Friday, August 09, 2013

Pleading Defect

            We have all heard that bad facts make bad law.  We have also heard that discretion is the better part of valor.  Sometimes, faced with bad alleged facts, a bad decision may be the predictable result of a motion to dismiss.  We are all for knocking out non-existent claims and making plaintiffs plead properly, but it is important when deciding to file 12(b)(6) to keep an eye on the relief being sought and how that relief would affect the case.  Although we do not have any knowledge of the considerations that went into the filing of this motion to dismiss, sitting here with our retrospectoscope, we see Williamson v. Stryker Corp., No. 12 Civ. 7083 (CM), 2013 U.S. Dist. LEXIS 104445 (S.D.N.Y. July 23, 2013), as a predictably bad decision.  We also feel a little like the venerable plaintiffs’ expert from whom we first heard the term retrospectoscope,” smug, in our office, blathering on about something with which we were not involved.  Unlike him, however, we will resist the temptation to ascribe motive to why the court ruled as it did, as we believe motive is a subject for juries/readers to decide not for experts/blawgers to opine.

            When testing the adequacy of a complaint, the “facts” at issue are those plaintiff chooses to include in the complaint, how she chooses to include them.  As such, it is not surprising that the facts analyzed in Williamson seem incomplete to anyone who knows medical device product liability cases.  It seems likely that a knee replacement plaintiff with at least four prior failed replacement surgeries, who was told her only treatment options were amputation or surgery with defendants’ product, had something else going on besides her basic allegation that:  1) the product broke not long after an uneventful implant, 2) then she was assured by defendants’ representative that there had been no prior reports of such breaks, 3) then she proceeded to have the broken product explanted and replaced with another of the same product, and 4) then the second implant also broke.  The court cannot assume additional facts to help direct its analysis to a different result, though.  (Hint, hint.)  So, the court analyzed a motion to dismiss every count other than failure to warn and loss of consortium.  The central allegations were of “manufacturing defect and/or design defect”:
(1) it relies exclusively on “the fixation in place of the femur and tibia bones”; (2) the knee revision system “is not properly designed or constructed to constrain motion between the femur and tibia bones so as to promote fusion by compression of the bones”; and (3) motion by the bones causes excess force on the device, leading to breakage and injury.
Id. at **6-7.  Plus, plaintiff’s review of defendants’ website before the first surgery and discussions with their representatives before the second allegedly created liability under theories of fraud, warranty, negligent misrepresentation, and deceptive business practice.  As you might expect from this lead in, defendants’ motion was denied across the board, although the court directed plaintiff to flesh out her fraud and negligent misrepresentation counts.

            We have said before that the court’s recitation of the legal standard on motions to dismiss, much like whether a “presumption against preemption” is mentioned, can be a strong indicator of where the decision is heading.  Here, while there is a paragraph on the TwIqbal standards and another on the heightened pleading standards of Fed. R. Civ. P 9(b), the overall message is that the “facts” from the complaint will be “assumed to be true” without any analysis of whether they are “facts,” legal conclusions, or opinions.  Id. at ** 3 & 8-9.  First up was manufacturing defect, which New York law predicates on a product that “was not built to specifications or did not conform to the manufacturer’s intended design.”  Id. at *10.  The court excused the complaint’s lack of alleged “facts about the manufacturing process” based on the practical consideration that plaintiffs have limited knowledge about such details when they sue and the acceptance of “formulaic recitation[s] of elements of a cause of action” (as the court had noted were rejected in Twombly) as if they were proper factual assertions.  Id. at **11-12.  There was apparently nothing in the complaint about plaintiff’s particular devices deviating from specs, even from some pleading in the alternative, despite the New York law the court cited.
            On top of this, the court applied an unlabeled res ipsa loquitur to infer that proximate causation for a manufacturing defect had been pled.  Going well beyond the complaint itself, the court read an allegation that plaintiff thought the first implant broke when she felt sudden pain as enough to “conclude that nothing she or her physician did caused the knee devices to break.”   Id. at *13.  This, in turn, “suggests that the breaks were due to a flaw in the manufacturing process, as opposed to some mishap in surgery or an expected risk of using the device.”  Id. at **13-14.  We just felt a sudden pain in our head as we tried to follow that bit of reasoning.  That must make the reasoning defective.  The sound bite that “[i]t is sufficient at this stage that Plaintiffs have alleged that the two knee devices broke, and that nothing [plaintiff] did caused the break” certainly makes a mess of causation, to say nothing of proximate causation.  Id. at *14.  Even if defendants had been successful in their motion on the manufacturing defect claim, the plaintiff would have been allowed to amend and most of her claims would have been unaffected.

