Thursday, May 08, 2008

Warning Causation – Greatest Hits

As we’ve mentioned previously, we were both involved in the briefing in Ackermann v. Wyeth Pharmaceuticals, 2008 WL 1821379 (5th Cir. Apr. 24, 2008). That case is an excellent example of how defendants can win inadequate warning cases (which is what probably 90% of our drug/device product liability cases are) under the learned intermediary rule on the ground that the allegedly inadequate warning did not affect the prescribing physician’s decision to prescribe the drug.

As we’ve also pointed out, there have been a lot of defense wins on the warning causation issue recently. We’re not about to repeat everything we said there about how warning causation works in learned intermediary cases – but we will repeat this: we’d much rather have the warning causation question turn on the testimony of an independent, trained professional than on that of a self-interested layperson like the plaintiff who’s suing our client.

What defense counsel wouldn’t? Anyway, the Ackermann victory (which for obvious reasons we’re not at liberty to discuss in any depth until the appeal is completely over) got us thinking about what we’d consider to be some of the “greatest hits” in this area – not just the most recent hits (which are in our prior post), but the best examples – reasonably recent ones anyway – of this summary judgment genre.

So if you’re a defense lawyer (like us) preparing a motion for summary judgment on warning causation where you’ve got favorable prescriber testimony on the “I would have prescribed anyway” issue, here are some cases you might want to look at.

At the top we’d put Motus v. Pfizer Inc., 358 F.3d 659 (9th Cir. 2004). Procedurally, this is the closest thing to Ackermann out there. They’re both anti-depressant suicide cases where, if the court of appeals didn’t affirm on the warning causation issue, it would have to address a preemption defense lurking in the background. The doctor testified that he neither read nor relied upon the defendant’s warnings in prescribing the product. Id. at 661. No matter how strong an alternative warning would have been, it wouldn’t have made a difference since the doctor didn’t read it. Id. at 661-62. Short and sweet. Summary judgment affirmed.

Eck v. Parke, Davis & Co., 256 F.3d 1013 (10th Cir. 2001), is an instructive case because the court affirmed summary judgment despite controlling state law allowing one of those “read and heed presumptions. Id. at 1018-19. This presumption is one of the ways that pro-plaintiff courts have tried to put the thumb on the scale, by switching the burden of proving warning causation from the plaintiff to the defendant. We “presume that the plaintiff would have heeded an adequate warning, had it been given,” some courts say – even though what they’re really “presuming” is that the existing warning is inadequate in the first place, which is what the plaintiff is supposed to have to prove.

Yeah, we know, it doesn’t make much sense, particularly in a prescription product case where the presumption involves a medical professional treating a sick person – but we’ve blogged about that before, so we’re not going to reinvent that wheel either. The defendant won in Eck because the prescriber knew of the claimed risk, and considered it outweighed in his risk/benefit analysis by the risk of the disease that the medication was used to treat. Id. at 1021. This prescriber testimony was uncontroverted, so the defendant won, despite the state-law presumption.

Odom v. G.D. Searle & Co., 979 F.2d 1001 (4th Cir. 1992), also illustrates some of the warning causation options that defendants have. Odom involved a relatively rare injury (sterility) from an IUD. The prescriber testified that he already knew of this risk, and furthermore that the defendant’s product was the best on the market, and thus he still would prescribe it. Id. at 1002. Due to this confluence of “already knew and chose not to warn patient” with “I would still do it today” testimony, summary judgment for the defendant was affirmed for lack of causation by the allegedly inadequate warning. Id. at 1003.

An absolutely essential case if you’re in a jurisdiction that hasn’t definitively determined the heeding presumption issue is Thomas v. Hoffman-LaRoche, Inc., 949 F.2d 806 (5th Cir. 1992). Thomas provides probably the best substantive discussion of why the heeding presumption doesn’t make sense as applied to products with inherent, unavoidable risks. Id. at 812-14. On the warning issue itself, Thomas is a prior knowledge case – the prescriber testified that he already knew about the risk (seizures) and that an additional warning would not have led him not to prescribe the drug. Id. at 817. That testimony destroyed any possibility that a different warning would have changed the result. Id.

