §811(a), which governs scheduling of the substances to be “controlled”:

The Attorney General shall apply the provisions of this subchapter to the controlled substances listed in the schedules . . . and to any other drug or other substance added to such schedules under this subchapter.

21 U.S.C. §821, which governs enforcement generally:

The Attorney General is authorized to promulgate rules and regulations and to charge reasonable fees relating to the registration and control of the manufacture, distribution, and dispensing of controlled substances and to listed chemicals.

21 U.S.C. §822(a), which governs registration of manufacturers and distributors generally:

(1) Every person who manufactures or distributes any controlled substance . . . shall obtain annually a registration issued by the Attorney General in accordance with the rules and regulations promulgated by him.

(2) Every person who dispenses, or who proposes to dispense, any controlled substance, shall obtain from the Attorney General a registration issued in accordance with the rules and regulations promulgated by him.

21 U.S.C. §823(a-b), which governs requirements imposed through registration

The Attorney General shall register an applicant to manufacture controlled substances in schedule I or II if he determines that such registration is consistent with the public interest. . . .

The Attorney General shall register an applicant to distribute a controlled substance in schedule I or II unless he determines that the issuance of such registration is inconsistent with the public interest.

Section 823 is particularly interesting because it also permits the federal government (through the attorney general) to register “practitioners” – that is, physicians – and contains a limited anti-preemption savings clause:

Notwithstanding section 903 of this title, nothing in this paragraph shall be construed to preempt any State law that . . . requires a qualifying practitioner to comply with additional requirements relating to . . . reporting requirements.

21 U.S.C. §823(l). Section 823 thus is further support for Buckman-type preemption arguments by manufacturers and distributors, since Congress only preserved state-law reporting requirements imposed on “practitioners,” but does not allow state law to impose reporting requirements on manufacturers or distributors of controlled substances.

21 U.S.C. §825(a), regarding labeling:

It shall be unlawful to distribute a controlled substance in a commercial container unless such container, when and as required by regulations of the Attorney General, bears a label . . . containing an identifying symbol for such substance in accordance with such regulations.

21 U.S.C. §825(c), regarding drug warnings:

The Secretary shall prescribe regulations under section 353(b) of this title which shall provide that the label . . . contain[s] a clear, concise warning that it is a crime to transfer the drug to any person other than the patient.

The “secretary” refers to the HHS and thus the FDA.  See 21 U.S.C. §802(24). Section 825(c) thus seamlessly integrates federal enforcement authority.  Go directly to Buckman; do not pass “Go.”

21 U.S.C. §826(a, b), governing amounts of controlled substances it is legal to make:

The Attorney General shall determine the total quantity and establish production quotas for each basic class of controlled substance. . . .

. . .[U]upon application therefor by a registered manufacturer, the Attorney General shall fix a manufacturing quota for the basic classes of controlled substances . . . that the manufacturer seeks to produce.

21 U.S.C. §827(b, d-e), dealing with reporting requirements:

Every inventory or other record required . . . shall be in accordance with, and contain such relevant information as may be required by, regulations of the Attorney General, . . . and [] shall be kept and be available . . . for inspection and copying by officers or employees of the United States authorized by the Attorney General.

Every manufacturer registered . . . shall, at such time or times and in such form as the Attorney General may require, make periodic reports to the Attorney General of every sale, delivery or other disposal by him of any controlled substance, and each distributor shall make such reports with respect to narcotic controlled substances. . . .

[E]ach manufacturer registered . . . shall, with respect to narcotic and nonnarcotic controlled substances manufactured by it, make such reports to the Attorney General, and maintain such records, as the Attorney General may require. . . .  The Attorney General shall administer the requirements of this subsection. . . .

That’s a lot of “attorney general” references, thus further demonstrating that Floyd reached the correct result.  It seems very clear from the statute that reports about controlled substances are not the kind of thing that states have traditionally required or administered.

21 U.S.C. §830(a-b), imposing record-keeping and additional reporting requirements:

(1) Each regulated person who engages in a regulated transaction . . . shall keep a record of the transaction for two years after the date of the transaction.  (2) . . . Such record shall be available for inspection and copying by the Attorney General.  (3) . . . The Attorney General shall specify by regulation the types of documents and other evidence . . . for purposes of this paragraph.

(1) Each regulated person shall report to the Attorney General, in such form and manner as the Attorney General shall prescribe by regulation–(A) any regulated transaction involving an extraordinary quantity of a listed chemical, an uncommon method of payment or delivery, or any other circumstance that the regulated person believes may indicate that the listed chemical will be used in violation of this subchapter. . . . A regulated person may not complete a transaction . . . unless the transaction is approved by the Attorney General. The Attorney General shall make available to regulated persons guidance documents describing transactions and circumstances for which reports are required. . . .

Again, all reports are to the federal government, about information specified by the federal government, and subject to the approval of the federal government.

21 U.S.C. §843, concerning fraudulent reporting:

It shall be unlawful for any person knowingly or intentionally . . . to furnish false or fraudulent material information in, or omit any material information from, any application, report, record, or other document required to be made. . . .

In addition to any penalty provided in this section, the Attorney General is authorized to commence a civil action. . . .

Who can sue over fraudulent reporting to the Attorney General? The federal government.

21 U.S.C. §871(b), regarding the role of the Attorney General generally:

The Attorney General may promulgate and enforce any rules, regulations, and procedures which he may deem necessary and appropriate for the efficient execution of his functions under this subchapter.

21 U.S.C. §880(b), concerning inspections of regulated facilities:

[T]he Attorney General is authorized, in accordance with this section, to enter controlled premises and to conduct administrative inspections thereof. . . .  [E]ntries and inspections shall be carried out through officers or employees (hereinafter referred to as “inspectors”) designated by the Attorney General.

21 U.S.C. §882(c), providing for state enforcement against internet pharmacies:

[T]he State may bring a civil action on behalf of such residents in a district court of the United States with appropriate jurisdiction. . . .  No private right of action is created under this subsection.

In this rare instance where the CSA authorizes independent state enforcement, the statute specifically prohibits a private right of action.

21 U.S.C. §902, preserving the FDCA:

Nothing in this chapter, except [additional reporting requirements to the Attorney General, and prescription restrictions] shall be construed as in any way affecting, modifying, repealing, or superseding the provisions of the Federal Food, Drug, and Cosmetic Act.

Nothing in the CSA impairs the basis for Buckman preemption.

21 U.S.C. §903, concerning “field” preemption:

No provision of this subchapter shall be construed as indicating an intent on the part of the Congress to occupy the field . . ., to the exclusion of any State law on the same subject matter which would otherwise be within the authority of the State, unless there is a positive conflict between that provision of this subchapter and that State law so that the two cannot consistently stand together.

States can enforce their own laws against illegal drugs.  Notably, this savings clause expressly limits its scope to “field” preemption, and does not address other types of implied preemption, such as Buckman or “purposes and objectives” preemption.  To avoid preemption, the “subject matter” of the state law must be “otherwise within the authority of the State,” which would seem to preclude ad hoc imposition of reporting requirements on interstate commerce.  No private right of action is provided.

Probably the provision of the CSA that is most analogous to Buckman’s 21 U.S.C. §337(a) is §871(b) concerning the Attorney General’s general authority, but as we’ve detailed above, a lot of other CSA provisions reject private enforcement or specify federal enforcement.  Thus, as did the court in Floyd, we think that the structure of the CSA fully supports a preemption rationale, similar to Buckman, that because there is no private right of CSA action, would be private enforcers of the CSA – and in particular its reporting requirements to the federal government – are preempted from doing so under state law.

Ending the year on a high note is one thing that the Blog tries to do – with the top ten drug/device product liability decisions of the year.  Occasionally, a court will do something that ruins the party, with an eleventh-hour awful decision (the infamous Bausch v. Stryker Corp., 630 F.3d 546 (7th Cir. 2010), was decided two days before Christmas), but barring that, we’re looking for nothing but pleasant news in going through our list of the best decisions of 2018.  And make no mistake about it, there’s plenty to celebrate this year.

Before we continue, let us explain one thing.  Our list is limited to cases involving drugs or medical devices. Every year has cases that significantly impact our sandbox, even though they’re not within the drug/device arena to which we’re devoted.  Thus, we’ll tip our hat to Forman v. Henkin, 93 N.E.3d 882 (N.Y. 2018) (discussed here), which resoundingly reaffirmed defendants’ rights to complete discovery of plaintiffs’ social media.  Another hat tip goes to Encompass Insurance Co. v. Stone Mansion Restaurant, Inc., 902 F.3d 147 (3d Cir. 2018) (discussed here), with a full-throated endorsement of removal before service in the first squarely on-point appellate decision.  Finally, we like DeLeon v. BNSF Railway Co., 426 P.3d 1 (Mont. 2018) (discussed here), which answered in the negative a consent-to-jurisdiction question the United States Supreme Court had left open in 2017.

