Interesting Application of State of the Art Principles

There’s a lot going on in Wendell v. Johnson & Johnson, 2012 WL 3042302 (C.D. Cal. July 25, 2012), but much of it depends on very case-specific facts about a particular prescriber’s knowledge, and some of the rest was put off essentially as premature.  There is one aspect of the Wendell decision that merits our attention, however, and that’s the court’s application of state of the art principles in the context of a defendant’s sale of its proprietary rights to a drug.


That’s significant because, while the opinion doesn’t mention it, the identities of some of the defendants suggest that Wendell involves (at least partially) generic drugs and thus generic preemption.  With generic plaintiffs looking for other defendants after PLIVA v. Mensing, 131 S. Ct. 2567 (2011), this situation might come up again for our branded clients.

Here’s what went on in Wendell:  At various times the plaintiff’s decedent (who was chronically ill with inflammatory bowel disease) underwent treatment with different combinations of three drugs.  The decedent allegedly died from an adverse effect of one or all of these drugs.  2012 WL 3042302, at *1.  In Wendell the alleged injury was a type of cancer, but for purposes of this post, it could have been any newly discovered risk.

It’s not explicit, but the relevant defendant appears to be the original branded manufacturer of one of the drugs, but who left the market in July, 2003, after the onset of generic competition.  Id.  Plaintiff may well have used the branded drug at some point, since he started taking that product in 1999 (we’re not sure exactly when generics entered the market).  Id. at *2.  However, the first case report of the type of injury at issue in the litigation, in connection with the type of combination treatment the decedent received, was not published until February, 2005.  Id. at *3.  This risk led to the addition of a black box warning in May, 2006.  Id. at *4.

The branded manufacturer successfully raised the state of the art defense.  The risk for which it was being sued was not discovered until a year and a half after it left the market.  Wendell was decided under California law − meaning that the branded manufacturer was at least potentially subject to liability under Conte v. Wyeth, Inc., 85 Cal. Rptr.3d 299 (Cal. App. 2008) − although as we have noted, Conte itself has probably been impliedly overruled by the California Supreme Court.  However, Conte aside, the law in almost every state is the same (at least as to prescription medical products) concerning the sort of after-discovered risk at issue in Wendell − “Drug manufacturers need only warn of risks that are actually known or reasonably scientifically knowable.”  2012 WL 3042302, at *12 (quoting Carlin v. Superior Court, 920 P.2d 1347, 1354 (Cal. 1996)) (emphasis original). Here’s another nice excerpt from Carlin on state of the art:

We reiterate that strict liability for failure to warn is not absolute liability. . . .  [D]drug manufacturers are not strictly liable for a risk that was not known or reasonably scientifically knowable.  In this context, it is significant that the FDA precludes drug manufacturers from warning about every conceivable adverse reaction; they may warn only if there exists significant medical evidence of a possible health hazard.

920 P.2d at 1352.

Apparently, the first known incidence of the risk in question actually occurred before the defendant’s withdrawal from the market, but critically nothing was either been published or reported to the manufacturer or to the FDA until the 2005 article appeared in the literature.  Wendell, 2012 WL 3042302, at *12.  Why it took so long to surface, we don’t know, but not even the plaintiff argued that this delay was the defendant’s fault.  So, as a practical matter, the risk was “unknowable” at the point when the defendant left the market.  Thus, summary judgment was appropriate:

Plaintiffs have not presented evidence that the risk was actually known or should have been known by the scientific or medical communities of which [defendant] is a part. . . .  [I]nformation concerning the occurrence of [the risk] in connection with [the defendant’s drug] was not reported to AERS or discussed in the medical literature until after [defendant] ceased to distribute the drug.  Furthermore, drug manufacturers are not required to warn of every conceivable adverse reaction.  Thus, [defendant] cannot be held strictly liable for failure to warn. . . .

Likewise, Plaintiffs’ negligence claim requires them to prove that [defendant] did not warn of a particular risk for reasons that fell below the acceptable standard of care; i.e., what a reasonably prudent manufacturer would have known and warned about.  Plaintiffs have presented no evidence, expert or otherwise, indicating that a reasonable manufacturer would have been in a position to discover the case that [was] reported, prior to [the date defendant left the market].  Thus, Plaintiffs cannot prevail on their negligence claim.

Id. at *12-13.

