Second Circuit Rejects Claims Over Scientific Literature

The Second Circuit recently held that scientific opinions voiced during debate of unsettled questions generally cannot support false advertising or deceptive business practices claims. See ONY Inc. v. Cornerstone Therapeutics Inc., No.12-2414 (2d Cir. June 26, 2013).

Defendants in this case were various corporate sponsors, authors, and publishers of a scientific journal article comparing the effectiveness of several types of medical surfactants -- biological substances that line the surface of human lungs. Surfactants are critical to lung function: they facilitate the transfer of oxygen from inhaled air into the blood stream.  Plaintiff ONY and defendant Chiesi Farmaceutici are two of the biggest producers of surfactants for medical treatment.  These parties vigorously contested the relative effectiveness of their products – in the marketplace, in the scientific literature, and in this lawsuit. The parties agreed that two variables are particularly relevant to this comparison: mortality rate and length of stay. Mortality rate means the percentage of patients (typically infants) treated with a particular surfactant who do not survive. Length of stay refers to the amount of time a patient infant remains in the hospital for treatment.

The dispute stemmed from a study funded by Chiesi, carried out by a private research firm, and then promoted and ultimately published by three doctors, which concluded that Chiesi's surfactant product allegedly led to lower infant mortality rates than ONY's product. The doctors submitted their article to the Journal of Perinatology, a leading journal in the field of neonatology. .After the article’s publication, Chiesi issued a press release touting its conclusions and distributed promotional materials that cited the article’s findings.  Plaintiff complained that the article – and the distribution of selections from it – violated the Lanham Act and New York General Business Law § 349, and also constituted tortious injurious falsehood and interference with prospective economic advantage.

The district court concluded that plaintiff failed to state a claim based on publication of the article because the challenged statements were protected scientific opinion. The Second Circuit agreed, concluding that, as a matter of law, statements of scientific conclusions about unsettled matters of scientific debate cannot give rise to liability for damages sounding in defamation.

Plaintiff claimed that the article made statements about scientific findings that were intentionally deceptive and misleading, and that it therefore constituted false advertising. Plaintiff’s theory was that scientific claims made in print purport to be statements of fact that are falsifiable, and such statements can be defamatory or represent false advertising if known to be false when made. Plaintiff argued on appeal that the district court based its conclusion that the article’s statements were non-actionable solely on its determination that the assertions were statements of opinion, without conducting a more vigorous inquiry. 

The court of appeals noted that scientific academic discourse poses several problems for the fact-opinion paradigm of traditional First Amendment jurisprudence. Many conclusions contained in a scientific journal article are, in principle, “capable of verification or refutation by means of objective proof,” Phantom Touring, Inc. v. Affiliated Publ’ns, 953 F.2d 724, 728 n.7 (1st Cir. 1992). Indeed, it is the very premise of the scientific enterprise that it engages with empirically verifiable facts about the universe. At the same time, however, it is the essence of the scientific method that the conclusions of empirical research are tentative and subject to revision, because they represent inferences about the nature of reality based on the results of experimentation and observation. Importantly, those conclusions are presented in publications directed to the relevant scientific community, ideally in peer reviewed academic journals that warrant that research approved for publication demonstrates at least some degree of basic scientific competence. These conclusions are then available to other scientists who may respond by attempting to replicate the described experiments, conducting their own experiments, or analyzing or refuting the soundness of the experimental design or the validity of the inferences drawn from the results.

In a sufficiently novel area of research, propositions of empirical “fact” advanced in the literature may be highly controversial and subject to rigorous debate by qualified experts. Needless to say, courts are ill-equipped to undertake to referee such controversies. Instead, the trial of ideas plays out in the pages of peer-reviewed journals, and the scientific public sits as the jury.

Where, as here, a statement is made as part of an ongoing scientific discourse about which there is considerable disagreement, the traditional dividing line between fact and opinion is not entirely helpful. While statements about contested and contestable scientific hypotheses constitute assertions about the world that are in principle matters of verifiable “fact,” for purposes of the First Amendment and the
laws relating to fair competition and defamation, they are more closely akin to matters of opinion, and are so understood by the relevant scientific communities.

In that regard, said the court, it was relevant that plaintiff did not allege that the data presented in the article were fabricated or fraudulently created. If the data were falsified, the fraud might not be easily detectable by even the most informed members of the relevant scientific community. Rather, plaintiff here alleged that the inferences drawn from those data were the wrong ones, and that competent scientists would have included other data that were available to the defendant authors but that were not sufficiently taken into account in their analysis.