But as to those cases, we’re like movie reviewers going to off-Broadway plays.  We can find plenty to like, but it’s not what we specialize in.  It’s time for us to start giving the old two thumbs up to some really good decisions.  Let’s start the show; the envelopes, please:

  1. McNair v. Johnson & Johnson, 818 S.E.2d 852 (W. Va. 2018).  After the T.H. (2017-1A) and Rafferty (2018-1) decisions, we were fearful that the innovator liability tide might be turning against litigation sanity and towards the “everybody should recover for anything” philosophy of our opponents.  Then, along came the West Virginia Supreme Court of Appeals in McNair with a victory for traditional product liability concepts, and also demonstrating this isn’t the same court it used to be.  Indeed, McNair reads like a decision by the California Supreme Court, that is, by Justice Traynor fifty years ago, before the craziness set in.  Like Traynor’s work, the McNair opinion is all about chain-of-sale, manufacturer-responsibility, and cost-spreading.  Indeed, West Virginia’s former off-the-deep-end approach to the learned intermediary rule in Karl (2007-1), haunted the plaintiff here, since the statute that legislatively overruled Karl was also phrased in terms of the liability of drug “manufacturers,” and was cited as such in McNair.  Moreover, McNair threw the plaintiffs’ “let’s all evade preemption” arguments back in their face, following the Supreme Court’s example, in the generic preemption cases, of refusing to “distort” the law to reach a certain result.  And you know what?  We keep an innovator liability scorecard (that is, both good and bad results), and there hasn’t been a single decision since McNair adopting innovator liability.  McNair broke a worrisome trend, concerning an even more worrisome theory of liability, and for that reason, it’s number one on our good list in 2018.  We marveled at McNair, here.
  2. Gustavsen v. Alcon Laboratories, Inc., 903 F.3d 1 (1st Cir. 2018).  Gustavsen gets our nod as the best preemption decision of 2018.  It was about eye drop dosage, but where Gustavsen will be felt is in design defect claims.  The First Circuit (often not our favorite court) framed the preemption argument straight-forwardly – was the product change that the plaintiffs (lots of plaintiffs; this was one of those bogus no-injury class action strike suits) sought a “major change” that required FDA pre-approval to implement?  If so, then the Mensing independence principle mandated preemption, because it was impossible to square the necessity for agency pre-approval with the common law’s demand for immediate action.  But what is a “major change”?  Gustavsen extensively reviewed the FDA regulatory scheme, and concluded that changes to the dosage of a drug (here, eye drops) were major changes.  But the decision relies on the same FDA material we’ve looked to, and that classifies as “major” anything that significantly impacts the safety or effectiveness of an FDA-regulated product.  Think about that.  A change that doesn’t significantly affect product safety won’t change the outcome and can’t be an alternative design.  The bright-line approach in Gustavsen, means “ha det” to pretty much all design defect claims, including those involving 510(k) medical devices.  Sometimes really bad cases make really good law, and Gustavsen is a prime example.  We gloried in Gustavsen here.
  3. In re Accutane Litigation, 191 A.3d 560 (N.J. 2018).  We remember a time, not so long ago, when the New Jersey Supreme Court had never affirmed the exclusion of an expert witness under a Daubert-like rationale – indeed New Jersey law didn’t even resemble Daubert, which is one reason it attracted litigation tourists from all over the country seeking places at mass tort feeding troughs spread around various New Jersey vicinages.  In many ways, the long-running Accutane litigation was for years the epitome of what was wrong with the state’s mass tort system.  Not now.  In Accutane, the court found “little difference” between federal Daubert standards and those New Jersey follows, at least after Accutane.  A terrible intermediate appellate Accutane decision (2017-6) letting everything in under a “flexible” standard and less-than abuse of discretion appellate review was reversed on practically every point, with the result that some 2,000 accumulated junk Accutane cases are finally dismissed as scientifically baseless – a determination made years ago by federal courts under Daubert.  Following the discussion in Accutane, defendants should have no problem relying on the Reference Manual on Scientific Evidence in New Jersey state cases.  Whether or not it will be called “Daubert” going forward, New Jersey now requires “rigorous” gatekeeping for expert witness testimony, including “proper” applications of “methodology,” skepticism of animal studies and “lower forms of [scientific] evidence.”  Because of unpleasant experiences, we’re particularly gratified by the court’s criticism and limitation on Bradford-Hill opinions.  Hello Daubert, and goodbye to Accutane litigation.  We adulated Accutane here.
  4. Conklin v. Medtronic, Inc., ___ P.3d ___, 2018 WL 6613311 (Ariz. Dec. 18, 2018).  In this late-breaking addition to our list, Arizona’s highest court unanimously gave the boot to the notion of a preemption-busting state-law “warning” claim for failure to report adverse events to the FDA – and thus consigns the Ninth Circuit’s abysmal contrary prediction of Arizona law in Stengel (2013-2) to the dustbin of history.  Relying heavily on the court’s recent adoption of the learned intermediary rule in Watts (2016+1), Conklin’s rejection of failure to report as a tort duty was simple, and sweet:  a manufacturer of a prescription medical product (Conklin’s rationale applies to both drugs and devices) is only obligated to provide adequate warnings to a plaintiff’s prescribing/treating physician.  The FDA isn’t a learned intermediary, therefore the plaintiff’s postulated duty does not exist.  Nor could there be causation, because the FDA is not obligated to do anything with the adverse reports it does receive.  Since there is no underlying state-law duty, plaintiffs were really attempting to enforce the FDA’s reporting requirements, which they can’t do under Buckman preemption and the corresponding absence of any private right to enforce the FDCA.  Another nice tidbit about Conklin is it being (we believe) the first state high court to recognize the abolition of the presumption against preemption in express preemption cases.  Conklin might have ranked even higher, but failure-to-report claims are something of a sideshow and we don’t know of any such claim that’s ever supported an adverse verdict.  We congratulated Conklin here.
  5. This entry is from the non-RS side of the blog.  Dolin v. GlaxoSmithKline LLC, 901 F.3d 803 (7th Cir. 2018).  One of the big casualties of Levine (2009-1) was the preemption arguments offered by manufacturers of selective serotonin reuptake inhibitors (“SSRIs”) on the risk of suicidality.  We always thought they had a pretty good FDA regulatory history to support preemption, but with one exception (Dobbs (2011+8)), SSRI preemption arguments simply got massacred in the wake of Levine.  That is until Dolin.  The Seventh Circuit focused on 2007 as the critical time period of FDA review.  2007 was when the FDA told the defendant to stop using a product specific warning in favor of class-wide labeling that the plaintiff argued was inadequate. Thereafter, the defendant tried several times unsuccessfully to get the FDA to change its mind.  That was clear evidence as a matter of law, the unanimous Seventh Circuit concluded, whether or not the pro-plaintiff standards of Fosamax (2017-1) were applied.  The preemption facts were that good:  “The FDA said no, repeatedly.”  Not only was Dolin a big win for preemption, but the Seventh Circuit’s ruling also precluded possible certification of another lurking issue – innovator liability – to the Illinois Supreme Court, something that would have been risky for innovators.  We don’t know how many similar SSRI cases are still out there, but Dolin validates defendants’ preemption arguments.  We (non-RS) discussed Dolin here.
  6. In re Zimmer, NexGen Knee Implant Products Liability Litigation, 884 F.3d 746 (7th Cir. 2018).  The decision In the district court, we described as “the best Wisconsin law decision we have ever seen.”  Well, the Seventh Circuit affirmance was just as good, and much more binding.  The learned intermediary rule is now controlling law in all Wisconsin federal courts (a couple of which had refused to apply it).  Not only that, the rule applies to medical devices, and there is no heeding presumption to help plaintiffs with Wisconsin warning claims.  Sounding like us, the Seventh Circuit declared that appellate authority supported the learned intermediary rule in 48 states (according to our headcount, everywhere save South Dakota and Vermont).  As a consequence of the learned intermediary rule applying, summary judgment was affirmed on the basis the surgeon using the device already knew about the risks in question and did not rely on (or even read) the purportedly inadequate warnings.  Adding another state firmly to the roster of learned intermediary adopters is a big deal, and NexGen does so in style.  We applauded Nexgen here.
  7. In re Lipitor (Atorvastatin Calcium) Marketing, Sales Practices & Products Liability Litigation (No II) MDL 2502, 892 F.3d 624 (4th Cir. 2018).  Aside from preemption, the only defense that can wipe out an entire MDL in one fell swoop is Daubert.  That was the fate of Lipitor MDL claims regarding diabetes risk in 2016, in a decision that was our #7 best of that year.  This year, the Fourth Circuit affirmed in all respects.  Co-morbidity is no substitute for causation.  We particularly enjoyed the circuit court’s refusal to allow reliance on non-statistically relevant epidemiology in a supposed Bradford-Hill based causation opinion.  After affirming the analysis that mowed down all of the plaintiffs’ experts, the court rejected plaintiffs’ attempt to prove causation without any experts as “farcical” and “steeped in speculation.”  Finally, plaintiffs’ procedural argument that summary judgment was only appropriate against the bellwether plaintiffs, and that all other cases should have been remanded.  The Fourth Circuit refused to countenance what would have amounted to use of one-way classes in MDLs.  Since all the MDL plaintiffs would have relied on the MDL plaintiffs’ experts, had they presented admissible opinions, the converse is also true – all of the MDL plaintiffs’ cases required expert testimony, and thus fail when their designated experts are excluded.  We lauded Lipitor here.
  8. Shuker v. Smith & Nephew, PLC, 885 F.3d 760 (3d Cir. 2018).  One recurrent, vexing issue in PMA preemption is what happens when the plaintiff claims injury from a multi-component device construct and not all of the components are pre-market approved.  In what we think is a decision of first impression at the appellate level, Shuker got the right answer – claims attacking the PMA components are preempted, while claims solely involving other, 510(k) cleared, components are not.  Plaintiff’s argument that the presence of any non-PMA components precludes any preemption got nowhere.  Off-label use of medical devices in this way is contemplated by the FDCA and does not displace preemption.  Shuker also made a favorable personal jurisdictional ruling, affirming dismissal of claims against another defendant because, after BMS (2017+1), the “stream of commerce” theory of personal jurisdiction was no longer viable.  That ended decades of Third Circuit avoidance of the issue.  So, with two good rulings on two of the Blog’s crown jewels, preemption and personal jurisdiction, why isn’t Shuker higher on our list?  Well, Shuker also dropped a gratuitous footnote essentially refusing to follow the Supreme Court’s abolition of the presumption against preemption.  We complained about that here.  If not a crown jewel, the presumption is, at minimum, one of our pet peeves.  So, eighth it is.  We shook Shuker out, here.
  9. In re DePuy Orthopaedics, Inc., Pinnacle Hip Implant Products Liability Litigation, 888 F.3d 753 (5th Cir. 2018).  If this case seems eerily familiar, it is – it also showed up as #5 on our worst of 2018 decisions.  Yes, the same case.  Why?  Well, for all of the problems with the court’s legal rulings that we discussed last week, the fact remains that the court reversed a half-billion dollar verdict, and it did so due to “the district court’s evidentiary errors and [Ps’ counsel Mark] Lanier’s deception.”  The bellwether trial had been “an almost uninterrupted flow of unduly prejudicial and irrelevant information to the jury.”  How prejudicial?  How about tying the defendant to Saddam Hussein?  Totally irrelevant, concluded the Fifth Circuit.  How about allegations of corporate racism?  A “spectacle,” held the court.  And more − suicide, cancer, tobacco industry, transvaginal mesh – a true smorgasbord of prejudicial error.  But that, alone, wouldn’t have overcome the later questionable legal rulings. Calling out Mark Lanier by name for “unequivocal decept[ion]” of the jury, did, however.  It seems that Lanier told the jury that certain experts were testifying for free when they were really being “bought” (the court’s word) by way of about $65,000 of contributions to their favored charities.  Because these “falsehoods marred plaintiffs’ victory,” a new trial was ordered.  Like Longfellow’s little girl, when Pinnacle Hip was bad, it was horrid, but when it was good it was very, very good indeed.  Thus, this decision has the singular distinction of placing on both our top and bottom ten lists simultaneously.  We parsed Pinnacle Hip here.
  10. In re Bard IVC Filters Products Liability Litigation, 289 F. Supp.3d 1045 (D. Ariz. 2018).  Prior to the Pelvic Mesh MDL, a defendant’s medical device clearance by the FDA under its 510(k) substantial equivalence clearance process was routinely admitted.  But arguments falsely equating preemption and admissibility have since gained traction, most notably in Eghnayem (2017-2) and- Cisson (2016-2), where the court affirmed as not an abuse of discretion the exclusion of all mention of the FDA in a medical device design defect case as more prejudicial than probative.  We think that such a result is unsupportable and gives juries an inaccurate perception of how medical devices are designed, but at this point, we’re fighting that battle all over again, with recent appellate authority against us.  Thus, it was a very big deal when we received a published victory on this issue in Bard IVC, expressly disagreeing with those decisions and their no preemption = inadmissibility rationale, and holding instead that 510(k) clearance is relevant both to risk/utility design defect and punitive damages.  Admission of FDA evidence would be less likely to confuse the jury than its exclusion.  For going against an erroneous tide, we consider Bard IVC to be the most significant drug/device trial court decision of 2018.  We evaluated Bard IVC here.