The situation in Wendell is not at all uncommon.  Branded manufacturers frequently sell their rights to drugs and exit the market after generic competition begins − and generic drugs enjoy the Mensing preemption defense.  So plaintiffs are going to be looking around for other entities to sue.  Conte liability is one example of that, and one of the many problems with that theory is its lack of any temporal limit.  So Conte claims are likely to arise well after branded defendants have left the market.  Besides Conte, branded manufacturers are likely to be sued no matter how remote actual use of their products is from onset of injury, as was the case in Wendell.  The state of the art defense is one weapon we have to defeat allegations of remotely related liability, and Wendell shows what the defense can do where the cut-off date is the defendant’s ceasing manufacture.

Significant State of the Art Win

Although we aren't familiar with all the details, we know that state of the art - that is when a particular risk was scientifically knowable - is currently a big deal in Pain Pump litigation.  That litigation isn't MDLed so it's being fought on a case-by-case basis in courts across the country.  Given this fact, we thought we'd point out to our readers the recent Ninth Circuit decision in Rosa v. Taser International, Inc., No. 09-17792, slip op. (9th Cir. July 10, 2012), in which the court affirmed (under California law) that the defendant (not a drug company) could not be liable as a matter of law (summary judgment) because the particular risk was not scientifically knowable at the time.  The court's refusal to give excessive credence to hints and untested hyopotheses in medical articles should be useful to pain pump defendants.

Breaking News - Pain Pump State Of The Art Affirmed

As anyone following the Pain Pump litigation knows, the state-of-the-art issue - whether it was possible for the defendant to know of the alleged risk of its products given the state of scientific knowledge at the time of a plaintiff's surgery - is critical.  Well, today the first court of appeals to consider that question has affirmed summary judgment.  Rodriguez v. Stryker Co., No. 11-5335, slip op. (6th Cir. May 21, 2012) (applying Tennessee law).

The court in Rodriguez examined some 13 articles published over the course of seven decades that the plaintiffs claimed that established a jury question on the state of the art (that is, that the risk was knowable).  It concluded (as had the district court) that these articles proved nothing about this particular device at the time in question:

When all is said and done, not one of [plaintiff's] thirteen articles shows that medical experts understood in 2004 that infusing a joint with [the drug] for two days could cause irreversible cartilage damage.  [Defendant] had no duty to understand what the relevant medical literature did not.

Rodriguez likewise founs that the plaintiff's expert couldn't make anything more out of the same 13 articles than the articles themselves:

[Plaintiff's] theory requires two speculative leaps.  It requires the inference that evidence of harm resulting from other solutions meant that anesthetics would cause the same harm.  And it requires the inference that evidence of transient harm to joints meant that irreversible cartilage damage was likely.  Both are far too conjectural and too many steps removed from the problem that developed.

The court went on to demolish plaintiff's regulatory theory - that because the FDA had denied §510k clearance to the device, that that somehow put the manufacturer on notice of unknowable risks.

The FDA’s action means only that no other device on the market carried that indication for use.  It does not mean that the pump was (or might potentially be) dangerous to use in the joint space.  The 510(k) process does not comment on safety.

Slip op. at 8.  Ironically, the defendants won because of the adverse preemption decision in Medtronic, Inc. v. Lohr, 518 U.S. 470 (1996).  Plaintiffs were thus prevented from having their cake and eating it, too.  That is, plaintiffs who deny preemption because a §510k clearance doesn't address safety, can't play fast and loose by turning around and claiming that lack of §510k clearance has safety connotations.

Rodriguez also becomes the last in a very long line of cases to confirm that an independent "duty to test" does not exist at common law.  Rather, such a duty "collapses into the failure-to-warn claim."   Slip op. at 8.  A plaintiff cannot resort to a "duty to test" to avoid the state-of-the-art defense.  "The law does not require a company to test for hidden risks that neither it nor the medical community had a reasonable basis to suspect."  Id.

Nor can a claim of off-label marketing succeed where plaintiff's other theories have failed.  The common law is unconcerned with mere off-label marketing.  It would be concerned with concealment of risks (regardless of on or off-label), but once again, the state of the art precluded the concealment claim:

[I]t makes no difference that [defendant] knew surgeons would use its pump in the joint space or even encouraged them to do so since [plaintiff] has failed to show that Stryker knew or should have known that the use was dangerous.

Slip op. at 9.

Next, putting state of the art and legal theorizing aside, the claim in Rodriguez failed independently on the facts of the case because the implanting physician knew all about this device (having used it many times) and thus did not rely on any of the defendant's warnings - so the opinion is also a good "already knew" learned intermediary warning causation case.  Slip op. at 10-11 (citing one of Bexis' Bone Screw summary judgment wins).

