In our post on the worst decisions of 2013, we highlighted two trilogies of dreck, the First Circuit’s Neurontin decisions and the Pennsylvania Superior Court’s Reglan decisions. Today, we discuss a rare pentalogy of drecky decisions from the same court in the same litigation issued over nine days. See In re Actos (Pioglitazone) Prods. Liab. Litig., MDL No. 6:11-md-2299, 2014 U.S. Dist. LEXIS 4808 (W.D. La. Jan. 10, 2014); In re Actos (Pioglitazone) Prods. Liab. Litig., MDL No. 6:11-md-2299, 2014 U.S. Dist. LEXIS 5289 (W.D. La. Jan. 14, 2014); and three slip ops. concerning experts Panigrahy, Southgate and Grossman.
Like the Lance decision in yesterday’s post, the five Actos MDL Daubert decisions make a strong early case to be included with the worst of whatever else comes down the pike in 2014. Last week, we posted on the same court’s overly narrow construction of conflict preemption. Before that, we noted some better—from our perspective—decisions from the same court involving smacking around a plaintiff lawyer for discovery nonsense, referring a state law question to the Arkansas Supreme Court, and even knocking out speculative medical monitoring opinions under Daubert. We infer from this progression, and from the repeated and vehement chastising of the Defendants in the five Daubert decisions, that there has been shift in how the court views the parties that is coming through in its decisions. We have been known to do some chastising of our own, but we never claimed to be neutral. We also have not seen so many “cautions” to parties—there is but one to plaintiff’s counsel (about framing questions at trial) amid dozen or more to Defendants—or so many characterizations that arguments in motions were false, disingenuous, anemic, borderline specious, or hyperbolic. You really have to read these decisions to get the full flavor, but it is hard to separate the substance from the delivery. It is also hard to discern a single explanation for all the strange rulings in the five decisions on five experts in different areas.
We start with the disclaimer that we did not read the briefs of the respective parties or review the studies and liability evidence. Like many with televisions or at least passing interest in the regulation of drugs, we have some understanding about how the evidence on the association of Actos and bladder cancer and the warnings about that issue have changed over time. We would not begin to say whether the Court’s characterizations of the evidence and argument from the Daubert briefs are correct, any more than we would say whether an opinion that Actos causes bladder cancer or that the drug’s sponsor should have taken some additional action is correct. As the Actos court and others before it got wrong, the admissibility of expert opinion testimony under Rule 702 and Daubert is a matter of procedure. Daubert is applied in all federal court cases, because it is about procedure, not substance. (Some requirements for expert testimony really are substantive state law, but that is a different issue.) Without knowing the substance of the litigation, let alone whether a particular proffered expert opinion is correct, we can tell that the decisions misapply the proper standards, including the ones they correctly outline in the matching “Applicable Law” section of each decision.
What really galls us, though, is the part of each decision—repeated with minor variation—starting the “Analysis” of the issues:
Plaintiffs bear the ultimate burden on the admissibility of their proffered evidence, thus, this court will first look to Plaintiffs’ prima facie showing on each issue. If a prima facie showing is made, this Court will proceed to a consideration of the Defendants’ specific challenges.
We agree that the burden of establishing admissibility of expert evidence is on the proponent of that evidence. However, there is nothing in Daubert, Kumho Tire, Rule 702, or the Advisory Committee Notes to Rule 702 characterizing this burden as being met by a prima facie showing. The Actos court provides no citation for such a standard and we cannot see why that would be the right standard. One of the earliest post-Daubert circuit decisions on the admissibility of expert evidence, in re Paoli, made clear prima facie was not the standard: “By holding that the admissibility of scientific testimony is governed by Rule 104(a), Daubert clearly holds that the party seeking admissibility must make out more than a prima facie case of reliability.” 35 F.3d 717, 743 n.9 (3d Cir. 1994). While we did not look exhaustively, the only mention by the Fifth Circuit of “prima facie” in the Daubert context comes from the well-known Moore v. Ashland Chemical case, which the Actos court cites on other propositions a few times. But Moore only says that Paoli—yes, the one from two sentences ago—“ concluded that a clinical physician’s performance of standard diagnostic techniques provides prima facie evidence that a doctor has considered alternative causes and has attempted to test his or her initial hypothesis as to cause.” 151 F.3d 269, 283 (5th Cir. 1998) (citing 35 F.3d at 758). This can hardly be construed as an endorsement of a particular standard of proof.
