TORTINI

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

LNT in Milward v. Acuity Specialty Products Group

September 28th, 2013

Professor Edward J. Calabrese previously has written about the shadowy origins of the linear no threshold (LNT) model of cancer causation.  See The Dubious Origins of the Linear No Threshold Model of Carcinogenesis (Jan. 10, 2013).  Recently, Calabrese has deepened his historical scholarship with two additional, interesting articles. SeeToxicologist Says NAS Panel ‘Misled the World’ When Adopting Radiation Exposure Guidelines” (Aug. 13, 2013).

These articles, now available online, are important tools in the work chest of lawyers who litigate health effect claims.  Edward J. Calabrese, “How the US National Academy of Sciences misled the world community on cancer risk assessment: new findings challenge historical foundations of the linear dose response,” 87 Arch. Toxicol. (2013) (in press); Edward J. Calabrese, “Origin of the linearity no threshold (LNT) dose–response concept,” 87 Arch. Toxicol. 1621 (2013). Professor Calabrese illuminates the exaggerations and ipse dixit in the origins of the linear-no threshold model, first applied to radiogenic cancers, and later to human carcinogenesis more generally.

On remand from the First Circuit, the trial judge in Milward, now the Hon. Douglas Woodlock, faced a renewed Rule 702 motion directed to Milward’s specific causation expert witnesses.  Milward v. Acuity Specialty Products Group, Inc., Civil Action No. 07–11944–DPW, 2013 WL 4812425 (D. Mass. Sept. 6, 2013). Plaintiffs attempted to invoke the dubious LNT concept to argue that benzene should be in the “differential” for ascertaining the specific cause of Mr. Milward’s APL. In performing a careful Rule 702 analysis, Judge Woodlock rule that, “[t]o the extent Butler [Milward’s expert witness] seeks to establish specific causation based on the argument that any level of benzene is sufficient to cause leukemia—a so-called “no safe level,” “no threshold,” or “linear” model—her opinion is inadmissibly unreliable.”  Id. at *8.

In recognizing Dr. Butler’s reliance upon LNT concepts in civil litigation as unreliable, Judge Woodlock followed the lead of other courts, within the First Circuit, which have previously rejected expert witness opinion testimony founded upon the LNT model.  See, e.g., Whiting v. Boston Edison Co., 891 F.Supp. 12, 25 (D.Mass.1995) (“[t]he linear non-threshold model cannot be falsified, nor can it be validated. To the extent that it has been subjected to peer review and publication, it has been rejected by the overwhelming majority of the  scientific community. It has no known or potential rate of error. It is merely an hypothesis.” ); Sutera v. Perrier Group of America Inc., 986 F.Supp. 655, 666 (D.Mass.1997) (“Accordingly, although there is evidence that one camp of scientists … believes that a non-linear model is appropriate basis for predicting the risks of low-level exposures to benzene, there is no scientific evidence that the linear no-safe threshold analysis is an acceptable scientific technique used by experts in determining causation in an individual instance.”).  Strong precedent outside the First Circuit also supports Judge Woodlock’s holding.  See Allen v. Pennsylvania Eng’g Corp., 102 F.3d 194, 198 (5th Cir.1996); Henricksen v. ConocoPhillips Co., 605 F.Supp. 2d 1142, 1166 (E.D.Wash.2009) (“[Plaintiffs’ expert witness’s] theory that any amount of exposure more than negligible should be considered substantial risk factor for AML flies in the face of the scientific literature reviewed and other expert testimony in this case that there is a threshold or dose below which you do not see a statistically significant risk of developing AML.”); In re W.R. Grace & Co. 355 B.R. 462, 476 (Bankr. D. Del. 2006) (the “no threshold model . . . flies in the face of the toxicological law of dose-response . . . doesn’t satisfy Daubert, and doesn’t stand up to scientific scrutiny”); Cano v. Everest Minerals Corp., 362 F. Supp. 2d 814, 853–54 (W.D. Tex. 2005) (even accepting the linear, no-threshold model for uranium mining and cancer, it is not enough to show exposure, you must show causation as well). See also McClain v. Metabolife Int’l, Inc., 401 F.3d 1233, 1244 (11th Cir. 2005) (“in evaluating the reliability of the experts’ opinions on general causation, it would help to know how much additional risk for heart attack or ischemic stroke Metabolife consumers have over the risks the general population faces”). National Bank of Commerce v. Assoc. Milk Producers, 22 F. Supp. 2d 942, 960 (E.D. Ark. 1998), aff’d, 191 F.3d 858 (8th Cir.1999). See generally Federal Judicial Center, Reference Manual on Scientific Evidence, at 643 n. 28 (3d ed.2011).

The district court in Milward held that because Dr. Butler “did not and could not quantify a threshold exposure level for benzene, Milward cannot posit that his cumulative exposure level crossed a relevant threshold.” Milward, 2013 WL 4812425, at 8.  In addressing Milward’s reliance upon LNT, Judge Woodlock rejected three specious arguments, which frequently recur in Rule 702 litigation.

First, the district saw through the argument that the claimed benzene-APL LNT model was good science because the United States Environmental Protection Agency (EPA) relies upon it.  The EPA applies the LNT model for benzene

“due to uncertainty about the shape of the dose-response curve below 40 ppm-years.”

Id. at 8[1]. The district court recognized that the EPA’s reasoning was a “classic example of a cautious prophylactic administrative rule” that “does not support the reliability of the linear, no-threshold model in establishing specific causation.”  Id.  In so ruling, the Milward district court joins a long line of courts that have distinguished administrative rulemaking from civil litigation standards for causation.  See, e.g., Allen v. Pa. Eng’g Corp., 102 F.3d 194, 198 (5th Cir. 1996)(“This methodology results from the preventive perspective that the agencies adopt in order to reduce public exposure to harmful substances. The agencies’ threshold of proof is reasonably lower than that appropriate in tort law, which traditionally makes more particularized inquiries into cause and effect.”)

Second, the Milward district court also saw through plaintiffs’ argument that the First Circuit’s embrace of its “weight of the evidence” general causation approach, which appears to enjoy support among federal bureaucrats, required approval of plaintiffs’ attempt to use a LNT prophylactic or precautionary approach. Milward, 649 F.3d at 18 & n. 9.  Plaintiffs have the burden of showing reliability of the LNT model, and the EPA’s acknowledged uncertainty about the model for benzene was an insuperable barrier to their success.  Milward, 2013 WL 4812425, at 8 & n.4.

Third, the district court rejected Dr. Butler’s attempt to “bait and switch,” by pointing to a study on hematotoxicity as opposed to carcinogenicity.  Butler argued that there was “no clear evidence of a threshold below which benzene does not cause hematotoxicity in humans.”[2] The court recognized that the study referred to the lack of a hematotoxicity threshold for low average doses of benzene. Hematotoxicity is not necessarily induction of APL; nor was the lack of clear evidence for a threshold evidence against a threshold.

Even in the regulatory realm, the LNT model is losing traction.  See Chlorine Chemistry Council v. EPA, 206 F.3d 1286, 1287 (D.C. Cir. 2000) (invalidating EPA regulation under the Safe Drinking Water Act, when the EPA persisted in using an LNT model, after it had concluded that chloroform, a contaminant in drinking water from chlorination exerted a “a nonlinear mode of carcinogenic action”).  In the scientific realm, researchers can merely watch in amazement at the “political science” that proceeds under a mistaken, outdated model of carcinogenesis.  See, e.g., Brant A. Ulsh, “Checking the Foundation: Recent Radiobiology and the Linear No-Threshold Theory,” 99 Health Physics 747 (2010) (“However, a large and rapidly growing body of radiobiological evidence indicates that cell and tissue level responses to this damage, particularly at low doses and/or dose-rates, are nonlinear and may exhibit thresholds. To the extent that responses observed at lower levels of biological organization in vitro are predictive of carcinogenesis observed in vivo, this evidence directly contradicts the assumptions upon which the microdosimetric argument is based.”); Bernard L. Cohen, “The Linear No-Threshold Theory of Radiation Carcinogenesis Should Be Rejected,” 13 J. Am. Physicians & Surgeons 70, 75 (2008) (“The conclusion from the evidence reviewed in this paper and more extensively elsewhere is that the linear-no threshold theory (LNT) fails very badly in the low-dose region, grossly overestimating the risk from low-level radiation. This means that the cancer risk from the vast majority of normally encountered radiation exposures is much lower than given by usual estimates, and may well be zero or even negative.”); Maurice Tubiana, Ludwig E. Feinendegen, Chichuan Yang, and Joseph M. Kaminski, “The Linear No-Threshold Relationship Is Inconsistent with Radiation Biologic and Experimental Data,” 251 Radiology 13, 13, 15-16, 18 (2009) (noting that LNT model is obsolete in view of known upregulation of cellular protective mechanisms against cancer; “LNT was a useful model half a century ago. But current radiation protection concepts should be based on facts and on concepts consistent with current scientific results and not on opinions. Preconceived concepts impede progress; in the case of the LNT model, they have resulted in substantial medical, economic, and other societal harm.”).



