Evan M. Tager is a Partner with Mayer Brown LLP in the firm’s Washington, DC office and is the WLF Legal Pulse’s Featured Expert Contributor on Judicial Gatekeeping of Expert Evidence. Jonathan S. Klein, an Associate, with the firm. Along with five other Mayer Brown attorneys, they co-authored a January 2020 WLF Monograph, Admissibility of Expert Testimony: Manageable Guidance for Judicial Gatekeeping.
* * *
Despite Daubert’s mandate that district courts must scrutinize expert testimony to ensure that juries hear only reliable and relevant expert testimony, in practice the courts often treat important questions of reliability as matters for cross examination. The courts often let juries hear unreliable expert testimony under the mistaken view that, in seemingly close cases, Daubert and Federal Rule of Evidence 702 effectively tip the scales in favor of admissibility (for example, through a so‑called “presumption in favor of admissibility”).
On October 23, 2020, the Ninth Circuit will hear oral argument in one such case, Hardeman v. Monsanto Corp., No. 19-16636, in which the plaintiff’s theory of liability hinges on expert testimony that the district court called “too equivocal” and tainted by “significant problems.” Even though the district court acknowledged that methodological flaws pervaded the opinions of the plaintiff’s experts, the court denied Monsanto’s Daubert motions—based largely on an erroneous view that the applicable precedent requires courts to decide “close cases” in favor of admissibility. In re Roundup Prods. Liab. Litig., 390 F. Supp. 3d 1102, 1113 (N.D. Cal. 2018).
In Hardeman, one of several thousand cases in the Roundup MDL, plaintiff Edwin Hardeman alleges that Monsanto failed to adequately warn that Roundup, which contains the pesticide glyphosphate as an active ingredient, could cause Non‑Hodgkin’s lymphoma. To prevail on his claim, Hardeman needs to prove both general causation (that glyphosphate is capable of causing Non‑Hodgkin’s lymphoma in humans generally) and specific causation (that the glyphosphate in Roundup actually caused Hardeman’s Non‑Hodgkin’s lymphoma). A jury trial yielded an $80.2 million verdict for Hardeman, which included a $75 million punitive‑damages award that the district court later reduced to $20 million.
Hardeman proffered six experts on general causation, and the district court correctly excluded several for failure to offer a relevant and reliable opinion. These experts largely borrowed putative findings from IARC, a branch of the World Health Organization, which in 2015 classified glyphosphate as a “probable carcinogen.” In re Roundup Prods. Liab. Litig., 390 F. Supp. 3d at 1113‑14, 1144‑49. But IARC itself stressed that this classification meant only that glyphosphate warranted further study; in fact, IARC expressly disclaimed any conclusion that glyphosphate likely causes Non‑Hodgkin’s lymphoma in humans. Id. at 1113‑14.
Several other defects tainted the IARC’s classification. For example, although the governing law required Hardeman to prove general and specific causation by a preponderance of the evidence, the IARC classification could not offer any assistance to a jury in a civil trial because IARC declined to assign any numeric value to the term “probable.” Id. And the IARC classification focused on studies involving mice, not humans. Id.
Hardeman proffered, and Monsanto moved to exclude as unreliable, the testimony of three other experts on general causation (Drs. Portier, Ritz, and Weisenberg). All three experts’ conclusions depended on analyses that the district court acknowledged were flawed. In re Roundup Prods. Liab. Litig., 390 F. Supp. 3d at 1131‑44. In finding a purported association between glyphosphate and Non‑Hodgkin’s lymphoma, the experts relied largely on a series of studies that used data from the late 1970s through the early 1980s. For instance, one of the key studies on which the plaintiff’s experts relied, De Roos 2003, examined data from 1979 to 1986. Id. at 1118. The study found a “relative risk” of 2.1, which translated into a 110% increase in prevalence of Non‑Hodgkin’s lymphoma among people who used glyphosphate compared to those who did not. Id.
But Monsanto first introduced Roundup to the market in 1974, and experts on both sides agreed that the “latency period” (the time between an exposure to a substance and the development of an illness) for Non‑Hodgkin’s lymphoma is no less than five to ten years and often as much as twenty (or more) years. In other words, the long latency period meant that data from 1979 to 1986 could not have shown much, if any, increased association between glyphosphate and Non‑Hodgkin’s lymphoma. Based partly on this discrepancy between the study’s findings and the latency period for Non‑Hodgkin’s lymphoma, the district court agreed with Monsanto’s argument that several studies on which the plaintiff’s experts relied failed to account for confounding factors, that is, potential causes of Non‑Hodgkin’s lymphoma other than the glyphosphate in Roundup. See, e.g., In re Roundup Prods. Liab. Litig., 390 F. Supp. 3d at 1140‑41.
