There was a time when we posted frequently about attempts to impose liability for injuries allegedly caused by the use of a generic prescription drug. Much of the attention has been directed to trying to pin liability on the company that developed the drug originally, even when the plaintiff took another company’s generic version. When it comes to the liability of the manufacturer of the generic drug that the plaintiff actually took, it seemed like things were pretty settled. The Hatch-Waxman Act requires generics to have labels matching the label of the NDA holder’s drug and does not allow the use of the CBE to go beyond it—cutting off the Levine rationale for limited preemption on warnings claims. Add these to the inability of a generic manufacturer to change the design of its drug, generic manufacturers have generally not been liable absent a failure to timely update its label to match the NDA holder’s label—and even that situation can be preempted in some courts. This combination of dead ends and hard roads has been a main driver for the push against NDA holders and innovators in the dozen years since the Conte abomination. Similarly, although there are still occasional decisions where states’ high courts officially join the ranks of the super-majority, the learned intermediary doctrine has been so widely accepted now that it seems like a long time since this was a hot topic.

Pennsylvania has been one of the strongest states on learned intermediary, so it is bit surprising to see the issue of whether there is an exception to the learned intermediary doctrine under Pennsylvania law as dispositive in a case. This is doubly so when the case involves claims against a generic drug manufacturer. That is what we have in Polt v. Sandoz, Inc., ___ F. Supp. 3d ___, No. CV 16-2362, 2020 WL 2730931 (E.D. Pa. May 26, 2020), where an earlier ruling on a motion to dismiss for preemption left plaintiff with enough of an opening to pursue “extensive discovery” and make the court decide the remaining issues on summary judgment three years later. (Thanks to our friends at Greenberg for passing on the decision.) The basic facts are as follows. The plaintiffs’ mother took defendant’s generic cardiac drug off-label for years and ultimately died from a pulmonary event that was disclosed as a risk in the label. The drug’s label matched the label of the NDA holder’s drug and there was no allegation that it was manufactured incorrectly. The drug, however, had a federal regulatory requirement for a medication guide, which the manufacturer is supposed to provide to pharmacies to be provided, in turn, to patients when they fill a prescription. One of the prescribers, although aware of the relevant pulmonary risk like the other two prescribers, claimed not to know the drug was being prescribed off-label or that it had a medication guide. Plaintiffs’ remaining claims thus turned on whether Pennsylvania law—not federal law—imposed a duty on the generic manufacturer to provide a warning to patients through a medication guide, even though Pennsylvania law otherwise focused the duty to warn inquiry for prescription drugs solely on the learned intermediaries. This ends up being a preemption issue, at least as the court framed it.

We can largely skip over the remaining warnings claims focused on the prescribers—the package insert matched the branded one and the prescribers knew the risk—to get to the purported claim based on warning the patient. If Pennsylvania did not recognize such a claim this circumstance—which would require creating a new exception to the learned intermediary doctrine—then a claim based on failing to comply with a duty would be preempted. (Frankly, the court’s lead-in on the preemption playing field was a bit off, missing the import of Mensing and injecting inapplicable law on parallel claims, but it got there.) Citing the McDaniel case out of the Sixth Circuit (discussed here and lauded as a 2018 honorable mention winner here), the court noted that most decisions on this issue had found preemption under Buckman because the applicable state law usually does not recognize “an underlying state tort law duty” to provide a medication guide for the benefit of patients.

As alluded to above, Pennsylvania’s history of recognizing the learned intermediary doctrine went back a long time—at least to the 1971 decision of the Pennsylvania Supreme Court in the Incollingo case. The court recapped some of that history—but not the more recent decisions—before acknowledging that Pennsylvania had only ever recognized one exception to the learned intermediary doctrine even though other states (or federal courts predicting state law) had made up others. Because plaintiff did not claim any recognized exception applied, we do not have to get into whether the court was right about there being any Pennsylvania exception. Instead, plaintiff urged the creating of a new court-created exception based on certain FDA regulations.

The Court will reject this suggestion for three reasons: (1) it would break new ground and significantly expand liability under Pennsylvania law; (2) even if adopted, it would be inapplicable in this case; and (3) it would be a misuse of negligence per se, which does not operate to create a new duty.

The first reason drew on a combination of the Erie restraint doctrine we tout from time to time—particularly when courts ignore it in creating new state court duties—and Pennsylvania courts’ own “reluctance” to expand tort liability without legislative action. Thus, “the Court will decline the invitation to alter Pennsylvania’s social policy calculus as it relates to the liability of manufacturers engaged in the sale of prescription drugs.”

For the second reason, the court acknowledged that the FDA regulations would need to essentially remove the prescriber from her typical role as learned intermediary. With this drug—unlike certain cases with contraceptives—there was no evidence to suggest the patient could receive it “without a learned intermediary weighing the risks and benefits.” For plaintiffs’ mother, three different physicians had done that in deciding to prescribe her the drug. Thus, without deciding that Pennsylvania would adopt any exception, the urged exception would not apply.

The last reason took the court some time to explain, but is actually fairly straightforward. Calling a claim negligence per se and claiming that a federal duty has been violated does not change what state duties do or not exist. Pennsylvania does not impose a duty to provide medication guides. Rather, it requires warnings for prescription drugs go to the learned intermediaries. Nothing about negligence per se—again, we will refrain from commenting on the larger issue of whether negligence per se based on FDA regulations can impose liability for claims like this—changes what duties exist under state law.

So, it looks like the scattered but continuing attempts to impose liability with this generic versions of this drug have failed again. Along the way, the strict application of the learned intermediary doctrine in Pennsylvania has been re-endorsed and preemption shows its strength. It would be nice if it had not taken an extra three years to get there. It seems to us that the court might have determined the non-existence of the purported claim under Pennsylvania law that plaintiffs needed to avoid preemption without three years of extensive, and presumably costly, discovery. Perhaps future courts facing similar issues will be emboldened to take, well, bold action at the 12(b)(6) stage.

 

Here’s another guest post by friend-of-the-blog, Dick Dean of Tucker Ellis.  Dick is a big fan of primary jurisdiction (if only more courts shared his enthusiasm), and this post deals with some recent Supreme Court precedent that is helpful to such arguments, and that might otherwise have been missed.  As always our guest bloggers deserve 100% of the credit (and any blame) for their ideas.

**********

Timing is everything in life.  On April 29, after I finished reading Steve McConnell’s post playing off of the NFL draft, I started reading Supreme Court cases decided within the last few days.  By chance the first one I looked at was Atlantic Richfield v. Christian, —S.Ct.—, 2020 WL 1906542 (April 20, 2020).  It does not have to do with our sandbox; it is a CERCLA Superfund case.  But when I read it, I immediately thought of primary jurisdiction (the 16th pick in in Steve’s post) though that term is never used in the case.  It is not a classic primary jurisdiction case which involves prudential deference by a court to an agency with specific expertise on an issue before the court.  However, the underlying rationale of the doctrine was implicated in Atlantic Richfield by statutory scheme, since Congress gave the EPA specific powers relevant to the issue before the court (remediation of Superfund sites).  The case was resolved on that statutory basis.

