TORTINI

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

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.

The Shmeta-Analysis in Paoli

July 11th, 2019

In the Paoli Railroad yard litigation, plaintiffs claimed injuries and increased risk of future cancers from environmental exposure to polychlorinated biphenyls (PCBs). This massive litigation showed up before federal district judge Hon. Robert F. Kelly,[1] in the Eastern District of Pennsylvania, who may well have been the first judge to grapple with a litigation attempt to use meta-analysis to show a causal association.

One of the plaintiffs’ expert witnesses was the late William J. Nicholson, who was a professor at Mt. Sinai School of Medicine, and a colleague of Irving Selikoff. Nicholson was trained in physics, and had no professional training in epidemiology. Nonetheless, Nicholson was Selikoff’s go-to colleague for performing epidemiologic studies. After Selikoff withdrew from active testifying for plaintiffs in tort litigation, Nicholson was one of his colleagues who jumped into the fray as a surrogate advocate for Selikoff.[2]

For his opinion that PCBs were causally associated with liver cancer in humans,[3] Nicholson relied upon a report he wrote for the Ontario Ministry of Labor. [cited here as “Report”].[4] Nicholson described his report as a “study of the data of all the PCB worker epidemiological studies that had been published,” from which he concluded that there was “substantial evidence for a causal association between excess risk of death from cancer of the liver, biliary tract, and gall bladder and exposure to PCBs.”[5]

The defense challenged the admissibility of Nicholson’s meta-analysis, on several grounds. The trial court decided the challenge based upon the Downing case, which was the law in the Third Circuit, before the Supreme Court decided Daubert.[6] The Downing case allowed some opportunity for consideration of reliability and validity concerns; there is, however, disappointingly little discussion of any actual validity concerns in the courts’ opinions.

The defense challenge to Nicholson’s proffered testimony on liver cancer turned on its characterization of meta-analysis as a “novel” technique, which is generally unreliable, and its claim that Nicholson’s meta-analysis in particular was unreliable. None of the individual studies that contributed data showed any “connection” between PCBs and liver cancer; nor did any individual study conclude that there was a causal association.

Of course, the appropriate response to this situation, with no one study finding a statistically significant association, or concluding that there was a causal association, should have been “so what?” One of the reasons to do a meta-analysis is that no available study was sufficiently large to find a statistically significant association, if one were there. As for drawing conclusions of causal associations, it is not the role or place of an individual study to synthesize all the available evidence into a principled conclusion of causation.

In any event, the trial court concluded that the proffered novel technique lacked sufficient reliability, that the meta-analysis would “overwhelm, confuse, or mislead the jury,” and that the proffered meta-analysis on liver cancer was not sufficiently relevant to the facts of the case (in which no plaintiff had developed, or had died of, liver cancer). The trial court noted that the Report had not been peer-reviewed, and that it had not been accepted or relied upon by the Ontario government for any finding or policy decision. The trial court also expressed its concern that the proffered testimony along the lines of the Report would possibly confuse the jury because it appeared to be “scientific” and because Nicholson appeared to be qualified.

The Appeal

The Court of Appeals for the Third Circuit, in an opinion by Judge Becker, reversed Judge Kelly’s exclusion of the Nicholson Report, in an opinion that is still sometimes cited, even though Downing is no longer good law in the Circuit or anywhere else.[7] The Court was ultimately not persuaded that the trial court had handled the exclusion of Nicholson’s Report and its meta-analysis correctly, and it remanded the case for a do-over analysis.

Judge Becker described Nicholson’s Report as a “meta-analysis,” which pooled or “combined the results of numerous epidemiologic surveys in order to achieve a larger sample size, adjusted the results for differences in testing techniques, and drew his own scientific conclusions.”[8] Through this method, Nicholson claimed to have shown that “exposure to PCBs can cause liver, gall bladder and biliary tract disorders … even though none of the individual surveys supports such a conclusion when considered in isolation.”[9]

Validity

The appellate court gave no weight to the possibility that a meta-analysis would confuse a jury, or that its “scientific nature” or Nicholson’s credentials would lead a jury to give it more weight than it deserved.[10] The Court of Appeals conceded, however, that exclusion would have been appropriate if the methodology used itself was invalid. The appellate opinion further acknowledged that the defense had offered opposition to Nicholson’s Report in which it documented his failure to include data that were inconsistent with his conclusions, and that “Nicholson had produced a scientifically invalid study.”[11]

Judge Becker’s opinion for a panel of the Third Circuit provided no details about the cherry picking. The opinion never analyzed why this charge of cherry-picking and manipulation of the dataset did not invalidate the meta-analytic method generally, or Nicholson’s method as applied. The opinion gave no suggestion that this counter-affidavit was ever answered by the plaintiffs.

Generally, Judge Becker’s opinion dodged engagement with the specific threats to validity in Nicholson’s Report, and took refuge in the indisputable fact that hundreds of meta-analyses were published annually, and that the defense expert witnesses did not question the general reliability of meta-analysis.[12] These facts undermined the defense claim that meta-analysis was novel.[13] The reality, however, was that meta-analysis was in its infancy in bio-medical research.

