Photo of Eric Alexander

We watched some of the recently concluded Winter Olympics.  When the temperature at the Black Sea-side resort exceeded the temperature of Washington, D.C., by 20° Fahrenheit (or 11.1° Celsius), we questioned the “Winter” part of the event’s title in a way we never expected.  Of course, we watched some action from familiar sports, like the hockey showdowns with our once-archrival and our northern neighbor.  We also watched some of the biathlon, a fascinating sport that involves cross country skiing with a rifle on your back and stopping periodically to shoot at targets.  Our initial thoughts on that sport ran like this:  “This is not a sport that people do casually.  Maybe this is to echo the Winter War.  Maybe they could make it a more militaristic triathlon by adding a phase where an edged weapon is used to decapitate a snowman while skiing.  These tortilla chips are making me thirsty.”  This being a blog on drug and device law—as its title hints—and our thirst having been quenched, it occurs to us that our experience with the cases we write about is like watching the biathlon.  (We may have quenched our thirst with a nice winter ale and/or a little Russian water, but we are sober as we write.)  We do not hear much about the exhausting cross-country grind of discovery.  We do not follow all the ups and downs of trial argument and evidence.  We check in, periodically, when the litigants have paused long enough to shoot (at each other, unlike in the Olympic version of the biathlon) and the court issues some ruling explaining the targets, hits, and misses.  Even then, we typically pay attention to the results of the shooting only when the decision is published (or sent to us by a participant) and it seems interesting.  Occasionally, we notice multiple decisions from a single case as it proceeds past discovery, dispositive motions, and pretrial motions, and through trial.  These may not be the highest stakes or most groundbreaking cases, though.  This is somewhat like watching NBC focus on the battle for 15th place in the Men’s Relay, just because the U.S. team was battling.  (Bulgaria, you are the ancestral homeland of the mighty Spartacus.  Need you grab more glory by 0.7 seconds?)

The Guenther case is one of many Aredia/Zometa cases involving allegations about osteonecrosis of the jaw (ONJ) whose decisions have been discussed in our posts.  Guenther, however, has achieved special puffy coat status due to its recurring role in six prior posts.  Over the last year, we have posted on decisions on case consolidationsummary judgment, punitive damages, limiting Dr. Parisian (as part of a compilation), motions in limine, and more motions in limine.  Yet, we know that there were other decisions and developments in the case that shaped its course.  We know from Guenther v. Novartis Pharm. Corp., No. 6:08-cv-456, 2014 U.S. Dist. LEXIS 21182 (Feb. 20, 2014), that its course took the case all the way to trial, where the plaintiff won on strict liability and negligent warnings causes of action and was awarded $1.3 million in compensatory damages, leading defendant to move for judgment notwithstanding the verdict, a new trial, and/or remittitur.  All we know about the trial evidence from the plaintiff, prescribing physicians, other treating physicians, and the experts comes from the decision itself, as the court identifies and characterizes the evidence it wants as it explains its particular findings.  See, this is just like watching biathlon.  No need to analyze the parallels any further.

The JNOV motion challenged whether plaintiff proved that the ONJ warnings were inadequate when the plaintiff took the drug and that such inadequacy caused her injury.  The first part was brushed aside without much analysis.  Plaintiff offered testimony from two recurring experts, Dr. Marx and, of course, Dr. Parisian, that the defendant “knew or should have known of the association between Zometa and ONJ prior to September 2003, when it altered the Zometa labels to first include information about ONJ.”  Id. at *5.  That tells us nothing about whether the warning that was added in 2003 was inadequate when plaintiff took the drug.  We get barely more information from the characterization that there was evidence that some of defendant’s employees “were much less equivocal about the association between Zometa and ONJ than the company was indicating publicly (both via the label and via other communications with the medical community).”  Id.  Without attempting a discourse on Florida warnings law, it is fair to say that prescription drug warnings are adequate or inadequate based on whether they convey the particular risk to physicians consistent with the actual scientific evidence.  That inquiry should not have much to do with how “equivocal” employees of the manufacturer are about scientific evidence in their internal communications.  This is particularly so under strict liability, where a warning could be adequate under objective standards even if many of the manufacturer’s employees subjectively deem it inadequate.  Without more information about the evidence presented or jury instructions, however, we cannot get too exercised about this part of the decision.