Atlantic Richfield bought the Anaconda Copper Smelter in Montana at a deep discount hoping to turn it into something profitable. That attempt failed miserably.  It turns out that what it had bought was soon to be a Superfund site.  Atlantic agreed to a remediation plan with the EPA which is to continue to 2025. Then a group of 98 Montana landowners sued Atlantic in Montana state court for restoration damages under Montana law. Plaintiff’s remediation demands went substantially beyond what the EPA had ordered, for example, plaintiffs proposed that the land have a maximum contamination level of 15 parts per million of arsenic. The EPA had prescribed 250. There were many other remedies sought beyond what was in the plan Atlantic challenged the lawsuit.

The first issue was whether CERCLA barred this action. The trial court and the Montana Supreme Court said no.  The U.S. Supreme Court agreed.  The Court held that CERCLA deprived state courts of jurisdiction of claims brought under the Act but that it did not “displace state court jurisdiction over claims brought under other sources of law.”  Since the claims for restoration was under state law, it was not displaced.  The second question was whether the landowners needed EPA approval to seek remedial action.  The Superfund law provides that when a remediation plan was underway “no potentially responsible party” could undertake separate remediation unless it was authorized by the EPA.  So this part of the case turned on a determination of whether the land owners were “potentially responsible parties” under the Act.  The Supreme Court found that they were.  In sum, the landowners could sue in state court and were not barred from doing so by the CERCLA remediation but they could not sidestep the EPA.  It was not exclusive jurisdiction because the suit could proceed if the EPA allowed it but it was the equivalent of primary jurisdiction by statute.

In Steve’s post he referenced the use of primary jurisdiction in our sandbox.  He embedded posts discussing a number of food labeling and advertising cases.  While those are clearly the most numerous primary jurisdiction cases implicating the FDA within the last ten years, there are over 500 other primary jurisdiction cases involving the FDA.  (You can’t say much when discussing a 16th round pick).  There have been a number of cases finding primary jurisdiction where the claim was related to or premised on the assumption that a drug was safe and effective.  Not only does the FDA have significant medical and scientific expertise to bring to bear on whether a drug is safe and effective, Weinberger v. Bentex Pharm., Inc., 412 U.S. 645, 654 (1973), it is the only entity that is authorized to make such decisions.  Weinberger v. Hynson, Westcott & Dunning, Inc. 412 U.S. 609, 627 (1973).  So it is not surprising that claims implicating that decision have been dismissed or side-tracked by primary jurisdiction.  National Ethical Pharmaceutical Assn. v. Weinberger, 503 F.2d 1051 (4th Cir. 1974) (attempt to challenge FDA determination of what was or was not a new drug could proceed only after a formal ruling from the FDA); Israel v. Baxter Laboratories, Inc., 466 F.2d 272 (D.C. Cir. 1972) (action by drug manufacturer that competitor and others conspired to keep drug off market by preventing fair consideration required remand and consideration by FDA to determine safety and efficacy).

From approximately 2000 to 2010, plaintiffs’ lawyers often tried to influence or modify the course of FDA generated recalls.  More specifically, they would request that courts change language in a recall already issued to be more expansive.  Courts generally declined those requests since the issue in a recall is “what is the appropriate dissemination of medical information to the consuming public.”  Clark v. Actavis Group hf, 567 F. Supp.2d 711 (D.N.J. 2008).  And since that is a medical issue it is within the province of the FDA, and courts should defer to that expertise.  Those cases are discussed in “Out of Your Jurisdiction: Why FDA Recalls and Courts Do Not Mix” in IADC Drug, Device and Biotechnology Newsletter (August 2011).

Most recently the doctrine has been advanced in state court opioid litigations against the pharmaceutical companies alleging improper marketing or advertising.  The companies argued that courts should stay proceedings until completion of studies ordered by the FDA regarding specific opioids.  Those cases are discussed in Stapleton “In Defense of the Hare: Primary Jurisdiction Doctrine and Scientific Uncertainty in State-Court Opioid Litigation,” 86 U. Chi. L. Rev. 1697 (Oct. 2019).

16th draft picks may not be as exciting as a top 5 pick, but they can make big plays helping to win games.

It’s a short work week, and we’ve got a short case for today’s discussion. Ebrahami v. Mentor Worldwide LLC, 2020 WL 2510760 (9th Cir. May 12, 2020), eats up only one page when we set our printer to double-sided (which we always do now, since the pandemic work-at-home routine gobbles up paper and ink cartridges). As with Cuban coffee, a Hemingway story, or an episode of Children’s Hospital, short here means good. Very good. (We’ve written about the Ebrahimi litigation before, including here and here and here. Ebrahimi is clearly a gift that keeps on giving.)

The plaintiff in Ebrahimi alleged injuries from a silicone breast implant. Such implants are class III medical devices, which means that they went through the premarket approval process of the Medical Device Amendments to the Food, Drug, and Cosmetic Act. And that means that any state law tort claim that tries to impose requirements that are different from or in addition to FDA requirements would be preempted.

That preemption is pretty iron clad for design defect claims. Lately, we’ve seen more and more plaintiffs attempt to circumvent that preemption by styling their claim as manufacturing defect rather than design defect. That’s a bit of a change from what had been standard practice. More often, manufacturing defect claims were mere makeweight – the plaintiffs just about never really sniffed any evidence that the product at issue wandered from specs, and they would ultimately not even oppose the inevitable summary judgment motion. But the times they are a changing.

And yet, a rose by any other name would still be preempted. In Ebrahimi, the plaintiff claimed that the breast implant failed because the manufacturer violated the FDA’s Current Good Manufacturing Practices (“CGMP”). That’s the usual maneuver to try to state a parallel claim, which would elude preemption by alleging that the defendant deviated from a particular Pre-market approval or other FDA requirement applicable to the class III medical device. Anybody can claim a CGMP violation. But is more than an incantation of those four letters required? Bad courts say no, while good courts say yes. The district court was a good court, and granted the defendant’s Rule 12(b)(6) motion to dismiss.

Would the Ebrahimi appellate court also be a good court? Yes, and it is the court where we clerked, the Ninth Circuit. That circuit has long been unfairly maligned. Sure, it can let loose the occasional stinker (E.g., Stengel), but its overall output is as sound as any other court. Sometimes we get used to thinking negatively about something long past the point when it stopped being true. Adam Sandler, it turns out, is capable of fine acting. American cars are well-made. Philly sports fans are actually … okay, so some things don’t change. But the Ninth Circuit has smart judges who get things right.

The Ebrahimi court got it right: “Ebrahimi essentially contends that the court can plausibly infer that Mentor must have violated at least one of the FDA’s CGMPs by not catching her allegedly defective implants. However, even if more general FDA’s requirements are sufficient for a parallel claim, mere allegations suggesting that [Ebrahimi’s] particular breast implant[s] w[ere] defective do not show that [Mentor] failed to comply with the FDA’s Current Good Manufacturing Practices.”

Let’s count the things the Ninth Circuit got right in Ebrahimi:

1. It rejected the parallel claim.
2. It rejected the attempt to recast the claim as manufacturing defect.
3. It rejected the vague reliance on violations of CGMP.
4. It rejected the plaintiff’s reliance on a quasi res ipsa loquitur theory (that the defendant must have done something wrong because the device injured the plaintiff).

That’s a nice litany of correctness for such a short opinion.

