Revisiting Daubert: Ensuring Equity and Integrity in the Admissibility of Scientific Testimony

In litigation, where justice hinges on the impartial examination of evidence, recent trends in federal courts raise significant concerns. Plaintiffs’ scientific and medical experts are increasingly restricted in their testimony, being limited to what is already published in the peer-reviewed literature, or precluded from drawing conclusions that exceed limitations identified in the literature.[i] This restriction not only tilts the scales of justice, but also exacerbates an inherent power differential, favoring defense experts who can and often rely on literature sponsored by defendants and industry generally. Such a trend increasingly undermines the pursuit of truth and jeopardizes the very foundation of fair legal proceedings.

Recent Daubert decisions in cases involving Zantac®[ii], Tylenol®[iii], and paraquat[iv] highlight a concerning trend in federal courts: the exclusion of highly qualified experts because their analyses have not been adopted or independently validated in the peer-reviewed literature. These cases underscore a critical flaw in the application of Daubert standards, related case law, and Federal Rule of Evidence 702, insofar as the admissibility of scientific testimony should turn on whether the expert reliably applied generally accepted methodologies to analyze the scientific data, not whether they were the first to do so. The courts have, alarmingly and increasingly, applied the peer-review prong of Daubert as a sine qua non. Rigid adherence to the peer-reviewed literature risks sidelining valuable insights so that they never reach the jury in its capacity as trier of fact, perpetuating systemic inequities and undermining the very spirit of the Daubert inquiry.

In the Zantac® case, a renowned toxicologist was barred from testifying, despite employing well-established methodologies to analyze data related to the drug’s safety. Similarly, in litigation involving Tylenol® and paraquat, experts at the forefront of their fields faced exclusion based on the lack of peer-reviewed validation of their analyses, notwithstanding their reliable application of generally accepted methodologies. These cases exemplify the arbitrary nature of excluding scientific testimony simply because it arises within the context of litigation, disregarding the expertise, reliability, and validity underlying the analyses.

The cornerstone of any legal dispute lies in the presentation of evidence and the testimony of experts. Experts play a critical role in elucidating complex scientific matters for the courts, providing insights that the judges, juries, and lawyers might struggle to grasp independently. However, recent judicial practices have imposed constraints on the scope of expert testimony, particularly affecting plaintiffs’ scientific and medical experts, perhaps indicating that the courts have overstepped their role as gatekeeper. Reliable but competing scientific opinions go to the weight of the evidence, not its admissibility, and should be explored through cross-examination at trial. The requirement to adhere strictly to the peer-reviewed literature or risk pre-trial exclusion, in whole or in part, erects formidable barriers for plaintiffs’ experts, while providing defense experts with a convenient pathway to reinforce their arguments.

The imbalance is stark. While defense experts can draw upon a vast array of literature often funded by corporate interests, plaintiffs’ experts find themselves hamstrung by the dearth of independent research in certain fields. Industries notorious for their influence over research and its funding can easily shape the narrative by, inter alia, steering studies toward outcomes favorable to their interests. For example, in a major litigation involving hormone replacement therapy and the development of breast cancer, documents produced in litigation included dozens of ghostwritten articles, reviews, and commentaries published in medical journals and supplements by vendors hired by the defendant to promote unproven benefits of the therapy, downplay its harms, and cast competing therapies in a negative light.[v] More recently, in a large scoping review of 75 papers from four major biomedical journal databases that compared primary studies of drugs or medical devices based on source of sponsorship, Lundh et al. found that industry sponsorship leads to more favorable efficacy results and conclusions than sponsorship by other sources.[vi] The authors also found that industry-sponsored studies statistically had less agreement between the results and the conclusions drawn from those results than non-industry-sponsored studies (RR = 0.83, 95% CI = 0.70-0.98).⁶ Historically, the courts have not adequately scrutinized the sponsorship bias associated with industry-sponsored studies, leading to a situation ripe for cherry-picking by defense experts. Consequently, plaintiffs’ experts are left with limited options, often unable to present comprehensive evidence that reflects the full spectrum of scientific inquiry.

