TORTINI

For your delectation and delight, desultory dicta on the law of delicts.

Counter Cancel Culture – Part II: The Fixing Science Conference

February 12th, 2020

So this is what it is like to be denounced? My ancestors fled the Czar’s lands before they could be tyrannized by denunciations of Stalin’s Soviets. The work of contemporary denunciators is surely much milder, but no more principled than the Soviet versions of yesteryear.

Now that I am back from the Fixing Science conference, sponsored by the Independent Institute and the National Association of Scholars (NAS), I can catch up with the media coverage of the event. I have already addressed Dr. Lenny Teytelman’s issues in an open letter to him. John Mashey is a computer scientist who has written critical essays on climate science denial. On the opening day of the NAS conference, he published online his take on the recent NAS’s conference on scientific irreproducibility.[1] Mashey acknowledges that the Fixing Science conference included “credible speakers who want to improve some areas of science hurt by the use of poor statistical methods or making irreproducible claims,” but his post devolves into scurrilous characterizations of several presenters. Alas, some of the ad hominems are tossed at me, and here is what I have to say about them.

Mashey misspells my name, “Schactman,” but that is a minor flaw of scholarship. He writes that I have “published much on evidence,” which is probably too laudatory. I am hardly a recognized scholar on the law of evidence, although I know something about this area, and have published in it.

Mashey tautologically declares that I “may or may not be a ‘product defense lawyer’ (akin to Louis Anthony Cox) defending companies against legitimate complaints.” Mashey seems unaware of how the rule of law works in our country. Plaintiffs file complaints, but the standard for the legitimacy of these complaints is VERY low. Courts require the parties to engage in discovery of their claims and defenses, and then courts address dispositive motions to dismiss either the claims or the defenses. So, sometimes after years of work, legitimate complaints are revealed to be bogus complaints, and then the courts will dismiss bogus complaints, and thus legitimate complaints become illegitimate complaints. In my 36 years at the bar, I am proud to have been able to show that a great many apparently legitimate complaints were anything but what they seemed.

Mashey finds me “worrying” and “concerning.” My children are sometimes concerned about me, and even worry about me, about I do not think that Mashey was trying to express solicitude for me.

Why worry? Well, David Michaels in his most recent book, Triumph of Doubt (2020), has an entire chapter on silica dust. And I, worrisomely, have written and spoken, about silica and silicosis litigation, sometimes in a way critical of the plaintiffs’ litigation claims. Apparently, Mashey does not worry that David Michaels may be an unreliable protagonist who worked as a paid witness for the lawsuit industry on many occasions before becoming the OSHA Administrator, in which position he ignored enforcement of existing silica regulations in order to devote a great deal of time, energy, and money to revising the silica regulations. The evidentiary warrant for Michaels’ new silica rule struck me then, and now, as slim, but the real victims, workers, suffered because Michaels was so intent on changing a rule in the face of decades of declining silicosis mortality, that he failed, in my view, to attend to specific instances of over-exposure.

Mashey finds me concerning because two radical labor historians do not like me. (I think I am going eat a worm, ….) Mashey quotes at length from an article by these historians, criticizing me for having had the audacity to criticize them.[2] Oh my.

What Mashey does not tell his readers was that, as co-chair of a conference on silicosis litigation (along with a co-chair who was a plaintiffs’ lawyer), I invited historian Gerald Markowitz to speak and air his views on the history of silica regulation and litigation. In response, I delivered a paper that criticized, and I would dare say, rebutted many of Markowitz’s historical conclusions and his inferences from an incomplete, selectively assembled, and sometimes incorrect, set of historical facts. I later published my paper.

Mashey tells his readers that my criticisms, based not upon what I wrote, but upon the partisan cries of Rosner and Markowitz, “seems akin to Wood’s style of attack.” Well, if so, nicely done, Woods.

But does Mashey believe that his readers deserve to know that Rosner and Markowitz have testified repeatedly on behalf of the lawsuit industry, that is, those entrepreneurs who make lawsuits?[3] And that Rosner and Markowitz have been amply remunerated for their labors as partisan witnesses in these lawsuits?

And is Mashey worried or concerned that in the United States, silicosis litigation has been infused with fraud and deception, not by the defendants, but by the litigation industry that creates the lawsuits? Absent from Rosner and Markowitz’s historical narratives is any mention of the frauds that have led to dismissals of thousands of cases, and the professional defrocking of any number of physician witnesses.  In re Silica Products Liab. Litig., MDL No. 1553, 398 F. Supp. 2d 563 (S.D.Tex. 2005). Even the redoubtable expert witness for the plaintiffs’ bar, David S. Egilman, has published articles that point out the unethical and unlawful nature of the medico-legal screenings that gave rise to the silicosis litigation, which Michaels, Rosner, and Markowitz seem to support, or at the very least suppress any criticism of.[4]

So this is what it means to be denounced! Mashey’s piece is hardly advertisement for the intellectual honesty of those who would de-platform the NAS conference. He has selectively and inaccurately addressed my credentials. As just one example, and in an effort to diminish the NAS, he has omitted that I have received a grant from the NASEM to develop a teaching module on scientific causation. My finished paper is published online at the NASEM website.[5]

I do not know Mashey, but I leave it to you to judge him by his sour fruits.


[1]  John Mashey, “Dark-Moneyed Denialists Are Running ‘Fixing Science’ Symposium of Doubt,” Desmog Blog (Feb. 7, 2020).