[1] The court EPA Office of Research and Development, Carcinogenic Effects of Benzene: An Update, at 38–39 (April 1998).

[2] Dr. Butler cited Richard B. Hayes, et al., “Benzene and Lymphohemaptopoietic Malignancies in Humans,” 40 Am. J. Indus. Med. 117, 120 (2001).

Differential Diagnosis in Milward v. Acuity Specialty Products Group

September 26th, 2013

Graffiti on the bathroom wall in the building that housed my undergraduate college’s philosophy department:

How does a philosopher treat constipation?

By using iterative disjunctive syllogism.

The joke is that this variety of syllogism is nothing other than reasoning by the process of elimination.

A or B or C

~A

B or C

~B

∴C

The syllogism works as a valid form of argument if the premises are all true.  So, if we start with three possible causes, A, B, and C, and we know that one or more of them caused an outcome, then we might proceed by the process of elimination to show that we can rule out all the others but the alleged cause.  The first line of the syllogism is true if at least one of the disjuncts is true.  As we rule out particular disjuncts upon learning that they are in fact false, we are left we a smaller set of disjuncts.  If we can proceed until we are left with the disjunct of interest, we may actually have succeeded in identifying a cause in fact of the particular case.

In the syllogistic argument above, we must be able to show that A and B are false before we can then conclude that C is true.

In differential etiology, we start with known causes, exposures or conditions that are known to be capable of causing a disease or disorder.  We do not know whether the potential causes were actually in play in a given case.  If we can use this syllogistic reasoning to conclude that the defendant’s product was a cause of the of the plaintiff’s harm, we might actually have shown specific causation in a reliable fashion.  If, however, we cannot proceed to a conclusion that unequivocally includes the defendant’s product, we are left with an indeterminate outcome, and the plaintiff must take nothing.

The Milward case was recently back in the news.  On remand from the First Circuit, the district judge, now the Hon. Douglas Woodlock, faced a renewed Rule 702 motion directed to Milward’s specific causation expert witnesses.  Milward v. Acuity Specialty Products Group, Inc., Civil Action No. 07–11944–DPW, 2013 WL 4812425 (D. Mass. Sept. 6, 2013).

Judge Woodlock wryly commented upon the First Circuit’s ignoring the statutory mandate of Rule 702, by its embracing caselaw that predated the 2000 statutory amendment of the Rule:

“While a 2000 amendment to Fed.R.Evid. 702 codified a rigorous reliability test, the Daubert line of cases has been read by the First Circuit as “demand[ing] only that the proponent of the evidence show that the expert’s conclusion has been arrived at in a scientifically sound and methodologically reliable fashion.” Ruiz–Troche v. Pepsi Cola of Puerto Rico Bottling Co., 161 F.3d 77, 85 (1st Cir.1998). “So long as an expert’s scientific testimony rests upon good grounds based on what is known, it should be tested by the adversarial process, rather than excluded for fear that jurors will not be able to handle the scientific complexities.” Milward, 639 F.3d at 15 (internal quotation and citation omitted).”

Milward, at *3.  After noting the statute’s “rigorous reliability test,” and the First Circuit’s having  diluted the statutory standard by drawing from pre-statute caselaw, Judge Woodlock got down to the business of gatekeeping, by examining the facts of record before him.

The defense’s first challenge was to Milward’s industrial hygienist’s opinion that quantified his benzene exposure.  The industrial hygienist, James Stewart, estimated Milward’s benzene exposure, both total and from individual products.  The defense challenge was interesting, given that plaintiffs have challenged defendants’ use of similar exposure recreations to advance apportionments that will defeat joint and several liability.  The district court denied the defense challenge, and turned its attention to the specific causation issue, which proved to be a good example of patho-epistemology .

The plaintiffs relied upon Dr. Sheila Butler, who was board certified in occupational medicine, pathology, and hematology, to opine that Brian Milward’s exposure was responsible for causing his Acute Promyelocytic Leukemia (“APL”), a rare subtype of Acute Myeloid Leukemia (“AML”). Butler’s opinion was simple if not simpistic:

“there is a ‘reasonable medical probability that there is a direct causal association between Mr. Milward’s APL and his excessive occupational exposure to benzene containing substances’ based primarily on

(1) the fact that his exposure to benzene preceded his development of APL, and

(2) a survey of studies showing increased AML risk following low average dose exposures to benzene.”

Milward at *6.

Simplistic and simply wrong. Butler had equated exposure and some risk, unquantified, with specific causation, an empty and unsupported assertion.  Judge Woodlock did not dignify this subjective opinion with further discussion, but turned his attention to Butler’s “differential diagnosis” analysis by which Butler claimed to have eliminated other potential causes of Milward’s APL such that she could say that benzene was a specific cause.

The district court started from the premise that so-called differential diagnosis is useful and accepted for assessing causation. Id. at *7 (citing Baker v. Dalkon Shield Claimants Trust, 156 F.3d 248, 253 (1st Cir.1998).  For some reason, however, the court emphasized that the differential etiology was particularly appropriate when the expert witness’s opinion lacks a foundation of epidemiologic studies or a “well-established threshold exposure levels at which disease occurs.”  The district court did not explain what it possibly could have meant by this emphasis, and I doubt that there is any basis for the court’s statement.

The real issue in Milward, on remand, was whether Dr. Butler applied the differential etiology in a reliable manner.  The defense argued that Dr. Butler failed to rule out competing risk factors, Milward’s prior smoking, and his morbid obesity, as causes of Milward’s APL.  The court dismissed this challenge with the recognition that plaintiffs might still prevail if Milward’s disease resulted from either benzene and smoking or benzene, smoking, and obesity. Sadly, the court did not address the quality or quantity of the evidence for smoking, or for obesity, and APL; nor did it address the magnitude of the associations that were being claimed by the defense, or by the plaintiffs.  The court did not explore the evidentiary base for the defense assertion that smoking or obesity causes APL such that it should be in the first line of the iterative disjunctive syllogism.

The problem, of course, was not the plaintiffs’ failure to rule out obesity or smoking, but their failure to rule out the unknown factors, which account for the solid majority of APL cases.  Indeed, in the first round of Rule 702 briefings and hearings, plaintiffs’ expert witness, Dr. Martin Smith, opined that between 70 and 80 percent of APL cases are idiopathic; that is, they have no known cause.  Id. at *7.  The syllogism thus becomes very difficult because one proposition in the first line of the argument is that the cause is unknown, and the plaintiff cannot arrive at the conclusion that his APL was caused by benzene unless and until he provides reliable evidence that more likely than not, his APL disease was not caused by one or more of the unknown causes.  In other words, plaintiffs must show that the APL was not a background case that would have occurred regardless of occupational benzene exposure, and perhaps regardless of occupational exposure with obesity and smoking.  Judge Woodlock, relying heavily upon the Restatement (Third) of Torts expressed the matter this way:

“When a disease has a discrete set of causes, eliminating some number of them significantly raises the probability that the remaining option or options were the cause-in-fact of the disease.  Restatement (Third) of Torts: Phys. & Emot. Harm § 28, cmt. c(4) (2010) (‘The underlying premise [of differential etiology] is that each of the[] known causes is independently responsible for some proportion of the disease in a given population.  Eliminating one or more of these as a possible cause for a specific plaintiff’s disease increases the probability that the agent in question was responsible for that plaintiff’s disease.’). The same cannot be said when eliminating a few possible causes leaves not only fewer possible causes but also a high probability that a cause cannot be identified. Id. (‘When the causes of a disease are largely unknown … differential etiology is of little assistance’.).”

In the face of this irrefutable logic of this part of comment c, Butler argued that she had “ruled out” idiopathic APL by “ruling in” benzene.  Of course, benzene had to be postulated as a general cause in order for it to be placed into the first line of the syllogism, but Butler’s assertion about ruling in benzene as a specific cause is truly an ipse dixit, a non sequitur, and a petitio principii, all rolled into one opinion.  After all, the APL case may have arisen out of benzene exposure and the unknown causes, or only the unknown (idiopathic) causes.  Butler cannot rule in benzene until she rules out idiopathic causes as the sole specific causes in this case. To be fair, the prevalence of idiopathic cases cited by Martyn Smith might be lower in a population with heavy benzene exposure, assuming Smith’s general causation were true, but again, such an acknowledgment would only raise the question of what the prevalence of idiopathic cases is in a population of exposure that looked like Mr. Milward’s.