So there they are, our top ten picks for the best drug/medical device decisions of 2018.  As good as these are, though, they’re not enough.  2018 was a very good year for the defense, and the Blog was pleased to present far more than ten noteworthy outcomes where the right side of the “v.” prevailed.  As is our custom, here are then next ten good decisions of 2018 that couldn’t quite make our top ten.

Honorable Mentions: (11) You know it’s been a good year for the defense when an MDL-wide Daubert win only scores an honorable mention, but we’re not complaining about In re Mirena IUS Levonorgestrel-Related Products Liability Litigation, 2018 WL 5276431 (S.D.N.Y. Oct. 24, 2018) (discussed here and here).  (12) McDaniel v. Upsher-Smith Laboratories, Inc., 893 F.3d 941 (6th Cir. 2018), was the first appellate decision that shot down the latest plaintiff preemption dodge, that a manufacturer somehow owes a duty not only to make FDA-mandated medication guides available, but to ensure that third parties (pharmacists or physicians) actually give them to patients (discussed here).  (13) There isn’t much generic drug product liability litigation anymore, but Guilbeau v. Pfizer, Inc., 880 F.3d 304 (7th Cir. 2018), was pretty darn good in deep-sixing one of the few arguments those plaintiffs had left, holding that a “reference listed drug” designation didn’t stop (or even slow down) preemption.  We put it on our A-list here.  (14) The New Jersey Supreme Court issued a second In re Accutane Litigation, decision − 194 A.3d 503 (N.J. 2018) – on choice of law, holding that by coming to New Jersey, mass tort litigation tourists must accept New Jersey law, including the presumption of adequacy of FDA-approved warnings.  Bye-bye to several hundred more plaintiffs.  Not ranked higher because all choice of law decisions can cut both ways (discussed here).  (15) A common plaintiff tactic to drive up a case’s nuisance value, unreasonable discovery demands, was sanctioned (in the sense of imposing costs) in Vallejo v. Amgen, Inc., 903 F.3d 733 (10th Cir. 2018), and the sanctions were affirmed.  It’s about time (discussed here).  (16) Although unusual, sometimes good things do happen in Pennsylvania, such as the unanimous affirmance of Buckman preemption of off-label promotion claims in Caltagirone v. Cephalon, Inc., 190 A.3d 596 (Pa. Super. 2018) (discussed here).  (17) Russell v. Johnson & Johnson, ___ S.W.3d ___, 2018 WL 5851101 (Ky. App. Nov. 9, 2018), affirmed IDE preemption, holding that contrary pre-Riegel precedent from the Kentucky Supreme Court was no longer valid (discussed here).  (18) Tutwiler v. Sandoz, Inc., 726 F. Appx. 753719024 (11th Cir. 2018), rejected several common plaintiff arguments on the learned intermediary rule under Alabama law (discussed here).  (19) Byrd v. Janssen Pharmaceuticals, Inc., 333 F. Supp.3d 111 (N.D.N.Y. 2018), is the first decision recognizing that, because the FDA must approve any attempt to discuss off-label use in the label, Mensing impossibility preemption bars challenges to the adequacy of warnings about off-label risks.  Would have ranked higher had the opinion been better written (discussed here).  (20) Henson v. Dep’t of Health & Human Services, 892 F.3d 868 (7th Cir. 2018).  Plaintiffs often resort to Freedom of Information Act (“FOIA”) requests to supplement ordinary discovery, and in Henson, denial of FOIA requests to the FDA was affirmed as to trade secrets, internal FDA memoranda, and medical/personnel records.  Such appellate decisions are uncommon (discussed here).

Right in the middle of our compiling this post, Conklin was decided, and we had to reorder things.  Thus, we also thought we’d acknowledge the following near misses:  In re Testosterone Replacement Therapy Products Liability Litigation Coordinated Pretrial Proceedings, 2018 WL 3303269 (N.D. Ill. July 5, 2018) (here); Blackburn v. Shire U.S., Inc., 2018 WL 2159927 (N.D. Ala. May 10, 2018) (here); Gravitt v. Mentor Worldwide, LLC, 289 F. Supp.3d 877 (N.D. Ill. 2018) (here).

Looking back over our previous best and worst decisions, first and foremost we’re watching the Supreme Court’s consideration of Fosamax implied preemption in Merck v. Albrecht (see here).  This appeal is from In re Fosamax (Alendronate Sodium) Products Liability Litigation, 852 F.3d 268 (3d Cir. 2017), which we ranked as the worst case in all of 2017.  We’d like to see it flip and become our best case of 2019.  If we’re really lucky, Albrecht might even repair some of the damage done by Levine (2009-1), the worst case in the entire history of the Blog.

We’ve already discussed how the sixth worst case of 2017 was reversed in Accutane and how Stengel v. Medtronic, our second worst case of 2013, was just gutted by Conklin.  The rest of the bottom ten of 2017, unfortunately look like they’re over with.  Of our 2017 top ten and honorable mentions, only Utts (2017+6) still faces a pending appeal.  Fingers crossed.  We’ve already discussed the Lipitor affirmance of a 2017 honorable mention decision.  Inge v. McClelland, the New Mexico wrongful conduct rule case (2017+15), was affirmed in a non-precedential opinion.  Inge v. McClelland, 725 F. Appx. 634 (10th Cir. 2018).  Everything else appears final.

We quickly reviewed our earlier top/bottom ten lists.  Nothing else looks changed, good or bad.

Going forward, other than Albrecht, we’re watching with interest to see if the Supreme Court reaches the cy pres issues in Frank v. Gaos, or whether they go off on a standing tangent, leaving cy pres for some other day (here).  A petition for certiorari was just filed in the Dolin case (2018+5) discussed above. Another pending appeal that we’re watching is Burningham v. Wright Medical (here), in which the Utah Supreme Court will decide if comment k applies to medical devices in the same way it applies to drugs.  Oral argument is January 9, 2019.  There is also another Pinnacle Hip appeal in the works, since the MDL judge proceeded with another consolidated trial in the face of a request from the Fifth Circuit that the practice cease.  There are also several pending Pennsylvania appeals (non-drug/device, for the moment) concerning general personal jurisdiction by consent/registration to do business, one of which has been accepted for en banc review.