[2]  David Rosner & Gerald Markowitz, “The Trials and Tribulations of Two Historians:  Adjudicating Responsibility for Pollution and Personal Harm, 53 Medical History 271, 280-81 (2009) (criticizing me for expressing the view that historians should not be permitted to testify and thereby circumvent the rules of evidence). See also David Rosner & Gerald Markowitz, “L’histoire au prétoire.  Deux historiens dans les procès des maladies professionnelles et environnementales,” 56 Revue D’Histoire Moderne & Contemporaine 227, 238-39 (2009) (same); D. Rosner, “Trials and Tribulations:  What Happens When Historians Enter the Courtroom,” 72 Law & Contemporary Problems 137, 152 (2009) (same). I once thought there was an academic standard that prohibited duplicative publication!

[3] I have been critical of Rosner and Markowitz on many occasions; they have never really responded to the substance of my criticisms. See, e.g., “How Testifying Historians Are Like Lawn-Mowing Dogs,” (May 15, 2010).

[4]  See David Egilman and Susanna Rankin Bohme, “Attorney-directed screenings can be hazardous,” 45 Am. J. Indus. Med. 305 (2004); David Egilman, “Asbestos screenings,” 42 Am. J. Indus. Med. 163 (2002).

[5]  “Drug-Induced Birth Defects: Exploring the Intersection of Regulation, Medicine, Science, and Law – An Educational Module” (2016) (A teaching module designed to help professional school students and others evaluate the role of science in decision-making, developed for the National Academies of Science, Engineering, and Medicine, and its Committee on Preparing the Next Generation of Policy Makers for Science-Based Decisions).

Counter Cancel Culture – The NAS Conference on Irreproducibility

February 9th, 2020

The meaning of the world is the separation of wish and fact.”  Kurt Gödel

Back in October 2019, David Randall, the Director of Research, of the National Association of Scholars, contacted me to ask whether I would be interested in presenting at a conference, to be titled “Fixing Science: Practical Solutions for the Irreproducibility Crisis.” David explained that the conference would be aimed at a high level consideration of whether such a crisis existed, and if so, what salutary reforms might be implemented.

As for the character and commitments of the sponsoring organizations, David was candid and forthcoming. I will quote him, without his permission, and ask his forgiveness later:

The National Association of Scholars is taken to be conservative by many scholars; the Independent Institute is (broadly speaking) in the libertarian camp. The NAS is open to but currently agnostic about the degree of human involvement in climate change. The Independent Institute I take to be institutionally skeptical of consensus climate change theory–e.g., they recently hosted Willie Soon for lecture. A certain number of speakers prefer not to participate in events hosted by institutions with these commitments.”

To me, the ask was for a presentation on how the so-called replication crisis, or the irreproducibility crisis, affected the law. This issue was certainly one I have had much occasion to consider. Although I am aware of the “adjacency” arguments made by some that people should be mindful of whom they align with, I felt that nothing in my participation would compromise my own views or unduly accredit institutional positions of the sponsors.

I was flattered by the invitation, but I did some due diligence on the sponsoring organizations. I vaguely recalled the Independent Institute from my more libertarian days, but the National Association of Scholars (NAS, not to be confused with Nathan A. Schachtman) was relatively unknown to me. A little bit of research showed that the NAS had a legitimate interest in the irreproducibility crisis. David Randall had written a monograph for the organization, which was a nice summary of some of the key problems. The Irreproducibility Crisis of Modern Science: Causes, Consequences,and the Road to Reform (2018).

On other issues, the NAS seemed to live up to its description as “an organization of scholars committed to higher education as the catalyst of American freedom.” I listened to some of the group’s podcasts, Curriculum Vitae, and browsed through its publications. I found myself agreeing with many positions articulated by or through the NAS, and disagreeing with a few positions very strongly.

In looking over the list of other invited speakers, I saw great diversity of view points and approaches, One distinguished speaker, Daniele Fanelli, had criticized the very notion that there was a reproducibility crisis. In the world of statistics, there were strong defenders of statistical tests, and vociferous critics. I decided to accept the invitation, not because I was flattered, but because the replication issue was important, and I believed that I could add something to the discussion before an audience of professional scientists, statisticians, and educated lay persons. In writing to David Randall to accept the invitation, I told him that with respect to the climate change issues, I was not at all put off by healthy skepticism in the face all dogmas. Every dogma will have its day.

I did not give any further consideration to the political aspect of the conference until early January, when I received an email from a scientist, Lenny Teytelman, Ph.D., the C.E.O. of a company protocols.io, which addresses reproducibility issues. Dr Teytelman’s interest in improving reproducibility seemed quite genuine, but he wrote to express his deep concern about the conference and the organizations that were sponsoring it.

Perhaps a bit pedantically, he cautioned me that the NAS was not the National Academy of Sciences, a confusion that never occurred to me because the National Academies has been known as the National Academies of Science, Engineering and Medicine for several years now. Dr. Teytelman’s real concern seemed to be that the NAS is a “‘politically conservative advocacy group’.” (The internal scare quotes were Teytelman’s, but I was not afraid.) According to Dr. Teytelman, the NAS sought to undermine climate science and environmental protection by advancing a call for more reproducible science. He pointed me to what he characterized as an exposé on NAS, in Undark,1 and he cautioned me that the National Association of Scholars’ work is “dangerous.” Finally, Dr. Teytelman urged me to reconsider my decision to participate in the conference.

I did reconsider my decision, but reaffirmed it in an email I sent back to Dr. Teytelman. I realized that I could be wrong, in which case, I would eat my words, confident that they would be most digestible:

Dear Dr Teytelman,

Thank you for your note. I was aware of the piece on Undark’s website, as well as the difference between the NAS and the NASEM. I don’t believe anyone involved in science education would likely to be confused between the two organizations. A couple of years ago, I wrote a teaching module on biomedical causation for the National Academies. This is my first presentation at the request of the NAS, and frankly I am honored by the organization’s request that I present at its conference.