Dr. Butler argued that Martyn Smith had previously ruled in benzene, but that was only as a general cause that can then be represented as one disjunct in the first line of the syllogism.  Here Judge Woodlock identified another gap between Smith’s general causation opinion and Dr. Butler’s attempt to use Smith’s opinion to place benzene into the differential etiology for Mr. Milward.  On this remand, the plaintiffs had to show that “the levels of exposure that are hazardous to human beings generally as well as the plaintiff’s actual level of exposure.” Id. (citing Westberry v. Gislaved Gummi AB, 178 F.3d 257, 263-64 (4th Cir.1999) (talcum powder undisputedly could cause sinus problems, and plaintiff was exposed at levels known to be causative).  The court suggested that Milward had not yet shown that exposure at the levels he experienced could cause APL.  Of course, even if Milward sustained cumulative exposures capable of causing APL, this fact sufficed only to place benzene into the differential diagnosis, and it did not advance the iterative disjunctive syllogism to a conclusion of either a single or multiple disjuncts that included benzene.

Judge Woodlock did a good job of saving the First Circuit from the notoriety of its general causation decision in the Milward case. The new trial court decision is a strong reminder that risk does not equal causation.  Differential etiology cannot rule out idiopathic cause(s) as the sole specific cause of a plaintiff’s disease unless there is a fingerprint of causation that makes the risk identifiable as a cause in a specific case.  Such a fingerprint or biomarker was apparently absent in the Milward case.  Similarly, the differential etiology might rule out putative specific causes on a probabilistic basis if the idiopathic cases made up a small number of all the cases in relation to the number of cases that arise from the exposure that is the subject of the litigation.

The Milward decision joins other soundly decided differential diagnosis cases coming out of the First Circuit.  See, e.g., Plourde v. Gladstone, 190 F. Supp. 2d 708, 722-723 (D. Vt. 2002) (excluding testimony where expert failed to rule out causes of plaintiff’s illness other than exposure to herbicides); Whiting v. Boston Edison Co., 891 F. Supp. 12, 21 n.41 (D. Mass. 1995) (noting that differential diagnosis cannot be used to support conclusion of specific causation when 90% disease cases are idiopathic).

But lest anyone get too comfortable with the notion that this issue has been mastered by the federal judiciary, keep in mind that there are some really poorly reasoned cases out there. See, e.g., Allen v. Martin Surfacing, 263 F.R.D. 47, 56 (D. Mass. 2008) (admitting general and specific causation testimony to be tested by adversary process, rather than excluded altogether, despite paucity of epidemiologic evidence and general acceptance that there are no known causes of amyotrophic lateral sclerosis).

The limits of the “process of elimination” approach has been addressed by some scientific organizations, such as the Teratology Society, in the particularly demanding context of determining a cause for a child’s congenital malformation:

“Biologic plausibility includes a consideration of alternative explanations for the outcome in an individual plaintiff. For example, if a plaintiff has a birth defect syndrome caused by a known genetic disorder, chemical exposure becomes implausible as a cause of the abnormality in that particular individual. The consideration of alternative explanations is sometimes misused by expert witnesses to mean that failure to find an alternative explanation for an outcome is proof that the exposure at issue must have caused the outcome. A conclusion that an exposure caused an outcome is, however, based on positive evidence rather than on lack of an alternative explanation.”

The Public Affairs Committee of the Teratology Society, “Teratology Society Public Affairs Committee Position Paper Causation in Teratology-Related Litigation,” 73 Birth Defects Research (Part A) 421, 423 (2005).

A brief, partial survey of differential etiology cases is set out below.


SECOND CIRCUIT

McCullock v. H.B. Fuller Co., 61 F.3d 1038, 1044 (2d Cir. 1995) (defining differential etiology as an analysis “which requires listing possible causes, then eliminating all causes but one”)

Prohaska v. Sofamor, S.N.C., 138 F. Supp. 2d 422, 439 (W.D.N.Y. 2001) (excluding expert’s opinion and granting summary judgment where expert “was unable to rule out, to a reasonable degree of medical certainty, [plaintiff’s] pre-existing condition, scoliosis, as a current cause of her pain”)

Zwillinger v. Garfield Slope Hous. Corp., 1998 WL 623589, at *20 (E.D.N.Y. Aug. 17, 1998) (excluding testimony and granting summary judgment where expert failed to rule out alternative causes of plaintiff’s immunotoxicity syndrome)

THIRD CIRCUIT

Magistrini v. One Hour Martinizing Dry Cleaning, 180 F. Supp. 2d 584, 608-610 (D.N.J. 2002) (excluding testimony of expert who sought to testify that dry cleaning fluid caused leukemia, but failed to rule out smoking as an alternative cause), aff’d, 68 F. App’x 356 (3d Cir. 2003)

In re Paoli R.R. Yard PCB Litig., 2000 WL 274262, at *5 (E.D. Pa. March 1, 2000) (expert’s opinion should be excluded “because she failed to rule out alternative causes” of plaintiff’s injuries)

Kent v. Howell Elec. Motors, 1999 WL 517106, at * 5 (E.D. Pa. July 20, 1999) (excluding expert testimony and granting summary judgment because expert could “not rule out reasonable alternative theories of what caused the retaining ring to fail”);

O’Brien v. Sofamor, 1999 WL 239414, at *5 (E.D. Pa. Mar. 30, 1999) (excluding expert’s testimony and granting summary judgment where plaintiff “offer[ed] no evidence that [plaintiff’s experts] performed a differential diagnosis, or even considered other potential causes” of plaintiff’s back condition)

Schmerling v. Danek Med., Inc., 1999 WL 712591, at *9 (E.D. Pa. Sept. 10, 1999) (excluding expert’s testimony and granting summary judgment on the grounds that expert’s failure to rule out alternative causes “alone warrants a determination that the expert’s methodology is unreliable”);

Turbe v. Lynch Trucking Inc., 1999 WL 1087026, at *6 (D.V.I. Oct. 7, 1999) (excluding expert’s testimony where expert “expressed awareness of obvious alternative causes” yet “did not investigate any other possible causes”);

Reiff v. Convergent Technologies, 957 F. Supp. 573, 582-83 (D.N. J. 1997) (excluding expert’s testimony and granting summary judgment where expert failed to rule out alternative causes of plaintiff’s carpal tunnel syndrome)

Rutigliano v. Valley Bus. Forms, 929 F. Supp. 779, 787 (D.N.J. 1996) (excluding expert’s testimony and granting summary judgment where the “record is replete with evidence, including [the expert’s] own admissions, that [plaintiff’s] symptoms could be attributable to medical conditions other than formaldehyde sensitization”)

Diaz v. Matthey, Inc., 893 F. Supp. 358, 376-377 (D.N.J. 1995) (excluding testimony and granting summary judgment where expert failed to rule out alternative causes for plaintiff’s asthma) (Irenas, J.)

Wade-Greaux v. Whitehall Labs., Inc., 874 F. Supp. 1441 (D.V. I.), aff’d, 46 F.3d 1120 (3d Cir. 1994) (excluding testimony of expert who failed to rule out alternative causes of plaintiff’s birth defects)

FOURTH CIRCUIT

Westberry v. Gislaved Gummi AB, 178 F.3d 257, 262-263 (4th Cir. 1999) (“Differential diagnosis, or differential etiology, is a standard scientific technique of identifying the cause of a medical problem by eliminating the likely causes until the most probable one is isolated”)

Shreve v. Sears, Robuck & Co., 166 F. Supp. 2d 378, 397-98 (D. Md. 2001) (excluding testimony where expert failed to rule out other causes of plaintiff’s injury other than an alleged defect in snow thrower)

Fitzerald v. Smith & Nephew Richards, Inc., 1999 WL 1489199 (D. Md. Dec. 30, 1999) (excluding expert’s testimony and granting summary judgment where expert “failed to rule out what could have been another cause of [plaintiff’s] condition”)

Aldridge v. Goodyear Tire & Rubber Co., 34 F. Supp. 2d 1010, 1024 (D. Md. 1999), vacated on other grounds, 223 F.3d 263 (4th Cir. 2000) (excluding testimony of plaintiffs’ experts where they “failed to adequately address possible alternative causes of plaintiffs’ illnesses”)