Finally, on the administrative front, the FDA’s 2013 attempt to blow up generic implied preemption was officially interred not long ago.  FDA initiatives to make the post-SMDA 510(k) process even more rigorous, and to formalize regulations for the “de novo” review process for Class II devices might eventually pay preemption-related dividends, but right now are only in the early stages of the administrative process.  We have not seen any recent FDA movement on truthful off-label promotion and the First Amendment.  Legislatively, Congress seems essentially paralyzed on anything having to do with prescription medical products and product liability, and with a divided Congress in 2019, that seems unlikely to change any time soon.

At this point 2018 is essentially in the books.  Happy New Year to all of our readers, and – for those of you on the right side of the “v.,” good luck and much success in 2019.  For those of you on the other side, check this out.

This guest post is from long-time friend of the blog Bill Childs, from Bowman & Brooke, who also wishes to thank Elizabeth Haley for research assistance.  It’s a reworking of a piece on bogus scholarly literature that Bill previously published here.  We thought it was both good and relevant enough that we approached Bill with a request to re-run it as a guest post on the Blog, and he graciously accepted.  As always, our guest bloggers are 100% responsible for the content of their posts (and here that disclaimer also extends to B&B and its clients), and deserve all the credit (and any blame).


The Daubert court, in interpreting Rule 702 of the Federal Rules of Evidence, laid out various non-exclusive criteria for consideration in evaluating proposed scientific evidence, one of them peer review. As the Court put it:  “The fact of publication (or lack thereof) in a peer reviewed journal…will be a relevant, though not dispositive, consideration in assessing the scientific validity of a particular technique or methodology on which an opinion is premised.”  Daubert v. Merrell Dow Pharms., 509 U.S. 579, 594 (1993).  Peer review, or the absence thereof, was mentioned repeatedly by the New Jersey Supreme Court in endorsing Daubert in the recent decision in In re: Accutane Litigation, 191 A.3d 560, 586, 592, 594 (N.J. 2018).  Among other things, the Court noted that the plaintiffs’ expert had not submitted “his ideas…for peer review or publication,” considering that failure to be a strike against his methodology. Id. at 572.

Compared to other Daubert factors (or those described in the subsequent comments to Rule 702), the presence or absence of peer review may seem more binary than other factors − i.e., easier for a court to evaluate − it’s either there or it’s not, it seems.  Not so, either in the traditional sense of peer review or the changing world of things that now get called peer review.  Given this perceived simplicity, though, it frequently gets less attention than it deserves.  Litigants should think about peer review as being more complex than it appears, and in some specific contexts, additional exploration − whether through discovery into your adversaries’ experts, or early investigation of your own potential experts − may make sense.

Daubert vs. Predator

One fascinating consequence of this consideration of peer review in the Daubert context is the potential for experts publishing litigation-related work in what are called “predatory journals” (sometimes also called “vanity publications).”  See Kouassi v. W. Illinois Univ., 2015 WL 2406947, at *10-11 (C.D. Ill. May 19, 2015); Jeffrey Beall, “Predatory Publishing Is Just One of the Consequences of Gold Open Access,” 26 Learned Pub’g 79-84 (2013); John Bohannon, “Who’s Afraid of Peer Review?” 342 Science 60-65 (Oct 4, 2013).

Predatory journals, like the eponymous Predator in the 1987 film and its 2018 reboot, camouflage themselves.  They make themselves look not like the Central American jungle background, but like legitimate medical or scientific journals.  Their publishers’ websites generally look like legitimate publishers’ websites (if sloppy at times), their PDFs look like “real articles,” and their submission process might even look normal.  They’ll even claim to have peer review and editorial boards and all the rest of what you expect from journals.  Like the Predator, they even try to manipulate their editorial voices to sound like real journals.

These journals are, however, just aping the façades of real journals.  They typically do not have legitimate peer review processes − or possibly any review processes at all.  Frequently, if an author pays the exorbitant fees, the submitted article will get published.

Myriad examples exist revealing such journals as frauds.  My favorite is probably the publication of a case report of “uromysitisis” an entirely fictional condition − first referenced in Seinfeld as a condition from which Jerry claims to suffer after being arrested for public urination − by the purported journal Urology & Nephrology Open Access Journal.  The author of the intentionally nonsensical article − not a urologist, nor a medical doctor at all − wrote about his experience here. After that article’s exposure as an obvious fake, and something that even the most casual of reviewers should have rejected, the article was removed, but the “journal” is still up and publishing on the MedCrave site, described, a bit awkwardly, as “an internationally peer-reviewed open access journal with a strong motto to promote information regarding the improvements and advances in the fields of urology, nephrology and research.”  A few years earlier, a computer scientist published an article consisting solely of the phrase “Get me off your [obscenity] mailing list,” with related graphs, repeated for eight pages.  That journal remains in existence as well.

Such journals are largely set up to entrap new (and naïve) scholars who are under tremendous pressure to publish for promotion and tenure purposes − but they also can provide an opportunity for dubious expert witnesses to get something published they can cite as “peer reviewed,” especially as courts more and more often note the presence or absence of peer review.  It isn’t news to many litigation experts that having peer review for some of their more outlandish assertions can increase the odds of their testimony being admitted.  If an expert in fact has published in a predatory journal (and it can be shown that the expert knew or should have known about that fact), that fact should count against the admissibility of the testimony.

Given the camouflage, it is fortunate that there are resources and strategies that can help identify such publications.  Retraction Watch, published by the Center for Scientific Integrity and headed by science writer Adam Marcus and physician and writer Ivan Oransky (full disclosure: Ivan and I are friends, based in large part on our shared love for power pop like Fountains of Wayne and western Massachusetts bands like Gentle Hen.  He should not be blamed for my Predator references) while not focused solely (or even largely) on predatory journals, is an accessible look at the world of retractions “as a window into the scientific process.”  They keep an eye out for interesting developments in the world of predatory journals, and scientific publications generally, and their coverage is what made me suspicious when, in one of my cases, an adversary’s expert’s article was published by a MedCrave journal (home to the Seinfeld article).  Retraction Watch’s coverage of that article led to what I assume will be the only time in my career I had the chance to ask a Ph.D./M.D. if he was familiar with Seinfeld and if the show is, in fact, fiction, based on him publishing − and in fact being listed as an editor of − another MedCrave journal.

There is also a list of suspected predatory journals archived at Beall’s List.  The appearance of a journal on that list is not conclusive evidence that it is predatory, but it is enough to raise questions.  The removal of a journal from the Directory of Open Access Journals for “editorial misconduct” or “not adhering to best practices” (see list, here) is another giveaway.  The Loyola Law School’s “Journal Evaluation Tool” can also provide a useful rubric, accessible to non-scientifically-trained lawyers, for evaluating whether a journal is likely legitimate or not. And your own experts can likely provide feedback to you about journals.

Most experts will not have published in predatory journals.  But it is still worth the time to explore the question, especially about pivotal articles on which the experts are relying − whether the expert is your adversary’s or your own.  Even if the publication offer was innocently accepted (i.e., even if the author did not realize she was publishing in a predatory journal), the lack of rigor in evaluating the article by the publisher should at a minimum eliminate any weight given to the peer review factor. And if an author has intentionally published in such a journal, that should be the equivalent of an intentionally false statement in a C.V.

Not All Peer Review Is the Same

Of course, these relatively new faux journals are not the only way experts get published.  Consider the most traditional form of peer review, where editors of a journal have outside reviewers, usually with their identities screened from the authors, evaluate the quality and originality of the work, confirming that the methodologies presented appear legitimate and that the conclusions reached are reasonable based on what’s described.  Given that those goals line up nicely with the goals of a Daubert analysis, it is sensible indeed for a court to look at that as a potential indicator of reliability − indeed, that’s why peer review is a factor in the first place.

But even if a proffered expert testifies to having followed a methodology that matches something in a peer-reviewed publication, it is often worth at least a few deposition questions about the review process and a line in your subpoena duces tecum requesting copies of any materials the author has received relating to the review, or to attempt some third party discovery on the journals in question − though some courts may limit or refuse that discovery.  See, e.g., In re Bextra & Celebrex Mktg. Sales Practices & Prod. Liab. Litig., 2008 WL 859207 (D. Mass. March 31, 2008) (granting protective order for non-party medical journal publisher, expressing concerns about a chilling effect).  The propriety of allowing such discovery is beyond the scope of this article, but I addressed it in more detail in The Overlapping Magisteria of Law and Science: When Litigation and Science Collide, 85 Neb. L. Rev. 643 (2007).