But when the conclusions reached by experiments are presented alongside an accurate description of the data taken into account and the methods used, the validity of the authors’ conclusions may be
assessed on their face by other members of the relevant discipline or specialty.The appeals panel therefore concluded that, to the extent a speaker or author draws conclusions from non-fraudulent data, based on accurate descriptions of the data and methodology underlying those conclusions, on subjects about which there is legitimate ongoing scientific disagreement, those statements are not grounds for a claim of false advertising. Even if the conclusions that the authors drew from the results of their data could be actionable, such claims would be suspect when, as here, the authors readily disclosed the potential shortcomings of their methodology and their potential conflicts of interest

This analysis properly reflected a worry about the chilling impact on crucial and valuable research, including comparative effectiveness research, of lawsuits by competitors who are unhappy with or disagree with the results of such studies. Such a debate belongs in the marketplace of scientific ideas, not in the court room. 

State Supreme Court Affirms Exclusion of Unsound Expert Opinion

The Arkansas Supreme Court upheld recently the decision of a state trial court to bar a methodologically unsound expert report that sought to link fertilizer to a child's cancer.  Green v. George's Farms Inc., No. 10-26 (Ark. 2/17/11).

Defendants/Appellees were engaged in the poultry-production business in northwest Arkansas. For decades, their feeds have included the additive 3-Nitro, an FDA-approved product that is used to promote growth and prevent disease. Feed turns into excrement. Growers typically remove the chicken litter from poultry houses once a year, and the litter is then applied as fertilizer -- in this case to the fields surrounding plaintiffs' home, including areas near several schools. The chicken litter is spread primarily in the spring and fall, commonly at a ratio of two tons per acre. Sounds like good recycling.  But, said the court, roxarsone, an organic derivative of arsenic, comprises twenty percent of the ingredients contained in 3-Nitro. Arsenic, said the court, is a carcinogen and is considered both a cancer-causing agent and a promoter of cancer. The roxarsone that is fed to the chickens degrades into an inorganic form of arsenic that is excreted by the chickens.

Plaintiff Michael Green, was born and raised in the neighborhood near the schools and fields. In the fall of 1999, he experienced unexplained bruising, and was eventually diagnosed with a rare form of leukemia called chronic myelogenous leukemia (CML). He sued, alleging that the arsenic-laced litter produced by roxarsone-fed chickens polluted the air as a result of ventilating the chicken houses, and that dust clouds formed when the litter is spread, and that exposure to this inorganic arsenic caused his leukemia. (Other plaintiffs joined in the complaint, but the trial court severed the claims of the other plaintiffs.)  

In pretrial rulings, the circuit court granted summary judgment in favor of defendants on the issue of causation. Additionally, the court excluded under the Daubert rule all testimony pertaining to certain parts of a report entitled “Exposures to Carcinogenic Arsenicals and Other Toxic Substances in Washington County, Arkansas,” prepared by plaintiffs’ expert, Dr. Rod O’Connor.

On the (first) appeal the supreme court held that a question of fact remained on the issue of causation. Green I, 373 Ark. at 396, 284 S.W.3d at 42–43. However, the court found no abuse of discretion in the circuit court’s decision to exclude the testimony. Id. at 408, 284 S.W.3d at 51. On remand, the trial court granted a directed verdict, and this, the second appeal, ensued.

Noting that the lack of publication and peer review had influenced the court’s earlier decision, appellants asserted that Dr. O’Connor’s work had since been peer reviewed and published, and thus it should have been admitted on remand. Defendants responded that the law-of-the-case
doctrine precluded reconsideration of this issue. The trial court found that the published article utilized the same unreliable methodology to estimate peak air exposure concentration that it had previously ruled inadmissible. The trial court also stated that the expert's calculations were based on unreasonable assumptions and scientifically unsound mathematical extrapolations from dust samples collected in the area, and that Dr. O’Connor continued to use a formula that the EPA had warned should only be used to determine air levels of lead. The trial court found that the theory advanced by the expert had never been tested and still had not been sufficiently tested by any other scientist.

The doctrine of law of the case prohibits a court from reconsidering issues of law and fact that have already been decided in a case. The doctrine also provides that a decision of an appellate court establishes the law of the case for the trial upon remand and for the appellate court itself upon subsequent review. The law-of-the-case doctrine in some forms also prevents consideration of an argument that could have been raised at the first appeal and is not made until a subsequent appeal.  The doctrine serves to effectuate efficiency and finality in the judicial process, and its purpose is to maintain consistency.  The law-of-the-case doctrine is conclusive only where the facts on the second case/appeal are substantially the same as those involved in the prior appeal, and it does not apply if there was a material change in the facts.