The Northern District of California did not mince words in its opinion in Rodman v. Otsuka America Pharmaceutical, Inc., 2020 WL 2525032 (N.D. Cal. May 18, 2020) and so neither will we.  Plaintiff had three theories of failure to warn and a design defect claim and they all failed in grand fashion due in some significant part to her reliance on Dr. Laura Plunkett’s bag of tricks – or to be more direct, Dr. Plunkett’s creating phony scientific evidence.

Plaintiff was prescribed the drug Abilify and later developed an involuntary movement disorder called tardive dyskinesia (“TD”); a known and warned of risk of the drug.  Id. at *1.  Plaintiff’s failure to warn claim was based on the following allegations:  the label did not accurately state the incidence rate of TD in Abilify users; the label did not address the risk specifically in patients taking lower doses of the drug; and the label failed to include instruction regarding screening for TD.  Id. at *2.

Dr. Plunkett was plaintiff’s sole expert on the first theory.  The Abilify label states that the incidence rate of TD was “greater than 0.1% but less than 1%.”  Dr. Plunkett testified that that was not accurate pointing to two studies that she opined demonstrated incidence rates of 3.4% and 5%.  This is where the phony science comes in.  Those are in fact not true incident rates.  Both studies “looked at a pool of those with TD and identified how many of them used Abilify, which does not result in a true incidence.”  Id. at *5.  Looking backwards from a population of persons with the condition rather than starting with the correct universe of all drug users leads to an exaggerated incidence rate and one that is not generally accepted by the scientific community.  In fact, both of the studies Dr. Plunkett relied on explicitly “cautioned that their sources cannot be used to calculate an incidence rate.”  Id.   The court was unwilling to allow Dr. Plunkett to “analyze data that was not [her] own and reinterpret[ ] it in a manner inconsistent with the conclusions of those who originally generated it.”  Id. (citation omitted).  The court noted that this was not the first time Dr. Plunkett’s testimony had been excluded for “too great an analytical gap between the available data and the conclusion.”  Id.  With Dr. Plunkett’s opinions excluded for “extrapolat[ing] conclusions beyond the scope of her resources,” id. at *7, plaintiff’s first failure to warn claim could not survive summary judgment.

Plaintiff’s second and third failure to warn claims failed on causation grounds.  With respect to the adequacy of the warning regarding lower dosages, plaintiff’s prescribing physician testified that he was aware of that risk already.  Further, he “unequivocally testified” that if plaintiff’s proposed label had been in effect, it would not have changed his prescribing decision.  Id. at *8-9.  Similarly, plaintiff’s prescriber testified that he knew how to monitor patients for TD, including using the very test plaintiff’s argued should have been part of the labeling.  In fact, the prescriber testified the test was something he learned about in medical school and he “didn’t need a drug company at the time to tell [him] about the [ ] test.”  Id. at *9.  Based on this testimony, plaintiff’s remaining failure to warn claims were dismissed.

That left negligent design defect – California does not recognize strict liability design defect claims for prescription drugs.  Id.  The adequacy of the product’s design is therefore measured under a risk-benefit analysis that includes as a factor whether there is a feasible safer alternative design.  Id. at *10.  Plaintiff’s opposition to defendant’s summary judgment motion was not to proffer evidence to meet her burden of proof but rather to argue that she had no obligation to offer proof at this stage because she was “not required to divulge the testimony of medical witnesses designated to testify at trial.”  Id.  She’s wrong.  Plaintiff is not entitled to defeat summary judgment based on “the hope that something can be developed at trial in the way of evidence to support [her] claim.”  Id. (citation omitted).  Plaintiff simply did not have the expert design evidence needed to support a negligence claim.  She cited testimony by one of plaintiff’s treaters that that doctor may prescribe different drugs instead of Abilify, but that testimony does not support that those drugs were safer alternatives.  Id.  Likewise, Dr. Plunkett’s report merely discussed the availability of other treatments, not whether they were safer alternatives.  Nor did plaintiff point to any evidence of the relative costs and benefits of Abilify versus other drugs or any evidence that defendant failed to act as a reasonably careful designer in similar circumstances.  Id.  Hiding behind the secrecy of trial strategy doesn’t cut it to defeat summary judgment.

Like the rest of you, we have enjoyed streaming new video offerings as we stave off social deprivation. But, more often than not, we find ourselves resorting to our “go to” stalwarts (we have mentioned Season 5, Episode 9 of Downton Abbey), among which we choose depending on the particular chord that needs to be struck.  A favorite old standby, whatever our mood, is the 2003 uber-chick flick, “Love, Actually.” It is the first film we purchased to reside permanently on our new DVR when we switched cable services (the second was “The American President” – back in the day we could recite Michael Douglas’s final press briefing speech verbatim) and it has it all.  It has comedy (Bill Nighy’s beyond-brilliant portrayal of aging rocker Billy Mack), tragedy (watch Liam Neeson’s face as he speaks at his character’s wife’s funeral), poignancy (Mark’s Christmas Eve flip chart declaration of hopeless love for Juliet, Karen’s realization that her gift box contains a Joni Mitchell CD, not the beautiful gold necklace she saw in husband Harry’s coat pocket) and romance (Jamie’s proposal to Aurelia, in hilariously subtitled Portuguese, at the restaurant where she is waiting on tables).  It is a deeply satisfying melting pot, and it never disappoints.

Similarly including several favorite themes – and doing justice to most of them, with one exception – is today’s case. (Shout out to the team at Nelson Mullins for the win and for sending us the decision.)  In Ebert v. C.R Bard, Inc., 2020 WL 2332060 (E.D. Pa. May 11, 2020), the plaintiff was implanted with the defendants’ inferior vena cava (“IVC”) filter to prevent pulmonary embolism. The filter eventually fractured – a risk described in the Instructions for Use (“IFU”) – and a piece of it migrated to the plaintiff’s pulmonary artery. The plaintiff underwent successful surgery to remove the filter, including the fractured portion. She asserted the usual product liability claims, including claims sounding in negligence (design defect and failure to warn), breach of express warranty, negligent misrepresentation, and strict liability, and the defendants moved for summary judgment.

Negligent Design Defect

As the court explained, the Pennsylvania Supreme Court’s 2014 Lance decision held, for the first time, that negligent design defect claims are cognizable in prescription drug cases, and Pennsylvania’s federal district courts broadened the holding to apply to prescription medical devices. But, as the defendants argued in Ebert, Lance held that the relevant duty of care was narrow: “[u]nder Pennsylvania law, pharmaceutical companies violate their duty of care [only] if they introduce a drug into the marketplace . . . with actual or constructive knowledge that the drug is too harmful to be used by anyone.” Ebert, 2020 WL 2332060 at *3 (citation to Lance omitted). The plaintiff disagreed, citing language from Lance to the effect that “the law of negligence establishes a duty, on the part of manufacturers, which can be viewed as a continuum” from graduated strengths of warnings “through non-marketing or discontinuance of marketing” when the manufacturer knows or should know “that the product should not be used in light of its relative risks.” Id.

The court wasn’t buying it. Noting that “Lance’s ‘continuum’ language is dicta explaining the range of duties of care applying to negligence claims generally,” id. at *4 (emphasis in original), the court “agree[d] with [the defendants’] reading of Lance,” which held “that a prescription medical device manufacturer violates its duty of care, as it applies to negligent design defect claims – when it tenders into the market a drug which it knows or should know is so dangerous that it should not be taken by anyone.” Id. (internal punctuation and citation to Lance omitted). The court held that there was no record evidence that the defendants placed their filter into the market with actual or constructive knowledge that it was too harmful to be used by anyone.  Further, the Plaintiff could not cure this void, and create an issue of fact, by attempting to demonstrate the existence of a safer alternative design.  As such, the court granted summary judgment in favor of the defendants on the negligent design defect claim.