If you get peer review notes, it’s possible you’ll find that a reviewer recommended the removal of a conclusion that the expert is now presenting, or that the reviewer warned against a particular inference from what is in the article.  Making it even easier, some journals, traditional and, more often, “open access,” are now posting their reviewers’ comments online.  Even if you do not find anything relevant, most experts will readily concede that peer review reflects at most an “approval” of the overall approach and is not a guarantee of correctness as to conclusions.  And sometimes you’ll be able to establish that the study in question was based on flawed data or that the work done for litigation did not, in fact, use the same methodology as that in the publication.  See, e.g., In re Mirena IUS Levonorgestrel-Related Prods. Liab. Litig., ___ F. Supp.3d ___, 2018 WL 5276431, at *11-13, *28, *34, 37-38, *50-51 (S.D.N.Y. Oct. 24, 2018) (rejecting expert’s reliance on “repudiated” open access journal article by author that did not disclose retention as a plaintiff’s litigation expert); In re Viagra Prods. Liab. Litig., 658 F. Supp. 2d 936, 945 (D. Minn. 2009) (reversing an initial denial of defendants’ Daubert motion after learning of flaws in underlying data and processing, noting that “Peer review and publication mean little if a study is not based on accurate underlying data.”); Palazzolo v. Hoffman La Roche, Inc., No. A-3789-07T3, 2010 WL 363834, at *5 (N.J. Super. App. Div. Feb. 3, 2010) (finding no abuse of discretion in excluding an expert’s conclusion based on conclusion that the expert did not in fact use the methodology claimed to have used in the underlying peer-reviewed study).

Sometimes, even in a more traditional context, the peer review that was performed was not what was likely pictured by the Daubert court, particularly when the work at issue is outside the so-called “hard sciences.”  In a publicized example, the review of a history-oriented book about the lead and vinyl chloride industries, authored by frequent plaintiffs’ experts and published by the University of California, involved reviewers known to − and in some cases recommended by − at least one of the authors . See 85 Neb. L.R. at 660-63 (describing this situation; original book website was removed).  Whether or not that review was adequate for the academic purpose, it was materially different from, say, the reviewers of a double-blind clinical trial, and the facts surrounding it seem plainly relevant to how much weight a court should give it under Rule 702 and Daubert.  Without that discovery, the court may well not have learned about what “peer review” meant in that context.

Consider also the scenario where an expert says that their methodology has gone through peer review but the article has not yet been published.  Again, it may be worth pursuing more details, especially if the expert seems likely to cite to that review in defending their position.  If it has not yet been accepted for publication, consider requesting a copy of the comments the expert received from the reviewers. If those comments are provided, they may be helpful; if their production is refused, the fact of that review should be rejected as a basis for admissibility.

What To Watch Out For

Fundamentally, the important thing is to look through your and your adversaries’ experts’ C.V.s with care, especially as to articles that are directly on point with the issue you’re addressing.  It is not enough to think about what the articles say, and it also is not enough to think to yourself, “Well, that sounds like a legitimate journal.”  Look at the publishers’ site; look for hints in the article itself; and do some searches.  Ask a few questions of the expert about author fees and what the peer review entailed and throw in a document request to see if there is something worth exploring further.  And if you are dealing with a situation with what you think is a predatory journal, be ready to teach a judge about what that means; as of this writing, no court has referenced “predatory journals” in a reported Daubert decision.

When we’re trying to decide whether to see a movie, one place we turn for advice is the Rotten Tomatoes website. We visited the site recently to see what those purveyors of overripe fruit had to say about “The Favourite,” which attracted our attention with an interesting trailer that preceded “The Crimes of Grindelwald.”  Hint – the reviewers liked it much better than the movie we had seen – maybe we will, too.  While on the site we noticed a feature, “24 Worst Movie Remakes,” which listed the most negative “Tomatometer” scores subtracting the remake from the original. Of the 24, thankfully, we’ve only seen one.

We thought about that when compiling this year’s bottom ten worst drug/medical device product liability decisions of 2018.  We sure don’t want any remakes of these!  Of course, here, there’s little positive to subtract from to start with.  These clunkers start out with such awful scores that it’s hard to imagine much worse – we wish.   Unfortunately, this marks our twelfth year of doing this, so we’ve previously reviewed 110 (actually 111 – 2017 had two worsts) other similarly bad decisions. So yes, actually we can imagine worse.

Anyway, let’s stop imagining things and get on with the dreary job of discussing the ten worst prescription medical product liability litigation decisions of 2018. If any of these fiascos were yours, feel free to cry on our shoulders.  It’s not like it’s never happened to us (for instance, see 2013-2), so we know what it can be like.  But don’t worry, we’ll all feel less depressed next week, when we say “Happy New Year” with our best decisions – no rotten tomatoes needed.

Here goes nothing:

