In several recent decisions, seven federal district courts adopted arguments made by a team of Drinker Biddle lawyers and found causation expert testimony regarding an over-the-counter homeopathic cold remedy, Zicam® Cold Remedy No Drip Liquid Nasal Gel (“Zicam”), unreliable and inadmissible.1 In each case, the plaintiffs alleged that they experienced impairment of their sense of smell following use of Zicam. Each court rejected the proffered causation testimony as unreliable, unsupported by the scientifi c method, and contrary to Federal Rule of Evidence 702 and Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579 (1993) and its progeny.

  • On September 29, 2006, in Hans, et al. v. Matrixx Initiatives, Inc., et al., 3:04-cv-540-R, the Western District of Kentucky excluded the testimony of plaintiffs’ expert, Bruce W. Jafek. The court concluded that his threshold opinion that the use of Zicam delivers zinc gluconate (the active ingredient in Zicam) to the smell tissue2 was not scientifi cally supported.
  • On November 7, 2006, in Sutherland v. Matrixx Initiatives, Inc., et al., 04-AR-0129-M, the Northern District of Alabama excluded Dr. Jafek’s causation testimony.
  • On December 27, 2006, the Middle District of Alabama similarly excluded Dr. Jafek’s expert testimony in Benkwith v. Matrixx Initiatives, Inc., et al., 2:04- cv-623-MEF, 467 F.Supp.2d 1316 (M.D. Ala. 2006).
  • On January 4, 2007, in O’Hanlon v. Matrixx Initiatives, Inc., et al., CV 04-10391 AHM (JTLx), the Central District of California excluded all three of the plaintiffs’ proffered causation experts — Dr. Jafek, Dr. Terence Davidson and Dr. Gerhard Schrauzer — on the grounds that their causation testimony was not based on reliable scientifi c evidence.
  • On February 20, 2007, in Hilton v. Matrixx Initiatives, Inc., et al., 4:04-cv-519-Y, the Northern District of Texas, after holding an evidentiary hearing, concluded that Dr. Jafek’s causation opinions were not suffi ciently reliable to be admitted.
  • On March 16, 2007, in Salden v. Matrixx Initiatives, Inc., et al, No. 06-10277, the Eastern District of Michigan concluded that the testimony of plaintiff’s expert, Dr. Alan Hirsch, did not qualify as “scientifi c knowledge” and was not admissible.
  • Most recently, on March 30, 2007, in Wyatt v. Matrixx Initiatives, Inc., et al., No. 2:04-cv-1230-UWC, another judge of the Northern District of Alabama concluded that Dr. Jafek’s testimony lacked the necessary reliability to be admitted under Rule 702. The court also excluded the general causation testimony of the plaintiff’s treating physician.

Each court took a slightly nuanced approach in rejecting the proffered causation expert testimony, but all seven courts came to the same conclusion. The thorough analysis and discussion provided by the Alabama District Courts in the Sutherland and Benkwith decisions, applying Eleventh Circuit law, are particularly instructive for evaluating scientifi c testimony under the standards of Rule 702 and Daubert.

Background

Dr. Bruce Jafek is a board-certifi ed otolaryngologist with the Rocky Mountain Smell and Taste Center at the University of Colorado. His specialty is smell and taste dysfunction. In September 2003, he and two colleagues published a case series abstract/poster reporting on the case of a patient seen at the Center, and referring to nine other patients in the Center’s fi les with histories of smell loss following Zicam use. Dr. Jafek later published the case series in a peerreviewed journal in May 2004.

Near the time Dr. Jafek submitted his poster, he began discussing potential litigation with attorneys planning to sue the makers of Zicam, and soon after the poster was presented he was retained to consult and testify. By the time his paper was published, he had already appeared in nationwide pop news programs to tout his conclusion that Zicam causes smell loss. But he did not conduct any testing or experiment of the foundations for his opinions until after he had been deposed in the litigation. These efforts seemed calcu-lated to produce a result, rather than meaningfully investigate a scientifi c hypothesis. A deposition also revealed that Dr. Jafek’s case series was fraught with analytical shortcuts, including heavy reliance on unverifi ed internet questionnaires.