When it came to the specific meta-analysis at issue, the court did not discuss or analyze a single pertinent detail of the Report. Despite its lack of engagement with the specifics of the Report’s meta-analysis, the court astutely observed that prevalent errors and flaws do not mean that a particular meta-analysis is “necessarily in error.”[14] Of course, without bothering to look, the court would not know whether the proffered meta-analysis was “actually in error.”

The appellate court would have given Nicholson’s Report a “pass” if it was an application of an accepted methodology. The defense’s remedy under this condition would be to cross-examine the opinion in front of a jury. If, on the other hand, the Nicholson had altered an accepted methodology to skew its results, then the court’s gatekeeping responsibility under Downing would be invoked.

The appellate court went on to fault the trial court for failing to make sufficiently explicit findings as to whether the questioned meta-analysis was unreliable. From its perspective, the Court of Appeals saw the trial court as resolving the reliability issue upon the greater credibility of defense expert witnesses in branding the disputed meta-analysis as unreliability. Credibility determinations are for the jury, but the court left room for a challenge on reliability itself:[15]

“Assuming that Dr. Nicholson’s meta-analysis is the proper subject of Downing scrutiny, the district court’s decision is wanting, because it did not make explicit enough findings on the reliability of Dr. Nicholson’s meta-analysis to satisfy Downing. We decline to define the exact level at which a district court can exclude a technique as sufficiently unreliable. Reliability indicia vary so much from case to case that any attempt to define such a level would most likely be pointless. Downing itself lays down a flexible rule. What is not flexible under Downing is the requirement that there be a developed record and specific findings on reliability issues. Those are absent here. Thus, even if it may be possible to exclude Dr. Nicholson’s testimony under Downing, as an unreliable, skewed meta-analysis, we cannot make such a determination on the record as it now stands. Not only was there no hearing, in limine or otherwise, at which the bases for the opinions of the contesting experts could be evaluated, but the experts were also not even deposed. All of the expert evidence was based on affidavits.”

Peer Review

Understandably, the defense attacked Nicholson’s Report as not having been peer reviewed. Without any scrutiny of the scientific bona fides of the workers’ compensation agency, the appellate court acquiesced in Nicholson’s self-serving characterization of his Report as having been reviewed by “cooperating researchers” and the Panel of the Ontario Workers’ Compensation agency. Another partisan expert witness characterized Nicholson’s Report as a “balanced assessment,” and this seemed to appease the Third Circuit, which was wary of requiring peer review in the first place.[16]

Relevancy Prong

The defense had argued that Nicholson’s Report was irrelevant because no individual plaintiff claimed liver cancer.[17] The trial court largely accepted this argument, but the appellate court disagreed because of conclusory language in Nicholson’s affidavit, in which he asserted that “proof of an increased risk of liver cancer is probative of an increased risk of other forms of cancer.” The court seemed unfazed by the ipse dixit, asserted without any support. Indeed, Nicholson’s assertion was contradicted by his own Report, in which he reported that there were fewer cancers among PCB-exposed male capacitor manufacturing workers than expected,[18] and that the rate for all cancers for both men and women was lower than expected, with 132 observed and 139.40 expected.[19]

The trial court had also agreed with the defense’s suggestion that Nicholson’s report, and its conclusion of causality between PCB exposure and liver cancer, were irrelevant because the Report “could not be the basis for anyone to say with reasonable degree of scientific certainty that some particular person’s disease, not cancer of the liver, biliary tract or gall bladder, was caused by PCBs.”[20]

Analysis

It would likely have been lost on Judge Becker and his colleagues, but Nicholson presented SMRs (standardized mortality ratios) throughout his Report, and for the all cancers statistic, he gave an SMR of 95. What Nicholson clearly did in this, and in all other instances, was simply divide the observed number by the expected, and multiply by 100. This crude, simplistic calculation fails to present a standardized mortality ratio, which requires taking into account the age distribution of the exposed and the unexposed groups, and a weighting of the contribution of cases within each age stratum. Nicholson’s presentation of data was nothing short of false and misleading. And in case anyone remembers General Electric v. Joiner, Nicholson’s summary estimate of risk for lung cancer in men was below the expected rate.[21]

Nicholson’s Report was replete with many other methodological sins. He used a composite of three organs (liver, gall bladder, bile duct) without any biological rationale. His analysis combined male and female results, and still his analysis of the composite outcome was based upon only seven cases. Of those seven cases, some of the cases were not confirmed as primary liver cancer, and at least one case was confirmed as not being a primary liver cancer.[22]

Nicholson failed to standardize the analysis for the age distribution of the observed and expected cases, and he failed to present meaningful analysis of random or systematic error. When he did present p-values, he presented one-tailed values, and he made no corrections for his many comparisons from the same set of data.

Finally, and most egregiously, Nicholson’s meta-analysis was meta-analysis in name only. What he had done was simply to add “observed” and “expected” events across studies to arrive at totals, and to recalculate a bogus risk ratio, which he fraudulently called a standardized mortality ratio. Adding events across studies is not a valid meta-analysis; indeed, it is a well-known example of how to generate a Simpson’s Paradox, which can change the direction or magnitude of any association.[23]

Some may be tempted to criticize the defense for having focused its challenge on the “novelty” of Nicholson’s approach in Paoli. The problem of course was the invalidity of Nicholson’s work, but both the trial court’s exclusion of Nicholson, and the Court of Appeals’ reversal and remand of the exclusion decision, illustrate the problem in getting judges, even well-respected judges, to accept their responsibility to engage with questioned scientific evidence.