  1. Rafferty v. Merck & Co., 92 N.E.3d 1205 (Mass. 2018). Last year’s extra number one worst case (2017-1A) was a state supreme court’s recognition of the liability uber alles theory of innovator liability. This purely result-oriented claim is contrary to the most fundamental precepts of product liability, since liability does not follow a defendant’s profit from the sale of the product as a “cost of doing business,” but rather follows the path of least legal resistance – given that product liability claims against generics are preempted − from the generic drug to the original innovator drugmaker that actually lost profits due to generic competition. Because innovator liability is so potentially dangerous – imposing liability for the 90% of the market that is generic on the 10% that is branded – a similar decision by the Massachusetts Supreme Court takes home the number one spot again this year. Innovator liability is sought only because of preemption of direct generic claims, and Rafferty quite frankly admitted that the cause of action it created was a preemption dodge. Unlike last year’s case, Rafferty concealed its tort radicalism with a fig leaf. Instead of negligence, Rafferty required “recklessness” on the part of the defendant as a prerequisite to innovator liability. While trumpeting this scienter requirement as “circumscribing” liability “for public policy reasons,” Rafferty actually gutted the limitation by choosing a dumbed-down definition of recklessness that really amounts to little more than negligence. We don’t doubt that plaintiffs will easily plead recklessness, and thereby jack up the prices of other drugs, since by definition the defendant didn’t sell the one that hurt anyone. After Rafferty, generic drug plaintiffs get a unique second bite at the apple that no other product liability plaintiffs enjoy when the real manufacturer is for any reason judgment-proof. For deliberately tossing aside decades of product liability principles so that generic drug users can have someone – the wrong someone – to sue, Rafferty is the biggest judicial rotten tomato of 2018. We ripped Rafferty here.
  2. Campbell v. Boston Scientific Corp., 882 F.3d 70 (4th Cir. 2018). We hate consolidated, multi-plaintiff trials almost as much as we hate innovator liability. Consolidated trials are such a favorite tool of settlement-bludgeoning MDL judges, that we have adopted the hard line of “just say no” to MDLs – and even more to having any trials at all in any MDL. The law used to reject multi-plaintiff consolidations as virtually inherent abuses of judicial discretion, but MDLs have the deleterious effect of sacrificing legal principle to settlement. Last year it was the Eghnayem (2017-2) decision from the Eleventh Circuit allowing 4 plaintiffs to be tried together. This year’s decision means that now two Courts of Appeals have sanctioned this deliberately prejudicial procedure. If anything, Campbell is worse, because the jury, plainly unable to keep the cases straight, returned almost identical verdicts for each of four plaintiffs. Identical verdicts used to be a Due Process red line, but not in Campbell. After Campbell, it looks like MDLs may freely sacrifice the American legal tradition of separate trials to whatever judges want to do in the name of efficiency. Because that’s simply not right, Campbell is the second worst decision of 2018 and a poster child for why the “pre-trial” restriction in the MDL statute should be enforced according to its terms. We criticized Campbell here.
  3. Hammons v. Ethicon, Inc., 190 A.3d 1248 (Pa. Super. 2018). Pennsylvania is well on its way towards being a national personal jurisdiction outlier. Hammons is one reason why. To start with, Hammons was the only decision we were aware of in the entire country – and that includes other Pennsylvania decisions that should have been binding – holding that a defendant, not the plaintiff asserting jurisdiction, bore the burden of proof of proving lack of personal jurisdiction. Beyond that, the Court sandbagged the defendant, resting its jurisdictional decision on “evidence” that had not even been considered when the issue was decided at the trial court level (later introduced for non-jurisdictional purposes at trial). Finally, Hammons based personal jurisdiction on contacts with Pennsylvania that were not relevant to the theories of liability that the non-resident, litigation tourist plaintiff was pursuing, making a mockery of any sort of causal relationship in the context of the post-BMS “arising from”/”relating to” test for specific personal jurisdiction. Thus, Pennsylvania now has the dubious distinction of having unconstitutionally expansive appellate decisions on both general and specific jurisdiction. “Grasping” and “exorbitant” jurisdictional theories, you’ve got a friend in Pennsylvania. We hammered Hammons here.
  4. Godelia v. Doe 1, 881 F.3d 1309 (11th Cir. 2018). Godelia continued the Eleventh Circuit’s string of lousy decisions from last year (see 2017-2, 2017-7, and 2017-8). The court reversed a finding of PMA preemption where the plaintiff offered only a warning letter that did not even involve the type of problem that the plaintiff’s decedent encountered. Godelia’s excuse was that the warning letter was “not all-inclusive.” Godelia completes the Eleventh Circuit’s descent from one of the strongest to one of the weakest TwIqbal jurisdictions, allowing an allegation of “violation of the federal regulations noted above” (previously described as “not all-inclusive”) to plead a parallel manufacturing defect claim. Vague regulatory allegations were okay, as long as the FDA said the regulations were violated in the warning letter, and so were purely speculative causation allegations. As long as the plaintiff alleged a manufacturing defect, warranty, or misrepresentation claim, PMA preemption was defeated, even if the claim wasn’t actually based on the alleged FDCA violations. Warranty claims, which usually fail Florida’s privity requirements where prescription medical products are involved, survived because of alleged oral statements by sales representatives. That the device required a prescription does not require dismissal for lack of privity where direct contacts with a sales rep are alleged. And, of course, the court doubles down on Mink’s (2017-8) deliberate misapplication of Florida negligence per se legislative intent requirement. We didn’t post on Godelia before, but it was indeed God awful.
  5. In re DePuy Orthopaedics, Inc., Pinnacle Hip Implant Products Liability Litigation, 888 F.3d 753 (5th Cir. 2018). Wait a minute. The court reversed a half-billion dollar verdict and roasted plaintiffs’ counsel for improper conduct. How could this case be on the bottom ten list? The answer lies in the court’s grant of only a new trial. The Pinnacle Hip decision also made a number of legal rulings that range from questionable to atrocious. The most serious error the court made was refusing to apply established Texas law that comment k precludes strict liability across the board. Pinnacle Hip ignored – really ignored − a half dozen prior decisions (including one of its own) on this issue. Even if there wasn’t any precedent (which there was) expanding state-law liability where the state courts have not is not the job of a federal court sitting in diversity. Another flaw was allowing plaintiffs to offer a different product as a purported alternative design. The Fifth Circuit used to be good on this, but allowing an alternative design claim that the product shouldn’t have been composed of what the FDA allowed it to be made of really pushes the boundary. Two oddball liability theories also received a green light: “nonmanufacturer seller” and “negligent undertaking.” We also don’t like allowing “objective” evidence (that is, paid expert testimony on what a “reasonable” physician would have done) to substitute for what the plaintiff’s prescriber actually did in evaluating warning causation under the learned intermediary rule. Finally, the court also rejected a preemption argument – but frankly it wasn’t the right preemption argument, so we don’t take off many points there. We pummeled Pinnacle Hip for these deficiencies here.
  6. In re Smith & Nephew Birmingham Hip Resurfacing (BHR) Hip Implant Products Liability Litigation, 300 F. Supp.3d 732 (D. Md. 2018). The worst trial court decision of 2018 came from an MDL. No surprise. One of our objections to MDLs is their bending over backwards to let questionable claims continue, in pursuit of settlement leverage against defendants. BHR was propelled higher (lower?) on our list since the contortions in this case were aimed against PMA preemption, perhaps our favorite defense (and certainly the strongest). Anything that so much as contained a whiff of an FDCA violation became an unpreempted parallel claim, so what began as a Supreme Court waiver comment has just about swallowed Riegel’s preemptive rule in this MDL. BHR also distinguished Buckman into almost nothingness – if an FDCA violation is not a “critical” element of a negligence per se claim, it becomes hard to see what would be. PMA products have largely avoided MDLs since Sprint Fidelis, but BHR appears determined to be exception. We bashed BHR here.
  7. Stange v. Janssen Pharmaceuticals, Inc., 179 A.3d 45 (Pa. Super. 2018). Our second bottom ten case from Pennsylvania affirmed a half-million dollar Philadelphia verdict for gynecomastia – a “condition” that another court this year found not “serious” as a matter of law. No wonder Philadelphia attracts litigation tourists from all over the country. The litigation tourist in Stange was from Wisconsin. That’s enough to knock Stange a notch or two, since out of state decisions screwing up other states’ laws rarely get much attention from the affected states’ judges. The chief Pennsylvania law goof in Stange was on expert witnesses – applying a “novel” science limitation that was eliminated by the Pennsylvania Supreme Court in an asbestos case called Betz. On top of this, the court reversed the Philadelphia mass tort judge’s holding that New Jersey law applied to punitive damages, since the defendant was headquartered there, and that’s where all corporate conduct would have happened. New Jersey has a statute that precludes punitive damages for FDA-approved products. The court sent the case back to the trial court for a comparison with Wisconsin law even though (as we discussed here) a prior ruling by the same court – binding under Pennsylvania appellate practice – had already determined that the law of the corporate domicile applies to punitive damages choice of law issues. As that prior decision involved a Pennsylvania-domiciled defendant, Stange really applied the choice of law rule, “heads, plaintiffs win; tails, defendants lose.” That alone would qualify for this list. Stange is indeed strange, as we pointed out, here.
  8. A.F. v. Sorin Group USA, Inc., ___ F. Supp.3d ___, 2018 WL 4680022 (S.D.N.Y. Sept. 28, 2018). A.F. comes in at number eight on our list because it ignores the role of federal courts sitting in diversity jurisdiction, and also because (unlike the decisions that follow) it’s published. As is so often the case with federal decisions embracing novel liability theories, A.F. is a preemption case. The theory that survived dismissal – failure to report adverse events to the FDA – is an ill-concealed preemption dodge. But that didn’t stop A.F. What makes the case worse is (as we discussed here) New York courts have affirmatively rejected tort liability based on a variety of claims based upon alleged failures to make reports to governmental agencies. Failure to report claims are not a matter of first impression in New York. Not only that, but the same FDA-related claim had been rejected by another New York court not long before. A.F. is on our list because it ignores both Erie principles and existing New York law. We explained the awfulness of A.F. here.
  9. Sumpter v. Allergan, Inc., 2018 WL 4335519 (E.D. Mo. Sept. 11, 2018). Sumpter is another court-assisted preemption dodge case. This time the dodge is “manufacturing defect.” In what way did the device fail to meet specifications? The complaint doesn’t say, and the court doesn’t care – as evidenced by its reliance on one of the most widely reviled (and not just by us) PMA preemption decisions, Hofts (2009-5). As long as the product departed from its “intended result,” that’s enough. A plaintiff can “extrapolate” from the alleged injuries. That amounts to res ipsa loquitur, and a rather loose sort, at that. But Sumpter’s worse than that. Res ipsa loquitur only allows inference of a “defect”; Sumpter uniquely allowed the plaintiff to get by with an inference of FDCA violation from the mere fact of an alleged device failure. Missouri has become the opposite of the “Show Me State.” Sumpter is more like “don’t need to show me anything.” We discussed how Sumpter is septic here and here.
  10. Tryan v. Ulthera, Inc., 2018 WL 3955980 (E.D. Cal. Aug. 17, 2018). The Tryan decision got preemption wrong on so many levels that we didn’t know what to make of it. The claims themselves were farcical. Plaintiffs weren’t even injured. They claimed only that they were told the device was “approved” rather than “cleared,” and wouldn’t have purchased it had they known it was a 510(k) device. Thus, Tryan was through and through a private claim for violations of the FDCA. Without the FDCA, the plaintiffs’ claimed distinctions would never have existed. Thus, it’s a private action totally dependent on the FDCA and barred by that act’s prohibition against private enforcement. But Tryan got sidetracked on express preemption – not an issue – and never recovered. Tryan reasoned as if there were a “parallel claim” exception, but no such thing exists in implied preemption. “Mirroring” the FDCA only makes implied preemption stronger, since the FDCA precludes private enforcement (except for an exception only relevant to food). The result? A class action alleging that “FDA approved” and “FDA cleared” are so different that otherwise uninjured consumers should receive damages. Tryan is thus a travesty, as we explained here.

That’s our bottom ten, and that’s quite enough as far as we’re concerned. Time for a shower, to feel clean again, and for some egg nog to get back in the holiday spirit. Our spirit will be back next week, as we award legal Oscars instead of rotten tomatoes when we review the ten best drug/device decisions of 2018.  Finally, we’ve heard it said that there is no war on Christmas, but there is a war on Thanksgiving – and Christmas has won.  In that spirit, we wish everyone Happy Holidays.


We’ve always hated the Ninth Circuit’s decision in Stengel v. Medtronic Inc., 704 F.3d 1224 (9th Cir. 2013) (applying Arizona law), holding that allegations of failure to provide adverse event reports (“AERs”) to the FDA created a viable, and unpreempted, state law claim.  Now our #2 worst case of 2013 is effectively gone.  Kaput.

Today the Arizona Supreme Court in Conklin v. Medtronic, Inc., No CV-17-0322-PR, slip op. (Ariz. Dec. 18, 2018), expressly – and unanimously − disagreed with Stengel’s misinterpretation of Arizona product liability warning law, and since the Arizona Supreme Court is “supreme” on state-law issues, Stengel is invalid.  We’ll get directly to the good stuff:

We disagree with Stengel. . . .  In Stengel, the Ninth Circuit held that the [Medical Device Amendments] did not expressly or impliedly preempt the plaintiffs’ Arizona common law failure-to-warn claim based on Medtronic’s alleged failure to submit adverse event reports to the FDA.  That holding, however, was based on the unsupported premises that “Arizona law contemplates a warning to a third party such as the FDA” and that, “[u]nder Arizona law, a warning to a third party satisfies a manufacturer’s duty if . . . there is ‘reasonable assurance that the information will reach those whose safety depends on their having it.’”  Neither premise comports with Arizona law. . . .  [E]stablished law does not recognize a claim merely for failing to provide something like adverse event reports (which may not qualify as “warnings” under Arizona law) to a government agency that has no obligation to relay the information to the patient.

Because Stengel incorrectly recited and applied Arizona law, we decline to follow it. . . .  [O]ur case law contemplates that a medical device manufacturer may satisfy its duty to warn consumers by properly warning a third party, such as a learned intermediary.  But the FDA is not a learned intermediary or other relevant third party in that analysis.  And we are not aware of any case that supports the proposition that a manufacturer is independently required under Arizona law to warn a governmental regulatory body.

Slip op. at 10 ¶¶30-31 (citations omitted) (emphasis added, except for final emphasis, which is original).  Put the red flag on Stengel.

