Appellate advocates are limited to trial court records. The judgment winner may raise arguments, based upon those factual records, which would support the judgment, whether they raised or made the arguments below, and regardless whether the court below actually considered the arguments.[1] The judgment loser, on the other hand, is strictly confined to the arguments presented to the court below.
Responsible appellate advocates closely observe the limits of what they may argue in their briefs or in oral argument. In the Acetaminophen MDL, the plaintiffs’ counsel decidedly stepped out beyond what lawyers for an appellant are permitted to argue.
The parties filed their appellate briefs in the Acetaminophen MDL, in 2024. On August 15, 2025, plaintiffs’ counsel, Ashley Keller, filed a letter brief to advise the appellate court that Baccarelli’s for-litigation navigation guide review had been published in the pro-lawsuit industry journal, Environmental Health (and to provide the court with a copy).[2] The propriety of this submission was questionable, in pointing to evidence de hors the record. Keller never as much as asked for leave to file. Perhaps trying to justifying the questionable letter brief, Keller acknowledged that the published article “draws extensively on analysis from the reports the District Court excluded.” While perhaps excusing some of what plaintiffs were submitting, outside the record, the acknowledgment was a damning admission of a fact Baccarelli had failed to disclose in the published Prada review itself.
In this letter brief, Keller wandered even further outside the record and misrepresented the nature of peer review by suggesting that the peer reviewers of the Prada review agreed substantively with Baccarelli’s views on biological plausibility. Keller also distorted the ability of the Prada review to account for genetic and other confounding, when the review methodology was set up to downgrade null studies and count studies with stronger associations more heavily. The letter brief misleadingly claims that the article’s analyses “provide strong evidence that the observed relationship was not confounded but in fact was likely [a] relationship….” Importantly, the quoted language failed to qualify what kind of relationship, causal or non-causal, was being claimed or asserted.[3]
Keller harped on the Prada review’s statement that it had shown that causation was “plausible,” but not definitive. The Prada review, however, never advanced a claim that the available evidence supported a causal association. The Prada review authors not only disclaimed a definitive showing of causation, it also never suggested or intimated that causation had been provisionally or probably shown. Baccarelli, in his tweet and his official statement the following month, on September 22, 2025, confirmed that his published review did not establish causation, even in his own mind.
On September 24, 2025, Keller filed yet another letter brief, in an attempt to inject additional materials, outside the trial court record, from the Trump news conference of two days earlier.[4] In this brief, Keller argued that the FDA had commenced a safety labeling change in the form of a “Dear Doctor” letter, which stated that “the use of acetaminophen by pregnant women may be associated with an increased risk of … autism and ADHD in children.”[5] Keller’s argument ignored the failure of Doctor Makary’s letter’s to support the plaintiffs’ causal claim. That the FDA commissioner believed only that there “may be an association” was actually a potent point in favor of the defense’s position.
The Keller letter brief also adverted to Makary’s quoting and citing Baccarelli’s for-litigation report, in which Baccarelli had used causal language. Keller did not, however, try to explain how and why Makary was quoting from a litigation document, when the published Prada review eschewed causal conclusions. Nor did Keller try to reconcile Baccarelli’s very public statements that similarly eschewed causal conclusions and couched his clinical recommendations as “precautionary.”
Keller’s letter brief, which violated the law and norms of appellate advocacy, then jumped the shark in advancing a pseudo-preemption argument. Keller claimed that “[e]xpert opinion that is sound enough to persuade every Senate-confirmed federal scientist easily clears Rule 702(d)’s bar.” Of course, Makary, in his official communications with physicians, spoke only of an association and explicitly disclaimed causation. Similarly, NIH director Jay Bhattacharya and Dr. Oz never asserted that there was a causal association.
Keller’s argument turned from improper to bizarre, as he claimed that sustaining the Rule 702 exclusion of Baccarelli’s opinion “would pose grave separation of powers concerns.” Keller’s attempted to explain:
“The executive branch safeguards public health from dangerous pharmaceutical interventions. A decision holding that a jury may not hear the same expert evidence that the executive branch credited will badly damage the public trust required for the executive to take care that the public-health laws are faithfully executed.”[6]
Keller did not cite any judicial or other authority to support his argument. None exists. The separation of powers does not mean that the executive branch can unilaterally determine scientific facts. Plaintiffs’ counsel had made sure that Dr. Makary had Baccarelli’s litigation report for the Trump news conference, and they then used Makary’s off-hand quotation from the litigation report to support their litigation efforts.
Keller’s lobbying appeared to have had the effect of delaying the oral argument.[7] Just as the lawsuit industry had injected itself into the political assessment of acetaminophen, that same industry then capitalized on its own lobbying efforts to mislead the judicial process.