An “oldie but goodie” – unilaterally determined by us to be more than 20 years old – is Plummer v. Lederle Laboratories, 819 F.2d 349 (2d Cir. 1987) (applying, oddly, California law). Plummer is a vaccine case, and the physician testified that he didn’t warn about this particular risk (catching the disease from the live vaccine) because it was small and doing so would discourage patients from using the vaccine, and thus obtaining the vaccine’s substantially greater benefits. Id. at 352. Moreover, even if the prescriber’s testimony were discounted, the plaintiff had nothing affirmative with which to prove causation. Id. at 359 (That’s an argument we can make when there’s not a heeding presumption). Hence, judgment for the defendant notwithstanding the verdict.

Another case along the Plummer line is Stanback v. Parke, Davis & Co., 657 F.2d 642 (4th Cir. 1981). Once again, the prescribing doctor said that, although knowing of a relatively rare vaccine-related risk (Guillain-Barre Syndrome), it was not “his practice” to warn about it because the benefits of vaccination far outweighed the risks. Id. at 644. “Whatever may be said about [the prescriber’s] policies and decisions from the standpoint of the patient, it is clear that they precluded [defendant’s] failure to warn from having any effect whatsoever on [plaintiff’s] injury.” Id. at 645. The prescriber’s knowing decision not to give a warning, in short, is not a product liability defendant’s problem.

In the district courts, we’re not going back as far, given the number of decisions, but you’ll want to read Koenig v. Purdue Pharma Co., 435 F. Supp.2d 551, 556 (N.D. Tex. 2006), if you’re in one of those jurisdictions where the law’s unclear about whether there’s a heeding presumption. What we like best about Koenig is that, before the opinion, there was a plausible argument that Texas might apply such a presumption in a learned intermediary case, but after Koenig there wasn’t anymore. See 435 F. Supp.2d at 556-57. That’s certainly the way the Ackermann court viewed it. 2008 WL 1821379, at *7 (citing Koenig on the heeding presumption). The doctor’s testimony that defeated warning causation in Koenig was simple – I knew all about that risk (addiction to a narcotic drug) but decided to prescribe it anyway because I thought the patient needed it. 435 F. Supp.2d at 555.

Vanderwerf v. SmithKlineBeecham Corp., 529 F. Supp.2d 1294 (D. Kan. 2008), is another really good suicide warning case – really good because Kansas is one of those states that has fallen for the heeding presumption, so the affirmative evidence (the prescriber’s testimony) also had to overcome the presumption for the defendant to win. In Vanderwerf the prescribers (there were two of them) gave some testimony that was bad and some that was good, but the good stuff made the bad stuff irrelevant. The key fact was that there was nothing in the patient’s medical history that suggested that he was suicidal. Id. at 1313. That fact meant that prescriber testimony about “increased monitoring” of the patient was irrelevant, because without some reason to monitor in the first place (e.g. a suicide risk), there wouldn’t have been any monitoring, “increased” or otherwise.

Vanderwerf illustrates an important point – plaintiffs love to engage in flights of “what if” fancy in warning causation situations. It’s our job to bring those flights back to earth. Earth, in this case (as in many) is represented by the facts in the medical records. Attention to such facts is what makes lots of the plaintiffs’ flights of fancy untenable. Know the facts, and use them to cut off speculative “might have beens.”

Another really good suicide case is Allgood v. GlaxoSmithKline, 2008 WL 483574 (E.D. La. Feb. 20, 2008). One of the many problems with such cases is that treating psychiatrists are often loath to discuss suicide risks with depressed patients, for fear of being interpreted by such patients as suggesting that course of action. That was the case in Allgood, where the prescriber testified that even “knowing what he knows today” – including various label changes, dear doctor letters, and medical studies – he would not have changed his treatment because he does not warn patients about suicide risks. Id. at *15-17. Allgood illustrates another place for defense counsel to look for favorable causation testimony. There are often affirmative reasons why doctors choose not to give particular warnings about particular risks.