So how did the Conklin court reach this exemplary conclusion?  The path begins with an excellent preemption ruling in the trial court, Conklin v. Banner Health, 2015 WL 10688305 (Ariz. Super. Oct. 30, 2015), in a case involving a pre-market approved infusion pump.  That decision held all negligence per se claims (including the failure to report claim at issue here), preempted under Buckman Co. v. Plaintiffs Legal Committee, 531 U.S. 341 (2001).  The intermediate Arizona appellate court reversed preemption of the failure to report claim, following Stengel.  See Conklin v. Medtronic, Inc., 418 P.3d 912, 919 (Ariz. App. 2017).  A further appeal produced today’s opinion.

The Arizona Supreme Court first recognized that, contrary to Stengel, there is no presumption against preemption in express PMA preemption cases.  “[A]lthough federal laws are presumed not to preempt state laws, courts do not invoke that presumption when the federal statute contains an express preemption clause.”  Slip op. at 3 ¶8 (citations and quotation marks omitted).  Interestingly, though, Conklin ultimately based its decision mostly on implied, not express, preemption, id. at 7 ¶21 (finding duty-to-report claim impliedly preempted and not reaching express preemption), so the presumption against preemption didn’t really figure in the result.

Conklin also utilized the “narrow gap” metaphor to describe the interaction of express preemption and implied preemption in PMA device cases:

Read together, these two types of preemption, operating in tandem, have created a “narrow gap” for pleadings.  To make it through, a plaintiff has to sue for conduct that violates a federal requirement (avoiding express preemption), but cannot sue only because the conduct violates that federal requirement (avoiding implied preemption).

Slip op. at 6 ¶18 (citations and quotation marks omitted).

The one aspect of the plaintiff’s duty-to-report claim that Conklin held to be expressly preempted was the assertion that the FDA was obligated to make AERs widely available to the public.  Federal law did not require that, so:

[T]o the extent [plaintiff] argues that the FDA either has or assumed a duty to convey information from adverse event reports to treating physicians, patients, or more broadly public consumers . . ., [that] claim is expressly preempted because it likewise would impose under state law a requirement that is “different from, or in addition to,” any applicable federal requirement.

Id. at 6 ¶20 (regulatory citations omitted).

The broader implied preemption ruling was based, as indicated above, on the absence of any reporting duty owed under state law.  Conklin “assume[d] without deciding” that AERs could be “warnings” under state law, but several “but cf.” citations indicated its disinclination to so rule.  Slip op. at 7 ¶22. The learned intermediary rule, which the court previously adopted in Watts v. Medicis Pharm. Corp., 365 P.3d 944 (Ariz. 2016) (we discussed Watts here and here), proved to be key.  Under Arizona law, prescription medical product warnings need only go to learned intermediaries, and the FDA didn’t qualify as a “learned intermediary.” Arizona’s learned intermediary rule “only extends . . . to prescribing and other health-care providers.”  Slip op. at 8 ¶26. There was never any state-law obligation to warn the FDA:

The FDA is not a health care provider and does not prescribe anything for patients. . . .  Accordingly, even if we assume that adverse event reports may constitute relevant warnings, Arizona law does not permit a manufacturer to satisfy its duty to warn end-user consumers by submitting adverse event reports to the FDA.  And conversely, a manufacturer does not breach its duty to warn end users under Arizona law by failing to submit adverse event reports to the FDA.  [Plaintiff] cites no authority, and we are aware of none, for the proposition that Arizona law requires a manufacturer to warn a federal agency.

Id. at 8 ¶¶26-27 (citations omitted).  The duty to warn in Arizona “has not been extended to require a manufacturer to submit warnings to a governmental regulatory body.”  Id. at 8 ¶28.  Submitting AERs to the FDA does not provide “reasonable assurance” that such information “will reach end users (or end users’ health care providers) because the FDA is not required to publicly release such reports.”  Id. at 9 ¶28 (citations omitted).

Thus, the lack of any Arizona state-law duty to report was fatal to plaintiff’s reporting-based claims.  While Conklin held that such claims don’t exist from the outset, the case was litigated and appealed on preemption grounds.  Thus followed the preemption ruling that plaintiff’s purported reporting claims were purely FDCA-related, and thus preempted under Buckman:

Because only federal law, not state law, imposes a duty on [defendant] to submit adverse event reports to the FDA, [plaintiff’s] failure-to-warn claim is impliedly preempted under 21 U.S.C. § 337(a).  Absent an independent state law duty to submit adverse event reports to the FDA, [plaintiff’s] failure-to-warn claim, at bottom, is an attempt to enforce a federal law requirement.  That claim is impliedly preempted under the MDA.

Slip op. at 9 ¶29 (Buckman citations omitted).

Then came the slicing, dicing, and pureeing of Stengel that we quoted at the outset.  Conklin closed by dispatching the other cases plaintiff relied on, thus performing the additional service of highlighting the questionable validity of Fiore v. Collagen Corp., 930 P.2d 477 (Ariz. App. 1996).  See Id. at 11 ¶33 (Fiore “adopted a minority view . . ., and in any event likely does not survive Riegel”).

On this blog, we have declared, until we are blue in the face, that it is a usurpation of state court power for a federal court sitting in diversity to make up new theories of liability under state law.  That’s what Stengel did, and thankfully, in Conklin, the state high court thereby bypassed by the Ninth Circuit’s novel ruling had both the opportunity and inclination to call out the trespassing federal court.  Once again:

We disagree with Stengel. . . . Stengel incorrectly recited and applied Arizona law.

Slip op. at 10 ¶¶31-32.

Mic drop.

We opposed the FDA’s ill-advised 2013 proposal to revamp the process for changing generic drug labeling from the outset.  We had legal objections – that an FDA regulation could not alter the statutory “sameness” requirement imposed on generic labeling.  We had practical objections – that the change was a sop to the plaintiffs’ bar, intended primarily to reduce generic preemption in civil litigation, rather than to pursue any legitimate FDA objective.  And we had procedural objections – that FDA had secretly colluded with the plaintiffs’ bar in coming up with the proposal.

We were hardly the only ones to object, and the proposal was repeatedly deferred.  Once the proposal was postponed until after the 2016 election, we pronounced it dead.  Either the Ds would win, and approach the preemption issue directly by changing the makeup of the Supreme Court (Mensing/Bartlett were 5-4 decisions), or the Rs would win and the FDA would stop carrying water for the plaintiffs’ bar.

The latter happened, with [fill in the blank] consequences for the country and the world, but with predictable consequences for the FDA’s generic labeling proposal.  After a reasonable interval, the FDA formally put an end to the lingering 2013 proposal last week.  Here is the bottom line, from a December 13 statement posted on the FDA’s website:

In November 2013, the FDA proposed a rule . . ., which, if finalized, would have allowed generic drug makers to independently – meaning, without prior FDA review and approval – update and promptly distribute new safety information in drug labels. This is something that currently only branded drug makers can do.

This rule, if implemented, would have allowed generic manufacturers to independently update their drug labels with new information. We heard from manufacturers that they believed this change would have imposed on them significant new . . . liabilities. We heard arguments that the proposed rule could impose new costs on generic manufacturers that might have raised the price of generic drugs. . . . And, among other challenges, the new policy would have resulted in labels for the same drug that varied between different generic manufacturers, for some period of time. This could have led to consumer and provider confusion.

Today, the FDA is withdrawing this proposed rule. . . .

Dick Tracy signing off!  The bottom line is that, after five years, nothing has changed, particularly the scope of and basis for generic preemption.  Nor will there be any change to the FDA’s requirement of agency preapproval to changes to the “highlights” section of branded drug labeling – another potential source of preemption in civil litigation (this isn’t mentioned in the FDA statement, but is discussed in the accompanying Federal Register notice).

This agency coup de grâce is accompanied by the usual regulatory word salad about “ensur[ing] that generic companies continue to engage in an appropriate level of post-market safety surveillance” and “hurdles that – if the rule was implemented – could compromise public health.”  The FDA admitted:

[A]dditional or different warnings [could] temporarily appear in generic drug labeling compared to the brand drug – depending on the availability of information to various manufacturers and the timing of updates.  Such differences, even if temporary, could undermine confidence in generic drugs and their therapeutic equivalence.  We understand that the proposed rule may have also led to confusing, conflicting generic labels that were crowded with redundant safety information. Individual generic manufacturers might have added additional and at times superfluous information to their individual labels to avoid the risk of liability for failure to warn.

This, of course, is a problem hardly limited to generic drugs.  Competing branded drugs have all of these issues, too, which is one reason that the FDA frequently imposes class-wide labeling. There’s a solution for it – called “Expedited Agency Review” (“EAR”) – that would involve the FDA at an earlier stage in all labeling changes to ensure that this kind of chaos from occurring.  EAR was proposed as an alternative to the 2013 proposal, but is not mentioned in the FDA’s statement, probably because it would require “additional resources and help from Congress.”

The FDA’s statement does contain a useful clarification of the responsibility for labeling changes, after a branded “reference listed drug” is withdrawn from the market:

If the brand drug manufacturer has voluntarily withdrawn their marketing application, generics that reference the brand medicine can still be approved and marketed.  But the brand drug manufacturer is no longer responsible for making any necessary label updates that generic applicants can follow.

See also Federal Register Notice.  We add emphasis to this statement, since today is also the first anniversary of T.H. v. Novartis Pharmaceuticals Corp., 407 P.3d 18 (Cal. 2017), a case in which the California Supreme Court failed to recognize this basic fact.