The plaintiff’s experts also consistently ignored a more scientifically sound “cohort” study (AHS 2005), which found no association between glyphosphate and Non‑Hodgkin’s lymphoma, in favor of “case‑control” studies that tended to show a slight association between glyphosphate and Non‑Hodgkin’s lymphoma. Scientists typically prefer cohort studies to case‑control studies; in a prospective cohort study, scientists separate subjects into two groups, one exposed to the substance, and the other unexposed, and track the occurrence of the illness over time. After tracking cohorts between 1993 and 1997, the AHS 2005 study found no statistically significant association between glyphosphate and Non‑Hodgkin’s lymphoma.
In contrast, the case‑control studies interviewed people already diagnosed with Non‑Hodgkin’s lymphoma (or a relative, if death or illness prevented interviewing the person diagnosed with Non‑Hodgkin’s lymphoma) to ask about their use of pesticides, among other items. As the district court recognized, case‑control studies suffer from “recall bias,” where “those who become ill are more likely to ruminate about the possible causes of their disease.” Id. at 1118. Despite the pronounced defects in De Roos 2003 and the other case‑control studies on which the plaintiff’s experts relied, the district court held that these defects went to weight, not admissibility, and consequently denied Monsanto’s Daubert motions. Id. at 1130‑44.
In denying the Daubert motions, the district court acknowledged profound concerns about the opinions of the plaintiff’s experts. It noted that the “evidence of a causal link between glyphosphate exposure and [Non‑Hodgkin’s lymphoma] in the human population seems rather weak” (id. at 1108) and that the evidence “seems too equivocal to support any firm conclusion that glyphosphate causes” Non‑Hodgkin’s lymphoma. Id. at 1109. After reciting the Daubert factors, the district court added that the “Ninth Circuit has placed great emphasis on Daubert’s admonition that a district court should conduct this analysis “with a ‘liberal thrust’ favoring admission.” Id. at 1112‑13 (internal citation omitted). The district court interpreted Ninth Circuit precedent to compel “slightly more room for deference to experts in close cases than might be appropriate in some other Circuits,” even acknowledging that “[t]his is a difference that could matter in close cases.” Id. at 1113.
At bottom, the district court treated the admissibility question as a close call and for that reason effectively tipped the scales in favor of admissibility. But nothing in Rule 702 authorizes this type of tacit presumption in favor of admissibility; in fact, the Advisory Committee’s notes to the 2000 amendment to Rule 702 make clear that the proponent of expert testimony “has the burden of establishing that the pertinent admissibility requirements are met by a preponderance of the evidence.” Fed. R. Evid. 702 advisory committee’s note (2000). Whatever its name, a presumption in favor of admissibility impermissibly circumvents the proponent’s burden of proving the reliability of expert testimony.
Likewise, nothing in Daubert relieves the proponent of expert testimony from proving that the expert more likely than not reliably applied sound data to the facts of the case. The district court in Hardeman, and the Ninth Circuit decisions on which the district court relied, cited a line from Daubert that mentions the “liberal thrust” of the Federal Rules, but in reality the excerpt from Daubert clarified only that the courts cannot use Frye’s general‑acceptance test as a prerequisite for admissibility. Daubert, 509 U.S. 579, 588 (1993) (“a rigid ‘general acceptance’ requirement would be at odds with the ‘liberal thrust’ of the Federal Rules and their ‘general approach of relaxing the traditional barriers to ‘opinion’ testimony”) (citations omitted). Elsewhere in Daubert and its progeny, the Supreme Court repeatedly made clear that district courts must take care to guard against spurious conclusions masquerading as expert opinions. Id at 589 (“under the Rules the trial judge must ensure that any and all scientific testimony or evidence admitted is not only relevant, but reliable”).
Hardeman gives the Ninth Circuit an opportunity to correct the misconception that a district court must admit “equivocal” expert testimony or other expert testimony that it cannot confidently deem unreliable. But this misconception has infiltrated other circuits as well. For that reason, the Advisory Committee on the Federal Rules of Evidence has been considering ways to correct this erroneous understanding of the courts’ role. Although the Advisory Committee decided this past Summer not to alter the language of Rule 702, it will consider this Fall whether to add a committee note clarifying the distinction between weight and admissibility. Both bench and bar would benefit if it does so.