Finally, we've commented on plaintiffs' attempts in other cases to create some new duty to send a "Dear Doctor" letter in cases where their warning claims fail.  Well, last and least, the plantiff tried that tack in Rodriguez - and got nowhere. 

[Plaintiff] argues that [defendant] should have known that it needed to “revise its instructions or at least circulate a Dear Doctor Letter” when the FDA rejected its requests to approve the pump specifically for use in a joint.  As shown, however, the FDA position did not obligate [defendant] to send any such letter.

Slip op. at 11.  "Dear Doctor" letters are a regulatory invention.  If the FDA doesn't require them in a particular situation, the court in Rodriguez wasn't about to create some new, broader common-law Dear Doctor letter theory.

We can only hope that other courts are as sensible as the Sixth Circuit was in Rodriguez. Slip op. at 7 (emphasis added).  The expert had improperly attempted to substitute "hindsight negligence" for "common law negligence.  Id.Slip op. at 5. Actually, the court had a lot more to say, but that's the bottom line.

The Pain Pump Litigation Pumps Out another Good Decision

To us, sometimes plaintiffs’ lawyers sound like Cary Grant.  You know, “Duty, Duty, Duty.”  (By the way, “Judy, Judy, Judy” is a famous misquote.  Cary Grant never once said it in a movie.  But he did say it here – at the 0:50 and 1:30 marks.)  Plaintiffs’ lawyers are often looking for new duties and how companies failed to satisfy them.  When a traditional duty, such as failure to warn of a risk, doesn’t work, new alleged duties seem to crop up.  Susan Burnett of Bowman & Brooke kindly sent us a recent pain pump decision that’s a good example of this – Todd v. Stryker Corp., 2012 U.S. Dist. LEXIS 60953 (E.D. Cal. May 1, 2012).  And the court was having none of it.

In Todd, the traditional claim that the defendant company failed to warn of a risk of which it knew or should have known didn’t work.  The state of the science at the time that the pain pump was used in the plaintiff didn’t support the risk.  There had only been a tiny number of adverse events, a discussion of them at a medical conference, and a fairly weak set of medical literature.  Id. at *13-15 & *20-12.  And, worse for plaintiff, the company seemed to react responsibly to that information, conducting investigations of its files and the medical literature, reporting events to the FDA and seeking advice from outside experts.  Id. at *15-16.

So what’s a plaintiff to do?  Well . . . create a new duty.  While the FDA had cleared the pain pump’s 510(k) application and thereby allowed it to be marketed, it had denied another 510(k) application (apparently more than once) that sought to market it for the particular indication for which the pain pump was used in plaintiff.  Aha!  While maybe there was insufficient scientific evidence of an actual risk, the company should have warned doctors about this regulatory history.  Then maybe doctors would have some basis to speculate about a risk.

This is just what it sounds like: an attempt to sidestep the science.  But in a failure to warn case the science is not something to be worked around.  It defines the very duty that the defendant is supposed to satisfy.  Regulatory history adds nothing.  There’s a reason that labels include sections for indications, warnings, precautions, and adverse events, and not for regulatory history.  As the Todd court put it, regulatory history is simply not a risk about which a warning should be given:

There is no duty to warn the patient’s physician of a prescription product’s regulatory history, because the history is not a "dangerous propensity" about which an adequate warning must be given . . . . .

Id. at *9-10. 

That makes sense.  As we all know, the FDA denies clearance of 510(k) applications for many reasons.  Maybe the FDA decided that there was no substantially equivalent predicate device on the market, or that additional information was needed to clear the new indication. 

Putting regulatory history in a label would simply invite doctors to speculate on what it means.  That’s the opposite of what labels are for.  Doctors are experts in medicine, not regulation.  And, so, labels are places for scientific and medical information, not regulatory dockets.  (For this reason, the rule in informed-consent cases is that doctors need only inform patients of a drug's benefits and risks, not regulatory status.  See our post on this here.)  If the particular regulatory history in which plaintiff was interested was the result of science that supported the presence of a risk (which apparently was not the case), the science, not the regulatory history, should have been put in the label.  The court properly rejected this new duty. 

Plaintiff didn’t abandon regulatory history so easily, though.  She also argued that the mere fact that the FDA denied clearance for the indication signaled the need to warn of a risk.  But this thinking has the same deficiency as plaintiff’s “regulatory history” argument.  The FDA denies clearance for many reasons.  The denial isn’t what matters.  It’s the science.  And, the science didn’t support the risk. 

