Photo of Eric Alexander

It is common to characterize an opponent’s discovery request as a “fishing expedition.”  Of course, all discovery seeks something it may not find—much like our own experience with actual fishing for, you know, fish.  However, Federal Rule of Civil Procedure 26 (like various state court equivalents) requires that “the discovery appears reasonably calculated to lead to the discovery of admissible evidence,” focusing on the level of expectation that the discovery will turn up something relevant to the issues in the case.

Throwing a line in a dirty pond in Nebraska may be reasonably calculated to find you some fish, but not salmon.  In prescription drug product liability cases, it is the prescribing physician whose actions and testimony help define what marketing evidence is admissible.  (OK, so we got tired of talking about fish.)  When the prescriber has already testified what she did and did not rely on in making her decision to prescribe to the plaintiff, there is a good indication of what discovery on marketing practices will or will not be reasonably calculated to lead to the discovery of admissible evidence.  Like a big sign that says “there are fish here” or “there are no fish here.”  (We gave the fish thing another shot.)

Peetz v. Genentech, Inc., No. 8:10-CV-297, 2013 U.S. Dist. LEXIS 126803 (D. Neb. Sept. 5, 2013), is a case where the court ignored such a sign in the context of a motion for protective order on 30(b)(6) depositions on general marketing practices.  The plaintiff used defendants’ immunosuppressant drug for something called thrombotic thrombocytopenic purpura (TTP), an off-label autoimmune disorder—the drug was approved only for lymphoma—and claimed to have developed a paralyzing viral infection.  He claimed that defendants had not adequately warned in their marketing materials, and, presumably, label, that the immunosuppressant drug “carried a risk of severe immunosuppression.”  Id. at *6.

We pause from our discussion to note two things.  First, the plaintiff’s injury is surely tragic, as was the condition for which he sought medical treatment.  TTP apparently has up to a 95% mortality rate if left untreated and can produce paralysis itself.  There is nothing in the discovery rules that provides that plaintiffs with more significant injuries get more leeway on discovery, but that is what happens. Second, it is hard for us to understand a claim against a prescription drug manufacturer based on failure to warn that the intended therapeutic effect could have other consequences for the patient.  Explaining such things to the patient is entrusted to treating physicians and their individualized judgment about their patients.  The reasoning for this is at least as strong as why treating physicians are obligated to explain the risks—unintended effects—of a drug being prescribed.  As we have said a few times, other than spelling out Indications and Contraindications in the label, drug companies are not required to explain to licensed physicians which patient should get its marketed prescription drug.  As will be discussed below, a focus on what is “omitted” from marketing materials makes even less sense in a case where the claim is failure to warn doctors of the risk of successful treatment.

We return from our segue to the issue decided in Peetz, first by the magistrate judge and then by the district judge.  Plaintiff, in typical fashion, served three notices for representative depositions on various aspects of the marketing of the drug.  Because the prescriber had testified (the opinion says “said”) that “she had not been exposed to that marketing before prescribing” the product to the plaintiff, the defendants moved for a protective order.  Id. at *4.  Without any possible link between the general marketing of the product and claims in the case, producing witnesses on the general marketing of the product would be unduly burdensome.  The court never cites Rule 26—from which both the “reasonably calculated” and “undue burden” standards derive—or any case law about the Rule.  Instead, the court created a new “possibility” standard for discovery—if it is possible that “relevant evidence” can come from the discovery sought, then the discovery should proceed.
While we do not see that facts here satisfied even this standard, provided that the burdens were considered correctly, we do see how the posture may have contributed to the result.  This was, after all, an appeal of a magistrate judge’s decision.  The magistrate had allowed the depositions to proceed, but only as to the marketing of the drug for the particular indication for which plaintiff had used it, an off-label indication.  Then the plaintiff appealed that there should be no such limit and it does not appear that the defendants cross-appealed to block the depositions completely.  The standard of review on a magistrate’s discovery ruling allows reversal of a ruling that is clearly erroneous or contrary to law.  Id. at *8.  So, the district court did not take a de novo look at the original motion and discovery sought.
If it had, it might have seen that the prescribing physician’s testimony negating reliance on marketing efforts should have precluded broad discovery on generic marketing.  Just yesterday, we went on at length on primacy of direct evidence from a prescribing physician on her decision-making with a particular patient over indirect evidence of influence of all physicians on general prescribing practices.  The Eighth Circuit and district courts within it have followed the majority position in rejecting that indirect evidence of proximate cause for failure to warn can suffice in cases involving prescription medical products.  See, e.g., In re St. Jude Medical, Inc., 522 F.3d 836, 839-40 (8th Cir. 2008) (rejecting generalized proof of causation where the defense could challenge “the reliance or non-reliance of individual physicians and patients); Bruzer v. Danek Medical, Inc., No. 3-95-971/RHKJMM, 1998 WL 1048225, *7 (D. Minn. Oct. 1, 1998) (rejecting proximate cause based on allegedly misleading medical education on off-label use without evidence of reliance by the implanting surgeons).  This means that plaintiff’s argument that he should get discovery on defendants’ marketing practices because it might have “indirectly[] influenced the physicians who participated in the decision to prescribe [the drug] to the plaintiff” should have been rejected.  Admissible evidence must make a fact of consequence in the case more or less likely.  Fed. R. Evid. 401-02.  In a prescription drug warnings case, proximate cause—whether an adequate warning would have avoided the claimed injury—is not more or less likely based on indirect evidence of general marketing evidence unless there is some link through the prescribing physician.  Here, the link had been broken so the discovery sought was not reasonably calculated to lead to the discovery of admissible evidence.  That is our analysis based on the law, at least.