Negligent Failure to Warn

Regular readers of this blog know that we have a soft spot for a good warnings causation decision.  In Ebert, the court began its discussion by emphasizing that, even if a defendant had a duty to provide adequate warnings and didn’t provide them, “plaintiffs must further establish proximate causation by showing that had the defendant issued a proper warning to the learned intermediary, he would have altered his behavior and the injury would have been avoided.”  Id. at *5 (citations omitted).  We thought this was a very good start.

So a failure-to-warn plaintiff must satisfy two tests: the warning must have been inadequate, and the inadequacy must have proximately caused the plaintiff’s injury.  In Ebert, it was undisputed that the defendants’ IFU warned of the risk of filter fracture.  Nevertheless, “notwithstanding the fact that the IFU warned of the specific complication that [the plaintiff] experienced, she contend[ed] that the warnings were still inadequate” because the defendants did not provide comparative failure rates between their filter and other devices.Id. at *6.

But the plaintiff’s doctor testified that he never read the IFU in its entirety, and that he couldn’t recall if he had read any of it before the plaintiff’s implant surgery. “Thus,” the court held, “even assuming that the warnings were inadequate, more detailed warnings, such as comparative failure rates, would have made no difference” in the doctor’s decision to implant the defendants’ filter in the plaintiff. Simple stuff: if the doctor didn’t read the warnings, it didn’t matter what they did or didn’t include – they couldn’t have affected his prescribing decision.  No proximate cause, no failure-to-warn claim – summary judgment for the defendants.  Hard to believe so many courts get this wrong.

Breach of Express Warranty/Negligence Misrepresentation

The plaintiff alleged that an express warranty was created by statements in the IFU representing that most filter fractures did not have adverse clinical consequences.  In Pennsylvania, an express warranty “is created by any affirmation of fact or promise made by the seller to the buyer which relates to the goods and becomes part of the basis of the bargain.”  Id. at *7 (citations omitted).

The court emphasized that there was no evidence that the plaintiff ever read or saw the IFU or received any information from the defendants before her implant surgery, so no such information could have been part of the “basis of the bargain” when she agreed to the surgery.  Nor could the plaintiff defeat summary judgment by arguing that she “indirectly relied on the warranty” by relying on the information conveyed to her by her doctor in the informed consent process.  Even if such a warranty-once-removed could form the basis of a breach of warranty claim, there was no evidence that the doctor conveyed the information to her, let alone that it was a “basis of the bargain.”

Similarly, the court held that the plaintiff could not prevail on her negligent misrepresentation claims.  She argued that the defendants’ sales training manual misrepresented the filter’s risks, but there was no evidence that she or her doctor “justifiably relied” on the statements, or was even aware of the manual’s existence.  She also argued that the doctor “justifiably relied” on misrepresentations in the IFU, but, like the warnings claim, the doctor’s admitted failure to read the IFU defeated this argument.

Strict Liability

And so we come to the court’s only misstep, significant in helping to create bad law though it did not affect the defendants’ good result.

Most of you are familiar with comment k to Section 402(a) of the Restatement (Second) of Torts.  Comment k bars strict liability claims against the manufacturers of “unavoidably unsafe products,” explicitly including prescription drugs.  In its 1996 Hahn decision, the Pennsylvania Supreme Court adopted comment k.  Since then, decades of decisions have conceded that prescription medical devices are just like prescription drugs, for comment k purposes.

As we have reported in several recent posts (like this one and this one and this one), Pennsylvania’s federal courts have wrestled of late with this most obvious of propositions, some declining to extend comment k protection to the manufacturers of prescription medical devices.  Misconstruing the holdings of both Tincher and Lance, the Ebert court jumped on this bandwagon, predicting that the Pennsylvania Supreme Court would analyze comment k’s applicability to prescription medical devices on a case-by-case basis “determined by each case’s developed factual record,” id. at *10, and would not apply it to medical devices across the board.  (As we have explained in these pages, neither Lance nor Tincher limited the application of comment k to medical devices.  Lance held only that comment k did not apply to negligence actions, and Tincher did not deal with pharmaceutical products at all.)  But just as the bowling ball careened into the gutter, the bumpers deflected it back, at least for purposes of the defendants’ motion. Conducting its “analysis” on the basis of “a fully developed factual record,” the Court “conclude[d] that the . . . filter is an ‘unavoidably unsafe product’ such that the Pennsylvania Supreme Court would apply comment k to [it], thereby shielding [the defendants] from a strict liability claim.”  Id. at *11.  It reached this conclusion by reviewing literature confirming that the device carries risks.  As do all medical devices, right? We know – none of this, except the defendants’ result, makes any sense.  And the decision’s “case by case analysis” edict falls resoundingly on the wrong side of the teetering comment k scale.

But the bottom line is that the court granted summary judgment in the defendants’ favor on all of the plaintiff’s claims.  And that is a very good result.  Time to fire up the DVR – stay safe out there.

There’s a reason plaintiffs hate removal before service – “snap removal.”  It has the potential to wreak havoc on their mass tort business models, which are largely based on confronting defendants with as many cases as possible in the worst jurisdictions possible.  While federal courts are hardly perfect, they are usually better than the state-court fora that attract litigation tourists from all over the country.

Snap removal makes is much harder for litigation tourist plaintiffs to trap diverse defendants in the plaintiffs’ chosen fora, as is exemplified by the recent Castro v. Colgate-Palmolive Co., 2020 WL 2059741 (Mag. W.D.N.Y. April 29, 2020), which defense counsel in the case helpfully forwarded to us a little while ago.  Castro is an asbestos/talc case filed in New York state court by Virginia litigation tourists.  Id. at *1.  Unfortunately, New York is notorious as a plaintiff haven in such cases.  As is typical in such cases, plaintiffs sued multiple defendants, including some with principal places of business in New York.  Id.

One (at least) of the defendants was technologically equipped to effectuate snap removal, and did so:

[Defendant] removed the case to this District by filing removal papers . . . just one day after the date of electronic filing stamped by the state-court system on the verified complaint.  The 24-hour timing of the removal – [defendant] almost certainly was not served within 24 hours of filing and likely had means to monitor court dockets daily for activity against it − plus the language that no other defendant had “been properly joined or served” indicated that this removal was what some call a “snap removal.”

Id. at *2 (citations omitted).  Crying “gamesmanship,” plaintiffs moved to remand.  Id. at *1.

