Trump’s Tylenol warning cited a Harvard dean’s research. But a judge called his shifting conclusions ‘unreliable’

Epidemiologist’s Controversial Testimony on Acetaminophen Excluded in Lawsuit

A prominent Harvard dean and leading epidemiologist, Andrea Baccarelli, who gained attention for his research linking prenatal acetaminophen use to autism, has faced intense scrutiny after providing expert testimony in a high-profile lawsuit against the drug’s manufacturer. This testimony, which U.S. District Judge Denise Cote deemed “unreliable,” has raised questions regarding the interpretation of scientific evidence in legal contexts.

Baccarelli, the dean of the Harvard T.H. Chan School of Public Health, submitted a written statement in 2023 asserting that his literature review indicated a connection between prenatal exposure to acetaminophen, the active component in Tylenol, and the development of neurodevelopmental disorders, including autism and ADHD. According to Harvard, Baccarelli received approximately $150,000 for his involvement in the case.

Judge Cote criticized Baccarelli’s analysis, stating, “Dr. Baccarelli downplays studies that undermine his causation claims and highlights those that support his thesis.” She pointed out that none of the expert witnesses, including Baccarelli, had published research that directly supported their assertions in this particular case. Remarkably, just the previous year, Baccarelli co-authored a study cautioning against any change in clinical practices regarding acetaminophen use during pregnancy.

As the Trump administration prepared to announce recommendations to limit acetaminophen usage among pregnant women, Baccarelli had communicated a statement emphasizing the need for further research while advising caution with acetaminophen use, especially in cases of heavy or prolonged consumption.

In the same week, FDA Commissioner Marty Makary referenced a study by Baccarelli and his colleagues during a press briefing, noting, “A Mt. Sinai-Harvard study reviewed all existing literature and found overwhelming evidence of an association.” He quoted Baccarelli’s assertion, stating, “There is a causal relationship between prenatal acetaminophen use and neurodevelopmental disorders like ADHD and autism spectrum disorder.” However, Baccarelli’s most recent review, which encompassed 46 human studies, acknowledged an association rather than definitive causation.

This review concluded that immediate actions should be taken to advise pregnant women to limit acetaminophen intake to safeguard their children’s neurodevelopment. However, other autism researchers and epidemiologists have scrutinized these findings. Psychiatrist David Mandell expressed concerns regarding Baccarelli’s methodology and highlighted potential biases influencing the study’s outcomes.

Critics have noted the unconventional decision to include both adjusted and unadjusted results in the analysis, which could misrepresent the association between acetaminophen and autism. One notable study published in 2024 by JAMA utilized extensive genetic controls, revealing that once genetics were accounted for, the link between acetaminophen and autism essentially disappeared.

Mandell stated, “I’ve never encountered a review where someone includes unadjusted results deliberately to amplify an association. This raises valid concerns surrounding reliability.” Researchers also remarked that the qualitative assessment of each study’s evidence by Baccarelli and his team lacked the rigor typically associated with comprehensive literature reviews.

Adding to the debate, another recent study involving over 217,000 children in Japan found no significant relationship between acetaminophen use during pregnancy and autism rates. As the scientific community continues to grapple with the implications of these findings, the consensus among many researchers remains that pregnant individuals generally should not hesitate to use acetaminophen when necessary for managing pain or fever—conditions which, if untreated, could also pose risks to fetal neurodevelopment.

Mandell succinctly summarized the dilemma, stating, “Intelligent discourse exists, but when government officials adopt one perspective, it can significantly impact public health across large segments of the population.”

STAT’s reporting on disability issues is facilitated by grants from the Robert Wood Johnson Foundation and The Commonwealth Fund. Our financial supporters do not influence our editorial decisions.