Even in Paoli, no amount of ketchup could conceal the unsavoriness of Nicholson’s scrapple analysis. When the Paoli case reached the Court Appeals again in 1994, Nicholson’s analysis was absent.[24] Apparently, the plaintiffs’ counsel had second thoughts about the whole matter. Today, under the revised Rule 702, there can be little doubt that Nicholson’s so-called meta-analysis should have been excluded.


[1]  Not to be confused with the Judge Kelly of the same district, who was unceremoniously disqualified after attending an ex parte conference with plaintiffs’ lawyers and expert witnesses, at the invitation of Dr. Irving Selikoff.

[2]  Pace Philip J. Landrigan & Myron A. Mehlman, “In Memoriam – William J. Nicholson,” 40 Am. J. Indus. Med. 231 (2001). Landrigan and Mehlman assert, without any support, that Nicholson was an epidemiologist. Their own description of his career, his undergraduate work at MIT, his doctorate in physics from the University of Washington, his employment at the Watson Laboratory, before becoming a staff member in Irving Selikoff’s department in 1969, all suggest that Nicholson brought little to no experience in epidemiology to his work on occupational and environmental exposure epidemiology.

[3]  In re Paoli RR Yard Litig., 706 F. Supp. 358, 372-73 (E.D. Pa. 1988).

[4]  William Nicholson, Report to the Workers’ Compensation Board on Occupational Exposure to PCBs and Various Cancers, for the Industrial Disease Standards Panel (ODP); IDSP Report No. 2 (Toronto, Ontario Dec. 1987).

[5]  Id. at 373.

[6]  United States v. Downing, 753 F.2d 1224 (3d Cir.1985)

[7]  In re Paoli RR Yard PCB Litig., 916 F.2d 829 (3d Cir. 1990), cert. denied sub nom. General Elec. Co. v. Knight, 111 S.Ct. 1584 (1991).

[8]  Id. at 845.

[9]  Id.

[10]  Id. at 841, 848.

[11]  Id. at 845.

[12]  Id. at 847-48.

[13]  See, e.g., Robert Rosenthal, Judgment studies: Design, analysis, and meta-analysis (1987); Richard J. Light & David B. Pillemer, Summing Up: the Science of Reviewing Research (1984); Thomas A. Louis, Harvey V. Fineberg & Frederick Mosteller, “Findings for Public Health from Meta-Analyses,” 6 Ann. Rev. Public Health 1 (1985); Kristan A. L’abbé, Allan S. Detsky & Keith O’Rourke, “Meta-analysis in clinical research,” 107 Ann. Intern. Med. 224 (1987).

[14]  Id. at 857.

[15]  Id. at 858/

[16]  Id. at 858.

[17]  Id. at 845.

[18]  Report, Table 16.

[19]  Report, Table 18.

[20]  In re Paoli, 916 F.2d at 847.

[21]  See General Electric v. Joiner, 522 U.S. 136 (1997); NAS, “How Have Important Rule 702 Holdings Held Up With Time?” (March 20, 2015).

[22]  Report, Table 22.

[23]  James A. Hanley, Gilles Thériault, Ralf Reintjes and Annette de Boer, “Simpson’s Paradox in Meta-Analysis,” 11 Epidemiology 613 (2000); H. James Norton & George Divine, “Simpson’s paradox and how to avoid it,” Significance 40 (Aug. 2015); George Udny Yule, Notes on the theory of association of attributes in Statistics, 2 Biometrika 121 (1903).

[24]  In re Paoli RR Yard Litig., 35 F.3d 717 (3d Cir. 1994).

N.J. Supreme Court Uproots Weeds in Garden State’s Law of Expert Witnesses

August 8th, 2018

The United States Supreme Court’s decision in Daubert is now over 25 years old. The idea of judicial gatekeeping of expert witness opinion testimony is even older in New Jersey state courts. The New Jersey Supreme Court articulated a reliability standard before the Daubert case was even argued in Washington, D.C. See Landrigan v. Celotex Corp., 127 N.J. 404, 414 (1992); Rubanick v. Witco Chem. Corp., 125 N.J. 421, 447 (1991). Articulating a standard, however, is something very different from following a standard, and in many New Jersey trial courts, until very recently, the standard was pretty much anything goes.

One counter-example to the general rule of dog-eat-dog in New Jersey was Judge Nelson Johnson’s careful review and analysis of the proffered causation opinions in cases in which plaintiffs claimed that their use of the anti-acne medication isotretinoin (Accutane) caused Crohn’s disease. Judge Johnson, who sits in the Law Division of the New Jersey Superior Court for Atlantic County held a lengthy hearing, and reviewed the expert witnesses’ reliance materials.1 Judge Johnson found that the plaintiffs’ expert witnesses had employed undue selectivity in choosing what to rely upon. Perhaps even more concerning, Judge Johnson found that these witnesses had refused to rely upon reasonably well-conducted epidemiologic studies, while embracing unpublished, incomplete, and poorly conducted studies and anecdotal evidence. In re Accutane, No. 271(MCL), 2015 WL 753674, 2015 BL 59277 (N.J.Super. Law Div., Atlantic Cty. Feb. 20, 2015). In response, Judge Johnson politely but firmly closed the gate to conclusion-driven duplicitous expert witness causation opinions in over 2,000 personal injury cases. “Johnson of Accutane – Keeping the Gate in the Garden State” (Mar. 28, 2015).