Unfortunately for plaintiffs’ tourism plans, “their motion to remand came just four days before the Second Circuit issued an opinion that condoned [defendant’s] procedural maneuver.”  Id.  That decision of course, was Gibbons v. Bristol-Myers Squibb Co., 919 F.3d 699 (2d Cir. 2019), which we blogged about here, and then named as our #2 best case of 2019.  Thus, these litigation tourists were stranded in federal court.  “[D]espite [plaintiffs’] valiant efforts to explore the penumbras of the language in Gibbons, the holding of Gibbons compels denial of their motion [to remand]”:

  • The statute plainly provides that an action may not be removed to federal court on the basis of diversity of citizenship once a home-state defendant has been ‘properly joined and served.’
  • By its text, then, Section 1441(b)(2) is inapplicable until a home-state defendant has been served in accordance with state law.
  • Put simply, the result here − that a home-state defendant may in limited circumstances remove actions filed in state court on the basis of diversity of citizenship − is authorized by the text of Section 1441(b)(2) and is neither absurd nor fundamentally unfair.
  • [S]nap removal is, in itself, the limited circumstance that allows a home-state defendant to do what otherwise would be prohibited.

Castro, 2020 WL 2059741, at *3 (citations to Gibbons omitted) (emphasis original).

Plaintiffs’ “policy argument” thus went by the boards.  Modern technology is what it is.

In the era of 24-hour electronic filing, the Court is vaguely aware of third-party services that exist to monitor docket activity nationwide on a daily basis for new cases involving specified defendants.  Whatever Congress intended with Section 1441(b)(2), modern technology appears to have created a gap between what was intended and what is now technically possible under the literal language of the statute.

Id. at *4.  But, “updating statutes and regulations in the face of categorically new technological advances is for Congress and not for the judiciary.”  Id. (citation omitted).

Assuming, of course, that diversity of citizenship exists (snap removal cannot create diversity where it does not exist), Castro demonstrates why snap removal is particularly suited to pattern litigation mass torts.  A lot of pattern litigation (not all, of course), particularly in asbestos, is commenced by lawyers who don’t expect to do much more than fill in the blanks on some complaint form in their word processor and then seek settlements.  Indeed, in Castro, even filling in the blanks proved a serious challenge:

After identifying the parties, the verified complaint became somewhat confusing, possibly due to language borrowed from complaints filed in other cases.  Although all of the defendants appear to be cosmetics manufacturers, pharmaceutical companies, or retailers, the verified complaint contained a reference to distributing and selling “raw asbestos fibers of various kinds and grades.”  The verified complaint also referred to coming into contact with “asbestos products while working in various shipyards, steel mills, refineries, paper mills, chemical plants, industrial sites and facilities, construction sites and other facilities or was exposed to the defendants’ products through the normal use of these products.”

2020 WL 2059741, at *1 (citations omitted).  Lawyers who can’t be bothered even to file coherent complaints aren’t likely to invest in sophisticated technological countermeasures to combat snap removal.  So in asbestos/talc cases, and in other mass torts where the other side concentrates more on the “mass” than the “tort,” snap removal is very likely to succeed in reducing litigation tourism.  Indeed, in the three circuits (Second, Third, Fifth) where courts of appeals have explicitly blessed snap removal, removing defendants should consider responding to the kind of knee-jerk remand petition that was filed in Castro with motions seeking counsel fees.  Turnabout is fair play.

So, our advice is to “make it snappy.”  Also, fight any COVID-19 cloaked attempts by the other side to replace personal service with something electronic.

Now, if we can only be as successful with our personal jurisdiction arguments….

Missouri is central to America – geographically, culturally, and politically. Some of our greatest literature came from Missouri authors (Twain, Eliot, Angelou). Media figures as unifying as Walter Cronkite and as divisive as Rush Limbaugh at one time called Missouri home. American music wouldn’t be the same without tenor saxophonist Coleman Hawkins (listen to the 1939 version of “Body and Soul” to hear the essence of Jazz improvisation) and Chuck Berry (who invented rock and roll without any help from Marty McFly). Two of the country’s deepest philosophers, Reinhold Niebuhr and Yogi Berra, hailed from the Show Me state. We are grateful for Kansas City barbecue, especially accompanied by the best selling beer in the country, brewed on the other side of the state. (Don’t sneer; superchef David Chang says that Bud Light is his favorite beer to accompany food). We also would have found these last few weeks of the Coronoavirus lockdown much more difficult to get through without binge-watching Ozark, the Netflix series about a Chicago family relocating to a lovely part of Missouri to engage in some very unlovely money-laundering and mayhem. (Query why the local authorities in the Lake of the Ozark area don’t pay more attention when their resort area suddenly becomes a lot more explosive and murdery. Season 3 ended recently and – semi-spoiler alert – it didn’t turn out great for a certain holder of a law degree.)

In politics, Missouri has long occupied the crucial middle. The Missouri Compromise put off the Civil War … for a while. Harry Truman was a consequential President in the middle of the 20th Century. Currently, Missouri has a colorable claim to be one of the few genuinely purple states in a land too often split into red and blue. (But recently it does seem to be getting redder. The 538.com website plants Missouri firmly in the pro-Trump column.)

Missouri is also at the center of things in the legal arena. Border areas on both sides of the state harbor predatory plaintiff firms. At the same time, some of the fellow defense hacks we admire most have Kansas City and St. Louis addresses. But if we had to choose where to visit, we’d pick Kansas City, and not just because of Arthur Bryant’s. The courts in Kansas City are not bad, though not quite as good (yeah, we mean pro-defense) as the ones across the river in Kansas. But the courts in St. Louis (we mean the city, not the county), are just about the worst in America. (And by worst we mean pro-plaintiff; we mean unfair; we mean crazy.)

Did you ever watch the old Rumpole of the Bailey series, written by former barrister John Mortimer, and played on tv by the masterful Leo McKern? It should be mandatory viewing by put-upon lawyers everywhere. In one of his tougher cases, Rumpole, after a series of adverse rulings, invites the presiding judge to climb down from the bench and take a seat at his opponent’s table, since the judge’s bias was so palpable. Well, we heard a story about something like that actually happening in St. Louis. In the midst of a product liability litigation, a judge retired, and then went to work for the plaintiff firm. At that point, we would have a very depressing conversation with the client, and then commence dreaming of a retirement filled with watercolor painting and vodka gimlets.

Missouri for a while had a reputation as one of the big litigation tourist jurisdictions. Plaintiff firms would gin up consolidated cases, joining one or two Missouri plaintiffs with riverboatloads of non-Missourians, file them in St. Louis, and then let the madness begin. St. Louis juries doled out several multi-million and even multi-billion-dolar awards in cases that, under any set of dispassionate eyes, did not come close to warranting such, er, generosity. Research by Citizens Against Lawsuit Abuse demonstrated that excessive tort litigation in Missouri resulted in a loss of $2 billion in personal income annually and a loss of 32,205 jobs. That “tort tax” cost $505.21 per person. Thankfully, the SCOTUS opinions in Bauman and BMS made litigation tourism in Missouri more difficult, and Missouri courts have actually done a pretty good job of changing their ways. We’ve written about that evolution frequently, including here. Personal jurisdiction doctrine now permits corporate plaintiffs to face jurors at home or where the alleged injury occurred, rather than alwaysalwaysalways in the shadow of the Gateway Arch. It’s an improvement.

Anyway, putting aside the upside-down world of St. Louis city courts, Missouri seems to be heading back toward the center of Anglo-American jurisprudence. Maybe some of that is due to the state’s reddening, which we alluded to earlier. We have no desire to dive into the mud pit of politics. We’d be out of our league. But having once occupied the position of federal prosecutor and now working in an AmLaw 100 office (if we are ever allowed to return to the actual office), we are a bit sensitive about everyone assuming that we are a hard-core, atavistic right-winger. That assumption would be wide of the mark. One can be a social liberal but still believe corporate defendants deserve a fair shake. More to the point, one can occasionally color in the dots next to Democrat candidates while still favoring tort reform.