The proximate cause analysis, though, does chafe us like ill-fitting ski pants.  Part of the discomfort is because manufacturers can and do win prescription drug cases based on proximate cause for failure to warn.  Sometimes, the analysis is pretty straightforward.  Like in yesterday’s post about a different Zometa case, where the Ninth Circuit found no chance of plaintiff establishing proximate cause where “[t]he record makes clear that the doctor understood the connection between bisphosphonates and the risk of [ONJ], and that in his medical opinion the benefits of the treatment for the patient outweighed those risks.”   The “doctor” in that quote was the prescribing physician.  (We ask you to suppress your inclination to say “duh” at this point.)  Which doctor’s understanding and opinion matters was less clear in Guenther.  The defendant made the usually strong argument that it should win because “neither of the physicians who prescribed Zometa to Guenther testified that they would have declined to prescribe it had they known, at the time, of the association between that drug and ONJ.”  2104 U.S. Dist. LEXIS 21182, *6.  Defendant argued “Florida law requires such a showing to establish probable [sic] cause in a failure to warn prescription drug case.  But this is not an accurate statement of Florida law.”  Id.  Keeping in mind that this is a federal court sitting in diversity, the analysis of what Florida law on proximate cause for failure to warn is—as opposed to is not—was lacking.

That is not to say that the only acceptable framing of the prescription drug failure to warn proximate cause question—like in the jury question posed at the end of this trial but not discussed in the decision—is “did the plaintiff prove that an adequate warning as to the risk of [plaintiff’s injury] would have caused [plaintiff’s prescribing physician(s)] not to prescribe [the drug] to him/her such that her/his injury was avoided?”  Although changing a decision to prescribe to a decision not to prescribe may be the most logical way to satisfy proximate cause, the key is that a different warning to the prescribing physician changes that individual’s behavior and avoids the injury.  It may be that a doctor would have prescribed a different drug but the drug has a similar risk of the same injury (as with many classes of drugs that have been in litigation), and there is no evidence that the plaintiff’s injury would have been avoided.  No proximate cause.  It may be that the doctor would prescribe less of the drug and there is competent evidence that the different dosage or duration would have avoided the injury.  Potentially proximate cause.  It may be that the doctor would have prescribed the same amount of the same drug but told the plaintiff to stop the drug at the first sign of a problem.  With additional evidence that the plaintiff would have stopped the drug and stopping the drug would have prevented her injury, then there might be proximate cause for failure to warn.  There are other permutations, but they all involve a change in the prescribing physician’s behavior from a different/adequate warning and evidence linking that change to the plaintiff avoiding the injury she is suing over.  Keep that in mind; the Guenther court did not.

Rather than analyze what Florida cases on failure to warn claims for prescription drugs have actually required, the court tackled the straw man—snowman?—of whether the cases cited in the defendant’s motion foreclosed the possibility of establishing proximate cause other than through the prescriber testifying “I would not have prescribed the drug to the plaintiff.”  We checked Guarino v. Wyeth, 719 F.3d 1245 (11th Cir. 2013), an appeal from this same court on prescription drug warnings claims under Florida law, and there are decisions from Florida state appellate courts addressing what has to be shown to succeed in a prescription drug warnings claim under Florida law.  These cases not only say that the duty to warn runs to doctors, but that the particular doctor who should be the focus of the inquiry is the one who “weighs the drug’s benefits against its potential harms in deciding whether it is appropriate to the patient’s course of treatment.”  Id. at 1250 (citing Felix v. Hoffmann-LaRoche, Inc., 540 So. 2d 102, 104 (Fla. 1989)).

The Guenther court did not cite Guarino or any of the Florida warnings cases cited in Guarino.  Rather, it argued that the Second Circuit’s interpretation of Florida warnings law in In re Fosamax, 707 F.3d 189, 193 (2d Cir. 2013), that proximate cause requires proof that “a treating physician would have recommended that patient cease taking the drug if a different, adequate warning had been provided” was inapplicable or dicta.  It also argued that the decision of Hoffmann-LaRoche, Inc. v. Mason, 27 So.3d 75, 77 (Fla. 1st Dist. 2009)—an actual Florida court decision—which held that there was no proximate cause where the prescribing physician understood the risk from the label and would not have done anything different with a different label, did not foreclose the possibility that proximate cause might be satisfied if there was evidence that “a different warning could have prevented the plaintiff’s injury in some manner other than by changing the physician’s decision to prescribe – e.g., by prompting the physician to pass along a more detailed warning, or to reduce the dosage.”  2104 U.S. Dist. LEXIS 21182, *8.  Like we said above, we agree that a change in the prescribing physician’s behavior short of not prescribing the drug at all can, in some cases, create proximate cause for failure to warn.