And that was one of the issues here. The record reflected that the rejected parts of Dr. O’Connor’s report included his calculations relating to inhalation exposure reconstruction that were based on a formula for converting measurements of arsenic in dust to measurements of arsenic in air.  Plaintiffs argued that publication and peer review of his methodology represented a material change in circumstances that would permit reconsideration of this issue. However, the court recognized that publication is not a talisman or "get out of jail free" card from our favorite old board game. It was not apparent to the court whether the specific inhalation exposure reconstruction reflected in the report was indeed subjected to peer review in the article. The fact of publication (or lack thereof) in a peer reviewed journal thus is a relevant, though not dispositive, consideration in assessing the scientific validity of a particular technique or methodology on which an opinion is premised. 

Here, the absence of peer review and publication was but one factor in the overall assessment of the validity of Dr. O’Connor’s methodology. Given the serious flaws exposed in the expert's methods, the court was not persuaded that the publication of the article in a peer-reviewed journal constitutes a new fact that is sufficiently material to overcome the law-of- the-case doctrine. Further, plaintiffs’ argument that criticisms of his methods should only affect the weight but not the admissibility of the evidence was also barred by law of the case.


Peer Review (Part II)

In yesterday’s post, MassTortDefense began discussion of a peer review issue commented on in this week's New England Journal of Medicine. The editorial comments on two cases in which Pfizer served discovery on the journal, seeking peer-review and other editorial documents on manuscripts concerning Pfizer's Bextra and Celebrex, which of course have been the subject of litigation. Pfizer sought peer-review documents, including the critiques prepared by reviewers for the authors. The logic is that the critiques might help reveal flaws in methodology in the research. This was seen as particularly important given significant imprimatur of the NEJM and the use of its published studies in the cases.

In the case In re Bextra and Celebrex Marketing Sales Practices and Product Liability Litigation,
2008 WL 859207 (D.Mass. March 31, 2008) the court rejected the discovery, and granted the journal’s motion for a protective order. The court began by noting that the NEJM is the world's oldest and most frequently-cited medical journal. Physicians rely on information presented in the NEJM to follow medical developments and improve patient care. Manuscripts submitted for publication by the NEJM are subject to “peer review,” i.e., to screening and analysis conducted by experts in the subject matter discussed within the submitted articles. Peer reviewers are given a draft manuscript and are asked to provide a confidential, frank, honest evaluation of the manuscript's scientific validity and to provide an overall opinion with respect to a manuscript's “worthiness for publication.” Information from peer reviewers which is communicated to the authors does not reveal the identity of the peer reviewer. The NEJM informs prospective peer reviewers that it will maintain the confidentiality of their identities, unless the reviewer grants specific permission to the contrary. None of the peer reviewers for the eleven Bextra and Celebrex articles published had given the NEJM permission to disclose his or her identity.

The published articles were relevant to the MDL litigation and to plaintiffs’ allegations that Pfizer failed to act upon the results of studies described in the scientific literature, and that it was on notice from those studies of the alleged risks presented by the drugs.

At a hearing on the motion, Pfizer apparently narrowed the scope of its request, and withdrew its requests seeking certain documents reflecting the peer reviewer comments which were not shared with the authors, internal editorial comments or processes

Rule 26 and Rule 45 require application of a balancing test looking at whether the information can be obtained from some other source that is more convenient, less burdensome, or less expensive, the burden or expense of the proposed discovery, and its likely benefit, considering the needs of the case, the amount in controversy, the parties' resources, the importance of the issues at stake in the action, and the importance of the discovery in resolving the issues. Specifically here that would include the objector's interest in confidentiality and the potential injury to the free flow of information that disclosure may create. Another factor entitled, said the court, to special weight in conducting this balancing test is that the NEJM was a non-party to the MDL.

The district court concluded that the materials were relevant, but their probative value was limited. The reasoning was that plaintiffs' claims focus on what Pfizer knew, or should have known, via published articles in the scholarly literature. The peer reviewers' confidential comments don’t speak directly to that issue, the court thought. Of course, what a product seller should take from a published article does depend in great measure on the quality of the article, its methodology, is biases, its power, and other aspect peer reviewers are likely to comment upon. But the court felt that Pfizer's own experts were able to review and analyze the articles for flaws in methodology, and the peer reviewer comments contain no new data.

Peer review comments could also potentially be useful to impeach any causation experts of plaintiffs who rely on the studies, and to provide ammunition for defense causation experts. The court however, worried that this reflected a “growing problem” with the use of subpoenas to compel the giving of evidence and information by unretained experts. That seems a somewhat narrow and uniformed view of what defendants typically do when confronted with a mass tort or MDL: they retain as many of the best experts they can find.