The court’s analysis was different.  As noted, the magistrate had limited plaintiff to depositions on marketing efforts toward TPP, the off-label condition for which plaintiff was prescribed the drug.  We assume, as there is no discussion of it in the decision, that these would have been very brief depositions per the magistrate’s guidelines.  What drug company not only markets its drugs for off-label use but will put up a representative to detail such efforts in deposition?  But, ignoring that indirect evidence is irrelevant, general marketing efforts can downplay the risk plaintiff claims manifested itself with his injury. Efforts to tout that an immunosuppressant can effectively immunosuppress probably cannot really downplay the chance that you can get immunosuppression, but that is more of a merits issue.  As we have said, the issue was whether the discovery sought was reasonably calculated to lead to the discovery of admissible evidence.  The court held that “discovery into [defendants’] marketing efforts, beyond the use of [the drug] to treat TTP, may produce relevant evidence” and “to the extent that [defendants’] promotional efforts may have conveyed the idea that [the drug] was safe for long-term use, they might prove relevant to the plaintiff’s claim that [defendants] concealed a risk of immunosuppression.”  Id. at *7 (our italics).  That falls short of “reasonably calculated to lead to the discovery of admissible evidence” in two ways.  First, “may” and “might” are just above “cannot.”  Second, the hypothetical “relevant” evidence on marketing cannot be admissible evidence on marketing unless we ignore that whole indirect evidence issue.  Together, this is closer to fishing than what Dexter Morgan does on his boat.

What really gets us, though, is the court’s footnoted suggestion as to what the allegedly relevant evidence might look like.

That idea [of the drug’s safety] might be conveyed through commission or omission.  If the long-term use of [the drug] was known by [defendants] to be dangerous, then any promotion that failed to explain that risk might be seen as a missed opportunity to warn potential users.  An article that said, “[the drug] controls the symptoms of arthritis, but, oh, by the way, it might kill you” would seem unlikely to escape the attention of a physician researching the drug, even if the physician was interested in treating a different condition.

Id. at **7-8 (the court’s italics).  Let us see if we can unpack this briefly.  Setting aside both the court’s misplaced attempt at humor and the FDA, which is usually a bad idea in drug cases, we have:  (1) a court advocating broad discovery on marketing of a drug, (2) prescribed for an off-label use by (3) a physician who disavowed prior exposure to any of defendants’ marketing efforts, (4) on the hopes of discovering information provided to (5) other physicians that (6) did not explain that suppressing a patient’s immune system for a particular hoped-for therapeutic benefit could have other consequences.  Even in cases without an allegedly inadequately warned of risk that is close to or the same as the hoped for benefit, it makes much more sense to look to what the physician actually considered in prescribing to the plaintiff.

Here, even if the prescriber had read the hypothetical article that failed to say “oh, by the way, it might kill you,” we strongly suspect that she would have testified that she gained nothing from the absence of those words and that she would have lost her medical license if she did not know immunosuppression therapy could make it harder for the body’s immune system to fight infection.  That suspicion is reasonably calculated based on hundreds of depositions of learned intermediaries, not just some abstract recognition of a possibility.  We would not want to make up some fish story.