But then Guenther went off the tracks in addressing two Tennessee federal court cases interpreting Tennessee law on prescription drug warnings claims.  In one case, the court found there was a genuine issue as to proximate cause because “[t]he prescribing physician and his nurse had testified that they would have passed any additional warnings along to the patient, and there was testimony that the patient had expressed misgivings about [the drug’s] side effects while taking the drug.”  Id. at *9 (discussing Smith v. Pfizer, Inc., 688 F. Supp. 2d 735, 746 (M.D. Tenn. 2010)).  In the other, the court entertained the possibility that proximate cause could be satisfied by testimony of the prescribing physician that “he had changed his procedures since learning of a link between bisphosphonates and ONJ” but granted summary judgment based on the absence of reliable evidence linking the prescriber’s change to plaintiff’s injury.  Id. at *10 n.5 (discussing Payne v. Novartis Pharms. Corp., 2013 WL 4779571, *9 (E.D. Tenn. Sept. 6, 2013)).  From these cases, the Guenther court distilled that there had been a rejection of the proposition that “the only decision that matters for purposes of proximate cause is that of the prescribing physician.”  Id. at *10.  What?  These cases all focused on the decision of the prescribing physician, which is exactly what Florida warnings cases have done in providing their rationale for adopting the learned intermediary doctrine.  A patient only gets the prescription drug because Dr. Whoever prescribes it to her based on independent medical judgment gained by standing between the manufacturer and the patient.

So, what evidence was there about the learned intermediary in Guenther to satisfy proximate cause?  Nothing that we can see.

[T]here was testimony from [plaintiff’s expert] Dr. Marx that reducing dosages of a drug reduces its side effects, as well as testimony about the cumulative effect of Zometa based on its 11-year half life, and his effort to encourage Novartis to determine whether a lower recommended dosage of Zometa could accomplish the same beneficial results. (Doc. 295-6 at 38-39). There was also evidence that Guenther did not exhibit any symptoms of ONJ until after she had been given numerous doses of Zometa; from this, the jury could have concluded that Guenther would not have developed ONJ at all if she had ceased taking the drug at an earlier time.

Id. at **10-12. The prescriber is missing from this alleged causal chain.  In the context of denying the motion for new trial, we learn that the jury instructions reflected “this Court’s determination that Florida courts, if called upon to decide the question, would determine that a drug manufacturer’s duty to warn runs not just to prescribing physicians, but also to other health care providers in a position to reduce the risk of harm to the patient.”  Id. at *13.  No citation.  No explanation.  No sense.  But this instruction could have been harmless error because defendant “has not even described a plausible scenario, given the facts of this case, in which the instruction could lead this particular jury to reach an erroneous conclusion.”  Id.  How about “the jury found there was proximate cause for failure to warn without any evidence that the prescribing physicians would have done anything different with this patient with different information about the risk of ONJ,” because that seems pretty plausible.

We really should not have been surprised that the decision missed the mark on proximate cause.  It followed the same mistakes we highlighted in our post on the denial of summary judgment:  “To satisfy proximate causation testimony as to the learned intermediary doctrine, some of the plaintiffs’ evidence should at least address . . .  learned intermediary.”  That sentence doubled the number of times “learned intermediary” was mentioned in the post-trial decision.  We did not think that the “intermediary” part of “learned intermediary” was such a hard concept.  The term “intermediary” has a comparable Latin root “intermedius.”  “Inter” means what you think.  “Medius” was a reference to the middle finger.  Raising one’s middle finger as an insulting gesture came into vogue sometime after the Fall of the Roman Empire, so this could all be a big coincidence to Guenther’s counter-intuitive interpretation of the learned intermediary doctrine.

Speaking of gestures, the end of the decision made two in knocking off a total of $288,000 of the award by way of remittitur.  $195,000 of future medical expenses was taken away because there was no evidence of future expenses.  Id. at **14-15.  About another $93,000 of past medical expenses was taken away because the Florida collateral source statute precludes recovery on expenses paid by someone else unless there is a right of subrogation or reimbursement, which plaintiff apparently failed to prove.  Id. at **15-16.  So, plaintiff had just short of $12,000 economic damages and the original $1,000,000 for pain and suffering—which is arguable disproportionate now—and no punitive damages.  One million dollars may still seem like a big award, but there were clearly high costs of litigation this case even before the two week trial where plaintiff called her pricey experts.  We assume an appeal is coming too.  With the costs off the top and then her lawyers taking their cut, this is not much of a haul for the plaintiff, even if she gets to keep it.