Zicam treats cold symptoms. The most common cause of persistent smell impairment is upper respiratory infection, which includes the cold. Dr. Jafek purports to rule out the cold and the numerous other conditions and medications that can cause smell loss, and selects Zicam as the cause, based largely on temporality and distortion of the patients’ histories. His general causation opinion is based on analogy to studies involving a different zinc-containing compound, zinc sulfate, in various animal species and in a series of experiments on humans performed in the 1930s in search of a cure for polio. Dr. Jafek contends that “zinc is zinc” and disregards the signifi cant structural differences between zinc sulfate and zinc gluconate, which result in lower zinc fractions and more limited disassociation characteristics in the latter and, consequently, less available zinc. He also ignores the substantial differences in application methodology and dose volume in the zinc sulfate studies, which further distance the amount of available zinc.

The diffi culty of delivering a viscous gel like Zicam to the upper reaches of the nasal cavity, where the small and sheltered patch of smell tissue resides, was also ignored by Dr. Jafek. Various studies, including studies on the Zicam gel administered with the Zicam pump, demonstrate that Zicam cannot reach the smell tissue when the product is used as directed; even egregious violation of the package directions is unlikely to deliver any substantial amount of the gel to the smell tissue. Instead of performing his own in vivo deposition study and then reaching a conclusion, Dr. Jafek conducted a cadaver experiment long after he formed his opinion and during the course of litigation to prove that Zicam can reach the smell tissue. The procedure he adopted is both unprecedented and unreliable.

Finally, Dr. Jafek ignored and then misapplied basic principles of dose-response. He initially testifi ed that he did not have any opinions concerning the amount of gel that would need to reach the smell tissue to cause impairment and how much of the gel reaches the tissue under applicable conditions of use. He later offered an opinion based on reanalysis of rodent studies performed by other scientists, but it was based on an extrapolation methodology unsupported in the scientifi c community. Beyond those unsupported distribution and dose-response opinions, Dr. Jafek essentially relied on case reports (including his own fl awed case series) and adverse event reports of individuals experiencing smell loss after acquiring cold symptoms and using Zicam. Dr. Jafek’s testimony was challenged by Drinker Biddle lawyers and local counsel representing the Zicam manufacturers in the series of cases set forth above.

Sutherland v. Matrixx Initiatives, Inc., et al., United States District Court, Northern District of Alabama, No. 04-AR-0129-M, Memorandum Op. (November 7, 2006)

The Sutherland court evaluated the admissibility of Dr. Jafek’s opinions based on the Eleventh Circuit’s decision in McClain v. Metabolife Int’l, Inc. 401 F.3d 1220 (11th Cir. 2005), where the same trial judge, Hon. William M. Acker, had been reversed after allowing a pharmacist, Dr. James O’Donnell, to testify on causation in a case involving ephedra. The Sutherland court concluded that the methods and procedures employed by Dr. Jafek were not suffi ciently reliable under Rule 702, and were therefore inadmissible. Memorandum Op. at 4-5.

The court summarized the gatekeeping inquiry demanded by Rule 702 as requiring the court to: engage in a rigorous inquiry to determine whether: (1) the expert is qualifi ed to testify competently regarding the matters he intends to address; (2) the methodology by which the expert reaches his conclusion is suffi ciently reliable as determined by the sort of inquiry mandated by Daubert; and (3) the testimony assists the trier of fact, through the application of scientifi c, technical, or specialized expertise, to understand the evidence or to determine a fact in issue.

Id. at 8-9 (quoting Rink v. Cheminova, Inc., 400 F.3d 1286 (11th Cir. 2005)). The court determined that Dr. Jafek was a highly qualifi ed clinician and otolaryngologist, but these disciplines were not quite on point for his causation opinions. General causation investigation depends primarily on principles of toxicology, and a clinician may or may not have the background and employ the methodology needed to generate reliable causation opinions. As the court put it, Dr. Jafek “is wearing several hats in this litigation, but not all of them fi t.” Id. at 12.

As a threshold matter, the Eleventh Circuit in McClain had held that rigorous examination of the reliability of an expert’s general causation analysis is necessary only where the asserted causal relationship is not generally accepted in the scientifi c community. Sutherland argued that zinc, a metabolite of Zicam, is generally accepted to cause smell loss when applied to the smell tissue. The Sutherland court focused instead on the product as a whole and concluded that there was no general consensus in the scientifi c community that use of Zicam causes smell loss, and therefore a reliable foundation for Dr. Jafek’s general causation conclusion needed to be established. Id. at 13. This would require reliable scientifi c evidence that (1) Zicam can reach the smell tissue when used as Sutherland used it (exposure), and (2) it does so at a dose capable of causing permanent anosmia (toxic exposure). Id. at 13-14.

