A significant decision from the U.S. Court of Appeals for the Second Circuit has revived hundreds of product liability claims alleging that prenatal exposure to acetaminophen—the active ingredient in Tylenol—caused children to develop autism spectrum disorder (ASD) or attention deficit hyperactivity disorder (ADHD). Although the appellate court made clear that it was not determining whether such a causal relationship exists, its judgment represents an important development in the law governing expert evidence and judicial gatekeeping under the Daubert standard.
The multidistrict litigation (MDL), centred before the U.S. District Court for the Southern District of New York, involves claims against Kenvue, the manufacturer of Tylenol following its separation from Johnson & Johnson in 2023, together with several retailers that marketed acetaminophen products. The plaintiffs contend that manufacturers failed to warn pregnant women of alleged neurodevelopmental risks associated with acetaminophen use during pregnancy.
In 2024, U.S. District Judge Denise Cote dismissed approximately 550 cases after excluding all of the plaintiffs' general causation experts under Rule 702 of the Federal Rules of Evidence and the principles established in Daubert v. Merrell Dow Pharmaceuticals. Judge Cote concluded that the experts' methodologies were insufficiently reliable to be presented to a jury, leaving the plaintiffs without admissible evidence capable of establishing general causation.
The plaintiffs appealed, arguing that the district court had exceeded its proper gatekeeping role by evaluating the correctness of scientific conclusions rather than the reliability of the methodologies employed.
In a unanimous judgment, the Second Circuit agreed in part. The appellate court emphasised that it was not deciding the underlying scientific question:
"We are not deciding whether there is a general causal relationship between acetaminophen and ADHD and/or ASD."
The judges continued:
"And we are certainly not deciding the approach that policymakers concerned with protecting public health should take to regulating the use of acetaminophen."
Instead, the court focused exclusively upon the admissibility of expert opinion evidence. It concluded that "the district court exceeded its discretion" by excluding portions of the plaintiffs' expert testimony. The litigation has therefore been remitted to the district court for further proceedings.
The appeal centred upon several expert witnesses retained by the plaintiffs to address the issue of general causation.
Among the most prominent was Dr Andrea Baccarelli, M.D., Ph.D., an internationally recognised physician-scientist and epidemiologist who serves as Dean of the Harvard T.H. Chan School of Public Health. Dr Baccarelli has published extensively in environmental epidemiology, toxicology and public health, with particular expertise in analysing environmental exposures and their relationship to human disease.
In the litigation, Dr Baccarelli reviewed the available epidemiological literature examining prenatal acetaminophen exposure and childhood neurodevelopmental outcomes. His opinion was that the published evidence, when evaluated collectively, supported the existence of a causal relationship between prenatal acetaminophen exposure and increased risks of ADHD and ASD. His methodology relied upon established epidemiological techniques, including systematic review of observational studies and application of recognised scientific principles used to assess causation.
Judge Cote rejected his conclusions, finding that he had selectively interpreted the scientific literature. The Second Circuit, however, reached a different procedural conclusion. While expressly declining to endorse Dr Baccarelli's scientific opinions, the appellate judges held that his methodology reflected approaches accepted within his scientific discipline and therefore should not have been excluded solely because reasonable scientists disagreed with his conclusions.
The appellate court reached a similar conclusion regarding Dr Eric Hollander, M.D., a psychiatrist and internationally recognised authority in autism spectrum disorders, obsessive-compulsive disorder and neuropsychiatric research. Dr Hollander evaluated the medical literature concerning neurological development and the biological mechanisms potentially linking prenatal acetaminophen exposure with neurodevelopmental disorders. The Second Circuit concluded that his opinions likewise reflected accepted scientific methodologies and should have been admitted for consideration by the trier of fact.
The third expert whose testimony was reinstated was Dr Brandon Pearson, Ph.D., a neuroscientist and pharmacologist whose research focuses upon developmental neurobiology and experimental toxicology. Dr Pearson assessed laboratory, animal and mechanistic studies addressing the biological plausibility of acetaminophen-induced neurodevelopmental injury. Again, the Court of Appeals held that the district court had gone beyond its permissible role in rejecting his evidence.
Not all of the plaintiffs' experts succeeded on appeal.
The Second Circuit upheld the exclusion of Dr Robert Cabrera, Ph.D., whose opinion relied upon application of the Bradford Hill criteria for assessing causation. The appellate judges agreed with the district court that his analysis failed to weigh and synthesise the relevant factors in a sufficiently reliable manner.
Similarly, the court affirmed the exclusion of Dr Stan Louie, Pharm.D. Although Dr Louie addressed dose-response issues relating to acetaminophen exposure, the judges concluded that he had not adequately explained how he extrapolated from the available scientific studies to reach his conclusions regarding clinically significant exposure thresholds.
The judgment therefore draws an important distinction between scientific disagreement and methodological unreliability. The appellate court observed that courts applying Daubert should not exclude expert evidence simply because competing scientists disagree on the interpretation of evolving research. Provided that qualified experts employ accepted methodologies, differing scientific conclusions may properly become matters for cross-examination and evaluation by the jury rather than grounds for wholesale exclusion.
Kenvue welcomed the opportunity to continue challenging the plaintiffs' scientific case. In a statement following the judgment, the company maintained:
"Science matters, and we stand with the many public health and medical professionals who have reviewed the science on this topic and agree."
The company added that "credible, independent science shows no proven link" between prenatal acetaminophen exposure and either autism or ADHD, and indicated that it intends once again to challenge the admissibility and reliability of the plaintiffs' expert opinions during further proceedings.
For expert witnesses, the decision reinforces several enduring principles of American evidence law. First, qualification alone does not guarantee admissibility; methodology remains central to Rule 702 analysis. Secondly, judges exercising their Daubert gatekeeping function must distinguish between assessing the reliability of an expert's methodology and resolving genuine scientific controversies. Finally, where respected experts apply recognised scientific methods but reach competing interpretations, the adversarial process—rather than judicial exclusion—may often provide the appropriate mechanism for testing those opinions.
The Second Circuit's decision therefore represents not a scientific endorsement of any alleged association between acetaminophen and neurodevelopmental disorders, but a reaffirmation of the proper limits of judicial gatekeeping. Whether the plaintiffs can ultimately establish medical causation remains an issue for future proceedings. Nevertheless, the case is likely to become a leading authority on the admissibility of expert epidemiological evidence in complex pharmaceutical litigation and an important reference point for expert witnesses practising in toxic tort and product liability cases.
By Edward Price
https://hsph.harvard.edu/news/using-acetaminophen-during-pregnancy-may-increase-childrens-autism-and-adhd-risk/
https://witnessdirectory.com/signup.php