I have read other materials that have been critical of the NAS and its publications on climate change and other issues. I know that there are views of the organization from which I would dissent, but I do not see my disagreement on some issues as a reason not to attend, and present at a conference on an issue of great importance to the legal system.

I am hardly an expert on climate change issues, and that is my failing. Most of my professional work involves health effects regulation and litigation. If the NAS has advanced sophistical arguments against a scientific claim, then the proper antidote will be to demonstrate its fallacious reasoning and misleading marshaling of evidence. I should think, however, as someone interested in improving the reproducibility of scientific research, you will agree that there is much common ground for discussion and reform of scientific practice, on a broader arrange [sic] of issues than climate change.

As for the political ‘conservatism’, of the organization, I am not sure why that is a reason to eschew participation in a conference that should be of great importance to people of all political views. My own politics probably owe much to the influence of Michael Oakeshott, which puts me in perhaps the smallest political tribe of any in the United States. If conservatism means antipathy to post-modernism, identity politics, political orthodoxies, and assaults on Enlightenment values and the Rule of Law, then count me in.

In any event, thanks for your solicitude. I think I can participate and return with my soul intact.

All the best.

Nathan

To his credit, Dr. Teytelman tenaciously continued. He acknowledged that the political leanings of the organizers were not a reason to boycott, but he politely pressed his case. We were now on a first name basis:

Dear Nathan,

I very much applaud all efforts to improve the rigour of our science. The problem here is that this NAS organization has a specific goal – undermining the environmental protection and denying climate change. This is why 7 out of the 21 speakers at the event are climate change deniers. [https://docs.google.com/spreadsheets/d/136FNLtJzACc6_JbbOxjy2urbkDK7GefRZ/edit?usp=sharing] And this isn’t some small fringe effort to be ignored. Efforts of this organization and others like them have now gotten us to the brink of a regulatory change at the United States Environmental Protection Agency which can gut the entire EPA (see a recent editorial against this I co-authored). This conference is not a genuine effort to talk about reproducibility. The reproducibility part is a clever disguise for pushing a climate change denialism agenda.

Best,

Lenny

I looked more carefully at Lenny’s spreadsheet, and considered the issue afresh. We were both pretty stubborn:

Dear Lenny,

Thank you for this information. I will review with interest.

I do not see that the conference is primarily or even secondarily about climate change vel non. There are two scientists, Trafimow and Wasserstein with whom I have some disagreements about statistical methodology. Tony Cox and Stan Young, whatever their political commitments or views on climate change may be, are both very capable statisticians, from whom I have learned a great deal. The conference should be a lively conversation about reproducibility, not about climate change. Given your interests and background, you should go.

I believe that your efforts here are really quite illiberal, although they are in line with the ‘cancel culture’, so popular on campuses these days.

Forty three years ago, I entered a Roman Catholic Church to marry the woman I love. There were no lightning bolts or temblors, even though I was then and I am now an atheist. Yes, I am still married to my first wife. Although I share the late Christopher Hitchins’ low view of the Catholic Church, somehow I managed to overcome my antipathy to being married in what some would call a house of ill repute. I even manage to agree with some Papist opinions, although not for the superstitious reasons’ Papists embrace.

If I could tolerate the RC Church’s dogma for a morning, perhaps you could put aside the dichotomous ‘us and them’ view of the world and participate in what promises to be an interesting conference on reproducibility?

All the best.

Nathan

Lenny kindly acknowledged my having considered his issues, and wrote back a nice note, which I will quote again in full without permission, but with the hope that he will forgive me and even acknowledge that I have given his views an airing in this forum.

Hi Nathan,

We’ll have to agree to disagree. I don’t want to give a veneer of legitimacy to an organization whose goal is not improving reproducibility but derailing EPA and climate science.

Warmly,

Lenny

The business of psychoanalyzing motives and disparaging speakers and conference organizers is a dangerous business for several reasons. First motives can be inscrutable. Second, they can be misinterpreted. And third, they can be mixed. When speaking of organizations, there is the further complication of discerning a corporate motive among the constituent members.

The conference was an exciting, intellectually challenging event, which took place in Oakland, California, on February 7 and 8. I can report back to Lenny that his characterizations of and fears about the conference were unwarranted. While there were some assertions of climate change skepticism made with little or no evidence, the evidence-based presentations essentially affirmed climate change and sought to understand its causes and future course in a scientific way. But climate change was not why I went to this conference. On the more general issue of reform of scientific procedures and methods, we had open debates, some agreement on important principles, and robust and reasoned disagreement.

Lenny, you were correct that the NAS should not be ignored, but you should have gone to the meeting and participated in the conversation.


1 Michael Schulson, “A Remedy for Broken Science, Or an Attempt to Undercut It?Undark (April 18, 2018).

Judicial Gatekeeping Cures Claims That Viagra Can Cause Melonoma

January 24th, 2020

The phosphodiesterases 5 inhibitor medications (PDE5i) seem to arouse the litigation propensities of the lawsuit industry. The PDE5i medications (sildenafil, tadalafil, etc.) have multiple indications, but they are perhaps best known for their ability to induce penile erections, which in some situations can be a very useful outcome.