Missouri has recently hopped aboard the tort reform bandwagon. 2019 was a good year in that respect. 2020 looks good, too. Last week, the Missouri General Assembly enacted Senate Bill 591, which does a couple of things that restore rationality and fairness to lawsuits that can involve our drug and device clients. First, it reins in the much abused Missouri consumer fraud statute. It imposes a “reasonable consumer” standard on plaintiffs and requires “definitive and objective evidence” of “individual” damages. Think how different this damages proof would be as compared to plaintiff economic experts offering counterfactual scenarios containing the scientific rigor of a Rick and Morty episode. These requirements are also added to the class action section. Medical malpractice (chapter 538) claims are exempted. In addition, consumer fraud actions accrue on the date of purchase. The bill also limits class action attorney fees. The result is that consumer class actions now have a clear, concrete time-frame and are less likely to look like a potential jackpot to enterprising plaintiff lawyers. For corporate defendants, there is now less uncertainty, less overall exposure, and more predictability.

Second, the bill reforms punitive damages in a number of enlightened ways. Missouri law will require clear and convincing evidence that the defendant “intentionally harmed the plaintiff without just cause or acted with a deliberate and flagrant disregard for the safety of others.” The confusing, prejudicial, and ultimately useless “willful and wanton standard” is now gone. Here is the kicker: “Evidence of negligence including, but not limited to, indifference to or conscious disregard for the safety of others shall not constitute intentional conduct.” That’s potentially a game-changer. Let’s face it: against all original intention and sound jurisprudence, plaintiff attorneys have hoodwinked courts into turning straight-on negligence claims into cases with potential punitive damages. That doctrinal detour has increased exposure and uncertainty, placed a premium on plaintiff efforts to ramp up jury anger, and turned litigation from an engine of truth into a settlement meat-grinder.

But the goodies in the new Missouri tort reform bill do not stop there. Punitive damages cannot be based on nominal damages in the kind of cases we defend. Further, the bill limits vicarious punitive damages. Conduct must have been authorized by the principal. Oh, discovery on punitive damages is now limited. Punitive damages cannot be sought in the initial complaint, and there will no punitive damages discovery, except on amendment supported by evidence, which would be needed to secure leave of the court. If plaintiffs do manage to inject punitive damages into the case, trial will be bifurcated. Punitive damages cannot be based in any way on harm to non-parties.

All these excellent amendments are prospective only, applying to claims filed on or after August 8, 2020. Perhaps we will see a rush to the courthouse before then. But afterwards, a new litigation day will dawn in Missouri.

Two and half years ago we posted about a favorable California Superior Court ruling in the Risperdal and Invega Product Liability Cases litigation finding plaintiffs’ claims were preempted because there was no newly acquired information on which to base a CBE label change and because the FDA had rejected the proposed label change already.  We titled that post Never Too Late to Celebrate Preemption.  Apparently it’s also never too late for the California Court of Appeals to be a party pooper, a killjoy, a spoilsport, a wet blanket, a sourpuss, a downer, a real drip.  We’re sure you can see which way this is headed – reversal.

In what appears to be a de novo review in Risperdal and Invega Product Liability Cases, 2020 WL 2300213 (Cal. Ct. App. May 8, 2020) the court traces the history of pediatric labeling for Risperdal up through the FDA’s October 2006 approval for use in children and adolescents to treat autism.  The 2006 label that plaintiffs claim is inadequate included language about prolactin elevation and the reported rate of gynecomastia (enlargement of male breasts) among Risperdal users.  Id. at *1-3.  Because this is a well-known side effect of the drug that is warned about, you can search our blog and find a multitude of decisions granting summary judgment in defendant’s favor on causation grounds and indeed the last part of this decision upholds the dismissal of one plaintiff’s claims because his prescriber testified a changed warning would not have changed her decision to prescribe the drug.  See id. at *11-13.

Plaintiffs argue that the defendant could have used the CBE process to add a warning of a direct correlation between the drug and gynecomastia and to add a recommendation for regular monitoring of prolactin levels.  Id. at *9.  A CBE label change however must be based on newly acquired information.  The court agreed with defendant that there were no new studies or previously undisclosed studies that constituted newly acquired information.  Id.  Plaintiffs’ entire argument came down to a single table that was in a draft of study but not the final published results and therefore was not included when the study was provided to the FDA.  Defendant argued that the table could not be newly acquired information because “it did not reveal risks of a different type or greater severity or frequency and the analysis was based on the studies submitted to the FDA.”  Id.  In fact, the table did not change the overall rate of gynecomastia that was reported on defendant’s label.  But, it did show different rates at different time periods after use at least one of which was higher and which was not reported on the label.  This was enough for the court to conclude that the table “demonstrated a risk of greater frequency” and therefore could support a CBE label change.  Id.

Defendant argued that the table could not support a label change because the FDA would not have allowed it.  First, defendant pointed to a statement by the FDA in a brief filed in a different case in which it said that defendant had “submitted all the necessary data and information to conclude that risperidone was appropriately labeled.”  Id. at *10.  It is important to note that in that other case, the FDA was a defendant, not writing as an amicus.  Because the brief was “made in a wholly different context” and was not an “agency action taken pursuant to the FDA’s congressionally delegated authority,” it could not be used as “clear evidence” that the FDA would have rejected the label change.  Id.

But that was not defendant’s only evidence.  In 2012, the FDA denied a citizens petition filed by plaintiffs’ counsel challenging that the Risperdal label did not adequately address elevated prolactin levels, the need to monitor for elevated prolactin levels, or the rates of gynecomastia.  For the trial court, this was clear evidence that the FDA considered and rejected the argument that the Risperdal label was deficient with respect to prolactin levels and gynecomastia.  The appellate court parsed the issue differently.  It was important to the court that the FDA did not have the above-mentioned table when it denied the citizens petition.  Then, in denying the petition, the FDA found “there was no evidence that risperidone was unsafe or anything else that warranted revoking the pediatric indication of the drug,” and “no basis for requiring a black box warning about the lack of long-term safety data associated with pediatric use of risperidone.”  Id. at *4.  The denial also said that because the petition’s only specific labeling request was for a boxed warning, that was the only labeling request to which the FDA was responding, despite the petition “includ[ing] an extensive discussion of the current labeling.”  Id.  That was key for the appellate court.  It viewed the citizens petition as making a “broader request” to either take the drug off the market or include a black box warning as compared to plaintiffs’ argument that the unconsidered table supports a label change to include prolactin monitoring at certain periods of time.  Id. at *10.  For this court, the charges in the citizen petition while similar to plaintiffs’ allegations were distinct enough to not be “clear evidence” to support a preemption defense.  Never mind that the petition put the adequacy of the drug’s labeling specific to prolactin levels and gynecomastia before the FDA who determined that no changes were required.

We were enjoying our Risperdal preemption party and while this uncitable decision has put a damper on the celebration, as we note the litigation as a whole has much more good than bad.  Enough to keep us smiling.