Oglesby v. General Motors Corp., 190 F.3d 244, 250 (4th Cir. 1999) (affirming exclusion of testimony where “as a matter of logic, [the expert witness] could not eliminate other equally plausible causes” of cracked plastic inlet);

Driggers v. Sofamor, S.N.C., 44 F. Supp. 2d 760, 765 (M.D.N.C. 1998) (excluding expert’s testimony and granting summary judgment where “expert failed to rule out other possible causes of [plaintiff’s back] pain”);

Higgins v. Diversey Corp., 998 F. Supp. 598, 603 (D. Md. 1997), aff’d, 135 F.2d 769 (4th Cir. 1998) (excluding expert’s testimony that the accidental inhalation of a bleach caused plaintiff’s injuries, where expert “admit[ted] that he [could] not rule out several other possible causes”)

FIFTH CIRCUIT

Michaels v. Avitech, Inc., 202 F.3d 746, 753 (5th Cir. 2000) (excluding testimony when “plaintiff’s experts wholly fail[ed] to address and rule out the numerous other potential causes” of an aircraft disaster)

Black v Food Lion, Inc, 171 F3d 308 (5th Cir 1999) (expert witness, purporting to use a differential diagnosis, testified that plaintiff’s slip in the supermarket caused fibromyalgia, which is largely idiopathic) (“This analysis amounts to saying that because [the physician] thought she had eliminated other possible causes of fibromyalgia, even though she does not know the real ‘cause,’ it had to be the fall at Food Lion. This is not an exercise in scientific logic but in the fallacy of post-hoc propter-hoc reasoning, which is as unacceptable in science as in law.”)

Conger v. Danek Med., Inc., 1998 WL 1041331, at *5-6 (N.D. Tex. Dec. 14, 1998) (excluding expert’s testimony and granting summary judgment when expert “had not attempted to rule out [other potential sources] as causes for [plaintiff’s back] pain”);

Leigh v. Danek Med., Inc., 1998 WL 1041329, at *4-5 (N.D. Tex. Dec. 14, 1998) (excluding expert’s testimony and granting summary judgment where expert failed to rule out alternative causes of plaintiff’s back pain)

Bennett v. PRC Public Sector, 931 F. Supp. 484, 492 (S.D. Tex. 1996) (excluding testimony of expert who failed to consider and rule out alternative causes of plaintiff’s repetitive motion disorders)

SIXTH CIRCUIT

Nelson v. Tennessee Gas Pipeline Co., 1998 WL 1297690, at *6 (W.D. Tenn. Aug. 1, 1998) (excluding testimony of expert who “failed to engage in adequate techniques to rule out alternative causes and offers no good explanation as to why his opinion is nevertheless reliable in light of other potential causes of the alleged injuries”);

Downs v. Perstorp Components, 126 F. Supp. 2d 1090, 1127 (E.D. Tenn. 1999) (excluding expert testimony as to whether exposure to chemicals caused plaintiff’s injuries where expert failed to rule out alternative causes)

EIGHTH CIRCUIT

Jisa Farms, Inc. v. Farmland Indus., No. 4:99CV3294, 2001 U.S. Dist. LEXIS 26084 (D. Neb. 2001)

Thurman v. Missouri Gas Energy, 107 F. Supp. 2d 1046, 1058 (W.D. Mo. 2000) (expert’s opinion “that the pipeline failed because of corrosion” was excluded and summary judgment granted where expert reached the conclusion “without eliminating other causes”)

Bruzer v. Danek Med., Inc., 1999 WL 613329, at *8 (D. Minn. Mar. 8, 1999) (excluding expert’s testimony and granting summary judgment where expert did “not attempt to rule out any alternative potential causes for [plaintiff’s] continuing and increasing [back] pain”)

National Bank of Commerce v. Assoc. Milk Producers, 22 F. Supp. 2d 942, 963 (E.D. Ark. 1998), aff’d, 191 F.3d 858 (8th Cir.1999) (excluding testimony and granting summary judgment where expert did “not successfully rule out other possible alternative causes” for cancer)

TENTH CIRCUIT

In re Breast Implant Lit., 11 F. Supp. 2d 1217, 1234 (D. Colo. 1998) (excluding expert testimony where expert failed to “explain what alternative causes he considered, or how he ruled out other possible causes” of plaintiffs’ auto- immune disease)

Stover v. Eagle Products, 1996 WL 172972, at *11 (D. Kan. Mar. 19, 1996) (excluding testimony of expert who “[did] not explain in any meaningful detail how he [was] able to exclude the numerous multiple alternative causes” of injury to plaintiff’s dogs)

ELEVENTH CIRCUIT

Rink v. Cheminova, Inc., 400 F.3d 1286, 1295 (11th Cir. 2005) (“[I]n the context of summary judgment . . . differential diagnosis evidence by itself does not suffice for proof of causation.”)

STATE COURT CASES

Blanchard v. Goodyear Tire & Rubber Co.,  2011 Vt. 85, 30 A.3d 1271 (2011) (holding that plaintiff’s claim that his NHL was caused by benzene was not reliably supported by differential diagnosis when a large percentage of NHL cases have no known cause)

Bradford Hill on Statistical Methods

September 24th, 2013

I am indebted to the article by Dr. Frank Woodside and Allison Davis on the so-called Bradford Hill criteria, for reminding me about the distorted view that some plaintiffs’ counsel advance in litigation about Bradford Hill’s view of statistical testing.  Frank C. Woodside, III & Allison G. Davis, “The Bradford Hill Criteria: The Forgotten Predicate,” 35 Thomas Jefferson L. Rev. 103 (2013).  Dr. David Schwartz has also written an insightful blog post on Bradford Hill.  See David Schwartz, “5 Reasons to Apply the Bradford Hill Criteria in Your Next Case” (Sept. 20, 2013).

Here is where Bradford Hill postulates the position of a research question before his famous nine factors come into the analysis:

“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).

The starting point, before the Bradford Hill nine factors come into play, requires a “clear-cut” association, which is “beyond what we would care to attribute to the play of chance.”  What is “clear-cut” association?  The most reasonable interpretation of Bradford Hill is that the starting point is an association that is not the result of chance, bias, or confounding.

I parted company with Woodside and Davis over whether Bradford Hill was somehow dismissive of the role of assessing chance in explaining an association.  In acknowledging any validity in the plaintiffs’ interpretation of Bradford Hill’s 1965 paper, Woodside and Davis, do an injustice, in my view, to Bradford Hill’s careful articulation of his position.

The starting position, quoted above, seems very clear, but Woodside and Davis note that later on in his speech, Bradford Hill suggested that tests of significance do not contribute to proof of the hypothesis.  Bradford Hill’s actual words are, however, fairly precise:

“No formal tests of significance can answer those questions. Such tests can, and should, remind us of the effects that the play of chance can create, and they will instruct us in the likely magnitude of those effects. Beyond that they contribute nothing to the ‘proof’ of our hypothesis.”

Bradford Hill at 299.

Plaintiffs’ counsel sometimes argue that this passage means that significance testing contributes “nothing” to proving the hypothesis, but this ignores two key points.  First, the argument ignores where in the text the passage occurs:  after Bradford Hill’s discussion of the nine factors.  Bradford Hill’s statement can be understood only as a reflection back on the nine factors.  The phrase “those questions” refers back to the nine factors, and this is the limitation that Bradford Hill is placing upon “formal tests of significance.” The starting point, before the nine factors are examined, is, after all, a “clear-cut” association, “beyond what we would care to attribute to the play of chance.”

Second, plaintiffs’ counsel’s argument ignores the clear meaning of the “[b]eyond that” phrase.  Beyond what?  Well, the limited role is nothing other than quantifying the play of chance in the observed results.  This role is hugely important, and of course, is incorporated into the starting point before the nine factors are examined.  In modern analyses, the role of random variability would actually be explored in the analysis of the exposure-outcome gradient, and perhaps in some of the other nine factors as well.  Bradford Hill implied that a statistically significant association was a preliminary step, after which the really hard work began.

It would be unfair to Bradford Hill to read into his statement much about “strict” testing versus a more flexible inferential approach in selecting or interpreting a Type I error rate.  By the time he presented his Presidential Address to the Royal Society of Medicine in 1965, much fur had flown in the disputes between Neyman and Fisher.  Resolving Bradford Hill’s view on the dispute is not a pressing issue because on either account, the quantification of the p-value is an extremely important step in evaluating scientific data.