The Latin phrase prima facie means “at first face,” which is like saying an initial impression. In the context of summary judgment and directed verdict, the standard makes sense. Except when based on a defense, these motions basically ask the judge to decide if there is/was admissible/admitted evidence supporting each element of her case that the plaintiff has to prove. The judge does not weigh competing evidence or credibility or decide facts in dispute. This is quite a different task than is charged to a court under Fed. R. Evid. 104(a), which governs Daubert challenges: “The court must decide any preliminary question about whether a witness is qualified, a privilege exists, or evidence is admissible. In so deciding, the court is not bound by evidence rules, except those on privilege.” The judge must decide these questions. This is certainly emphasized in the many references in Daubert and its progeny to the role of the court as gatekeeper, something we discussed just the other day. By applying the inapplicable prima facie standard, something mimicked in dozens of “threshold” (in Latin, limen, so you can stand in limine) findings about the challenged experts, the Actos court may have taken many pages to relay its self-characterized “thorough review” of the record on the motions, but it abdicated its proper role as gatekeeper.
The misapplication of the standard of proof, however, was not the only misstep by the court in these decisions. All five decisions include a cut-and-paste section saying that each of the well-known Daubert factors support admissibility (at least “as a threshold matter”), but they are entirely conclusory and reveal nothing of the methodology actually used by these experts. The decisions say that the “testability” factor is satisfied because the experts’ reports cite studies that tested something, the “peer review” factor is satisfied because the experts’ reports cite peer-reviewed publications, and the “rate of error” factor is satisfied because the reports cite studies that include statistics with error rates. We could go on, but it is apparent that the court did not understand that the Daubert factors are meant to provide a framework for evaluating whether the methodology that the expert used to come up with his opinion was reliable. An utterly unreliable methodology could have been used an expert whose report cites a bunch of peer-reviewed studies with statistical measures of uncertainty, as where the studies have nothing to do with the opinion being offered because they relate to a different exposure, exposure level, disease, or organism. (That is what Ashland is really about.) For one of the challenges, the court mocks the defendant for saying that the expert did not base his causation opinions (for humans) on actual human studies when his report said he considered some in vitro (“in glass” in our continuing certamen game). A proper Daubert analysis would have looked to see if the expert employed a reliable methodology in trying to extrapolate from such studies to what happens in actual people taking the drug. Or, rather, whether the plaintiff proved—not just offered prima facie evidence—that he had. (The one time in the five decisions that the court actually dug down and looked at how an expert tried to get to his opinion from studies about something else, it excluded the particular opinion based on the failure to “bridge . . . the analytical between liver and lymphocytes and bladder or bladder cells.”)
A similar misunderstanding of what it means to prove that a reliable methodology was used plays out in rulings relating to whether certain challenged testimony amount to mere “personal opinion.” The Federal Rules of Evidence only allow opinions to be offered by fact witnesses under relatively narrow circumstances under Rule 701 and by expert witnesses who meet the requirements of Rules 702 and 703 as spelled out in Daubert and progeny. Particularly for witnesses who try to opine on conduct, it is easy for opinions to become “personal” rather than “expert.” Expert opinions would be those that measure conduct against objective and discernable standards. Opinions are personal, even when offered by someone with lots of experience, when they essentially say “in my experience, the defendant should have done ____.” This latter opinion is not rooted in any standard and often cannot be contested. The Actos court’s decision on the Daubert challenge to former FDA Commissioner Kessler flubbed this distinction repeatedly. While stating that “it is insufficient for an expert to based his or her opinions on education and experience alone, especially in the face of evidence to the contrary,” 2104 U.S. Dist. LEXIS 4808, *30 (citing Brown v. Illinois Cent. RR. Co., 705 F.3d 531, 536 (5th Cir. 2013)), the court dismissed several challenges to specific “opinions” offered by Dr. Kessler because he had “knowledge, training and experience” in dealing with similar issues. Having relevant experience can qualify a witness as an expert on a subject, but it does not mean he is offering an opinion derived from a reliable methodology. Ironically, the court attempts to buttress
its reasoning for rejecting a challenge to some of Dr. Kessler’s opinions as legal conclusions by citing the Advisory Committee Notes for Fed. R. Evid. 704, which basically say that getting away with the old ultimate issue rule does not mean experts can just say whatever they want. These notes contain a sentence we have been citing for many years, that the rules “afford ample assurances against the admission of opinions which would merely tell the jury what result to reach, somewhat in the manner of the oath-helpers of an earlier day.” In a case like this, a former FDA Commissioner testifying that the drug company defendant should have done what plaintiff contends is the height of oath helping, unless the opinions he offers are tied to standards rather than just his personal experience.