The launch of Viagra in 1998 was followed by litigation that claimed the drug caused heart attacks, and not the romantic kind. The only broken hearts, however, were those of the plaintiffs’ lawyers and their expert witnesses who saw their litigation claims excluded and dismissed.[1]

Then came claims that the PDE5i medications caused non-arteritic anterior ischemic optic neuropathy (“NAION”), based upon a dubious epidemiologic study by Dr. Gerald McGwin. This litigation demonstrated, if anything, that while love may be blind, erections need not be.[2] The NAION cases were consolidated in a multi-district litigation (MDL) in front of Judge Paul Magnuson, in the District of Minnesota. After considerable back and forth, Judge Manguson ultimately concluded that the McGwin study was untrustworthy, and the NAION claims were dismissed.[3]

In 2014, the American Medical Association’s internal medicine journal published an observational epidemiologic study of sildenafil (Viagra) use and melanoma.[4] The authors of the study interpreted their study modestly, concluding:

“[s]ildenafil use may be associated with an increased risk of developing melanoma. Although this study is insufficient to alter clinical recommendations, we support a need for continued investigation of this association.”

Although the Li study eschewed causal conclusions and new clinical recommendations in view of the need for more research into the issue, the litigation industry filed lawsuits, claiming causality.[5]

In the new natural order of things, as soon as the litigation industry cranks out more than a few complaints, an MDL results, and the PDE5i – melanoma claims were no exception. By spring 2016, plaintiffs’ counsel had collected ten cases, a minion, sufficient for an MDL.[6] The MDL plaintiffs named the manufacturers of sildenafil and tadalafil, two of the more widely prescribed PDEi5 medications, on behalf of putative victims.

While the MDL cases were winding their way through discovery and possible trials, additional studies and meta-analyses were published. None of the subsequent studies, including the systematic reviews and meta-analyses, concluded that there was a causal association. Most scientists who were publishing on the issue opined that systematic error (generally confounding) prevented a causal interpretation of the data.[7]

Many of the observational studies found statistically significantly increased relative risks about 1.1 to 1.2 (10 to 20%), typically with upper bounds of 95% confidence intervals less than 2.0. The only scientists who inferred general causation from the available evidence were those who had been recruited and retained by plaintiffs’ counsel. As plaintiffs’ expert witnesses, they contended that the Li study, and the several studies that became available afterwards, collectively showed that PDE5i drugs cause melanoma in humans.

Not surprisingly, given the absence of any non-litigation experts endorsing the causal conclusion, the defendants challenged plaintiffs’ proffered expert witnesses under Federal Rule of Evidence 702. Plaintiffs’ counsel also embraced judicial gatekeeping and challenged the defense experts. The MDL trial judge, the Hon. Richard Seeborg, held hearings with four days of viva voce testimony from four of plaintiffs’ expert witnesses (two on biological plausibility, and two on epidemiology), and three of the defense’s experts. Last week, Judge Seeborg ruled by granting in part, and denying in part, the parties’ motions.[8]

The Decision

The MDL trial judge’s opinion is noteworthy in many respects. First, Judge Richard Seeborg cited and applied Rule 702, a statute, and not dicta from case law that predates the most recent statutory version of the rule. As a legal process matter, this respect for judicial process and the difference in legal authority between statutory and common law was refreshing. Second, the judge framed the Rule 702 issue, in line with the statute, and Ninth Circuit precedent, as an inquiry whether expert witnesses deviated from the standard of care of how scientists “conduct their research and reach their conclusions.”[9]

Biological Plausibility

Plaintiffs proffered three expert witnesses on biological plausibility, Drs. Rizwan Haq, Anand Ganesan, and Gary Piazza. All were subject to motions to exclude under Rule 702. Judge Seeborg denied the defense motions against all three of plaintiffs’ plausibility witnesses.[10]

The MDL judge determined that biological plausibility is neither necessary nor sufficient for inferring causation in science or in the law. The defense argued that the plausibility witnesses relied upon animal and cell culture studies that were unrealistic models of the human experience.[11] The MDL court, however, found that the standard for opinions on biological plausibility is relatively forgiving, and that the testimony of all three of plaintiffs’ proffered witnesses was admissible.

The subjective nature of opinions about biological plausibility is widely recognized in medical science.[12] Plausibility determinations are typically “Just So” stories, offered in the absence of hard evidence that postulated mechanisms are actually involved in a real causal pathway in human beings.

Causal Association

The real issue in the MDL hearings was the conclusion reached by plaintiffs’ expert witnesses that the PDE5i medications cause melanoma. The MDL court did not have to determine whether epidemiologic studies were necessary for such a causal conclusion. Plaintiffs’ counsel had proffered three expert witnesses with more or less expertise in epidemiology: Drs. Rehana Ahmed-Saucedo, Sonal Singh, and Feng Liu-Smith. All of plaintiffs’ epidemiology witnesses, and certainly all of defendants’ experts, implicitly if not explicitly embraced the proposition that analytical epidemiology was necessary to determine whether PDE5i medications can cause melanoma.

In their motions to exclude Ahmed-Saucedo, Singh, and Liu-Smith, the defense pointed out that, although many of the studies yielded statistically significant estimates of melanoma risk, none of the available studies adequately accounted for systematic bias in the form of confounding. Although the plaintiffs’ plausibility expert witnesses advanced “Just-So” stories about PDE5i and melanoma, the available studies showed an almost identical increased risk of basal cell carcinoma of the skin, which would be explained by confounding, but not by plaintiffs’ postulated mechanisms.[13]

The MDL court acknowledged that whether epidemiologic studies “adequately considered” confounding was “central” to the Rule 702 inquiry. Without any substantial analysis, however, the court gave its own ipse dixit that the existence vel non of confounding was an issue for cross-examination and the jury’s resolution.[14] Whether there was a reasonably valid association between PDE5i and melanoma was a jury question. This judicial refusal to engage with the issue of confounding was one of the disappointing aspects of the decision.