Aside from resolving over 2,000 pending cases, Judge Johnson’s judgment was of intense interest to all who are involved in pharmaceutical and other products liability litigation. Judge Johnson had conducted a pretrial hearing, sometimes called a Kemp hearing in New Jersey, after the New Jersey Supreme Court’s opinion in Kemp v. The State of New Jersey, 174 N.J. 412 (2002). At the hearing and in his opinion that excluded plaintiffs’ expert witnesses’ causation opinions, Judge Johnson demonstrated a remarkable aptitude for analyzing data and inferences in the gatekeeping process.

When the courtroom din quieted, the trial court ruled that the proffered testimony of Dr., Arthur Kornbluth and Dr. David Madigan did not meet the liberal New Jersey test for admissibility. In re Accutane, No. 271(MCL), 2015 WL 753674, 2015 BL 59277 (N.J.Super. Law Div. Atlantic Cty. Feb. 20, 2015). And in closing the gate, Judge Johnson protected the judicial process from several bogus and misleading “lines of evidence,” which have become standard ploys to mislead juries in courthouses where the gatekeepers are asleep. Recognizing that not all evidence is on the same analytical plane, Judge Johnson gave case reports short shrift.

[u]nsystematic clinical observations or case reports and adverse event reports are at the bottom of the evidence hierarchy.”

Id. at *16. Adverse event reports, largely driven by the very litigation in his courtroom, received little credit and were labeled as “not evidentiary in a court of law.” Id. at 14 (quoting FDA’s description of FAERS).

Judge Johnson recognized that there was a wide range of identified “risk factors” for irritable bowel syndrome, such as prior appendectomy, breast-feeding as an infant, stress, Vitamin D deficiency, tobacco or alcohol use, refined sugars, dietary animal fat, fast food. In re Accutane, 2015 WL 753674, at *9. The court also noted that there were four medications generally acknowledged to be potential risk factors for inflammatory bowel disease: aspirin, nonsteroidal anti-inflammatory medications (NSAIDs), oral contraceptives, and antibiotics. Understandably, Judge Johnson was concerned that the plaintiffs’ expert witnesses preferred studies unadjusted for potential confounding co-variables and studies that had involved “cherry picking the subjects.” Id. at *18.

Judge Johnson had found that both sides in the isotretinoin cases conceded the relative unimportance of animal studies, but the plaintiffs’ expert witnesses nonetheless invoked the animal studies in the face of the artificial absence of epidemiologic studies that had been created by their cherry-picking strategies. Id.

Plaintiffs’ expert witnesses had reprised a common claimants’ strategy; namely, they claimed that all the epidemiology studies lacked statistical power. Their arguments often ignored that statistical power calculations depend upon statistical significance, a concept to which many plaintiffs’ counsel have virulent antibodies, as well as an arbitrarily selected alternative hypothesis of association size. Furthermore, the plaintiffs’ arguments ignored the actual point estimates, most of which were favorable to the defense, and the observed confidence intervals, most of which were reasonably narrow.

The defense responded to the bogus statistical arguments by presenting an extremely capable clinical and statistical expert witness, Dr. Stephen Goodman, to present a meta-analysis of the available epidemiologic evidence.

Meta-analysis has become an important facet of pharmaceutical and other products liability litigation[1]. Fortunately for Judge Johnson, he had before him an extremely capable expert witness, Dr. Stephen Goodman, to explain meta-analysis generally, and two meta-analyses he had performed on isotretinoin and irritable bowel outcomes.

Dr. Goodman explained that the plaintiffs’ witnesses’ failure to perform a meta-analysis was telling when meta-analysis can obviate the plaintiffs’ hyperbolic statistical complaints:

the strength of the meta-analysis is that no one feature, no one study, is determinant. You don’t throw out evidence except when you absolutely have to.”

In re Accutane, 2015 WL 753674, at *8.

Judge Johnson’s judicial handiwork received non-deferential appellate review from a three-judge panel of the Appellate Division, which reversed the exclusion of Kornbluth and Madigan. In re Accutane Litig., 451 N.J. Super. 153, 165 A.3d 832 (App. Div. 2017). The New Jersey Supreme Court granted the isotretinoin defendants’ petition for appellate review, and the issues were joined over the appropriate standard of appellate review for expert witness opinion exclusions, and the appropriateness of Judge Johnson’s exclusions of Kornbluth and Madigan. A bevy of amici curiae joined in the fray.2

Last week, the New Jersey Supreme Court issued a unanimous opinion, which reversed the Appellate Division’s holding that Judge Johnson had “mistakenly exercised” discretion. Applying its own precedents from Rubanick, Landrigan, and Kemp, and the established abuse-of-discretion standard, the Court concluded that the trial court’s ruling to exclude Kornbluth and Madigan was “unassailable.” In re Accutane Litig., ___ N.J. ___, 2018 WL 3636867 (2018), Slip op. at 79.3

