Third Circuit follows Second in protecting medical journals against trade libel claims

Pacira Biosciences, Inc. v. American Soc’y of Anesthesiologists, Inc., --- F.4th ----, 2023 WL 2621131, No. 22-1411 (3d Cir. Mar. 24, 2023)

Courts in particular kinds of false advertising cases say that scientific claims are not falsifiable, even as the majority of workaday false advertising claims involving scientific fact are (correctly) treated as falsifiable. The rule seems to be that scientific claims directed at sophisticated audiences, to whom details of the claims are disclosed, are treated as opinion, whereas scientific claims in standard consumer advertising are treated as factual. This may well be justified because of the different ways in which such claims are perceived, but it also smacks a bit of those old cases that held that educated men could be allowed access to things that would be obscene as to their footmen and housemaids.

Anyway, Pacira sued ASA, the editor-in-chief of its medical journal, and the authors of three articles for statements made about one of Pacira’s drug products, for trade libel under New Jersey law. The court of appeals affirmed a finding that the statements were nonactionable opinion.

Pacira makes Exparel, a local anesthetic administered at the time of surgery to control post-surgical pain; Pacira claims that it offers longer lasting pain relief than standard local anesthetics. The challenged statements conveyed defendants’ view that Exparel was not superior or was even inferior to standard analgesics for pain relief. E.g., the cover of the February 2021 issue of Anesthesiology stated that “Liposomal Bupivacaine Is Not Superior to Standard Local Anesthetics.” Pacira challenged three articles: (1) a meta-analysis of studies of Exparel, which concludes that the drug is “not superior” to standard anesthetics; (2) a narrative review of clinical trials involving Exparel, which reaches a similar conclusion; and (3) an editorial based on the meta-analysis and narrative review.

Pacira had specific complaints about each article, including that the first one cherry-picked studies; employed a “flawed method” known as “crude pooling,” in which results from different studies using different popoulations are grouped together; and violated “the standards of medical research” by failing to account for the statistical heterogeneity in population, medication type, outcome definition, and design of the studies on which it relied. Similarly, the narrative review allegedly failed to discuss the “most relevant anesthesia procedure,” ignored favorable studies with minimal explanation, and two of its authors failed to disclose financial conflicts of interest.

The ASA’s Continuing Medical Education program allowed participants to answer questions based on the articles and receive credit to satisfy medical licensure requirements. Pacira alleged that these questions restated as fact the articles’ conclusions, including that Exparel is “inferior” to standard anesthetics and that studies favoring it are biased.

In sorting potentially actionable fact statements from opinion, courts consider the (1) content, (2) verifiability, and (3) context of the statements.

Content: “Not superior” and “inferior”  were the type of “loose” or “figurative” language that the New Jersey Supreme Court has said is “more likely to be deemed non-actionable as rhetorical hyperbole.” (This seems like a overgeneralization of what those terms mean in the abstract, unaccompanied by more specifics—those types of words are generally opinion on their own, but here the context is one of analyzing relative efficacy; the court is atomizing the claim in order to reject it.) In a footnote, the court relied on past precedent: “If a statement could be construed as either fact or opinion, however, we must construe it as an opinion. A contrary presumption would ‘tend to impose a chilling effect on speech.’” Also, that most of the cases it cited were false advertising cases

strengthens, not undermines, our conclusion that the statements here are nonactionable opinions. Commercial advertisements are directed at consumers, whereas peer-reviewed academic journals are generally “directed to the relevant scientific community.” ONY, Inc. v. Cornerstone Therapeutics, Inc., 720 F.3d 490, 496-97 (2d Cir. 2013). If New Jersey courts have concluded that statements of relative superiority would not mislead an average consumer, then similar statements made in an academic journal will not mislead the experts reading the journal. See Eastman Chem. Co. v. Plastipure, Inc., 775 F.3d 230, 236 (5th Cir. 2014) (holding disputed statements made in sales brochures were actionable because “Eastman did not sue Appellants for publishing an article in a scientific journal. Rather, Eastman sought to enjoin statements made in commercial advertisements and directed at customers”).