In his textbook on medical statistics, Bradford Hill expands on the role of statistical analysis in medicine:

“Are simple methods of the interpretation of figures only a synonym for common sense or do they involve an art or knowledge which can be imparted? Familiarity with medical statistics leads inevitably to the conclusion that common sense is not enough. Mistakes which when pointed out look extremely foolish are quite frequently made by intelligent persons, and the same mistakes, or types of mistakes, crop up again and again. There is often lacking what has been called a ‘statistical tact, which is rather more than simple good sense’. That tact the majority of persons must acquire (with a minority it is undoubtedly innate) by a study of the basic principles of statistical method.”

Austin Bradford Hill, Principles of Medical Statistics at 2 (4th ed. 1948) (emphasis in original).

Even in this early work though, Bradford Hill acknowledges the limits of statistical methods:

“It is a serious mistake to rely upon the statistical method to eliminate disturbing factors at the completion of the work.  No statistical method can compensate for a badly planned experiment.”

Id. at 4 (emphasis in original).  That statistical method cannot save a poorly planned experiment (or observational study) does not, however, imply that statistical methods are not needed to interpret a properly planned experiment or study.

In the summary section of the first chapter, Bradford Hill removes any doubt about his view of the importance, and the necessity, of statistical methods:

“The statistical method is required in the interpretation of figures which are at the mercy of numerous influences, and its object is to determine whether individual influences can be isolated and their effects measured.”

Id. at 10 (emphasis added).

Conflicts of Interest in Asbestos Studies – the Plaintiffs’ Double Standard

September 18th, 2013

Conflicts of interest disclosures have become the stuff of “criminal” accusations.  In Weitz & Luxenberg P.C. v. Georgia–Pacific LLC, 2013 WL 2435565 (N.Y. App. Div., 1st Dep’t June 6, 2013), the court recited a defense expert witness’s failure to disclose his expert witness status in articles as part of a determination that a prima facie showing of “crime-fraud” had been made to justify the trial court’s in camera review of materials claimed to be protected from discovery under the attorney-client privilege:

“Nor did the articles reveal that Dr. Bernstein has been disclosed as a GP expert witness in NYCAL since 2009, that he had testified as a defense expert for Union Carbide Corporation in asbestos litigation, or that he had been paid by, and spoken on behalf of, the Chrysotile Institute, the lobbying arm of the Quebec chrysotile mining industry.”

Id. at *2. In “A Cautionary Tale on How Not to Sponsor a Scientific Study for Litigation,” I wrote about how the First Department of the New York Appellate Division went off the rails in considering the crime-fraud exception without first determining that the privilege applied in the first place.  The appellate court’s reasoning as to why the trial court should look for an exception was equally vacuous.  If failure to disclose consulting or testifying for attorneys is a “crime” or a “fraud,” then the playing field should be level and the indictment should apply to all sides.

Steve Markowitz is a physician with Queens College, City University of New York.  Dr. Markowitz testifies for plaintiffs, both here in New York City, and abroad, in asbestos personal injury cases.  See, e.g., Wannall v. Honeywell International Inc., 2013 WL 1966060 (D.D.C. May 14, 2013) (excluding Markowitz’ testimony as unreliable).  So the plaintiffs’ bar, which would equate failure to disclose consulting with “crime” or “fraud,” should be on the alert that Dr. Markowitz does not disclose his consulting arrangements in publications that bear on the issues covered by his testimony.

I blogged previously about an in-press version of Markowitz’s publication of an update of the epidemiologic study of North American insulators. SeeThe Mt. Sinai Catechism” (June 7, 2013). The paper is now out in final form, although behind a paywall.  Steven B. Markowitz, Steven M. Levin, Albert A. Miller, and Alfred Morabia, “Asbestos, asbestosis, smoking, and lung cancer. New findings from the North American insulator cohort,” 188 Am. J. Respir. Crit. Care Med. 90 (2013).  What is publicly available, however, are the disclosure statements for each of the authors. Lo, and behold, Dr. Markowitz declared no consultations that could be a potential conflict of interest.

This is a remarkable double standard.  Consulting for a defendant is a “crime,” if not disclosed, but plaintiffs’ testifying expert witnesses do not feel the need to disclose their consultancies at all.  What is more remarkable, however, is that the authors of this article strained and stretched their data to try to save their synergy theory.  Even the editorial that accompanied the article, while generally reciting the Mt. Sinai catechism, noted that the synergistic, multiplicative interaction was no longer so clear: “asbestos exposure and smoking together are associated with an at least additive increased risk.  …” John R. Balmes, “Asbestos and Lung Cancer: What We Know,” 188 Am. J. Respir. Crit. Care Med. 8,9 (2013).

The only reason that I harp on conflicts is that the Third Edition of the Reference Manual on Scientific Evidence (2011) improvidently started down this road, as have several federal district and state court judges, including the judges who sat on the First Department panel, which decided Weitz & Luxenberg P.C. v. Georgia–Pacific LLC.  I believe that the focus should be on the data and the analysis, not on the speaker.  If the courts insist upon creating this toxic environment for scientists who “consult,” then the toxicity should be visited on all parties’ expert witnesses equally.

Do English Judges Diss Epidemiology?

September 13th, 2013

As noted the other day, Claire McIvor, a senior lecturer, at the Birmingham Law School, has published an interesting U.K. perspective on the use of epidemiologic and statistical evidence in health-outcome litigation. SeeDebunking some judicial myths about epidemiology and its relevance to UK tort law,” in 21 Med. Law Rev. (2013), in press.

Ms. McIvor criticizes one case in particular for what she argues is an inappropriate dismissal of epidemiologic evidence as presented by an epidemiologist. Novartis Grimsby Ltd. v. Cookson, [2007] EWCA Civ 1261.

The pursuer, Cookson, worked for Novartis Grimsby, at its factory that manufactured dyes, including azo dyes, from 1964, until 2001, when he developed bladder cancer.  Cookson also chose to be exposed to various carcinogens as a personal lifestyle; he smoked cigarettes, 1/2 to one pack per day, for about 20 years, before quitting around 1980.

Cookson sued Novartis on allegations that he was overexposed to various aromatic amines[1], some of which are known to cause bladder cancer.  Novartis had previously paid such claims, but it contested Mr. Cookson’s case because of its belief that his workplace exposures had not been excessive, and that his past smoking habit more likely explained his cancer.  Both sides called physician expert witnesses, urologists, who both agreed that smoking and the aromatic amines could cause bladder cancer, but disagreed as to what caused Mr. Cookson’s disease.

Given the contest on causation, the two urologists agreed that the input of an epidemiologist, jointly instructed, would be helpful.  Now how quaint is that, for both sides to agree upon an expert witness?  Most lawyers in the United States would think it malpractice to engage in such a practice.

Professor Ray Cartwright, an epidemiologist who had published on the causes of bladder cancer, was the jointly instructed witness. A PubMed search for articles written by Cartwright on bladder cancer is set out below, and suggests that he was an appropriate choice, ex ante, at any rate.

Cartwright reviewed the epidemiologic literature, including some of his own studies. Cartwright’s report disappointed the plaintiff, however, when he opined that the workplace aromatic amine exposure was slight and posed only a low risk compared to the smoking. In assessing Cookson’s workplace exposure, Cartwright relied upon the exposure estimates of the parties’ industrial hygienists, and based his causal attribution upon an assessment that exposures were low.  Later, when plaintiff’s counsel showed that Cartwright misinterpreted some of the exposure data, Cartwright revised his report, but maintained that Cookson’s cancer was caused by smoking.

Professor Cartwright’s misstep on exposure probably diminished the strength of his opinion in the eyes of the trial judge, who ruled for the plaintiff.  Ms. McIvor seems to believe that this ruling improperly elevated clinical testimony over epidemiologic testimony, and credited “personalized probabilities” of the plaintiff’s testifying urologist, who attributed the cancer 20–25% to smoking, versus 70–75% to workplace exposures, and who opined that the workplace more than doubled the risk level that Cookson would have had had he never worked at the Novartis factory.  Novartis Grimsby Ltd. at 48.

Novartis appealed, on grounds that included an allegation of error in equating fact of exposure with causation of the bladder cancer.  Speaking for a unanimous England and Wales Court of Appeal, Lady Justice Smith dismissed the appeal, including its challenge to the medical causation issues.  Contrary to Ms. McIvor, however, the appellate court’s decision gave due weight to the epidemiologist, but found that the epidemiologic evidence was accessible to, and interpretable by, the clinicians. Although neither the appellate decision nor McIvor reviewed the actual epidemiologic evidence, several studies suggest that the relative risks for benzidine-derived dyes are greater than for smoking, and especially the risk for former smokers.  The judicial decision flowed not from improvidently dismissing epidemiologic evidence, or testimony by an epidemiologist, but from relying upon epidemiologic evidence marshaled by the plaintiff, through his urologist.[2]

Both sides agreed that smoking could cause bladder cancer, but they also had to agree that the risk of bladder cancer wanes after smoking cessation. Unfortunately, the Court of Appeal did not review the evidence, but the Surgeon General’s Reports note that cessation reduces risk by half after only a few years.  Wynder and Stellman (1977) and Wynder and Goldsmith (1977) suggest that the risk returns to baseline after 15 years of abstinence.  A study by Cartwright himself suggested the return to baseline in six years, although other studies (by Iscovich; Howe; Vineis; Hartge; and Burch) suggested an initial decline, followed by a persistent increased risk even beyond 15 years of abstinence.