The high court graciously acknowledged that defendants and amici had “good reason” to seek clarification of New Jersey law. Slip op. at 67. In abandoning abuse-of-discretion as its standard of review, the Appellate Division had relied upon a criminal case that involved the application of the Frye standard, which is applied as a matter of law. Id. at 70-71. The high court also appeared to welcome the opportunity to grant review and reverse the intermediate court reinforce “the rigor expected of the trial court” in its gatekeeping role. Id. at 67. The Supreme Court, however, did not articulate a new standard; rather it demonstrated at length that Judge Johnson had appropriately applied the legal standards that had been previously announced in New Jersey Supreme Court cases.4

In attempting to defend the Appellate Division’s decision, plaintiffs sought to characterize New Jersey law as somehow different from, and more “liberal” than, the United States Supreme Court’s decision in Daubert. The New Jersey Supreme Court acknowledged that it had never formally adopted the dicta from Daubert about factors that could be considered in gatekeeping, slip op. at 10, but the Court went on to note what disinterested observers had long understood, that the so-called Daubert factors simply flowed from a requirement of sound methodology, and that there was “little distinction” and “not much light” between the Landrigan and Rubanick principles and the Daubert case or its progeny. Id at 10, 80.

Curiously, the New Jersey Supreme Court announced that the Daubert factors should be incorporated into the New Jersey Rules 702 and 703 and their case law, but it stopped short of declaring New Jersey a “Daubert” jurisdiction. Slip op. at 82. In part, the Court’s hesitance followed from New Jersey’s bifurcation of expert witness standards for civil and criminal cases, with the Frye standard still controlling in the criminal docket. At another level, it makes no sense to describe any jurisdiction as a “Daubert” state because the relevant aspects of the Daubert decision were dicta, and the Daubert decision and its progeny were superseded by the revision of the controlling statute in 2000.5

There were other remarkable aspects of the Supreme Court’s Accutane decision. For instance, the Court put its weight behind the common-sense and accurate interpretation of Sir Austin Bradford Hill’s famous articulation of factors for causal judgment, which requires that sampling error, bias, and confounding be eliminated before assessing whether the observed association is strong, consistent, plausible, and the like. Slip op. at 20 (citing the Reference Manual at 597-99), 78.

The Supreme Court relied extensively on the National Academies’ Reference Manual on Scientific Evidence.6 That reliance is certainly preferable to judicial speculations and fabulations of scientific method. The reliance is also positive, considering that the Court did not look only at the problematic epidemiology chapter, but adverted also to the chapters on statistical evidence and on clinical medicine.

The Supreme Court recognized that the Appellate Division had essentially sanctioned an anything goes abandonment of gatekeeping, an approach that has been all-too-common in some of New Jersey’s lower courts. Contrary to the previously prevailing New Jersey zeitgeist, the Court instructed that gatekeeping must be “rigorous” to “prevent[] the jury’s exposure to unsound science through the compelling voice of an expert.” Slip op. at 68-9.

Not all evidence is equal. “[C]ase reports are at the bottom of the evidence hierarchy.” Slip op. at 73. Extrapolation from non-human animal studies is fraught with external validity problems, and such studies “far less probative in the face of a substantial body of epidemiologic evidence.” Id. at 74 (internal quotations omitted).

Perhaps most chilling for the lawsuit industry will be the Supreme Court’s strident denunciation of expert witnesses’ selectivity in choosing lesser evidence in the face of a large body of epidemiologic evidence, id. at 77, and their unprincipled cherry picking among the extant epidemiologic publications. Like the trial court, the Supreme Court found that the plaintiffs’ expert witnesses’ inconsistent use of methodological criteria and their selective reliance upon studies (disregarding eight of the nine epidemiologic studies) that favored their task masters was the antithesis of sound methodology. Id. at 73, citing with approval, In re Lipitor, ___ F.3d ___ (4th Cir. 2018) (slip op. at 16) (“Result-driven analysis, or cherry-picking, undermines principles of the scientific method and is a quintessential example of applying methodologies (valid or otherwise) in an unreliable fashion.”).

An essential feature of the Supreme Court’s decision is that it was not willing to engage in the common reductionism that has “all epidemiologic studies are flawed,” and which thus privileges cherry picking. Not all disagreements between expert witnesses can be framed as differences in interpretation. In re Accutane will likely stand as a bulwark against flawed expert witness opinion testimony in the Garden State for a long time.


1 Judge Nelson Johnson is also the author of Boardwalk Empire: The Birth, High Times, and Corruption of Atlantic City (2010), a spell-binding historical novel about political and personal corruption.

2 In support of the defendants’ positions, amicus briefs were filed by the New Jersey Business & Industry Association, Commerce and Industry Association of New Jersey, and New Jersey Chamber of Commerce; by law professors Kenneth S. Broun, Daniel J. Capra, Joanne A. Epps, David L. Faigman, Laird Kirkpatrick, Michael M. Martin, Liesa Richter, and Stephen A. Saltzburg; by medical associations the American Medical Association, Medical Society of New Jersey, American Academy of Dermatology, Society for Investigative Dermatology, American Acne and Rosacea Society, and Dermatological Society of New Jersey, by the Defense Research Institute; by the Pharmaceutical Research and Manufacturers of America; and by New Jersey Civil Justice Institute. In support of the plaintiffs’ position and the intermediate appellate court’s determination, amicus briefs were filed by political action committee the New Jersey Association for Justice; by the Ironbound Community Corporation; and by plaintiffs’ lawyer Allan Kanner.