Sutherland testifi ed she used the product in the manner specifi ed by the package directions. To prove that Zicam reaches the smell tissue when used as directed, Dr. Jafek relied on his observations that Zicam can be discharged in the open air a distance far greater than the short linear distance between the nasal opening and the smell tissue, and that this route is a “straight path.” The court found the straight path conclusion dubious, noting Dr. Jafek’s apparent “mixed thoughts on the matter” – in an earlier paper he had written that “the neuroepithelium is ‘almost anatomically inaccessible.’” Id. at 14-15. The court also criticized Dr. Jafek’s reliance on his open air observation in lieu of testing his theory with a reasonable scientifi c experiment such as examining the distribution pattern of a similarly viscous gel in a live human being. Id. at 15-16. Additional methodological concerns fl owed from Dr. Jafek’s attempt to rely on a study he had “heard of . . . peripherally” and a case report by another anti-Zicam expert witness which actually reached conclusions at odds with his own.

Dr. Jafek also relied on a cadaver study he conducted after his opinions had been challenged in litigation. The court found the study was infected with bias, since Dr. Jafek had previously and publicly reached his conclusion that Zicam causes anosmia:

The study was conducted in between Zicam-related depositions and . . . after Dr. Jafek’s public appearances on the subject. One of the hallmarks of the scientifi c method is an unbiased pursuit of knowledge. Scientists are to observe and interpret the results of their experiments; they are not to engage in agenda-driven fact fi nding for the purposes of litigation. The sequence of events implicates not only credibility, a matter for the jury, but a scientifi c method, a matter for the court as gatekeeper.

Id. at 17. The court also found major methodological fl aws in the way Dr. Jafek conducted the cadaver study. Consequently, the study failed to support Dr. Jafek’s conclusion that Zicam can reach the smell tissue in a living person. Id. at 17-18. The court concluded that the experiment “was a post hoc attempt to justify a prior conclusion.” Id.

Finally, the court rejected Dr. Jafek’s resort to circular reasoning, that Zicam must have reached Sutherland’s smell tissue, because it caused her smell loss. Like the other premises for his distribution opinion, this offered “no recognizable scientifi c basis.” Id. at 19-20.

The court next turned to Dr. Jafek’s opinion that product level doses of Zicam are capable of causing smell loss. The court rejected Dr. Jafek’s reliance on studies involving zinc sulfate, fi nding too much dissimilarity in the way the compounds were administered in the studies compared to Zicam use. The court also found that Dr. Jafek had not supported extrapolation as to the compounds, as he failed to “demonstrate experimentally that the two compounds have an identical effect” and failed to “address the many differences between the two substances.” Id. at 21. A test tube experiment purporting to show that many zinc salts precipitate protein in a similar manner was insuffi cient because it did no more than imply that the biological effects might be similar. Id. at 21-22.

Dr. Jafek also relied on his own case series to support his causation opinion, but McClain teaches that “[a] case study does not prove causation.” Id. at 23. The court further found the case series plagued by “serious methodological fl aws.” Id. at 24.

The court also concluded that Dr. Jafek had failed to establish the dose-response relationship – that Zicam use delivers a dose capable of causing permanent smell loss – a foundation which “is fundamental to scientifi c toxicology.” Id. at 25. His analysis included “a series of implausible analogies.” Id. For example, though there was no proof that any amount of Zicam reaches the smell tissue, he assumed the entire metered dose did so for the purpose of calculating an effects threshold. “Such a leap defi es logic and enters the never-never land of conjecture and wishful thinking.” Id. at 26.

Dr. Jafek’s dose-response extrapolation from a mouse model was found to be unwarranted and unsupported. He offered no scientifi c support for his assumption that toxic effects in a mouse would be similar to that in humans, nor for the assumption that minor effects in the mouse can equate to serious and permanent effects in humans. Dr. Jafek’s analysis also failed to distinguish between contact with the walls of the nasal cavity and contact with the smell tissue. Id. His inferential leaps, lack of testing and overall lack of toxicological evidence convinced the court that “his conclusions remain no more than an ‘intuitive, clinical hunch.’” Id. at 14, 27.