Lady Justice Smith declared herself perplexed by these data, which seemed to be at odds with the notion that bladder cancer develops after 20 or more years latency:

“I myself have found it hard to understand how the passage of time after stopping smoking could result in a reduced risk of developing the disease if the aetiology of the disease is that the cancer begins at the time of exposure but does not manifest itself until later. However, as I have said, this issue was not fully explored in evidence and both experts agreed that the risk of developing bladder cancer from smoking decreased after smoking ceased.”

Novartis Grimsby Ltd. at 45.

Clearly though, it was not helpful to have Cartwright contradicted by the data in his own study.  Although the higher aromatic amine exposures occurred early in the plaintiff’s work career, Cookson continued to have some exposure up until the time of his diagnosis in 2001.  Professor Cartwright may well have been further undermined by the lack of any “time windows” in the occupational epidemiology, which would have supported a similar argument of declining risk from the more intense occupational exposure in the 1960’s.  The absence of such evidence for benzidine, compared with the evidence of latency and post-cessation declining risk for smoking, clearly hurt the employer’s case.  This imbalance in the evidence clearly helps to explain and support the courts’ rejection of Cartwright’s testimony.

Given the epidemiologic evidence, it is not at all clear that the plaintiff’s testifying urologist’s opinion that smoking contributed 25%-30%, whereas aromatic amines contributed 70%-75%, was merely a subjective or personal probability.  Smoking is associated with a two- to three-fold increase in risk in prospective studies, but Cookson was 20 years post-cessation.  His aromatic amine dyestuff exposure, which carries a much higher relative risk for bladder cancer, continued through till the end of his work tenure.  See “Dyes metabolized to benzidine,” in IARC Monographs on the Evaluation of Carcinogenic Risks to Humans Volume 100F (WHO 2012).

Cookson’s bladder cancer might have been a “background” case, or a result of both smoking and aromatic amine exposure, or a result of one or the other contested causes.  There appeared to be no serious evidence of synergy.  Given the studies at issue, the plaintiff’s testifying urologist’s opinion may well have been a reasoned analysis of the epidemiologic evidence.  The epidemiologist’s opinion, on the other hand, was clearly undermined by the facts of smoking cessation, and an initial error in exposure estimation.  Novartis’ counsel argued that Cartwright was the “real expert” on the issue of attribution, but Cartwright’s opinion was lacking important foundational facts, and there was no argument that Mr. Barnard, the plaintiff’s urologist, had erred in interpreting the epidemiologic data.  Novartis Grimsby Ltd. at 56.  The real “expert” was in the data, and there was no showing (at least in the published opinion) that the clinician, Mr. Barnard, misunderstood or distorted the epidemiologic data.  In this respect, the Novartis Grimsby case is very different from the Milward case, in which a plaintiff’s toxicologist mistreated, misanalyzed, and misrepresented epidemiologic studies on benzene.

Lady Justice Smith rejected the appellant’s criticism of the trial judge’s weighting Mr. Barnard’s opinion over Professor Cartwright’s:

“The proposition that a clinician is not capable of fully understanding the published epidemiological literature on the causation of a condition within his own specialty seems unsustainable and would, I think, surprise many clinicians and epidemiologists. In my view, it was clear from his detailed reports on causation that Mr. Barnard was familiar with the published work and he was also able to discuss it intelligently when giving evidence. The Recorder was plainly of that view. As for the suggestion that Mr. Barnard was too ready to assume that working for the appellant created an increased risk, this was a good ‘jury point’ but, if it did not appeal to the Recorder, that was an end to it.”

Novartis Grimsby Ltd. at 57.

Although Ms. McIvor is correct to be concerned with the court’s eager over-generalization about the ability of clinicians to understanding of epidemiologic studies, there was little suggestion that Mr. Barnard had tripped up, and there was a good deal to suggest that Professor Cartwright’s opinion was lacking on essential issues.  Admittedly, this impression may have been created by selective reporting by the Court of Appeal.  I have not seen the record or the briefs, but Ms. McIvor has not cited anything from those sources.

Mr. Barnard, the plaintiff’s urologist, further testified that the “occupational exposure had more than doubled the risk due to smoking.”  Novartis Grimsby Ltd. at 53.  The Court of Appeal thus found it easy to affirm the verdict that Cookson had shown that his workplace exposure was the “but for” cause of his cancer.  Of course, the Court of Appeal here accepted evidence of risk and relative risk as showing causation, a dubious proposition. Novartis Grimsby Ltd. at 67. And the Court of Appeal, distinguishing a pneumoconiosis case, further pronounced that the bladder cancer injury was “indivisible,” and thus not capable of an apportionment because neither exposure could be said to make the disease more severe.  The Court could have said, if it yielded to its own risk as causation rationale, that both exposures made the cancer more likely, and the occupational exposure contributed to this overall risk three times as much as the plaintiff’s smoking.  In Justice Lady Smith’s words:

“The natural inference to draw from the finding of fact that the occupational exposure was 70% of the total is that, if it had not been for the occupational exposure, the respondent would not have developed bladder cancer. In terms of risk, if occupational exposure more than doubles the risk due to smoking, it must, as a matter of logic, be probable that the disease was caused by the former.”

Novartis Grimsby Ltd. at 74.

The Court of Appeal’s opinion was thus consistent with its own commitment to the conflation of risk with causation, a conflation that may well be objectionable, but does not seem to be the basis for Ms. McIvor’s objections to the Novartis decision.  Of course, a remand with directions to apportion would have a perfectly logical and consistent result with the insistence that risk be substituted for causation in supporting the verdict below.


Publications of Professor Cartwright on Bladder Cancer from National Library of Medicine Database

1: Subramonian K, Cartwright RA, Harnden P, Harrison SC. Bladder cancer in patients with spinal cord injuries. BJU Int. 2004 Apr;93(6):739-43. PubMed PMID: 15049983.

2: Cartwright RA. Bladder cancer screening in the United Kingdom. J Occup Med. 1990 Sep;32(9):878-80. PubMed PMID: 2074512.

3: Cuzick J, Babiker A, De Stavola BL, McCance D, Cartwright R, Glashan RW. Palmar keratoses in family members of individuals with bladder cancer. J Clin Epidemiol. 1990;43(12):1421-6. PubMed PMID: 2147716.

4: Philip PA, Fitzgerald DL, Cartwright RA, Peake MD, Rogers HJ. Polymorphic N-acetylation capacity in lung cancer. Carcinogenesis. 1988 Mar;9(3):491-3. PubMed PMID: 3345587.

5: Cartwright RA. Screening workers exposed to suspect bladder carcinogens. J Occup Med. 1986 Oct;28(10):1017-9. PubMed PMID: 3772536.

6: Boyko RW, Cartwright RA, Glashan RW. Bladder cancer in dye manufacturing workers. J Occup Med. 1985 Nov;27(11):799-803. PubMed PMID: 4067684.

7: Cartwright RA, Philip PA, Rogers HJ, Glashan RW. Genetically determined debrisoquine oxidation capacity in bladder cancer. Carcinogenesis. 1984 Sep;5(9):1191-2. PubMed PMID: 6467507.

8: Cartwright RA, Glashan RW. Palmar keratoses and bladder cancer. Lancet. 1984 Mar 10;1(8376):563. PubMed PMID: 6142276.

9: Cartwright RA, Adib R, Appleyard I, Glashan RW, Gray B, Hamilton-Stewart PA, Robinson M, Barham-Hall D. Cigarette smoking and bladder cancer: an epidemiological inquiry in West Yorkshire. J Epidemiol Community Health. 1983

Dec;37(4):256-63. PubMed PMID: 6655413; PubMed Central PMCID: PMC1052920.

10: Cartwright RA, Adib R, Appleyard I, Glashan RW, Richards B, Robinson MR, Sunderland E, Barham-Hall D. ABO, MNSs and rhesus blood groups in bladder cancer. Br J Urol. 1983 Aug;55(4):377-81. PubMed PMID: 6411162.