3 Nothing in the intervening scientific record called question upon Judge Johnson’s trial court judgment. See, e.g., I.A. Vallerand, R.T. Lewinson, M.S. Farris, C.D. Sibley, M.L. Ramien, A.G.M. Bulloch, and S.B. Patten, “Efficacy and adverse events of oral isotretinoin for acne: a systematic review,” 178 Brit. J. Dermatol. 76 (2018).

4 Slip op. at 9, 14-15, citing Landrigan v. Celotex Corp., 127 N.J. 404, 414 (1992); Rubanick v. Witco Chem. Corp., 125 N.J. 421, 447 (1991) (“We initially took that step to allow the parties in toxic tort civil matters to present novel scientific evidence of causation if, after the trial court engages in rigorous gatekeeping when reviewing for reliability, the proponent persuades the court of the soundness of the expert’s reasoning.”).

5 The Court did acknowledge that Federal Rule of Evidence 702 had been amended in 2000, to reflect the Supreme Court’s decision in Daubert, Joiner, and Kumho Tire, but the Court did not deal with the inconsistencies between the present rule and the 1993 Daubert case. Slip op. at 64, citing Calhoun v. Yamaha Motor Corp., U.S.A., 350 F.3d 316, 320-21, 320 n.8 (3d Cir. 2003).

6 See Accutane slip op. at 12-18, 24, 73-74, 77-78. With respect to meta-analysis, the Reference Manual’s epidemiology chapter is still stuck in the 1980s and the prevalent resistance to poorly conducted, often meaningless meta-analyses. SeeThe Treatment of Meta-Analysis in the Third Edition of the Reference Manual on Scientific Evidence” (Nov. 14, 2011) (The Reference Manual fails to come to grips with the prevalence and importance of meta-analysis in litigation, and fails to provide meaningful guidance to trial judges).

Slemp Trial Part 4 – Graham Colditz

July 22nd, 2017

The Witness

Somehow, in opposition to two epidemiologists presented by the plaintiff in Slemp, the defense managed to call none. The first of the plaintiffs’ two epidemiology expert witnesses was Graham A. Colditz, a physician with doctoral level training in epidemiology. For many years, Colditz was a professor at the Harvard School of Public Health. Colditz left Harvard to become the Niess-Gain Professor at Washington University St. Louis School of Medicine, where he is also the Associate Director for Prevention and Control at the Alvin J. Siteman Cancer Center.

Colditz is a senior epidemiologist, with many book and article publications to his credit. Although he has not published a causal analysis of ovarian cancer and talc, Colditz was an investigator on the well-known Nurses’ Health Study. One of Colditz’s publications on the Nurses’ cohort featured an analysis of talc use and ovarian cancer outcomes.

Although he is not a frequent testifying expert witness, Colditz is no stranger to the courtroom. He was a regular protagonist in the estrogen-progestin hormone replacement therapy (HRT) litigation, which principally involves claims of female breast cancer. Colditz has a charming Australian accent, with a voice tremor that makes him sound older than 63, and perhaps even more distinguished. He charges $1,500 per hour for his testimonial efforts, but is quick to point out that he has given thousands to charity. At his hourly rate, we can be sure he needs tax deductions of some kind.

In discussing his own qualifications, Colditz was low-key and modest except for what seemed like a strange claim that his HRT litigation work for plaintiffs led the FDA to require a boxed warning of breast cancer risk on the package insert for HRT medications. This claim is certainly false, and an extreme instance of post hoc ergo propter hoc. Colditz gilded the lilly by claiming that he does not get involved unless he believes that general causation exists between the exposure or medication and the disease claimed. Since he has only been a plaintiffs’ expert witness, this self-serving claim is quite circular.

The Examinations

The direct and cross-examinations of Dr. Colditz were long and tedious. Most lawyers are reluctant to have an epidemiologists testify at all, and try to limit the length of their examinations, when they must present epidemiologic testimony. Indeed, the defense in Slemp may have opted to present a clinician based upon the prejudice against epidemiologists testifying about quantitative data and analysis. In any event, Colditz’s direct examination went not hours, but days, as did the defense’s cross-examination.

The tedium of the direct examination was exacerbated by the shameless use of leading, loaded, and argumentative questions by plaintiff’s counsel, Allen Smith. A linguistic analysis might well show that Smith spoke 25 to 30 words for every one word spoken by Colditz on direct examination. Even aside from the niceties of courtroom procedure, the direct examination was lacking in aesthetic qualities. Still, it is hard to argue with a $110 million verdict, which cries out for explanation.

There were virtually no objections to Smith’s testifying in lieu of Colditz, with Colditz reduced to just “yes.” Sometimes, Colditz waxed loquacious, and answered, “yes, sir.” From judicial responses to other objections, however, it was clear that the trial court would have provided little control of the leading and argumentative questions.

Smith’s examination also took Colditz beyond the scope of his epidemiologic expertise in to ethics, social policy, and legal requirements of warnings, again without judicial management or control. We learned, over objection, from Colditz of all witnesses that the determination of causation has nothing to do with whether a warning should be given.

The Subject Matter

Colditz was clearly familiar with the subject matter, and allowed Smith to testify for him on a fairly simplistic level. The testimony was a natural outgrowth of his professional interests, and Colditz must have appeared to have been a credible expert witness, especially in a St. Louis courtroom, given that he was in a leadership role at the leading cancer center in that city.