One of the advantages that the FDA (and other government agencies) have over other litigants is that it gets to ignore court decisions it doesn’t like, in hopes of trying again later in what the Agency considers a more favorable forum.  Here’s how one court described the same policy by a different agency:

Understood in the most charitable light, not acquiescing to a given circuit’s diverging legal interpretation until the Supreme Court has the last word puts two roles in harmony − the [administrative agency’s] role of national say in what [administrative] law should be, and the judicial department’s emphatic province and duty to say what the law is. . . .

To that end, nonacquiescence allows for an issue’s “percolation” among the circuits; generating a circuit split that can improve the likelihood of certiorari being granted.  But, nonacquiescence is justifiable only as a means to judicial finality, not agency aggrandizement. . . .  [N]onacquiescence is divorced from its purpose when an agency asserts it with no stated intention of seeking certiorari.

Heartland Plymouth Court MI, LLC v. NLRB, 838 F.3d 16, 21-22 (D.C. Cir. 2016) (citations, quotation marks, and footnote omitted).  FDA uses nonaquiescence as do other federal agencies.  We believe that the FDA’s failure to appeal United States v. Caronia, 703 F.3d 149 (2d Cir. 2012), is an example of this policy.  So far the FDA’s gotten away with it.  The First Amendment/off-label promotion issue hasn’t reached any other circuit, and the FDA can do what it wants in all other circuits.

However, it doesn’t always work out that way, and FDA took one on the chin recently in Eagle Pharmaceuticals, Inc. v. Azar, 952 F.3d 323 (D.C. Cir. 2020).  The issue in Eagle Pharmaceuticals was marketing exclusivity under an FDA “orphan drug” approval.  Orphan drugs are those that treat rare diseases.  Development of such drugs is problematic “because the comparatively small demand for treatment left little motive for research and development.”  Id. at 325 (citation and quotation marks omitted).

Congress passed the Orphan Drug Act to provide incentives to overcome this barrier, one of which is “a seven-year period of exclusive approval rights during which time the FDA may not approve another” competing drug “for a person who is not the holder of” the original application.  Id. (citing 21 U.S.C. §360cc(a)).  The FDA added its own gloss to the statute:

[T]he FDA considers a drug the same as a previously-approved drug if it shares the same active moiety and is not otherwise clinically superior; it considers the drug to be different − and thus entitled to its own seven-year exclusivity period upon designation and approval − if it does not have the same active moiety or is clinically superior.

Id. at 326.  For those not familiar with FDA-speak, or chemistry, “moeity” is simply the drug’s “active ingredient.”  Id.  Another FDA gloss – and the one on which decision in Eagle Pharmaceuticals turns − is on this concept of “clinical superiority.”  At the beginning of the entire process – the “designation stage” – an applicant can obtain orphan drug status simply by “present[ing] a plausible hypothesis that its drug may be clinically superior to the first drug.”  Id. at 327 (quoting 21 C.F.R. 316.20(a)) (emphasis original with the court).  Later on, after the FDA acts, a person holding an approved application must “demonstrate . . . that the drug is clinically superior to the previously approved drug.”  Id. (quoting 21 C.F.R. §316.34(c)) (emphasis original with the court).

The reason for the FDA’s distinction is to prevent a process known as “evergreening” – which is the original applicant being able to use superiority claims to tack of one exclusivity period after another so as to “[o]btain infinite, successive 7-year periods of exclusivity for the same drug for the same use.”  Id. (quoting 78 Fed. Reg. 35117, 35127 (FDA June 12, 2013)).

That sounds like a laudable goal for FDA to pursue.  The only problem is that there’s no basis for it in the statute.  And that was what a court held back in 2014.

[T]he language of the Act’s exclusivity provision does not permit or invite any discretion on the part of the FDA regarding whether or not to continue authorizing new such drug marketing applications once an orphan drug has been so designated and approved.  Indeed, Congress has specifically established the only two situations in which the FDA can carry on regardless [and neither involves this situation].

Depomed, Inc. v. United States Dep’t of HHS, 66 F. Supp.3d 217, 231 (D.D.C. 2014).  However, rather than appeal the Depomed decision, FDA close to ignore it through nonacquiescence.  “The FDA initially appealed the Depomed decision but ultimately withdrew its appeal, opting instead to nonacquiesce to the decision in future cases.”  Eagle Pharmaceuticals, 952 F.3d at 328 (citation omitted).

We’d love it if our clients could do that.  But one problem with nonacquiescence is that it’s a recipe for procrastination.  It’s also why federal courts have increasingly resorted to nationwide injunctions to force the government to obey court orders.  Here, with no nationwide injunction having issued, the FDA (and Congress) did nothing to address the Depomed holding for the next six years – until the District of Columbia Circuit decided Eagle Pharmaceuticals.

First, the Chevron doctrine didn’t help the FDA, since (like Depomed) the appellate court in Eagle Pharmaceuticals found no ambiguity in the controlling statute:

Under the plain language of this provision, the FDA is barred from approving another application for “such drug” for the same disease for seven years once it approves an orphan drug for marketing.  Based on this language, the seven-year marketing exclusivity period applies automatically − the text leaves no room for the FDA to place additional requirements on a drug that has been designated and approved before granting its manufacturer the right to exclusivity.

952 F.3d at 331.  Eagle Pharmaceuticals thus rejected the FDA’s contrary arguments.  While the statute did not speak to “whether one or multiple manufacturers can win a period of exclusive approval,” that was not an ambiguity, merely an application of the statute’s requirements to a particular fact pattern.  Id.  Given the statutory context in which it was used “expiration” was not ambiguous, since it was modified by the phrase “of seven years.”  Id. at 332.  In sum:

We conclude that the text of §360cc(a) is unambiguous:  if the FDA approves a previously-designated orphan drug, it cannot approve another such drug for the same condition for seven years.  This language leaves no room for the FDA to add an after-the-fact requirement that a designated and approved drug prove clinical superiority before receiving that exclusive approval benefit.  Nothing in the statute’s text, structure or purpose limits this benefit to only one drug manufacturer.

Id. at 340.

The other escape hatch from plain statutory language is – as we have seen in removal before service – is to argue that the statute as written leads to an “absurd” result.  The DC Circuit was no friendlier to that argument than the circuits recently evaluating it in the removal context:

Granted, the Congress’s goal in enacting the [Act] was to reduce the cost of and incentivize orphan drug development but the fact that following the text of a statute may conflict with the statute’s larger purpose alone does not warrant departing from the text.  It is not our job to say how the Congress should accomplish its goals; rather, we will ignore what the Congress has written only if we are so convinced by a conflict between the text and the purpose that we think the Congress almost surely could not have meant what it said.

Id. at 334-35 (citations omitted).  Further, the issue that the FDA identified as an “absurdity” “could occur only if the FDA allowed it to happen.”  Id. at 334.

Finally, as to the FDA’s desire to prevent consecutive exclusivity periods the court pointed out “that the FDA may have created the self-evergreening problem itself.”  Id. at 336.  The FDA only needs to get off its collective duff (which nonacquiescence encouraged) and act:

[T]his result assumes that the FDA’s regulations for designation are correct and static.  Not so.  The self-evergreening problem is within the FDA’s power to manage and, if needed, alter. . . .  Were the FDA to change its regulations for designation to take into account both the active moiety and the formulation, for example, that would prevent a manufacturer from obtaining successive, automatic exclusivity periods for various formulations of the same drug simply by relying on its original designation.