            The court next addressed the argument that plaintiff’s design defect claims should be dismissed under comment k.  In a proper show of Erie restraint, the court noted that New York courts had not adopted the rule that all medical devices (or Class II devices like the one in this case) are “unavoidably unsafe,” so it could not determine if comment k applied on a motion to dismiss.  Id. at **17-19.  Whether considered an “exception” or a “defense”—the court called it both—comment k was not implicated by the allegations plaintiff put forward.  (The court afforded the same analysis to implied warranty, which really should be subsumed within strict liability design defect.)  The court followed this with some brutally bad dicta that comment k would have been unavailable because plaintiff had an unchallenged warnings claim.  This is based on a perverse, although not unprecedented, reading of the comment’s language that an unavoidably unsafe product that is “properly prepared, and accompanied by proper directions and warning, is not defective, nor is it unreasonably dangerous.”  While the comment language could be clearer, Restatement (Second) § 402A sets forth three types of defect triggering strict liability:  warnings, design, and manufacturing.  The “incapab[ility]”  of unavoidably unsafe products “of being made safe for their intended and ordinary use” relates to how the products are designed.  Any product can be defective if it lacks proper warnings or is not made to spec.  So, having an allegation about the product’s warnings should not affect whether the product is unavoidably unsafe and excepted from a design defect claim.  Says us, at least.
            Next up were the claims for express warranty, negligent misrepresentation, fraud, and deceptive business practices, which involve somewhat different elements but were all based on the same set of communications, through the defendants’ website before the first implant and calls and emails with defendants’ representatives before the second.  Of course, these are all claims being made for a medical device that was available only through a learned intermediary.  The court completely ignores whether any of these informational claims can be made for a prescription medical device or hinge solely on allegations about information relayed to plaintiff and other potential patients.  Yet, the learned intermediary doctrine adopted by New York courts is never mentioned in the decision.  That is quite an omission.

            Furthermore, the court’s analysis of each informational claim was hampered by its failure to analyze each implant separately.  Clearly, plaintiff cannot recover for her injuries related to the first implant based on her alleged reliance on misrepresentations in her direct contact with defendants’ representatives a year later.  This truism can play out in partial summary judgment, motions in limine, jury instruction, or in evidence and argument to the jury, but, on a motion to dismiss, dealing with the implants together just made a sloppy record.  So did the lack of detailed allegations about the content of the website allegedly reviewed and relied upon by plaintiff in deciding to have the first recommended surgery with defendants’ product—as opposed to doing nothing or having her leg amputated—which the court later excused because “the pages that were on the website at the relevant time . . . are no longer available.”  Id. at *32.  Unlike knowledge of the details of defendants’ manufacturing of the device, the plaintiff can and should be expected to allege what she personally considered and relied upon in having her surgery with defendants’ product.  Allowing plaintiff to be vague about what was warranted or represented to her made it easier for the court to find that these statements were the requisite “affirmation[s] of fact or promise[s],” as opposed to non-actionable “opinion[s] or commendation[s].”  Id. at 23.  Even the reasonableness of reliance is not an element of express warranty (as opposed to negligent misrepresentation) and none of this was alleged in the complaint, the court detoured to its views about what matters to patients:
You do not need to be a medical professional to know that a patient’s primary concern going into any surgery involving the use of a medical device is the safety of the product, and it is entirely reasonable that patients would rely upon such statements in deciding to go forward with surgery.  Patients do not seek information about the safety of a medical device in order to get a mere opinion; they want to know whether the device is actually safe.
Id. at *24.  We could go on for a while about these two sentences in the context of a case where a plaintiff faced amputation before her first surgery, but it seems to us that the express warranty claims were not going to be dismissed given the allegations about the plaintiff’s reliance on direct statements made to her by defendants’ representatives before the second surgery.  The court actually used plaintiff’s reliance on those statements to “reinforce[] the sufficiency of her pleading” as to the first surgery.  Ouch.  That pain in our head in back.