Balancing against that probative value, the court analogized the peer review process to the level of protection offered academics engaged in pre-publication research. Because the peer review contributes to the advancement of medicine and science by helping to ensure that faulty, incomplete, or misleading results are not published, and because a journal’s ability to advance medical knowledge in this way might be impaired, resulting in adverse consequences for physicians, patient care, and for society as a whole, the Journal’s objections seemed persuasive to the court. If reviewers thought their names or reviews would be subject to disclosure in unrelated litigation, there would be “chilling effect” on the peer review process and as a result, upon the medical community. The comments are both part of scholarly research efforts as well as part of the editorial process of a print publication. The court concluded that any wholesale disclosure by the NEJM of the peer reviewer comments communicated to authors will be harmful to the NEJM's ability to fulfill both its journalistic and scholarly missions, and by extension harmful to the medical and scientific communities, and to the public interest. Thus the subpoena was quashed.

The Northern District of Illinois, similarly, declined to compel compliance with similar subpoenas served by Pfizer in the same MDL on the Journal of the American Medical Association and the Archives of Internal Medicine. In re Bextra and Celebrex, C.A. No. 08C 402 (N.D.Ill. March 14, 2008). The district court there found that the any probative value of the requested information was outweighed by the burden imposed on the journals in invading the sanctity of the peer review process. In short, it was not unreasonable to believe that compelling production of peer review documents would compromise the process.

But it raises an interesting, perhaps test-able, question whether disclosure of comments would discourage people from submitting articles to the journal, or peer reviewers from commenting on manuscripts.  Isn't a prestigious thing to be able to say "I am a peer reviewer for the NEJM"? After all, a large part of the the academic world involves defending one's scholarly work in public, in symposia, in class rooms, etc.  And scientists frequently publish comments, letters to the editor, publicly critiquing others' work. Wonder if any respected journal conduct peer review in which complete confidentiality is not promised the reviewers?

Yesterday’s editorial commented that confidential peer review is a key component of the manuscript selection process. The Journal’s stand was consistent with the views of the International Committee of Medical Journal Editors. The ICMJE is a group of about 500 general medical journal editors whose participants meet annually and has produced a number of editions of uniform requirements for manuscripts submitted to biomedical journals. See Uniform Requirements for Manuscripts Submitted to Biomedical Journals: Writing and Editing for Biomedical Publication.

In its Uniform Requirements, the ICMJE states that unbiased, independent, critical assessment is an intrinsic part of all scholarly work, including the scientific process. Peer review is the critical assessment of manuscripts submitted to journals by experts who are not part of the editorial staff. Peer review can therefore be viewed as an important extension of the scientific process. Reviewers have rights to confidentiality, which must be respected by the editor. Editors must not disclose information about manuscripts (including their receipt, content, status in the reviewing process, criticism by reviewers, or ultimate fate) to anyone other than the authors and reviewers. “This includes requests to use the materials for legal proceedings.”

It will be interesting to watch and see if the issue arises again in a case in which the party seeking the information -- and might it be plaintiffs under the proper facts? -- can show a higher degree of probative value.

Peer Review, Daubert, and the New England Journal of Medicine

The new England Journal of Medicine weighed in today with an editorial on a topic that should be important for readers of MassTortDefense – peer review. Curfman, et al., Peer Review in the Balance, 358 NEJM 2276 (May 22, 2208).

The editorial weighs in on two recent decisions in the Celebrex litigation involving product liability defendant’s attempts to go beyond the veil of the published article and obtain discovery on information created in the peer review process as manuscripts were assessed. Those manuscripts related to studies of science relevant to the litigation, and the argument was advanced that commentary, discussion, and analysis of those studies by expert peer reviewers likely also contained relevant information about that pertinent science.

Legal Setting

Expert testimony is essential in virtually every mass tort or complex product liability litigation. Under the gate-keeping function of the trial judge, designed to keep junk science out of the courtroom, the Daubert opinion lists important factors to consider in admission of expert testimony, including whether the theory or technique has been subjected to peer review and publication. This factor is relevant, but not dispositive, because sometimes well-grounded but innovative theories will not yet have been published, and sometimes a topic may be of too limited interest to be published. But the Court believed that peer review increases likelihood that substantive flaws will be detected in an article, and publication places the work within the test of the marketplace of ideas. The theory is available to be scrutinized, and has been scrutinized by experts in the relevant field. In essence, submission of the method, theory, reasoning, to scrutiny of scientific community is considered component of good science.