11: Cartwright RA, Adib R, Appleyard I, Coxon JG, Glashan RW, Richards B, Robinson MR, Sunderland E, Barham-Hall D. Ten genetic polymorphisms in bladder cancer. J Med Genet. 1983 Apr;20(2):112-6. PubMed PMID: 6221102; PubMed Central PMCID: PMC1049011.

12: Cartwright RA. Historical and modern epidemiological studies on populations exposed to N-substituted aryl compounds. Environ Health Perspect. 1983 Mar;49:13-9. PubMed PMID: 6339220; PubMed Central PMCID: PMC1569142.

13: Cartwright RA, Robinson MR, Glashan RW, Gray BK, Hamilton-Stewart P, Cartwright SC, Barham-Hall D. Does the use of stained maggots present a risk of bladder cancer to coarse fishermen? Carcinogenesis. 1983;4(1):111-3. PubMed PMID: 6821882.

14: Cartwright RA, Glashan RW, Rogers HJ, Ahmad RA, Barham-Hall D, Higgins E, Kahn MA. Role of N-acetyltransferase phenotypes in bladder carcinogenesis: a pharmacogenetic epidemiological approach to bladder cancer. Lancet. 1982 Oct 16;2(8303):842-5. PubMed PMID: 6126711.

15: Garner RC, Mould AJ, Lindsay-Smith V, Cartwright RA, Richards B. Mutagenic urine from bladder cancer patients. Lancet. 1982 Aug 14;2(8294):389. PubMed PMID: 6124790.

16: Cartwright R. Occupational bladder cancer and cigarette smoking in West Yorkshire. Scand J Work Environ Health. 1982;8 Suppl 1:79-82. PubMed PMID: 7100861.

17: Glashan RW, Cartwright RA. Occupational bladder cancer and cigarette smoking  in West Yorkshire. Br J Urol. 1981 Dec;53(6):602-4. PubMed PMID: 7317749.

18: Cartwright RA, Gadian T, Garland JB, Bernard SM. The influence of malignant cell cytology screening on the survival of industrial bladder cancer cases. J Epidemiol Community Health. 1981 Mar;35(1):35-8. PubMed PMID: 7264531; PubMed Central PMCID: PMC1052117.

19: Cartwright RA, Adib R, Glashan R, Gray BK. The epidemiology of bladder cancer in West Yorkshire. A preliminary report on non-occupational aetiologies. Carcinogenesis. 1981;2(4):343-7. PubMed PMID: 7273315.

20: Cartwright RA, Glashan RW, Gray B. Survival of transitional cell carcinoma cases in 2 Yorkshire centres. Br J Urol. 1980 Dec;52(6):497-9. PubMed PMID: 7459578.

21: Cartwright RA, Bernard SM, Glashan RW, Gray BK. Bladder cancer amongst dye users. Lancet. 1979 Nov 17;2(8151):1073-4. PubMed PMID: 91807.

22: Cartwright RA. Genetic association with bladder cancer. Br Med J. 1979 Sep 29;2(6193):798. PubMed PMID: 519209; PubMed Central PMCID: PMC1596415.

23: Williams DR, Cartwright RA. The esterase D polymorphism in patients with diabetes or carcinoma of the bladder and a matched sample of non-dono. Ann Hum Biol. 1978 May;5(3):281-4. PubMed PMID: 686669.


[1] α-naphthylamine, some of which was contaminated with β-napthylamine, benzidine, dianisidine and o-tolidine

[2] There was a suggestion that the plaintiff’s urologist had invoked his clinical experience in treating men from the factory with bladder cancer, but the courts did not seem to give dispositive weight to this irrelevant argument for causation. Novartis Grimsby Ltd. at 56.

Daubert Bewigged

September 11th, 2013

Claire McIvor, a senior lecturer, at the Birmingham Law School, has published an interesting U.K. perspective on the use of epidemiologic and statistical evidence in health-outcome litigation. See “Debunking some judicial myths about epidemiology and its relevance to UK tort law,” in 21 Med. Law Rev. (2013), in press.

McIvor argues that British judges have failed to engage with epidemiologic evidence, and have relegated epidemiologic evidence to a status inferior to clinical evidence, even when testifying clinicians have little to offer the fact finder.  If the be-wigged judges have done this shame on them, but McIvor suggests that a pre-trial hearing is necessary to address the proper (and improper) range of methodologies and inferences:

“The very fact that methodologically problematic evidence can end up before a trial court is indicative of the need for a pre-trial admissibility test for scientific evidence in UK civil law. Such a test would afford the court an opportunity to evaluate the scientific reliability of any epidemiological evidence that the parties wish to introduce at trial.”

McIvor at 22.  In advancing this recommendation, McIvor expands upon a recent Law Commission recommendation for what she describes as “a pre-trial admissibility test for scientific evidence in criminal litigation, similar to that which is used in the USA.”  McIvor at 32 (citing Law Commission, Expert Evidence in Criminal Proceedings in England and Wales (Law Comm’n No. 325, 2011)).

This recommendation and discussion, however, are confusing and perhaps confused.  The test in the United States is not a pre-trial test, although a party may ask for a determination in advance, either in conjunction with a motion for summary judgment, or to limit the evidentiary display at trial.   Nonetheless, objections to expert witness opinion testimony can certainly be made at trial.  Indeed, if the pre-trial motion is denied, the moving party may well have to renew its objection at trial in any event.

MacIvor’s recommendation is puzzling for other reasons.  First, most civil cases are tried to the bench, and the need to challenge the expert opinion pre-trial is certainly less pressing.  Lengthy, methodological challenges are virtually impossible before a jury but they would be made in front of the presiding judge, in any event.  Second, having recommended the pre-trial procedure, and the substantive standard for reliability and validity, McIvor proceeds to tell us that it [the Daubert standard] has “proven to be a rather controversial test in practice.”  Id. at 32 n.84 (citing no less of an authority than Carl Cranor, Toxic Torts: Science, Law and the Possibility of Justice 62-90 ( 2006)).  Cranor is hardly an unbiased, reliable source, but if McIvor accepts his pronouncements, her recommendation is hard to understand.  Third, McIvor gives us an example of a class of cases, which at first blush, suggest that judges on the other side of the Atlantic just do not understand science.  In McTear v. Imperial Tobacco, [2005] 2 SC 1, the trial judge, Lord Nimmo Smith, ruled in favor of a tobacco company in a lung cancer personal injury case.  His ruling was largely based upon a rejection of the epidemiologic evidence, which McIvor suggests is unreasonable, but then tells us that the rejection might have resulted from the plaintiffs’ reliance upon reports without the benefit of an epidemiologist to explain and teach the trial judge about the meaning of the evidence.

Indeed, McIvor tells us that Lord Nimmo Smith complained in his opinion that he had not been:

“‘sufficiently instructed by the expert evidence about this discipline’ to be able to form his own judgment of the evidence. This was not an unreasonable point, at least as regards the issue of individual causation.”

McIvor at 32 (quoting Lord Nimmo Smith).  Well, it does suggest that the good Lord may have been a stubborn Scot, who was not going to give any weight to the common wisdom, but rather insist that the plaintiff make his case in court.  Even McIvor goes on to characterize the plaintiff’s counsel’s strategy as “unwise, in hindsight.”  Id.

Rule 702 is an extremely important part of the law of evidence in federal courts, and in many state courts.  The U.K. would do well to adopt it, with allowance for the very different role of judges in civil cases on the other side of the Atlantic.

Urging Review and Reversal, Scientists File Amicus Brief in the Harkonen Case

September 7th, 2013

Earlier this week, Professors Kenneth Rothman and Timothy Lash, and I, filed our Brief by Scientists And Academics as Amici Curiae, in the case, Harkonen v. United States.  As noted previously, Dr. Harkonen has petitioned the Supreme Court for review of the Ninth Circuit’s affirmance of his conviction for Wire Fraud.  Other amici will likely file on Monday, September 9, 2013.

Aaron Kesselheim’s Presentation on FDA Regulation of Manufacturer Speech

September 3rd, 2013

On August 5, 2013, Dr. Scott Harkonen filed his petition for a writ of certiorari with the United States Supreme Court. As noted in some previous posts, Dr. Harkonen was acquitted of misbranding, but convicted of wire fraud, for his role in issuing a press release about the results of a clinical trial of interferon gamma 1b, in patients with idiopathic pulmonary fibrosis.  (See Multiplicity versus Duplicity – The Harkonen Conviction; The Matrixx Motion in U.S. v. Harkonen; The (Clinical) Trial by Franz Kafka).