Moreover, a troubling dissonance arises when the courts impose on scientists a rigid and legally contrived framework by which to analyze issues of causation using observational data. Scientists are typically conservative linguists. They are generally slow to address causation in their writing, because it is not their primary focus and, when they do, tend to be overly cautious with respect to drawing ultimate conclusions. They focus on publishing significant data and securing research funding. They do not publish with the intent of meeting the legal requirements to establish causation. Scientists prefer to raise questions that lead to further research, rather than provide definitive answers. Judges, therefore, should not expect scientists to write scientific articles that reflect judicial opinions, and should scrutinize, not embrace, scientific writing that does so.

For example, a Bradford Hill analysis, the traditional means by which causation is addressed in many jurisdictions, is seldom published or publishable because it is more often a practitioner’s exercise than an academic one. In academia, Bradford Hill has fallen by the wayside to newer causal inference models by Rothman, VanderWeele, et al. that dominate modern causal inference literature.[vii],[viii],[ix] Outside of this literature, scientists are inconsistent with respect to how they address causation, if they address it at all. A systematic review of causal language used in over 1,000 high-profile observational studies across 18 high-impact journals found that nearly half (48%) of the studies had “no” or “weak” causal language, while fewer than 20% had “strong” causal language.[x] Importantly, nearly half (44.5%) of the studies with causal language demonstrated inconsistency between causal language used and recommendations implying causality.¹⁰

This lopsided dynamic not only undermines the integrity of legal proceedings, but also perpetuates systemic inequities. By restricting experts to existing literature, the courts effectively endorse a status quo that privileges the powerful and well-funded, while marginalizing the voices of those seeking redress for grievances. The erosion of trust in the legal system is an inevitable consequence, as the perception of bias becomes increasingly difficult to dispel. The implications extend beyond individual cases, casting a shadow over public health, environmental protection, and consumer safety. Litigation presents an opportunity to right the ship in a direction toward fairness and justice when all other checks and balances have failed in the preclinical, clinical, and postmarket timelines. Racial and ethnic minorities, and individuals lower on the socioeconomic spectrum, are often steered toward less expensive, lower quality interventions and products with higher risk profiles that can give rise to litigation when coupled with corporate liability. For example, an ongoing national litigation implicates preterm infants fed bovine-derived infant formula who developed necrotizing enterocolitis,[xi] a severe and often fatal injury that disproportionately affects Black infant mortality. In the United States between 1999 and 2020, Black infants were 2.5 times more likely to die from necrotizing enterocolitis compared to White infants,[xii] and prevalence data from the CDC shows that Black infants are fed infant formula over breastfeeding more than any other racial group.[xiii] It is ultimately society’s most marginalized and underserved populations, who’s plight was ignored or exploited in the first place, that are disproportionately negatively impacted if Plaintiffs’ experts are improperly excluded. When scientific testimony is shackled by arbitrary limitations, the pursuit of accountability and the safeguarding of public health are compromised. The ramifications are especially pronounced in cases involving emergent safety issues, for which the science is still developing and an expert’s original analysis is even more important.

It is imperative that federal courts reassess their approach and uphold the equity and integrity of the scientific process. Rather than perpetuating a system that amplifies the voices of the privileged few, the courts should strive to level the playing field. This necessitates a recalibration from rigid adherence to the peer-reviewed literature as the primary driver for admissibility of scientific testimony, toward a more nuanced and inclusive framework that recognizes the diversity of research methodologies and sources. Experts routinely apply established methodologies to underlying scientific data when performing original research in their ordinary practice. Disqualifying credible research based on its origin in litigation undermines the very spirit of the Daubert inquiry, and stifles the advancement of scientific knowledge.

The exclusion of highly qualified experts based on the lack of peer-reviewed validation perpetuates a system that favors the interests of defendants and industry generally, while marginalizing plaintiffs’ ability to present compelling evidence. The courts should embrace scientific inquiry as scientists do, by prioritizing methodological reliability and validity, and turning to the peer-reviewed literature as a starting, not an ending, point. Only then can the promise of justice be realized, and public trust in the legal system be restored.