The Daubert Court was aware from amicus briefing of another view of peer review, emphasizing that the peer review system is not a litmus test for truth, not a guarantor of scientific certainty, and does not guarantee that the particular data, research and analytical methodologies, or conclusions of the accepted papers represent the consensus opinion of the relevant community. Brief of Amici Curiae Chubin et al., Daubert, 1992 U.S. S.Ct. Briefs LEXIS 938. But the Court generally rejected this view.

There is some lower court variation in application of Daubert and Frye on peer review; some courts emphasize the mere fact of peer review, Doe v. Ortho-Clinical Diagnostics Inc., 440 F.Supp. 2d 465 (M.D.N.C. 2006)(thimerosal), with others offering a more nuanced view of peer reviewed publication serving as a validity enhancer. The presence of the limited view encourages plaintiffs to seek to get theory published somewhere to “check the box” of peer reviewed publication. The more rigorous view reads the Supreme Court as clearly intending that judges use their own evaluation of peer review and publication as a tool for exploring whether substantive flaws in an expert's methodology have been or could have been exposed. Black v. Rhone-Poulenc, Inc., 19 F.Supp.2d 592 (S.D.W.Va. 1998) (depends upon how the peer review was performed; who did it; plaintiffs failed to provide sufficient information to assess).

Now, before getting into the two recent decisions and the editorial, a brief background on peer review.

Brief History of Peer Review

Modern medical journals came into existence during the 19th century. They were originally a vehicle for personal journalism, with the models for medical editors including the crusading newspaper writer. Publication was a personal vehicle for the editor. Over time, newspapers came to resemble newspapers, with correspondents to cover medicine in different geographic and specialty areas. But in many medical journals, the editors wrote much or all of content. A second type of journal also arose, in addition to those journals that followed the newspaper model, and that was the official publications of research institutes, typically specializing in one kind of research.

In the 19th century editors often could not fill their columns. The problem was finding publishable manuscripts, not turning down original contributions. Editorial board members would help round up material for the editor to publish, rather than review and reject manuscripts. In the 20th century, the supply/demand ratio gradually reversed. The number of passable manuscripts began to increase greatly, so that choosing among good articles became an option. Between 1913 and 1925, the Journal of the American Medical Association received 1500-2000 manuscripts for consideration. And the end of World War II marked the beginning of a dramatic increase in the volume of research and the subsequent reporting of research results. So by the 1990’s, JAMA was receiving 4500 papers/year, plus 5500 letters to editor. Yet JAMA publishes about 500 articles a year. Or the NEJM, which was getting more than 4000 article submissions/year, and publishing less than 10%.

Today there are more than 10,000 medical and scientific journals, and more than 2 million scientific and medical articles published each year. Most are never cited again.

There are examples of historical scientific journals in France and England with well thought-out peer review process to select reports to be published in journals and memoirs. In 1752, the Royal Society of London, a medical society, began publishing Philosophical Transactions and established a "Committee on Papers" to review articles submitted for publication. But in the U.S., most journals were personal fiefdoms of editors. (E.g., James McKeen Cattell, who for 50 years edited (and owned) Science and other major journals at the same time. ) As noted, editors wrote much of the content; they considered themselves as expert as anyone to assess content. The only mechanism that resembled modern peer review, perhaps, was book publishers seeking outside advice regarding textbook manuscripts in technical fields.

Peer review developed in part to deal with situations in which an editor lacked the specialized knowledge to make decisions about highly technical articles. Particularly after WWII, more specialists in subject matter, in terms of technique and, especially, in medicine, and laboratory materials were needed. And when physicians began to specialize in particular fields, it also spawned a demand for specialized journals catering to that specialty.  The supply/demand ratio gradually reversed and there was a growing need to discriminate high-quality articles and turn down others in the growing volume of at least plausibly good submissions that they had to consider.

Fast Forward to Today
An example of rigorous peer review is the New England Journal of Medicine, which publishes less than 10% of the unsolicited manuscripts it receives. There is a multi-step process including an initial screen by and editor; and internal review by an associate editor; submission to at least 2 peer reviewers/referees; presentation of the article to an editorial meeting; a statistical/math review, etc. There is a step for an additional review and revision by the author.

Today, scientific and medical journal have a significant role in shaping the views and behavior of scientists, clinicians, and policy makers. Contrary to the theory that journals act as passive conduits for scientific content, studies show the impact of an article is strongly influenced by which journal published it, and authors know that. There is a potential for a few high-impact journals to shape heavily the use of scientific information. At the same time, the increasing number of biomedical journals, the advent of online publication, and the increasingly interdisciplinary nature and impact of research, are forcing journals to compete for the best research.

With that background, the current controversy in the next posting.