Another possible line of attack is illustrated in Nix v. SmithKline Beecham Corp., 2007 WL 4219157 (D. Ariz. Sep. 5, 2007), reconsideration denied, 2007 WL 4219157 (D. Ariz. Nov. 28, 2007), where the prescriber testified that he still continued prescribing the drug, and still believed it is a safe and effective treatment for condition plaintiff suffered. Id. at *3. Thus even though the prescriber also testified that he would have liked to have known more about the risk in question, ultimately that testimony didn’t matter because there was nothing to indicate that the prescription or treatment would have been different. Id. at *3-4. On reconsideration the court rejected the plaintiff’s attempt to substitute “reasonable physician” testimony in place of what his own prescriber actually did. 2007 WL 4219157, at *2-3. We’ve blogged about why allowing “reasonable physician” testimony to controvert causation is a bad idea before.

We also like Stafford v. Wyeth, 411 F. Supp.2d 1318 (W.D. Okla. 2006), because it’s another of these affirmative reason not to warn cases. The case involved fen-phen, a weight loss drug. The risk in question was potentially fatal heart disease. The doctor, however, testified “unequivocally” that the risk to the patient’s health (including her heart) from her obesity was significantly greater than the relatively remote risk from the drug. Id. at 1321. Even knowing everything he later found out about the drug’s risks, the prescriber reiterated that he still would prescribe drug to this patient. Id. Even in the face of a heeding presumption, this testimony defeated causation. Id. The “knowing everything you now know, would that have made a difference” question can be a powerful way of defeating warning causation in a learned intermediary case. Stafford also rejected the ploy of attempting an end run around what the prescriber actually did with “reasonable physician” expert testimony about what somebody else might have done. Id. at 1322.

The defendant in Madsen v. American Home Products Corp., 477 F. Supp.2d 1025 (D. Mo. 2007), defeated causation when the prescriber testified that he continued to prescribe the drugs in question even after he learned about the increased risk that the defendant had not warned about. Id. at 1036. The only reasonable inference was that if the doctor had been warned about the same risk at the time he treated plaintiff, the prescription decision would have been the same. Id. at 1-36-37.

In Ferguson v. Proctor & Gamble Pharmaceuticals Inc., 353 F. Supp.2d 674 (E.D. La. 2004), the physician had already looked into the drug risk in question and determined that it was “extraordinarily rare.” Id. at 676. In light of that, he testified that he would still prescribe the drug for the patient based upon his overall risk/benefit analysis. Id. at 677. The court granted summary judgment “[b]ased upon [the prescriber’s] representation that he would have prescribed the drug knowing of the risk of developing the condition of which plaintiff suffers….” Id. at 679.

In Miller v. Pfizer Inc., 196 F. Supp.2d 1095 (D. Kan. 2002), aff’d on other grounds, 356 F.3d 1326 (10th Cir. 2004), still another suicide case, the prescriber knew of reports of suicidal ideation, but at the time he prescribed the drug he believed it was a safe and effective treatment. Indeed, he testified that he “knew more” about the drugs than anyone the defendant could send to visit him. Id. at 1100. Even if he had received the additional warnings urged by plaintiffs, he would have prescribed the drug anyway. Id. at 1101. That testimony “br[o]k[e] the causal chain as to [defendant’s] allegedly inadequate warnings” and justified summary judgment. Id. at 1130.

Several prescribers in several cases testified in In re Norplant, 955 F. Supp. 700 (E.D. Tex. 1997), aff’d, 165 F.3d 374 (5th Cir. 1999), that none of the warning changes that subsequently appeared on the defendant’s labeling would have led them not to prescribe the drug to the plaintiffs. Id. at 711 & n.63. They also testified that they continued to believe the drug was safe and effective and continued to prescribe it. Id. In all the cases, the court granted summary judgment because this testimony precluded proof of warning causation. Id. at 711.