With Smith’s lead, Colditz broached technical issues of bias evaluation, meta-analysis and pooling, which would never be addressed later by a defense expert witness at an equal level of expertise, sophistication, and credibility. Colditz offered criticisms of the Gonzalez (Sister Study) and the latency built into the observation period of that cohort, and he introduced the concept of Berkson bias in some of the case-control studies. Neither of these particular criticisms was rebutted in the defense case, again raising the question whether the defense expert witness, Dr. Huh, a clinician specializing in gynecologic oncology, was an appropriate foil for the line up of plaintiffs’ expert witness. Dr. Colditz was able to talk authoritatively (and in some cases misleadingly) about issues, which Dr. Huh could not contradict effectively, even if he were to have tried.

Colditz characterized his involvement in the talc cases as starting with his conducting a systematic review, undertaken for litigation, but still systematic. As a professor of epidemiology, Colditz should know what a systematic review is, although he never fully described the process on either direct or cross-examinations. No protocol for the systematic review was adduced into evidence. Sadly, the defense expert witness, Dr. Huh, never stated that he had done a systematic review; nor did he offer any criticisms of Dr. Colditz’s systematic review. Indeed, Huh admitted that he had not read Colditz’s testimony. In general, observing Colditz’s testimony after having watched Dr. Huh testify shouted MISMATCH.

The Issues

Statistical Significance

The beginning point of a case such as Slemp, involving a claim that talc causes ovarian cancer, and that it caused her ovarian cancer, is whether there is supporting epidemiology for the claim. As Sir Austin Bradford Hill put it over 50 years ago:

Disregarding then any such problem in semantics we have this situation. Our observations reveal an association between two variables, perfectly clear-cut and beyond what we would care to attribute to the play of chance. What aspects of that association should we especially consider before deciding that the most likely interpretation of it is causation?”

Austin Bradford Hill, “The Environment and Disease: Association or Causation?” 58 Proc. Royal Soc’y Med. 295, 295 (1965). Colditz, and plaintiff’s counsel, did not run away from the challenge; they embraced statistical significance and presented an argument for why the association was “clear-cut” (not created by bias or confounding).

In one of his lengthy, leading questions, plaintiffs’ counsel attempted to suggest that statistical significance, or a confidence interval that excluded a risk ratio of 1.0, excluded bias as well as chance. Colditz to his credit broke from the straight jacket of “yes, sirs,” and disagreed as to bias. Smith, perhaps chastised then took a chance and asked an open-ended question about what a confidence interval was. With the bit in his mouth, Colditz managed to describe the observed confidence interval incorrectly as providing the range within which the point estimate would fall 95% of the time if the same study were repeated many times! There is a distribution of 95% confidence intervals, which cover the true parameter 95% of the time, assuming a correct statistical model, random sampling, and no bias or confounding. For the observed confidence interval, the true value is either included or not. Perhaps Colditz was thinking of a prediction interval, but Smith had asked for a definition of a confidence interval, and the jury got non-sense.

Dose Response

Colditz parsed the remaining Bradford Hill factors, and opined that exposure-gradient or dose response was good to have but not necessary to support a causal conclusion. Colditz opined, with respect to whether the statistical assessment of a putative dose-response should include non-exposed women, that the non-exposed women should be excluded. This was one of the few technical issues that Dr. Huh engaged with, in the defense case, but Dr. Colditz was not confronted with any textbooks or writings that cast doubt on his preference for excluding non-users.

Plausibility

Plaintiff’s counsel spent a great deal of time, mostly reading lengthy passages of articles on this or that plausible mechanism for talc’s causing human ovarian cancer, only to have Colditz, with little or no demonstrated expertise in biological mechanism, say “yes.” Some articles discussed that talc use was a modifiable risk and that avoiding perineal talc use “may” reduce ovarian cancer risk. Smith would read (accurately) and then ask Colditz whether he agreed that avoiding talc use would reduce ovarian cancer in women. Colditz himself catches and corrects Smith, some times, but not others.

Smith read from an article that invokes a claim that asbestos (with definition as to what mineral) causes ovarian cancer. Colditz agreed. Smith testified that talc has asbestos in it, and Colditz agreed. Smith read from an article that stated vaguely that talc is chemically similar to asbestos and thus this creates plausibility for a causal connection between talc and cancer. Colditz agreed, without any suggestion that he understands whether or not talc is morphologically similar to asbestos. It seems unlikely that Colditz had any real expertise to offer here, but Smith could not resist touching all bases with Colditz; and the defense did not object or follow up on these excesses.

Smith and Colditz, well mostly Smith, testified that tubal ligation reduces the otherwise observed increased risk of ovarian cancer from talc use. Smith here entrusts Colditz with providing the common-sense explanation. There is no meaningful cross-examination on this “jury friendly” point.

Consistency

Colditz testifed that the studies, both case-control and cohort studies, were consistent in showing an increased risk of ovarian cancer in association with talc use. Indeed, the studies are mostly consistent; the issue is whether they are consistently biased or consistently showing the true population risk. The defense chose to confront Colditz with the lack of statistical significance in some studies (with elevated risk ratios) as though these studies were inconsistent with the studies that found similar risk ratios, with p-values less than 5%. This confrontation did not go well for the defense, either on cross-examination of Colditz, or on direct examination of Dr. Huh. Colditz backed up his opinion on consistency with the available meta-analyses, which find very low p-values for the summary estimate of risk ratio for talc use and ovarian cancer.