The MDL court was less forgiving when it came to the plaintiffs’ epidemiology expert witnesses’ assessment of the association as causal. All the parties’ epidemiology witnesses invoked Sir Austin Bradford Hill’s viewpoints or factors for judging whether associations were causal.[15] Although they embraced Hill’s viewpoints on causation, the plaintiffs’ epidemiologic expert witnesses had a much more difficult time faithfully applying them to the evidence at hand. The MDL court concluded that the plaintiffs’ witnesses deviated from their own professional standard of care in their analysis of the data.[16]

Hill’s first enumerated factor was “strength of association,” which is typically expressed epidemiologically as a risk ratio or a risk difference. The MDL court noted that the extant epidemiologic studies generally showed relative risks around 1.2 for PDE5i and melanoma, which was “undeniably” not a strong association.[17]

The plaintiffs’ epidemiology witnesses were at sea on how to explain away the lack of strength in the putative association. Dr. Ahmed-Saucedo retreated into an emphasis on how all or most of the studies found some increased risk, but the MDL court correctly found that this ruse was merely a conflation of strength with consistency of the observed associations. Dr. Ahmed-Saucedo’s dismissal of the importance of a dose-response relationship, another Hill factor, as unimportant sealed her fate. The MDL court found that her Bradford Hill analysis was “unduly results-driven,” and that her proffered testimony was not admissible.[18] Similarly, the MDL court found that Dr. Feng Liu-Smith similarly conflated strength of association with consistency, which error was too great a professional deviation from the standard of care.[19]

Dr. Sonal Singh fared no better after he contradicted his own prior testimony that there is an order of importance to the Hill factors, with “strength of association,” at or near the top. In the face of a set of studies, none of which showed a strong association, Dr. Singh abandoned his own interpretative principle to suit the litigation needs of the case. His analysis placed the greatest weight on the Li study, which had the highest risk ratio, but he failed to advance any persuasive reason for his emphasis on one of the smallest studies available. The MDL court found that Dr. Singh’s claim to have weighed strength of association heavily, despite the obvious absence of strong associations, puzzling and too great an analytical gap to abide.[20]

Judge Seeborg thus concluded that while the plaintiffs’ expert witness could opine that there was an association, which was arguably plausible, they could not, under Rule 702, contend that the association was causal. In attempting to advance an argument that the association met Bradford Hill’s factors for causality, the plaintiffs’ witnesses had ignored, misrepresented, or confused one of the most important factors, strength of the association, in a way that revealed their analyses to be result driven and unfaithful to the methodology they claimed to have followed. Judge Seeborg emphasized a feature of the revised Rule 702, which often is ignored by his fellow federal judges:[21]

“Under the amendment, as under Daubert, when an expert purports to apply principles and methods in accordance with professional standards, and yet reaches a conclusion that other experts in the field would not reach, the trial court may fairly suspect that the principles and methods have not been faithfully applied. See Lust v. Merrell Dow Pharmaceuticals, Inc., 89 F.3d 594, 598 (9th Cir. 1996). The amendment specifically provides that the trial court must scrutinize not only the principles and methods used by the expert, but also whether those principles and methods have been properly applied to the facts of the case.”

Given that the plaintiffs’ witnesses purported to apply a generally accepted methodology, Judge Seeborg was left to question why they would conclude causality when no one else in their field had done so.[22] The epidemiologic issue had been around for several years, and addressed not just in observational studies, but systematically reviewed and meta-analyzed. The absence of published causal conclusions was not just an absence of evidence, but evidence of absence of expert support for how plaintiffs’ expert witnesses applied the Bradford Hill factors.

Reliance Upon Studies That Did Not Conclude Causation Existed

Parties challenging causal claims will sometimes point to the absence of a causal conclusion in the publication of individual epidemiologic studies that are the main basis for the causal claim. In the PDE5i-melanoma cases, the defense advanced this argument unsuccessfully. The MDL court rejected the defense argument, based upon the absence of any comprehensive review of all the pertinent evidence for or against causality in an individual study; the study authors are mostly concerned with conveying the results of their own study.[23] The authors may have a short discussion of other study results as the rationale for their own study, but such discussions are often limited in scope and purpose. Judge Seeborg, in this latest round of PDE5i litigation, thus did not fault plaintiffs’ witnesses’ reliance upon epidemiologic or mechanistic studies, which individually did not assert causal conclusions; rather it was the absence of causal conclusions in systematic reviews, meta-analyses, narrative reviews, regulatory agency pronouncements, or clinical guidelines that ultimately raised the fatal inference that the plaintiffs’ witnesses were not faithfully deploying a generally accepted methodology.

The defense argument that pointed to the individual epidemiologic studies themselves derives some legal credibility from the Supreme Court’s opinion in General Electric Co. v. Joiner, 522 U.S. 136 (1997). In Joiner, the SCOTUS took plaintiffs’ expert witnesses to task for drawing stronger conclusions than were offered in the papers upon which they relied. Chief Justice Rehnquist gave considerable weight to the consideration that the plaintiffs’ expert witnesses relied upon studies, the authors of which explicitly refused to interpret as supporting a conclusion of human disease causation.[24]

Joiner’s criticisms of the reliance upon studies that do not themselves reach causal conclusions have gained a foothold in the case law interpreting Rule 702. The Fifth Circuit, for example, has declared:[25]

“It is axiomatic that causation testimony is inadmissible if an expert relies upon studies or publications, the authors of which were themselves unwilling to conclude that causation had been proven.”

This aspect of Joiner may properly limit the over-interpretation or misinterpretation of an individual study, which seems fine.[26] The Joiner case may, however, perpetuate an authority-based view of science to the detriment of requiring good and sufficient reasons to support the testifying expert witnesses’ opinions.  The problem with Joiner’s suggestion that expert witness opinion should not be admissible if it disagrees with the study authors’ discussion section is that sometimes study authors grossly over-interpret their data.  When it comes to scientific studies written by “political scientists” (scientists who see their work as advancing a political cause or agenda), then the discussion section often becomes a fertile source of unreliable, speculative opinions that should not be given credence in Rule 104(a) contexts, and certainly should not be admissible in trials. In other words, the misuse of non-rigorous comments in published articles can cut both ways.