Oral Argument in the Second Circuit
The parties appeared before a panel of the United States Court of Appeals for the Second Circuit for oral argument, on November 17, 2025.[8] Judges Guido Calabresi, Gerard E. Lynch, and Eunice C. Lee made up the panel. The argument did not seem to go well for the defense. Keller made up facts, such as when he asserted falsely that no one had said that the Prada review was unreliable.[9]
When the defense counsel was at the podium, Judge Calibresi, 93 years old, wandered into a revealing but irrelevant story about his father who had been a cardiologist:
“Let me just tell you one thing. Years ago, my father, who was a doctor, a cardiologist, and smoked two packs of cigarettes a day, I showed him that first study that linked cigarettes and cancer, and he went through the whole thing saying there’s this wrong with it, there’s that wrong with it. It doesn’t demonstrate the answer. And that evening at supper, he said, I’m giving up smoking. And I asked him why, and he said, I don’t need to wait for the perfect study. I don’t need to wait for it. It’s enough for me to want to make that decision. Isn’t that the level at which we are supposed to be deciding?”[10]
The answer was simple: No!
Judge Calebresi had described a personal, subjective, precautionary assessment based upon a single study. It was not a conclusion that the father, Dr. Calebresi, ever defended in a scientific journal. The defense counsel wandered into a technical answer with citations to case law, but this response missed the important point. Judge Calebresi had illegimately invoked the precautionary principle, which is quite different from well understood concepts of causation in science and in law.
From Judge Calebresi’s personal anecdote, the audience was left guessing when that “first study” was reviewed by the judge and his father. Studies generally do not “demonstrate” a conclusion, with a quod erat demonstrandum at the end. The first controlled epidemiologic study was probably one published by Franz Hermann Müller, in German, in 1939.[11] Judge Calebresi was seven at the time; his family and he left Italy that same year to escape from fascist anti-Jewish legislation. A more likely candidate was the famous paper, published in English, in 1950, by Sir Austin Bradford Hill and Sir Richard Doll.[12] The English language studies of the 1950s generally reported risks of lung cancer increased by 1,000 percent and more, a very different scenario from the small, inconsistent associations relied upon by Baccarelli.
Regardless of the vague, ambiguous personal anecdote, Judge Calebresi presented defense counsel with a compellingly easy hypothetical question: is the law interested in only satisfying itself that there is one single study that supports an association when it must evaluate a claim of causation? The answer is no, that hypothetical is not a showing of causation, but of an association. It may be later supported so that a body of similar studies leads to a conclusion of causation, but Judge Calebresi’s hypothetical does not satisfy the requirements of law or science, not even remotely.
(to be continued)
[1] Williams v. Norris, 25 U.S. (12 Wheat.) 117, 120 (1827) (“If the judgment should be correct, although the reasoning, by which the mind of the Judge was conducted to it, should be deemed unsound, that judgment would certainly be affirmed in the superior Court.”); U.S. v. American Railway Express Co., 265 U.S. 425, 435 (1924) (“the appellee may, without taking a cross-appeal, urge in support of a decree any matter appearing in the record, although his argument may involve an attack upon the reasoning of the lower court or an insistence upon matter overlooked or ignored by it.”). See also Lee v. Porter, 63 Ga. 345 (1879) (the “tipsy coachman” doctrine).
[2] Ashley Keller, Letter Brief on behalf of Plaintiffs-Appellants, In re Acetaminophen – ASD-ADHD Prods. Liab. Litig., Appeal Nos. 24-916(L), 24-1121 (con), 24-2360 (con), docket entry 217.1 (Aug. 15, 2025).
[3] Id., citing and quoting Prada Review at 29 (the interpolated “a” was correcting the text and should have been [sic]).
[4] Ashley Keller, Letter Brief on behalf of Plaintiffs-Appellants, In re Acetaminophen – ASD-ADHD Prods. Liab. Litig., Appeal Nos. 24-916(L), 24-1121 (con), 24-2360 (con), docket entry 226.1 (Sept. 24, 2025).
[5] Id., citing and quoting from Martin Makary, FDA Commissioner, Notice to Physicians on the Use of Acetaminophen During Pregnancy (Sept. 22, 2025).
[6] Id.
[7] Amanda Bronstad, 2nd Circuit Delays Arguments in Acetaminophen Appeal After Trump Targets Tylenol, LAW.COM (Sept. 25, 2025) at https://www.law.com/2025/09/25/2nd-circuit-delays-argumen ts-in-acetaminophen-appeal-after-trump-targets-tylenol/.
[8] Audio recording and transcript of oral argument, In re Acetaminophen – ASD-ADHD Prods. Liab. Litig., Appeal Nos. 24-916(L), 24-1121 (con), 24-2360 (con) (Nov. 17, 2025), at https://www.courtlistener.com/audio/101306/in-re-acetaminophen-asd-adhd-products-liability-litigation/.
[9] Id. at 15:02 – 15:22 (minutes, seconds of recording).
[10] Id. at 28:49 – 29:33 (Calibresi, J.).
[11] Franz Hermann Müller, Tabakmissbrauch und Lungencarcinoma, 49 ZEITSCHRIFT FÜR KREBSFORSCH 57 (1939).
[12] Richard Doll & Austin Bradford Hill, Smoking and carcinoma of the lung, 2 BRIT MED. J. 740 (1950). See Colin White, Research on smoking and lung cancer: a landmark in the history of chronic disease epidemiology, 63 YALE J. BIOL. & MED. 29 (1990).