Verifiability: The statements here weren’t verifiable. (I suspect the doctors targeted would be surprised to hear that their CME wasn’t actually telling them verifiable information. Arguably, this line of cases highlights one way in which the collapse of any kind of consensus reality has damaged the First Amendment.) First, they were “tentative scientific conclusions and were expressly disclosed as such.” Adopting the reasoning of ONY, the court reasoned that the statements, though theoretically provable, were nonetheless still “tentative and subject to revision, because they represent inferences about the nature of reality based on the results of experimentation and observation.”

["that’s just, like, your opinion, man" from The Big Lebowski]

Of possible interest: Cf. to Daubert’s statement: “Scientific conclusions are subject to perpetual revision.” But in Daubert, that conclusion is part of the reason that we let factfinders resolve factual disputes. Also, in a footnote: “One court has recently observed that if there is consensus on a scientific issue, then a statement about the issue may be deemed verifiable. Conformis, Inc. v. Aetna, Inc., 58 F. 4th 517, 533 (1st Cir. 2023).” [This is a product disparagement case that, in some significant tension with ONY and the case at bar, finds the statement that a treatment is “not clinically effective and not accepted by doctors and insurance providers as a standard treatment” is at least potentially factual for purposes of surviving a motion to dismiss. The court's references to Plastipure and this case strengthen my conviction that this is a special rule: statements that are in fact verifiable will nonetheless be treated as opinion when made in scholarly format with sufficient detail about their limitations. I don't think this is wrong as a result, I just wish the rule were stated more clearly.]

However, the broad statements above are probably limited by the following discussion of the “express[]” qualifications of the challenged materials. “For example, immediately before concluding that EXPAREL is not superior to standard analgesics, the Hussain Article enumerates five ‘notable limitations’ of its study, including ‘variabilities’ that ‘may have played a confounding effect,’ ‘publication bias’ in selecting studies, and statistical limitations due ‘to scarcity of data.’” The second article ended, “[h]owever, medicine is constantly evolving with ongoing research, and the use of [EXPAREL] for postoperative analgesia will certainly be no different.” It then identifies several “knowledge gaps for future research,” including improving comparative data for certain metrics.

Anyway, Pacira’s allegations “boil down to disagreements about the reliability of the methodology and data underlying the statements,” which isn’t the same thing as verifiability…. [M]ere disputes about the reliability of a scientific study’s disclosed methodology cannot create an actionable falsehood for trade libel, as such disputes do not address whether the statements themselves are verifiable.” [There were no allegations of faked data or the like.] Allegedly undisclosed conflicts of interest weren’t relevant to falsifiability, even if relevant to actual malice.

Another footnote tries to preserve ordinary Lanham Act cases: “For [establishment] claims, literal falsity may be established by showing that ‘the underlying studies upon which the representations are based are not sufficiently reliable to permit one to conclude with reasonable certainty that they established the claim made.’” [Um, the court just said above that reliability was different from verifiability—why wouldn’t those claims still be unverifiable even if unreliable? Immediately after this footnote, in text it says: “It is only after establishing the statements can be proven true or false that reliability of the underlying data and methodology may become relevant.” Separately, if the claim isn’t an establishment claim—it doesn’t expressly or by implication claim scientific backing—it has, historically, been possible to falsify it by proving the contrary with scientific evidence.]

The real justification for calling these statements opinion, which the court gets to next, is that these disputes should be kept out of the legal system in order to avoid chilling scientific research and discourse. [What happens when plaintiff’s competitors disseminate these articles as part of their own self-promotion? The commercial/noncommercial speech distinction can handle this, but defamation/trade libel may be ill-suited to do so.]

Context also pointed in favor of calling these statements opinion: This was “a peer-reviewed journal for anesthesiology specialists. While statements are not protected solely because they appear in a peer-reviewed journal, such journals are often ‘directed to the relevant scientific community.’ Their readers are specialists in their fields and are best positioned to identify opinions and ‘choose to accept or reject [them] on the basis of an independent evaluation of the facts.’” More generally, “statements directed at readers who are capable of performing an independent evaluation of the facts upon which an opinion is based support the conclusion that the opinion is nonactionable.”

The court pointed out that “the readers were provided with the data and methodology on which the statements were based.” [Note that if no one is likely to do their own research, that may not matter, just as no one really reads lots of other disclosures. Also, doctors are actually not statistics experts, in general, and are just as vulnerable as the rest of us to being spun.]