            This may not even be the worst part of the decision.  Negligent misrepresentation claims under New York law require a “special relationship” between the plaintiff and defendant, which was going to be the only potential sticking point given the court’s analysis of other claims.  Duties created by a special relationship allow an end-run on the learned intermediary doctrine, whereby the defendants’ duty to provide adequate information about the product’s risks runs to the surgeon.  The defendants here had no direct financial relationship with plaintiff and did not participate in her medical care.  They only provided information about their product, allegedly false, on their website and responded to unsolicited inquiries from plaintiff.  The court held that, “after [plaintiff’s] first surgery, she established a relationship of trust with Defendants,” which was enough to defeat the motion as to both surgeries based on curious Second Circuit law excusing inadequate pleading of a special relationship where the plaintiff has pled that the defendant had special expertise about its product and understood that someone might rely on its statements.  Id. at *30 (“a sparsely pled special relationship of trust or confidence is not fatal to a claim for negligent misrepresentation where the complaint emphatically alleges the other two factors”) (internal quotations omitted).  We foolishly thought separate elements of a cause of action were separate.  The larger problems, though, are allowing vague allegations about the content of the website to suffice based in part of the subsequent direct contact—despite the heightened pleading requirement acknowledged by the court—and failing to understand why the direct contact occurred.  A device manufacturer is required by regulation to process and report complaints about its products.  A patient (potential plaintiff) who contacts the manufacturer to say she was injured when a knee implant failed will have direct interaction with the potential defendant.  If she asks questions about the product’s performance or complaint history, then she will get “representations” on those subjects.  Under this Hobson’s choice, a manufacturer who complies with its complaint handling responsibilities will have a special relationship with potential plaintiffs that might extend backwards to its public statements about its product.  After all this, the court gave plaintiff an instruction to amend her complaint as to negligent misrepresentation and fraud—which got a “ditto” analysis—to include the detail the court had excused her from including in the first place.
            Lastly, the court allowed the statutory deceptive business practice claims to stand.  The only issue was whether the website was “consumer-oriented.”  Based on the same vague allegations about the website’s content—remember, plaintiff was excused from knowing the details of what she allegedly relied upon—the court deduced it was “clearly consumer-oriented” and made statements about the product’s safety without disclosing its risks.  Id. at *35.  We are not sure where the court got that information, as no cites to the complaint are given and the court was not allowed to consider extraneous material.  We did look at the 2013 version of website for the product, at least the portion you can view without entering your medical license number and hospital affiliation.  It included terms like “arthrodesis,” “likelihood and speed of union,” “transversal locking screws,” and “fatigue strength laboratory tested to 10 million normal gait cycles.”  Those may be consumer-oriented if the consumer is an orthopedic surgeon.  The website also disclaimed that the manufacturer “does not dispense medical advice and recommends that surgeons be trained in the use of any particular product before using it in surgery” and noted surgeons “must always rely on his or her own professional clinical judgment when deciding whether to use a particular product when treating a particular patient.”  Somehow that website content was not analyzed in the opinion, probably because plaintiff did not include it in her complaint.  Maybe she will add that when she amends her complaint to add the basic details about the representations she allegedly received.

No comments: