For those of us who spend large chunks of our professional lives in mass tort MDLs, expressions like “settlement inventory” are ubiquitous. It is too easy to lose sight of the fact that these expressions put the rabbit in the hat, assuming that resourceful plaintiff lawyers will necessarily queue up for settlement payments at the end of the day and disregarding the fact that plaintiffs are supposed to be able to prove injury and causation before money changes hands. And so we were especially pleased to read last week’s decisions in the Lipitor MDL, in which the cautious and relentlessly thorough MDL judge granted defendant Pfizer’s omnibus summary judgment motion, dismissing all but a single newly-filed case, because the plaintiffs could not meet their burden of proving that Lipitor caused their diabetes. (There are two decisions relating to different dosages of Lipitor. You can read them here and here.)
The Lipitor MDL, created in February 2014, centralized cases in which female plaintiffs alleged that Lipitor – prescribed in dosages of 10 mg., 20 mg., 40 mg., and 80 mg. – caused them to develop Type 2 diabetes. In March and April of 2015, as trials of two bellwether cases approached, the plaintiffs served common expert disclosures of general causation experts, followed, in May and June, 2015, by specific causation expert disclosure in the two bellwether cases. After full expert discovery, Pfizer moved to exclude all of these experts. Ultimately, the court excluded the plaintiffs’ expert testimony on general causation with respect to Lipitor dosages of 10, 20, and 40 mg. The court also excluded the testimony of the bellwether plaintiffs’ specific causation experts, finding that the experts failed to rule out other possible causes of diabetes and relied impermissibly on temporal relationships, and granted summary judgment for Pfizer on the claims of both bellwether plaintiffs. (You can read our posts on these Daubert decisions here and here.)
However, the court left one general causation “window” open, based on one study finding that it was more likely than not that patients with certain characteristics who took 80 mg of Lipitor would not have developed diabetes in the absence of Lipitor. Thus, though neither of the bellwether plaintiffs had the requisite characteristics, it was possible that a plaintiff existed who took Lipitor at the 80 mg dosage and who could proffer a specific causation opinion that would survive Daubert.
But, in a January 2016 hearing, the plaintiffs’ lead counsel confirmed that no plaintiff in the MDL satisfied these criteria. With the agreement of counsel for both sides, the court issued an order to show cause to see if any plaintiff could differentiate her case in a manner that would survive Daubert based on the decisions that had already been issued. If no such plaintiff came forward, the court would grant summary judgment in all of the cases pending in the MDL. The court’s order required any plaintiff who “disputes the position taken by Plaintiffs’ Lead Counsel and asserts that her case can survive summary judgment . . . [to] provide notice to the court within 15 days of this order and set forth with specificity how her case is distinguished from the Court’s [earlier Daubert rulings]. The Court will then promptly set a schedule in each case for identifying expert witnesses.”
Eleven months later, not a single plaintiff had come forward in response to the order to show cause. Instead, in opposition to Pfizer’s summary judgment motion, the plaintiffs argued that it was “theoretically possible that some unidentified plaintiffs may have some unidentified circumstantial non-expert evidence of specific causation.” In response, the court gave the plaintiffs yet another opportunity to come forward with any cases that could survive summary judgment. Again, not a single plaintiff came forward. Instead, the plaintiffs filed an omnibus response arguing that no plaintiff other than the two bellwether plaintiffs had had an opportunity to “hire experts” or “prepare expert reports.”
Obviously, the court rejected this argument, given the multiple previous opportunities the plaintiffs had not seized. The court did, however, provide the plaintiffs with a final opportunity to defeat summary judgment by submitting non-expert evidence meeting their burden of proving causation. Two sets of plaintiffs came forward. One group of plaintiffs submitted nothing but their Plaintiff Fact Sheets and some of their medical records. They contended that they were diagnosed with diabetes after taking Lipitor and they lacked some risk factors for diabetes, though they had others. At oral argument, their lawyer stated that this “was the best thing [he] could come up with non-expert evidence” and reiterated that the plaintiffs were not diagnosed with diabetes until after they took Lipitor.
The second group of plaintiffs initially submitted no evidence at all. They argued that summary judgment was precluded by the their Fact Sheets and health histories, relevant general causation evidence, and the substantive state law applying to each of their claims. The court entered an order requiring these plaintiffs to submit any evidence they wanted the court to consider. In response, the plaintiffs “literally dumped boxes upon boxes of documents on the Court, with no discernment or suggestion as to which documents they claimed precluded summary judgment.” And, the (saintly) court reviewed all of the documents, “almost all of which were completely irrelevant . . . [including] pictures from colonoscopies, EKGs, and pap smear results.”
Dosages Less Than 80 mg.: General Causation
As explained, the court had previously excluded the testimony of the plaintiffs’ general causation experts as to Lipitor dosages less than 80 mg. However, plaintiffs ingesting lower dosages of Lipitor argued that alleged admissions by Pfizer were sufficient to defeat summary judgment. The plaintiffs’ primary piece of evidence was a one-sentence e-mail by a Pfizer vice president including the comment that “atorvastatin increases the risk of diabetes.” The court held that the email was, “at best, evidence of an association, not causation. An association does not equal causation, and epidemiologists engage in a rigorous analysis of multiple factors to determine whether an association is causal.” (Citations omitted). Moreover, “even if the one-sentence e-mail indicated that [the Pfizer vice president] thought Lipitor caused diabetes, the Court finds such an email could not replace expert testimony when expert testimony is required by substantive state law. (You can read about another decision applying this principle here.) A single statement by a single employee (even a Vice President) in a single email about a single study is not the type of clear declaration” that is sufficient to defeat summary judgment. Nor was NDA data allegedly demonstrating an association between Lipitor and increased blood glucose levels, or a statement on the Lipitor web site about reports of elevated blood sugar levels, or a Japanese label insert acknowledging a possible association. In sum, the plaintiffs had no evidence sufficient to create an issue of material fact on general causation as to 10, 20, and 40 mg. dosages of Lipitor, given the court’s earlier exclusion of the plaintiffs’ general causation expert’s testimony as to these dosages, and plaintiffs taking these dosages of Lipitor could not survive summary judgment.
Specific Causation: Expert Testimony is Required
This left the question of whether an 80 mg. plaintiff could adduce sufficient specific causation evidence to capitalize on the open general causation “window.”
In response to the plaintiffs’ argument that some of the applicable states’ laws would permit them to survive summary judgment on specific causation without expert testimony, the court stated, “While the specific language used by courts vary to some degree, all jurisdictions require expert testimony at least where the issues are medically complex and outside common knowledge and lay experience.” (The court followed this statement with a fourteen-page string cite of precedents from all 50 states.) The court emphasized, “Here, expert testimony is certainly required. Diabetes is a complicated, progressive disease with a number of risk factors. Plaintiffs’ general causation experts cannot even figure out how to determine whether an individual’s diabetes was caused by Lipitor or other factors, and Plaintiffs’ specific causation expert cannot determine which people in a room of 100 people or 10 people had ‘statin-induced’ diabetes as opposed to non-statin-induced diabetes . . . . If these experts cannot make this determination, it is certainly not within the common knowledge of a lay person. A jury’s finding of causation in the absence of any expert testimony would be based on impermissible speculation or conjecture.” (Citations omitted.)
Expert Testimony on General Causation Combined with Non-Expert Evidence
But the plaintiffs argued that, in some jurisdictions, a plaintiff can survive summary judgment with a combination of expert testimony that a substance is a possible cause of the plaintiff’s injury and “non-expert” evidence. The court conceded that this statement is true “as far as it goes,” but held, “Plaintiffs have pointed to no such probative, non-expert evidence here.” The plaintiffs pointed to the fact that “numerous plaintiffs” had “no history of diabetes prior to their initial Lipitor exposures.” The court responded (we love this), “This is undoubtedly true. It is impossible that Lipitor would have caused a Plaintiff’s diabetes if she developed the disease prior to ever taking the drug. However, the converse of this statement is not true. Plaintiff may have developed diabetes after taking Lipitor, after having a grandchild, after tasting crème brulee for the first time, or after she turned 65. However the fact that Plaintiff developed diabetes after these events does not allow a reasonable jury to infer causation, without speculation and conjecture.” (Citations omitted.)
In short, the court concluded, while there are times when a temporal relationship can provide compelling evidence of causation, “such circumstances are not present here. Plaintiffs developed diabetes months or years after taking Lipitor and they had other substantial risk factors for the disease. The Court has already found that the temporal relationship at issue here is insufficient to form the basis of a reliable causation opinion under Daubert. . . . Therefore, it is necessarily insufficient to create an issue of fact as to causation.” (Citations omitted).
Finally, the court rejected the argument that it had overstepped its role as an MDL court by addressing specific causation and that it should remand the cases to their transferor courts. The court held, “This Court is familiar with the science and issues present and can dispose of the issues far more quickly and efficiently than dozens of courts spread across the country.”
And it did just that. For plaintiffs who ingested Lipitor dosages of less than 80 mg., the court held that the supposed ‘admissions’ did not create a genuine issue of material fact as to general causation, given the court’s earlier exclusion of the plaintiffs’ experts’ general causation testimony related to these dosages. For all plaintiffs, the court held that temporal relationships could not create a genuine issue of material fact as to specific causation. And so, despite being given many more chances than they deserved, the Lipitor plaintiffs found themselves, quite properly, out of court. We hope that the judge’s uncommon patience and precision ensures that these unassailable decisions will withstand appeal. We hope that you will read the decisions with appreciation, and we congratulate Pfizer on this hard-fought victory.