There have been, and will continue to be, occasions in which published studies contain data, relevant and important to the causation issue, but which studies also contain speculative, personal opinions expressed in the Introduction and Discussion sections.  The parties’ expert witnesses may disagree with those opinions, but such disagreements hardly reflect poorly upon the testifying witnesses.  Neither side’s expert witnesses should be judged by those out-of-court opinions.  Perhaps the hearsay discussion section may be considered under Rule 104(a), which suspends the application of the Rules of Evidence, but it should hardly be a dispositive factor, other than raising questions for the reviewing court.

In exercising their gatekeeping function, trial judges should exercise care in how they assess expert witnesses’ reliance upon study data and analyses, when they disagree with the hearsay authors’ conclusions or discussions.  Given how many journals cater to advocacy scientists, and how variable the quality of peer review is, testifying expert witnesses should, in some instances,  have the expertise to interpret the data without substantial reliance upon, or reference to, the interpretative comments in the published literature.

Judge Seeborg sensibly seems to have distinguished between the absence of causal conclusions in individual epidemiologic studies and the absence of causal conclusions in any reputable medical literature.[27] He refused to be ensnared in the Joiner argument because:[28]

“Epidemiology studies typically only expressly address whether an association exists between agents such as sildenafil and tadalafil and outcomes like melanoma progression. As explained in In re Roundup Prod. Liab. Litig., 390 F. Supp. 3d 1102, 1116 (N.D. Cal. 2018), ‘[w]hether the agents cause the outcomes, however, ordinarily cannot be proven by epidemiological studies alone; an evaluation of causation requires epidemiologists to exercise judgment about the import of those studies and to consider them in context’.”

This new MDL opinion, relying upon the Advisory Committee Notes to Rule 702, is thus a more felicitous statement of the goals of gatekeeping.

Confidence Intervals

As welcome as some aspects of Judge Seeborg’s opinion are, the decision is not without mistakes. The district judge, like so many of his judicial colleagues, trips over the proper interpretation of a confidence interval:[29]

“When reviewing the results of a study it is important to consider the confidence interval, which, in simple terms, is the ‘margin of error’. For example, a given study could calculate a relative risk of 1.4 (a 40 percent increased risk of adverse events), but show a 95 percent ‘confidence interval’ of .8 to 1.9. That confidence interval means there is 95 percent chance that the true value—the actual relative risk—is between .8 and 1.9.”

This statement is inescapably wrong. The 95 percent probability attaches to the capturing of the true parameter – the actual relative risk – in the long run of repeated confidence intervals that result from repeated sampling of the same sample size, in the same manner, from the same population. In Judge Seeborg’s example, the next sample might give a relative risk point estimate 1.9, and that new estimate will have a confidence interval that may run from just below 1.0 to over 3. A third sample might turn up a relative risk estimate of 0.8, with a confidence interval that runs from say 0.3 to 1.4. Neither the second nor the third sample would be reasonably incompatible with the first. A more accurate assessment of the true parameter is that it will be somewhere between 0.3 and 3, a considerably broader range for the 95 percent.

Judge Seeborg’s error is sadly all too common. Whenever I see the error, I wonder whence it came. Often the error is in briefs of both plaintiffs’ and defense counsel. In this case, I did not see the erroneous assertion about confidence intervals made in plaintiffs’ or defendants’ briefs.


[1]  Brumley  v. Pfizer, Inc., 200 F.R.D. 596 (S.D. Tex. 2001) (excluding plaintiffs’ expert witness who claimed that Viagra caused heart attack); Selig v. Pfizer, Inc., 185 Misc. 2d 600 (N.Y. Cty. S. Ct. 2000) (excluding plaintiff’s expert witness), aff’d, 290 A.D. 2d 319, 735 N.Y.S. 2d 549 (2002).

[2]  “Love is Blind but What About Judicial Gatekeeping of Expert Witnesses? – Viagra Part I” (July 7, 2012); “Viagra, Part II — MDL Court Sees The Light – Bad Data Trump Nuances of Statistical Inference” (July 8, 2012).

[3]  In re Viagra Prods. Liab. Litig., 572 F.Supp. 2d 1071 (D. Minn. 2008), 658 F. Supp. 2d 936 (D. Minn. 2009), and 658 F. Supp. 2d 950 (D. Minn. 2009).

[4]  Wen-Qing Li, Abrar A. Qureshi, Kathleen C. Robinson, and Jiali Han, “Sildenafil use and increased risk of incident melanoma in US men: a prospective cohort study,” 174 J. Am. Med. Ass’n Intern. Med. 964 (2014).

[5]  See, e.g., Herrara v. Pfizer Inc., Complaint in 3:15-cv-04888 (N.D. Calif. Oct. 23, 2015); Diana Novak Jones, “Viagra Increases Risk Of Developing Melanoma, Suit Says,” Law360 (Oct. 26, 2015).

[6]  See In re Viagra (Sildenafil Citrate) Prods. Liab. Litig., 176 F. Supp. 3d 1377, 1378 (J.P.M.L. 2016).