This segues to another recurring gaffe in these decisions. Even though every drug MDL since 2000 (going back to In re Diet Drugs) has excluded expert opinions as to the knowledge and intent of the defendant, this judge did not. Instead, she said it will be a matter of the form of the question at trial because there seemed to be general evaluation of the evidence going to the issue of whether the defendant knew some fact or had some motivation in doing or not doing something. The main reason the many other MDL decisions issued blanket prohibitions of purported expert opinion on what was in the “head” of the defendant or its actors is that juries are supposed to decide such issues based upon consideration of the evidence. Letting an expert—especially one who might be accorded significant weight because of his past positions—do it for the jury is wrong, and not simply a matter of how the question is phrased and what foundation is laid first. This is so whether the witness is a former FDA Commissioner, purported pharmaceutical marketing expert, oncologist, pathologist, or molecular biologist, as the five experts here are.
Lastly—not because there are not many more issues with these decisions—we return to Rule 104. As in many pharmaceutical cases, the plaintiff here wanted to offer evidence about defendant’s marketing practices even if there was no proof that the particular plaintiff’s prescription of the drug was influenced by those practices. The decision in 2014 U.S. Dist. LEXIS 5289 concerns the admissibility of one vehicle for such evidence, a purported expert in pharmaceutical marketing (who had never been involved in marketing pharmaceuticals). Strangely ignoring that “fit” is a Daubert issue and Rule 702 requires that the proposed expert opinion “will help the trier of fact to understand the evidence or to determine a fact in issue,” the court suggested that trying to exclude opinions on marketing not connected to the prescription of Actos to the trial plaintiff was not a proper Daubert challenge. In its own words:
Clearly, they are relevant; whether admissible or inadmissible, perhaps on other grounds, might be the more relevant inquiry under the objections made by Defendants within their motion as Defendants tend to conflate two separate legal inquiries as will be discussed below . . . . The Defendants’ error is in seeking to impose upon the Plaintiffs an obligation to absolutely prove the existence of such a link and establish before Dr. Fugh-Berman’s opinions may be found relevant and, therefore, admissible, in connection with the instant motion . . . . Absolute proof is not the legal inquiry at issue; proof by a preponderance of the evidence is the relevant standard as to general liability, and that inquiry is the province not of this Court as the gatekeeper, but of the trier of fact if the evidence exists to support a prima facie finding . . . . This Court, therefore, finds the Plaintiffs have sufficiently demonstrated, as a threshold matter, the potential relevance of Dr. Fugh-Berman’s opinions to meet the gatekeeper standard.
Id. at **35-37 (emphasis in original). Wow. Apparently, a prima facie showing of potential relevance counts as proof of fit.
We said this was a Rule 104 issue. It is. In addition, to the requirement in 104(a) that a court must decide preliminary questions affecting admissibility, 104(b) addresses conditional relevance, “Relevance that Depends on a Fact”:
When the relevance of evidence depends on whether a fact exists, proof must be introduced sufficient to support a finding that the fact does exist. The court may admit the proposed evidence on the condition that the proof be introduced later.
In a prescription drug product liability case, testimony about the defendant’s marketing conduct is not relevant (even for punitive damages given recent Supreme Court authority) unless there is a factual basis—the conditional fact—that the marketing affected the prescription of the drug to the plaintiff. Expert opinions about general marketing (or some specific marketing) do not help the jury determine facts in issue unless the link to the prescription to the plaintiff is established. A Daubert challenge like this one requires the judge to decide this preliminary question based on “proof.” Maybe the court here was just following the second part of 104(b) and conditionally admitting the expert opinions subject to certain proof at trial. It did, after all, deny the motion without prejudice. Maybe much of the expert evidence at issue in these five decisions will not get admitted at trial based on plaintiff’s inability to lay foundation or ask properly formed questions. Maybe. First impressions can be misleading. (That is why it is not the right standard for a Daubert motion.)