Dr. Harkonen’s petition presents two questions:

“1. Whether a conclusion about the meaning of scientific data, one on which scientists may reasonably disagree, satisfies the element of a “false or fraudulent” statement under the wire fraud statute, 18 U.S.C. § 1343?

2. Whether applying 18 U.S.C. § 1343 to scientific conclusions drawn from accurate data violates the First Amendment’s proscription against viewpoint discrimination, or renders the statute, as applied , unconstitutionally vague.”

Both questions are important given that the government has conceded that Dr. Harkonen’s press release accurately presented the raw data and calculated p-values.  The crime, if crime it be, lay in Dr. Harkonen’s drawing a causal inference from a subgroup, p = 0.004, which was not prespecified, in a specified secondary endpoint of survival (p = 0.08), when the subgroup was clearly based upon the goals of the trials, and there was other corroborative evidence in the form of two previous trials, clinical practice, and strong mechanistic evidence.

The government argued that NO inferences could be drawn from a trial that “failed” on its primary endpoint.  The government’s embrace of this statistical orthodoxy greatly misrepresented scientific practice to the courts below.  The only “failed” trial is one that is not conducted.

There are many who would go to great lengths to distort the facts of the Harkonen case in order to demonize the pharmaceutical industry, or to arm the Justice Department with a weapon that can shut down scientific speech about pharmaceutical interventions.  The expansion of the Wire Fraud Act, seen in the Harkonen case, to achieve these political goals will not only affect pharmaceutical company scientists, but also government and academic scientists.  The standard for falsity, drawn from an out-dated, tendentious, and overly rigid conception of hypothesis testing will apply equally to non-industry scientists in False Claim Act cases.  Perhaps in future posts, I can provide some good examples, on condition that any qui tam relators share their bounty with me.

Back in May, Aaron Kesselheim presented (by video) a paper, written with Michelle Mello, of the Harvard School of Public Health, on “The Prospect of Continued FDA Regulation of Manufacturer Promotion in an Era of Expanding Commercial Speech.”  Kesselheim went out of his way to misrepresent the facts of the Harkonen case, as part of his brief against off-market promotion.

By way of background, Aaron S. Kesselheim is a physician and a lawyer, and an Assistant Professor of Medicine at Harvard Medical School.  He is also a faculty member in the Division of Pharmacoepidemiology and Pharmacoeconomics in the Department of Medicine at Brigham and Women’s Hospital.   Given his position and his training in two professions, as well as the extraordinary stakes involved in allowing the government to prosecute scientists for drawing allegedly false conclusions about facts that the government concedes are accurate, Dr. Kesselheim should have exercised much greater care in checking his own assertions more closely.

Dr. Kesselheim focused primarily on the Second Circuit’s recent decision in United States v. Caronia, 703 F.3d 149 (2d Cir. 2012) , which reversed a judgment of conviction for off-label promotion, on First Amendment grounds.  About nine minutes into his presentation, Kesselheim turned to alternative strategies for the government to use to squelch off-label promotion.  One of his suggestions was to follow the model of the Harkonen prosecution, and to prosecute off-label promotion as false and misleading speech.

In his discussion of his suggested strategy, Kesselheim suggested that Dr. Harkonen had made misleading “conclusory, unsubstantiated claims for efficacy,“ and “without reference to supporting evidence.”  It is Kesselheim, however, who seriously mislead his listeners and readers by stating that Dr. Harkonen had made “conclusory, unsubstantiated claims for efficacy.”  The Press Release that was the subject of the government’s indictment set out accurately actual count data and calculated p-values.  No data were fabricated or falsified.  Within the limited space and the informal context of a Press Release, Dr. Harkonen had provided a substantial account of the data from InterMune’s clinical trial, as well as citing a previous, independent clinical trial and its extension, clinical experience, and mechanism research on the action of interferon γ-1b.  Unfortunately, it is Kesselheim who is speaking in conclusory sound bites when he ignores the context and content of the actual Press Release at issue.

Kesselheim went on to suggest that Harkonen’s statement was refuted by a “company-sponsored clinical trial showing that the drug was not effective.” This statement is not only false, but shows a flagrant disregard for statistical analysis and the data in the Harkonen case.  Kesselheim implies that a clinical trial that fails to show treatment efficacy thereby shows that the treatment was not effective.  His statement commits the fundamental error of equating a failure to reject the null hypothesis at a specified level of attained significance with acceptance of the null hypothesis.  This reasoning is fallacious and fundamentally flawed.

To be sure, the prespecified secondary survival endpoint in InterMune’s clinical trial did not meet the 0.05 cutoff (it was 0.08), although the per-protocol analysis for this endpoint came up at 0.055, on a preliminary analysis of the data. When the clinical trial was fully analyzed and written up for publication in the New England Journal of Medicine, the treatment-adherent analysis for survival in the entire clinical trial was 0.02, with a statistically significant hazard ratio for survival, favoring the therapy:

“Analysis of the treatment-adherent cohort of patients showed an absolute reduction in the risk of death of 9 percent in the interferon gamma-1b group, as compared with the placebo group, and a relative reduction in the risk of 66 percent (5 percent of 126 patients in the interferon gamma-1b group and 14 percent of 143 patients in the placebo group died, P=0.02). The hazard ratio for death in the interferon gamma-1b group, as compared with the placebo group, was 0.3 (95 percent confidence interval, 0.1 to 0.9).”

Ganesh Raghu, Kevin K. Brown, Williamson Z. Bradford, Karen Starko, Paul W. Noble, David A. Schwartz, and Talmadge E. King, Jr., for the Idiopathic Pulmonary Fibrosis Study Group, “A Placebo-Controlled Trial of Interferon Gamma-1b in Patients with Idiopathic Pulmonary Fibrosis,” 350 New Engl. Med. J. 125, 129-30 (2004).

Dr. Harkonen, in his Press Release, did focus on what seems like an eminently sensible subgroup, within the survival secondary endpoint, of mild- and moderate-cases, which, a priori, were believed to be the patients mostly likely to benefit from the interferon γ-1b therapy.  (What was not known before the trial was at what point in disease progression might patients no longer respond with greater survival, and hence the difficulty in setting the boundary between moderate and severe cases.)  Kesselheim might argue that the interferon γ-1b clinical trial, standing alone, was inconclusive, but he certainly cannot argue truthfully that the trial showed that the biological product to be ineffective.  Clinical trials do not neatly divide the world of possible results into demonstrations of efficacy and demonstrations of inefficacy.  Not only does the evidence come in degrees, but there is a range of “inconclusiveness” in between the two extremes. Given his background, training, and experience, Kesselheim certainly should know this, and he should apologize for his inaccurate statements.

Kesselheim might well have stopped there, but he went on to acknowledge that the company-sponsored clinical trial at issue did find, in post-hoc analyses, a non-significant trend of benefit in a subset of patients.  Talk of misleading speech!  The p-value at issue was 0.004, uncorrected for multiple comparisons, but no one, not Kesslheim, not the government or anyone else, has offered any appropriate adjustment for multiple comparisons that would inflate that 0.004 to over 0.05.  Kesselheim has no warrant for branding the subgroup finding “non-significant,” until he shows that the p = 0.004, when appropriately modified (if it can be), exceeds 0.05.

Kesselheim mangles other, less technical facts.  He claims that the company saw a ten-fold increase in sales of interferon γ-1b for idiopathic pulmonary fibrosis.  No such fact was ever, or could ever, be established in the Harkonen case.  Kesselheim claims that Dr. Harkonen admitted, in emails, that he did not really believe that the trial “demonstrated” benefit; no such emails were ever adduced at trial, and this seems to be part of a fictional narrative that Dr. Kesselheim has manufactured.  Finally, Kesselheim harrumphs that FDA declined to approve drug.  The company never filed a new drug application for the idiopathic pulmonary fibrosis indication; there was no application to reject.  Perhaps more important is that the Press Release was issued before InterMune had made any formal submission of data to the FDA, an event that did not take place until the following year.

Kesselheim sighs that the Harkonen prosecution will be a difficult act to follow because it requires a case-by-case showing of falsity, with the necessity of expert testimony, and heavy cognitive demands on lay jurors. How ironic that Kesselheim, a lawyer and a physician, and a Harvard Medical School faculty member, buckled under the cognitive demands of his topic. Indeed, Kesselheim’s confusion is a strong argument for why the Supreme Court should put a stop to the practice of asking jurors to second guess whether a scientist has incorrectly inferred causation from accurately presented facts.

Let’s hope Dr. Harkonen gets a fair hearing in the Supreme Court.