[7]  See, e.g., Jenny Z. Wang, Stephanie Le , Claire Alexanian, Sucharita Boddu, Alexander Merleev, Alina Marusina, and Emanual Maverakis, “No Causal Link between Phosphodiesterase Type 5 Inhibition and Melanoma,” 37 World J. Men’s Health 313 (2019) (“There is currently no evidence to suggest that PDE5 inhibition in patients causes increased risk for melanoma. The few observational studies that demonstrated a positive association between PDE5 inhibitor use and melanoma often failed to account for major confounders. Nonetheless, the substantial evidence implicating PDE5 inhibition in the cyclic guanosine monophosphate (cGMP)-mediated melanoma pathway warrants further investigation in the clinical setting.”); Xinming Han, Yan Han, Yongsheng Zheng, Qiang Sun, Tao Ma, Li Dai, Junyi Zhang, and Lianji Xu, “Use of phosphodiesterase type 5 inhibitors and risk of melanoma: a meta-analysis of observational studies,” 11 OncoTargets & Therapy 711 (2018).

[8]  In re Viagra (Sildenafil Citrate) and Cialis (Tadalafil) Prods. Liab. Litig., Case No. 16-md-02691-RS, Order Granting in Part and Denying in Part Motions to Exclude Expert Testimony (N.D. Calif. Jan. 13, 2020) [cited as Opinion].

[9]  Opinion at 8 (“determin[ing] whether the analysis undergirding the experts’ testimony falls within the range of accepted standards governing how scientists conduct their research and reach their conclusions”), citing Daubert v. Merrell Dow Pharm., Inc. (Daubert II), 43 F.3d 1311, 1317 (9th Cir. 1995).

[10]  Opinion at 11.

[11]  Opinion at 11-13.

[12]  See Kenneth J. Rothman, Sander Greenland, and Timothy L. Lash, “Introduction,” chap. 1, in Kenneth J. Rothman, et al., eds., Modern Epidemiology at 29 (3d ed. 2008) (“no approach can transform plausibility into an objective causal criterion).

[13]  Opinion at 15-16.

[14]  Opinion at 16-17.

[15]  See Austin Bradford Hill, “The Environment and Disease: Association or Causation?” 58 Proc. Royal Soc’y Med. 295 (1965); see also “Woodside & Davis on the Bradford Hill Considerations” (April 23, 2013).

[16]  Opinion at 17 – 21.

[17]  Opinion at 18. The MDL court cited In re Silicone Gel Breast Implants Prod. Liab. Litig., 318 F. Supp. 2d 879, 893 (C.D. Cal. 2004), for the proposition that relative risks greater than 2.0 permit the inference that the agent under study “was more likely than not responsible for a particular individual’s disease.”

[18]  Opinion at 18.

[19]  Opinion at 20.

[20]  Opinion at 19.

[21]  Opinion at 21, quoting from Rule 702, Advisory Committee Notes (emphasis in Judge Seeborg’s opinion).

[22]  Opinion at 21.

[23]  SeeFollow the Data, Not the Discussion” (May 2, 2010).

[24]  Joiner, 522 U.S. at 145-46 (noting that the PCB studies at issue did not support expert witnesses’ conclusion that PCB exposure caused cancer because the study authors, who conducted the research, were not willing to endorse a conclusion of causation).

[25]  Huss v. Gayden, 571 F.3d 442  (5th Cir. 2009) (citing Vargas v. Lee, 317 F.3d 498, 501-01 (5th Cir. 2003) (noting that studies that did not themselves embrace causal conclusions undermined the reliability of the plaintiffs’ expert witness’s testimony that trauma caused fibromyalgia); see also McClain v. Metabolife Internat’l, Inc., 401 F.3d 1233, 1247-48 (11th Cir. 2005) (expert witnesses’ reliance upon studies that did not reach causal conclusions about ephedrine supported the challenge to the reliability of their proffered opinions); Happel v. Walmart, 602 F.3d 820, 826 (7th Cir. 2010) (observing that “is axiomatic that causation testimony is inadmissible if an expert relies upon studies or publications, the authors of which were themselves unwilling to conclude that causation had been proven”).

[26]  In re Accutane Prods. Liab. Litig., 511 F. Supp. 2d 1288, 1291 (M.D. Fla. 2007) (“When an expert relies on the studies of others, he must not exceed the limitations the authors themselves place on the study. That is, he must not draw overreaching conclusions.) (internal citations omitted).

[27]  See Rutigliano v. Valley Bus. Forms, 929 F. Supp. 779, 785 (D.N.J. 1996), aff’d, 118 F.3d 1577 (3d Cir. 1997) (“law warns against use of medical literature to draw conclusions not drawn in the literature itself …. Reliance upon medical literature for conclusions not drawn therein is not an accepted scientific methodology.”).

[28]  Opinion at 14

[29]  Opinion at 4 – 5.

Statistical Significance at the New England Journal of Medicine

July 19th, 2019

Some wild stuff has been going on in the world of statistics, at the American Statistical Association, and elsewhere. A very few obscure journals have declared p-values to be verboten, and presumably confidence intervals as well. The world of biomedical research has generally reacted more sanely, with authors defending the existing frequentist approaches and standards.[1]

This week, the editors of the New England Journal of Medicine have issued new statistical guidelines for authors. The Journal’s approach seems appropriately careful and conservative for the world of biomedical research. In an editorial introducing the new guidelines,[2] the Journal editors remind their potential authors that statistical significance and p-values are here to stay:

“Despite the difficulties they pose, P values continue to have an important role in medical research, and we do not believe that P values and significance tests should be eliminated altogether. A well-designed randomized or observational study will have a primary hypothesis and a prespecified method of analysis, and the significance level from that analysis is a reliable indicator of the extent to which the observed data contradict a null hypothesis of no association between an intervention or an exposure and a response. Clinicians and regulatory agencies must make decisions about which treatment to use or to allow to be marketed, and P values interpreted by reliably calculated thresholds subjected to appropriate adjustments have a role in those decisions.”[3]

The Journal’s editors described their revamped statistical policy as being based upon three premises:

(1) adhering to prespecified analysis plans if they exist;

(2) declaring associations or effects only for statistical analyses that have pre-specified “a method for controlling type I error”; and

(3) presenting evidence about clinical benefits or harms requires “both point estimates and their margins of error.”