More information on the withdrawal of the FDA’s 2013 generic labeling proposal is available at 83 Fed. Reg. 64299 (FDA Dec. 14, 2018)

The FDA has recently released a proposed rule “to establish requirements for the medical device De Novo classification process” provided in 21 U.S.C. §360c(f)(2).  FDA, “Medical Device De Novo Classification Process,” 83 Fed. Reg. 63127 (Dec. 7, 2018).  This de novo classification option is a relatively recent addition to the FDCA (via the 1997 FDA Modernization Act), and provides:

(ii) In lieu of submitting a report under section 360(k) [a/k/a/ §510(k)] . . ., if a person determines there is no legally marketed device upon which to base a determination of substantial equivalence . . ., a person may submit a request under this clause for the Secretary to classify the device.

(iii) Upon receipt of a request . . ., the Secretary shall classify the device subject to the request under the criteria set forth in subparagraphs (A) through (C) of subsection (a)(1) [meaning Class I, Class II, or Class III] within 120 days.

(iv). . . .

(v) The person submitting the request for classification under this subparagraph may recommend to the Secretary a classification for the device and shall, if recommending classification in class II, include in the request an initial draft proposal for applicable special controls, as described in subsection (a)(1)(B), that are necessary, in conjunction with general controls, to provide reasonable assurance of safety and effectiveness and a description of how the special controls provide such assurance. Any such request shall describe the device and provide detailed information and reasons for the recommended classification.

360c(f)(2)(ii-v) (emphasis added).  The emphasized statutory language establishes the standard to which these “de novo” devices are held: “reasonable assurance of safety and effectiveness.”  That’s important to preemption because of what Medtronic, Inc. v. Lohr, 518 U.S. 470 (1996), had to say over twenty years ago:  “provid[ing] the FDA with a ‘reasonable assurance’ that the device is both safe and effective[,] [d]espite its relatively innocuous phrasing . . ., is a rigorous one.  Manufacturers must submit detailed information regarding the safety and efficacy of their devices, which the FDA then reviews.”  Id. at 477.

Famously, Lohr also declared that the 510(k) substantial equivalence process “[t]he 510(k) process is focused on equivalence, not safety.”  Id. at 493 (emphasis original).  That is certainly cannot the case with de novo devices, since by definition there is no equivalence determination to be made.  Indeed, the FDA “may decline to undertake a classification request submitted under clause (ii) if [it] identifies a legally marketed device that could provide a reasonable basis for review of substantial equivalence.”  21 U.S.C. §§360c(f)(2)(iv).

We’ve expounded at great length in our “Lohr Has Two Shadows” post about how Lohr had been an anachronism ever since it was decided because Congress toughened up the original 510(k) process that Lohr addressed with the Safe Medical Devices Act of 1990.  The SMDA imposed the same “reasonable assurance of safety and effectiveness” standard on substantial equivalence determinations that Lohr had praised as “rigorous in Lohr.  See 21 U.S.C. §§360c(a)(1)(B), 360c(f)(1)(A)(v).

Nevertheless, since Lohr, the judiciary has engaged in a conspiracy to ignore the SMDA and treat all 510(k) medical devices as if they were evaluated under minimalistic standard that the Supreme Court addressed in 1996.  With de novo devices, it’s going to be a lot harder for judges to continue sticking their heads in the sand and chanting “Lohr, Lohr, Lohr” every time a non-PMA device is at issue.

The FDA’s proposed rule makes clear that the de novo process is essentially the SMDA’s “special controls” regime without the fig leaf of substantial equivalence:

The De Novo classification process provides a pathway to ensure the most appropriate classification of a device consistent with the protection of the public health and the statutory scheme for device regulation. This pathway is intended to limit unnecessary expenditure of FDA and industry resources that may occur if devices for which general controls or general and special controls provide a reasonable assurance of safety and effectiveness are subject to a PMA due to a lack of a predicate.

83 Fed. Reg. at 63129 (emphasis added).  No longer can the SMDA standard for clearance of Class II devices be dismissed as involving “equivalence, not safety.”  It’s the same “rigorous” standard that Lohr equated with PMA – only extended to other classes of devices, in particular:

For any class II recommendation, the De Novo request must also provide an initial draft of proposed special controls along with a description of how the special controls provide reasonable assurance of safety and effectiveness.

Id. at 63130.

The proposed regulations firmly establish that de novo devices are to be evaluated for “safety and effectiveness,” not equivalence, in practically every aspect of their FDA review:

  • “If the submitter recommends that the device be classified as class II, FDA proposes that the recommendation must include a draft proposal for applicable special controls, and a description of how those special controls provide reasonable assurance of safety and effectiveness of the device.”  Id. at 63133.
  • “FDA proposes that the De Novo request include reference to any published standard relevant to the safety or effectiveness of the device.”  Id.
  • “For a clinical investigation involving human subjects, FDA proposes to require that a discussion of . . . safety and effectiveness data, adverse reactions and complications, patient discontinuation, patient complaints, device failures . . . and replacements. . . . FDA would use the summary of investigations in assessing safety and effectiveness of the device.”  Id.
  • Valid scientific evidence is evidence . . . from which it can fairly and responsibly be concluded by qualified experts that there is reasonable assurance of the safety and effectiveness of a device under its conditions of use.”  Id. at 63133-34.
  • “The proposed benefit and risk considerations section would expressly require that, “pursuant to the determination of safety and effectiveness section of the regulations, a discussion be included demonstrating that . . . the probable benefit to health from use of the device outweighs any probable injury or illness from such use (i.e., a discussion demonstrating the safety and effectiveness of the device) when the device is used according to its labeling.”  Id. at 63134.
  • “Any safety and effectiveness data to assist FDA in assessing whether the clinical investigation supports that a reasonable assurance of safety and effectiveness exists. FDA would assess reasonable assurance of safety and effectiveness by evaluating the valid scientific evidence submitted to support the De Novo request. FDA would review the data to assess whether the data supports the claims made in the indications for use and demonstrates that the probable benefits of the device outweigh the probable risks.”  Id.
  • “Discussion of data on any adverse reactions to the use of the device . . . or complications related to the use of the device. . . . Frequency data and severity data are particularly useful in safety and effectiveness determinations. FDA would review the rates of complications in clinical investigations in assessing the safety and effectiveness of the device.”  Id.
  • “FDA would need all discontinuation data in order to determine the safety and effectiveness of the device.” Id.
  • “Trends in complaints may point to possible risks posed by the device. FDA would review such trend analyses in assessing the safety and effectiveness of the device.”  Id. at 63135.
  • “In analyzing failures, factors such as location, user application, and repeat component failures may apply. FDA would review such analyses in assessing the safety and effectiveness of the device.”  Id.
  • “Statistical analysis of the results from each clinical investigation. The statistical analysis should specify and discuss all effects. FDA would review such analyses in assessing the safety and effectiveness of the device.”  Id.
  • Any “contraindication, precaution, warning, or other limiting statement relevant to the use of the device. . . . This includes information regarding any special care to be exercised by a practitioner or patient for the safe and effective use of the device.”  Id.
  • “[I]f a De Novo request relies primarily on data from a single investigator at one investigation site, the De Novo request must include a justification showing why these data and other information are sufficient to demonstrate the safety and effectiveness of the device.”  Id.
  • “FDA further proposes to require that a De Novo request include a discussion of the clinical significance of the results, pursuant to the determination of safety and effectiveness.”  Id.
  • “FDA proposes to require that the De Novo request include any other identification, discussion, and analysis of any other data, information, or report relevant to the safety and effectiveness of the device.”  Id.
  • “FDA proposes that the De Novo request must include other information that is necessary for FDA to determine whether general controls or general and special controls provide a reasonable assurance of safety and effectiveness of the device. Examples would include marketing experience outside the United States, medical device reporting (MDR) data . . ., and patient preference information.”  Id. at 63136.
  • “FDA proposes to require the De Novo requester to update its pending De Novo request with new safety and effectiveness information . . . as such information becomes available.”  Id.
  • “FDA proposes that FDA would be able to inspect relevant facilities prior to granting or declining a De Novo request. Such an inspection is intended to assist FDA in determining whether a reasonable assurance of safety and effectiveness can be provided by general or general and special controls.”  Id. at 63137.

Emphasis added in all cases.  Once all of these safety/effectiveness requirements are complied with, the FDA would issue an “administrative order” granting the request and specifying the de novo device’s classification.  Id.  By such orders “FDA would determine the safety and effectiveness of the device using the criteria specified in the determination of safety and effectiveness section of the regulations.”  Id. (citing 21 C.F.R. §§860.289(d), 860.7).

Given the proposed regulations’ emphasis on determinations of device “safety and effectiveness” throughout, as well as its reliance on “special controls” customized to each de novo request, we think that de novo devices should be protected by preemption under Lohr and Riegel v. Medtronic, Inc., 552 U.S. 312 (2008).  We have here device-specific requirements intended “to provide reasonable assurance of safety and effectiveness” of these devices

Oddly, there is no explicit discussion – or even mention – of “preemption” in the FDA’s proposed rule.  So that nobody ends up buying a pig in a poke, we advise that industry-side comments to this proposal should request the FDA to specify explicitly that it believes the de novo process provides preemptive protection under 21 U.S.C. §360k(a), for all products that the FDA concludes have “reasonable assurance of safety and effectiveness” under 21 U.S.C. §360c(f)(2), and in accordance with Lohr‘s analysis .  While, in the past, FDA opinions on preemption haven’t always been helpful, or even intelligible, this time it might be.