Id.  In short, the FDA “cannot avoid the Congressional intent clearly expressed in the text simply by asserting that its preferred approach would be better policy.”  Id. at 337 (citation and quotation marks omitted).

It appears to us that FDA’s nonacquiescence maneuver here left it worse off than if it had pursued an appeal in the Depomed case six years ago.  Not only has the Agency wasted six years, now it has lost in the District of Columbia Circuit, the preeminent circuit for administrative law issues such as this.  Does the FDA now appeal to the Supreme Court?  Does the FDA feel lucky?  Unlike six years ago, there may now be a majority on the Supreme Court that would seek to use an appeal like this one to knock down the entire Chevron, U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837 (1984), edifice under which Eagle Pharmaceuticals was decided.

Go ahead, make our day.

There is quite a bit of discussion these days about protocols. Using good judgment in setting how often you should wash your hands, what measures should be in place for a certain type of business to operate, how often to test for infection and/or antibodies, and many other protocols seems like a no-brainer. We will refrain from explaining our, shall we say, dismay at how often debate about such protocols seems to include considerations that should not be considered, as it were. We need not dance around our feelings about protocols for ex parte contact and depositions of treating physicians. Setting aside our own cases—for obvious reasons—we can say that the protocols in MDLs and various versions of coordinated proceedings tend to favor the plaintiffs, particularly those with counsel who are willing to push the limits or blow right by them without consequence.

McLaughlin v. Bayer Essure, Inc., No. 14-7315, 2020 U.S. Dist. LEXIS 79769 (E.D. Pa. Apr. 13, 2020), is not from an MDL but from a coordinated proceeding of related cases relating to a contraceptive device. That may be why a Special Discovery Master was appointed under Fed. R. Civ. P. 53 and was issuing a Report and Recommendation on what the protocol would be for treating physician depositions. The parties had submitted dueling proposals on ex parte contact with these expected deponents and the order of questioning in the depositions. What resulted was an R&R that was short and pretty sensible, something we do not always see when setting the rules for these depositions. Not to go out on a limb, but these depositions tend to be really important in medical product cases. This is particularly so when the doctors (or other healthcare providers) are not coming live to trial, so the depositions provide the universe of evidence from these witnesses. Juries may disregard what paid experts say about medical issues and go with what the “independent” treaters say. For prescription products, what the prescriber says about whether she knew enough to decide to use the product with the plaintiff and whether she would have done anything different with a different warn is critical to at least one claim. These are not novel or insightful observations. Nor is the observation that the woodshedding of a witness—“horseshedding” per McLaughlin, a term we had not heard before—can really matter. Who goes first—and last—can also matter, although less.

Plaintiffs proposed what you might expect: plaintiffs’ counsel get to talk to willing healthcare providers about whatever they want ex parte; defense counsel do not get to talk to them at all; and, when a deposition occurs, plaintiffs’ counsel get to ask questions first and defense counsel can ask about the ex parte contact without any prior production or identification of what documents were given/shown to the deponent as part of the ex parte contact. Id. at **2-3. They stopped short of proposing to preclude defense counsel from asking any questions at the deposition or make them pay for a share of the costs of the ex parte party, like the snacks or beverages. Defendants proposed that both sides should be allowed to have ex parte contact, that such contact could not address “liability issues” and documents beyond medical records and product warnings, that such contact be disclosed prior to deposition, and that whichever party noticed a deposition should get to go first. Id. at **3-4. The R&R came down somewhere in the middle, but that resulted in a fairer overall protocol than we often see.

The R&R started with one observation and three principles. First, these were cases with healthcare providers outside of the trial subpoena power of the ED Pa, so the depositions would be “trial depositions.” Id. at **1-2. Second, under Fed. R. Civ. P. 30(c)(1), “The examination and cross-examination of a deponent proceed as they would at trial under the Federal Rules of Evidence.” Id. at *2. Third, plaintiffs had waived their physician-patient privilege by bringing lawsuits about their physical and mental condition. Id. Fourth, there is nothing privileged about a lawyer talking to a third party. Id.

Recognizing that “the danger of excessive and unethical horseshedding exists in EVERY ex parte communication between counsel and any witness, particularly including the client” and the lack of consistency in court rulings on what to do about this risk, the Master attempted to be fair to both sides (and the potential deponents). Id. at *5 (emphasis in original).

The Special Discovery Master sees no reason for the court to depart from the usual practice of letting counsel for BOTH sides meet with prospective deponents pre-deposition where the deponents consent.

Id. at *6 (emphasis in original and internal citation omitted). This rejected plaintiffs’ argument that defense counsel could not talk to healthcare providers outside of a deposition, an argument relying in part on “a rather novel, physician-patient HIPPA [sic] privilege ‘penumbra’ (for lack of a better term) which [purportedly] applies even though the privilege is waived.” Id. at *7. (See this ancient post and this not-so-ancient one.)  Consideration of “equality of access to critical evidence, avoidance of surprise,” and the general purpose of Federal Rules of Civil Procedure was enough to allow defense counsel to “enjoy their traditional access to the treating physicians on the same terms . . . as Plaintiffs.” Id. at **7-8. This is a provision any drug or defense defendant would likely accept.

The next issue, at least as we are presenting them, was what those terms would be in terms of the scope and disclosure of any ex parte contact. In terms of scope,

the Special Discovery Master rejects the imposition of the limitations on such communications proposed by Defendants. Indeed the limitations proposed by Defendants would preclude Plaintiffs’ counsel from discussing key issues in the case to which the physicians are privy, including medical malpractice, negligent training, failure to warn, and alternative treatments that may have been considered to name a few.

Id. at *6. It does look like some of these topics would not have been precluded by the Defendants’ proposal, but we still hazard that many defendants would take the possibility of free-range ex parte contact by both sides over plaintiffs have limits on their exclusive contact that may be hard to police. Of course, it could be that a physician, presented with the choice of two ex parte meetings and a deposition or no ex parte meetings and a deposition might go the route of less time commitment. A witness who does not subject herself to any potential woodshedding—or horseshedding—may be disinclined to speculate in deposition about things like what corporate documents and snippets of corporate deposition testimony mean. Similarly, the requirement that any ex parte meetings with a treating physician be disclosed at least 72 hours before a deposition, including with “a list of all materials shared with the treating physician,” may both discourage bad behavior by counsel and encourage the deponent to play it straight.

The last issue to decide in the R&R was the order of questioning in depositions of treating physician depositions. In this instance, the usual practice that the party that notices the deposition gets to go first did not hold. The logic for this was that, at trial, the plaintiffs would be expected “to call every treating physician (even if less than 100% favorable) to make a prima facie case.” Id. at **8-9. Given what was noted about 30(c)(1) above, this was enough to propose that “Plaintiffs’ counsel has priority of questioning the depositions of the treating physicians as they would at trial.” Id. at *9. Our experience in trials in drug and device litigation, including of cases where treaters are within trial subpoena range, is that plenty of plaintiffs hope the jury never hears from certain treaters and, if they do testify, would prefer to go after them on cross. So, we think the premise of this particular recommendation is a bit flawed. On the other hand, many defense lawyers might prefer to go second in treater deposition for a range of reasons we will not share here. Even if that is not the case, the order of questioning will generally not be as important as whether plaintiffs’ counsel has had exclusive, unfettered access to the witness before the deposition starts.