With a hat tip to the ASA’s recent pronouncements on statistical significance,[4] the editors suggest that their new guidelines have moved away from bright-line applications of statistical significance “as a bright-line marker for a conclusion or a claim”[5]:

“[T]he notion that a treatment is effective for a particular outcome if P < 0.05 and ineffective if that threshold is not reached is a reductionist view of medicine that does not always reflect reality.”[6]

The editors’ language intimates greater latitude for authors in claiming associations or effects from their studies, but this latitude may well be circumscribed by tighter control over such claims in the inevitable context of multiple testing within a dataset.

The editors’ introduction of the new guidelines is not entirely coherent. The introductory editorial notes that the use of p-values for reporting multiple outcomes, without adjustments for multiplicity, inflates the number of findings with p-values less than 5%. The editors thus caution against “uncritical interpretation of multiple inferences,” which can be particularly threatening to valid inference when not all the comparisons conducted by the study investigators have been reported in their manuscript.[7] They reassuringly tell prospective authors that many methods are available to adjust for multiple comparisons, and can be used to control Type I error probability “when specified in the design of a study.”[8]

But what happens when such adjustment methods are not pre-specified in the study design? Failure to to do so do not appear to be disqualifying factors for publication in the Journal. For one thing, when the statistical analysis plan of the study has not specified adjustment methods for controlling type I error probabilities, then authors must replace p-values with “estimates of effects or association and 95% confidence intervals.”[9] It is hard to understand how this edict helps when the specified coefficient of 95% is a continuation of the 5% alpha, which would have been used in any event. The editors seem to be saying that if authors fail to pre-specify or even post-specify methods for controlling error probabilities, then they cannot declare statistical significance, or use p-values, but they can use confidence intervals in the same way they have been using them, and with the same misleading interpretations supplied by their readers.

More important, another price authors will have to pay for multiple testing without pre-specified methods of adjustment is that they will affirmatively have to announce their failure to adjust for multiplicity and that their putative associations “may not be reproducible.” Tepid as this concession is, it is better than previous practice, and perhaps it will become a badge of shame. The crucial question is whether judges, in exercising their gatekeeping responsibilities, will see these acknowledgements as disabling valid inferences from studies that carry this mandatory warning label.

The editors have not issued guidelines for the use of Bayesian statistical analyses, because “the large majority” of author manuscripts use only frequentist analyses.[10] The editors inform us that “[w]hen appropriate,” they will expand their guidelines to address Bayesian and other designs. Perhaps this expansion will be appropriate when Bayesian analysts establish a track record of abuse in their claiming of associations and effects.

The new guidelines themselves are not easy to find. The Journal has not published these guidelines as an article in their published issues, but has relegated them to a subsection of their website’s instructions to authors for new manuscripts:

https://www.nejm.org/author-center/new-manuscripts

Presumably, the actual author instructions control in any perceived discrepancy between this week’s editorial and the guidelines themselves. Authors are told that p-values generally should be two-sided. Authors’ use of:

“Significance tests should be accompanied by confidence intervals for estimated effect sizes, measures of association, or other parameters of interest. The confidence intervals should be adjusted to match any adjustment made to significance levels in the corresponding test.”

Similarly, the guidelines call for, but do not require, pre-specified methods of controlling family-wide error rates for multiple comparisons. For observational studies submitted without pre-specified methods of error control, the guidelines recommend the use of point estimates and 95% confidence intervals, with an explanation that the interval widths have not been adjusted for multiplicity, and a caveat that the inferences from these findings may not be reproducible. The guidelines recommend against using p-values for such results, but again, it is difficult to see why reporting the 95% confidence intervals is recommended when p-values are not recommended.


[1]  Jonathan A. Cook, Dean A. Fergusson, Ian Ford, Mithat Gonen, Jonathan Kimmelman, Edward L. Korn, and Colin B. Begg, “There is still a place for significance testing in clinical trials,” 16 Clin. Trials 223 (2019).

[2]  David Harrington, Ralph B. D’Agostino, Sr., Constantine Gatsonis, Joseph W. Hogan, David J. Hunter, Sharon-Lise T. Normand, Jeffrey M. Drazen, and Mary Beth Hamel, “New Guidelines for Statistical Reporting in the Journal,” 381 New Engl. J. Med. 285 (2019).

[3]  Id. at 286.

[4]  See id. (“Journal editors and statistical consultants have become increasingly concerned about the overuse and misinterpretation of significance testing and P values in the medical literature. Along with their strengths, P values are subject to inherent weaknesses, as summarized in recent publications from the American Statistical Association.”) (citing Ronald L. Wasserstein & Nicole A. Lazar, “The ASA’s statement on p-values: context, process, and purpose,” 70 Am. Stat. 129 (2016); Ronald L. Wasserstein, Allen L. Schirm, and Nicole A. Lazar, “Moving to a world beyond ‘p < 0.05’,” 73 Am. Stat. s1 (2019)).

[5]  Id. at 285.

[6]  Id. at 285-86.

[7]  Id. at 285.

[8]  Id., citing Alex Dmitrienko, Frank Bretz, Ajit C. Tamhane, Multiple testing problems in pharmaceutical statistics (2009); Alex Dmitrienko & Ralph B. D’Agostino, Sr., “Multiplicity considerations in clinical trials,” 378 New Engl. J. Med. 2115 (2018).

[9]  Id.

[10]  Id. at 286.