Unlike the Zoloft case1, in which consistency was generated across different end points by cherry picking, the consistency in the talc case was evidenced by a consistent elevation of risk ratios for the same end point, across studies. When subgroups of ovarian cell or tumor types were examined, statistical significance was sometimes lost, but the direction of the risk ratio above one was maintained. Meta-analyses generated summary point estimates with very low p-values.

The Gold Standard

Colditz further gilded the consistency lilly by claiming that the Terry study2, a pooled analysis of available case-control studies, was the “gold standard” in this area of observational epidemiology. Smith and Colditz presented at some length as to how the Cochrane Collaboration has labeled combined “individual patient data” (IPD) analyses as the gold standard. Colditz skimmed over the Cochrane’s endorsement of IPD analyses as having been made in the context of systematic reviews, involving primarily randomized clinical trials, for which IPD analyses allow time-to-event measurements, which can substantially modify observed risk ratios, and even reverse their direction. The case-control studies in the Terry pooled analysis did not have anything like the kind of prospectively collected individual patient data, which would warrant holding the Terry paper up as a “gold standard,” and Terry and her co-authors never made such a claim for their analysis. Colditz’s claim about the Terry study cried out for strong rebuttal, which never came.

The defense should have known that this hyperbolic testimony would be forthcoming, but they seemed not to have a rebuttal planned, other than dismissing case-controls studies generally as smaller than cohort studies. Rather than “getting into the weeds” about the merits of pooled analyses of observational studies, as opposed to clinical trials, the defense continued with its bizarre stance that the cohort studies were better because larger, while ignoring that they are smaller with respect to number of ovarian cancer cases and have less precision than the case-control studies. SeeNew Jersey Kemps Ovarian Cancer – Talc Cases” (Sept. 16, 2016). The defense also largely ignored Colditz’s testimony that exposure data collected in the available cohort studies was of limited value because lacking in details about frequency and intensity of use, and in some cases, collected on only one occasion.

Specific Causation

Colditz disclaimed the ability or intention to offer a specific causation opinion about Ms. Slemp’s ovarian cancer. Nonetheless, Colditz volunteered that “cancer is multifactorial,” which says very little because it says so much. In plaintiffs’ counsel’s hands, this characterization became a smokescreen to indict every possible present risk factor as playing a part in the actual causation of a particular case, such as Ms. Slemp’s case. No matter that the plaintiff was massively obese, and a smoker; every risk factor present must be, by fiat, in the “causal pie.”

But this would seem not to be Colditz’s own opinion. Graham Colditz has elsewhere asserted that an increased risk of disease cannot be translated into the “but-for” standard of causation3:

Knowledge that a factor is associated with increased risk of disease does not translate into the premise that a case of disease will be prevented if a specific individual eliminates exposure to that risk factor. Disease pathogenesis at the individual level is extremely complex.”

Just because a risk factor (assuming it is real and causal) is present does not put in the causal set.

Cross-Examination

The direct examination of Graham Colditz included scurrilous attacks on J & J’s lobbying, paying FDA user fees, and other corporate conduct, based upon documents of which Colditz had not personal knowledge. Colditz was reduced to nothing more than a backboard, off which plaintiff’s counsel could make his shots. On cross, the defense carefully dissected this direct examination and obtained disavowals from Colditz that he had suggested any untoward conduct by J & J. The jury could have been spared their valuable time by a trial judge who did not allow the scurrilous, collateral attacks in the first place.

The defense also tried to diminish Dr. Colditz’s testimony as an opinion coming from a non-physician. The problem, however, was that Colditz is a physician, who understands the biological issues, even if he is not a pathologist, toxicologist, or oncologist. Colditz did not offer opinions about Slemp’s medical treatment, and there was nothing in this line of cross-examination that lessened the impact of Colditz’s general causation testimony.

Generally, the cross-examination did not hurt Dr. Colditz’s strongly stated opinion that talc causes ovarian cancer. The defense (and plaintiff’s counsel before them) spent an inordinate amount of time on why Dr. Colditz has not updated his website to state publicly that talc causes ovarian cancer. Colditz blamed the “IT” guys, a rather disingenuous excuse. His explanation on direct, and on cross, as to why he could not post his opinion on his public-service website was so convoluted, however, that there was no clear admission or inference of dereliction. Colditz was permitted to bill his opinion, never posted to his institution’s website, as a “consensus opinion,” endorsed by several researchers, based upon hearsay emails and oral conversations.


1 See In re Zoloft Prod. Liab. Litig., No. 16-2247 , __ F.3d __, 2017 WL 2385279, 2017 U.S. App. LEXIS 9832 (3d Cir. June 2, 2017) (affirming exclusion of dodgy opinion, which involved changing subgroup end points across studies of maternal sertraline use and infant cardiac birth defects ).

2 Kathryn L. Terry, et al., “Genital powder use and risk of ovarian cancer: a pooled analysis of 8,525 cases and 9,859 controls,” 6 Cancer Prev. & Research 811 (2013).

3 Graham A. Colditz, “From epidemiology to cancer prevention: implications for the 21st Century,” 18 Cancer Causes Control 117, 118 (2007).