Undaunted, the plaintiff tried yet another new duty, one that we've seen tried before: the duty to test.  We already deconstructed here plaintiffs’ past attempts to create a claim out of the amorphous and open-ended concept of duty to test.  We even created a cheat sheet of plaintiffs’ many failed attempts.  Suffice it to say that the Todd court wasn’t buying it either.  It found, as have many other courts, that there is no such thing as an independent tort for failure to test:

When the warnings accompanying a prescription product adequately inform the prescribing physician of dangers inherent in its use, the manufacturer's alleged failure to test that product cannot, by itself, either cause injury or be a source of liability for the manufacturer.

Id. at *10.

Now, the Todd plaintiff’s claims had other, more ordinary deficiencies, including a proximate cause problem because the treating doctor decided to use the pain pump and medication on the basis of his own training and judgment, not any alleged influence by the company.  Todd, 2012 U.S. Dist. LEXIS 60953, at *23-25.  But what makes Todd most interesting, at least to us, was the court’s treatment of plaintiff’s less ordinary claims – the new duties, duties, duties.  The court wasn’t having any of it.

In Colorado, State-Of-The-Art Defense Is Up, Duty To Test Down

Once a mass tort becomes widely publicized, sometimes based on a combination of a bad study, a verdict or two for plaintiffs, internet sites, and late night television ads, lots of plaintiffs come out of the woodwork, including plaintiffs with weak, late, or marginal claims. Those plaintiffs and the clever lawyers who assist them believe that if they just get past the obstacles the law throws in their way – pesky things like the statute of limitations, product identification, causation, and the like – then they will be given the keys to the kingdom, or at least to the defendant’s vault. As a result, in the life cycle of a popular mass tort, courts often see and resolve almost every kind of claim and issue, including issues that have not been litigated in the jurisdiction for a while.

The pain pump litigation is a good example. It has generated tons of opinions, perhaps because it is not an MDL. The latest is Pavelko v. Breg, Inc., No. 09-cf-01461 (D. Colo. Feb. 28, 2011).

Plaintiff Pavelko had shoulder surgery in 2003 and had a Breg pain pump to alleviate post-surgical pain. Thirty-three months later, in March 2006, she went to a doctor complaining of shoulder pain, and he diagnosed her as having chondrolysis. She brought strict liability and negligence claims in 2009, alleging that defendant Breg was liable because it should have known and warned that pain pumps caused chondrolysis.

After discovery, Breg moved for summary judgment based primarily on the undisputed fact that no one reported any potential connection between shoulder pain pumps and chondrolysis until 2006, three years after plaintiff’s surgery. In other words, the state-of-the-art defense, although the court did not use that term. Plaintiff responded with the alleged duty to test. Plaintiff argued that Breg had a duty to disclose risks that would have been disclosed by reasonable testing and submitted a bunch of medical articles supposedly suggesting further tests.

The court smacked down this argument because plaintiff did not submit expert testimony supporting her argument that these articles put a reasonably prudent manufacturer on notice about the need for further testing, and the court’s own review of the articles found nothing to put the manufacturer on notice. Slip op. at 11-12. The court also bounced for lack of supporting expert evidence plaintiff’s claim that Breg’s general testing of the device was inadequate.

Plaintiff tried to get around these problems by arguing that Breg marketed its device for a purpose not cleared by the FDA, based on that popular notion among plaintiffs that allegations of off-label marketing cure all ills. The court disagreed, finding both that plaintiff failed to submit expert testimony supporting this claim (an unsworn report by the ubiquitous Suzanne Parisian about other manufacturers didn’t do it) and the evidence showed that use of the pain pump in the joint space was considered to be an approved use. Id. at 14-15.

All that stuff is good for our side. But the court put a blemish on this little gem by considering and denying summary judgment based on Colorado’s two-year statute of limitations. There was a factual dispute about whether plaintiff was told in 2006 that the pain pump could have caused her chondrolysis, and therefore summary judgment based on actual knowledge was denied. Fair enough. But Breg argued that plaintiff had a duty to ask her doctor about the potential cause of her chondrolysis in 2006. Her doctor testified that, if asked in 2006, he would have told her that the pain pump may have been the cause. The court, however, found that the plaintiff did not have knowledge of facts that would have put her on notice that her chondrolysis was caused by the wrongful conduct of another and therefore denied summary judgment on this ground.

Given that the court granted summary judgment based on the state-of-the-art defense, the court’s statute of limitations discussion was 100% unnecessary to its decision and therefore dicta. Despite the dicta, Pavelko is a good win for the defense based on the established legal principles that one bad apple don’t spoil the whole bunch and two out of three ain’t bad.