In Woulfe v. Eli Lilly & Co., 965 F. Supp. 1478 (E.D. Okla. 1997), still another suicide case, the prescriber already knew about the claimed risk and continued to prescribe the drug for patients, like the plaintiff, who had not reported suicidal thoughts. Id. at 1481. Since the plaintiff did not exhibit those signs, an additional warning would not have brought about a different treatment. Thus, the court granted summary judgment. Id. at 1482-83. Expert testimony that a “reasonable physician” would have done something different was rejected because it “ignore[d] the conduct of the actual prescriber. Id. at 1484.

Finally, a few relevant state court decisions. In a case near and dear to Bexis’ heart (because he worked on it), in Lineberger v. Wyeth, 894 A.2d 141 (Pa. Super. 2006), the prescriber testified that “I still would have prescribed” the drug even if the warning in question had been included. Id. at 150. The plaintiff’s predictable testimony that she would not have taken the drug if that same warning had been passed along to her was therefore immaterial, since the no different warning would have reached her. Id.

And a couple of Texas cases: In Ethicon Endo-Surgery, Inc. v. Meyer, ___ S.W.3d ___, 2007 WL 4462713 (Tex. App. Dec. 20, 2007), the prescriber knew all about the risk from previous experience, and could not affirmatively testify that an additional warning would have changed anything. Id. at *4. That, the court held, precluded the plaintiff from proving that a different warning would have made a difference in the result, id. at *5, since the only testimony the plaintiff had to establish causation was “speculative.” Id.

In Guzman v. Synthes, 20 S.W.3d 717 (Tex. App. 1999), there was affirmative prescriber testimony that at the time he treated the patient, he knew about the risk in question (fatigue fracture), that he would not have heeded any additional warnings about that risk, and that he would have treated the patient in the exactly the same when he testified as he did at the time of surgery. Id. at 720. The defendant was therefore entitled to judgment notwithstanding the verdict. Id.

In light of all these cases, and lots more we didn’t discuss, our advice to defense counsel seeking to defeat drug warning claims on causation grounds would be well advised to seek the following testimony from prescribers:
  • That the physician already knew about the risk from his/her prior experience and training, and knew more about the risk than any sales representative could have known.
  • That because the prescriber knew all about the risk, s/he did not need and would not have paid attention to, additional warnings.
  • That the prescriber still believes in the product and still prescribes it today to patients in the same position as the plaintiff.
  • That for one reason or another, the prescriber would not have passed an additional warning along to the plaintiff even if it had been included in the product’s labeling.
  • That, in the prescriber’s individual risk/benefit calculus for the plaintiff, s/he concluded that the benefit still outweighed the risk, even with new information.
  • That the prescriber simply didn’t read or rely upon the defendant’s warnings.
  • In a non-heeding presumption case, that the prescriber can’t say one way or the other whether an additional warning would have made a difference.

Also, in taking testimony and collecting other evidence from prescribers designed to set up a dispositive motion on warning causation, keep the following points in mind:

  • If the plaintiff gets some adverse speculative testimony from the prescriber, go to the specific facts of the case – in the medical records – to show that the speculative situation didn’t occur in this particular case.
  • Don’t let plaintiffs get away with arguing a heeding presumption unless you have to, and if there is a presumption, good affirmative testimony will rebut it anyway.
  • If there’s not a presumption, then if the plaintiff can’t get affirmative causation testimony from the prescriber, plaintiff loses.
  • Don’t let plaintiffs get away with trying to replace the prescriber’s actual testimony with some hypothetical “reasonable physician” (one of their own hired experts) who was never part of the plaintiff’s treatment.

1 comment:

Anonymous said...

Great post...a true service...thank you.
-Young Defense Atty