The court also identifi ed several disqualifying fl aws in Dr. Jafek’s specifi c causation analysis, fi nding his “specifi c causation testimony even more questionable than his general causation testimony” in that he failed to address the evidence undermining his diagnosis and opinions. Id. at 29- 33. The court found it unnecessary to further consider the problems with Dr. Jafek’s specifi c causation testimony. Id. at 33.

Having excluded Dr. Jafek, the court found it unnecessary to reach the Daubert motion fi led to exclude the causation testimony of plaintiff’s treating physician, who relied heavily on Dr. Jafek’s opinion and case series. Id. at 33. Benkwith v. Matrixx Initiatives, Inc., et al., 2:04- cv-623-MEF, 467 F.Supp.2d 1316 (M.D. Ala. 2006) Like its predecessor in Sutherland, the Benkwith court concluded at the outset that “the scientifi c community does not generally recognize that zinc gluconate is toxic to the human olfactory epithelium.” Id. at 1323.

Plaintiff Benkwith also testifi ed that her use of Zicam was consistent with the package directions. The court found that Dr. Jafek’s opinion that Zicam can reach the smell tissue when used as directed was not based on reliable scientifi c methodology and failed to meet the minimum level of scientifi c rigor required by Daubert. Id. at 1324.

First, the court observed that Dr. Jafek’s interpretation of the proper manner of use, based on the directions, was questionable. Id. at 1323. The court put that issue aside and concluded that Dr. Jafek’s reliance on personal observations of the manner of discharge of the Zicam gel and nasal anatomy, rather than conducting in vivo experimentation, violated the scientifi c method and failed to demonstrate the level of intellectual rigor of an expert in the fi eld. Id. at 1324. The court also found the cadaver study “not suffi ciently relevant or reliable to survive scrutiny under Rule 702. ” Id. at 1325. Like the Sutherland court, the court found the protocol unreliable and unrepresentative of actual use conditions, and was troubled that the experiment was conducted only after Dr. Jafek had formed his conclusions: it “appears to have been undertaken more to bolster a conclusion than to test a hypothesis.” Id. at 1325. Also consistent with Sutherland, the court found that Dr. Jafek’s reliance on studies he had not fully reviewed “does not fall within Rule 702’s standards for expert testimony,” and the case series by another expert which disagreed with several of Dr. Jafek’s conclusions did not support his distribution opinion. Id. at 1326.

The court next found that Dr. Jafek had no reliable scientific foundation for his opinion that zinc gluconate can cause permanent anosmia. First, the court rejected Dr. Jafek’s attempt to characterize his case series as an “epidemiological study,” as it did not employ control groups or other methods of epidemiology. Id. at 1326. The case series, as well as other uncontrolled anecdotal data and adverse event reports, failed to support his opinion. Id. at 1326-27. Dr. Jafek had impermissibly failed to factor the background rate of anosmia into his analysis. Id. at 1327. The polio studies relating the early experience with zinc sulfate were not proof of Zicam toxicity because there was no support for Dr. Jafek’s chemical analogy – his “reasoning ignores differences in chemical structure between [zinc sulfate and zinc gluconate]” and the radically different method of application. Id. at 1327-28, 1330. Dr. Jafek’s reliance on animal studies involving administration of zinc sulfate bore the same fl aw, as well as the unsupported species extrapolation. Id. at 1328, 1330. And even the premise that zinc sulfate is capable of causing permanent anosmia was based on “uncontrolled, anecdotal reports [that] do not meet the reliability standards of Daubert.” Id. at 1329-30.

The court also found that Dr. Jafek’s dose-response analysis relied on an unexplained animal-to-human extrapolation methodology and the unsupported assumption that the entire metered dose of Zicam would reach the sheltered smell tissue. Id. Application of the Daubert factors to all of the evidence proffered on general causation led the court to conclude that Dr. Jafek’s opinion did not meet the level of scientifi c rigor required by Rule 702. Id. at 1330-31.

Finally, the court also excluded Dr. Jafek’s specifi c causation opinion. In addition to the failure to demonstrate the prerequisite general causation, the court found Dr. Jafek’s heavy reliance on temporal relationship and a nonspecifi c burning sensation experienced by the plaintiff insuffi cient, and his dismissal of the cold and other potential causes inadequate. Id. at 1331-32. Moreover, Dr. Jafek had never adequately demonstrated how Zicam could have reached plaintiff’s smell tissue, given her manner of use. Id. at 1332.

Conclusion

Though they buttress many important points in the evergrowing area of Daubert jurisprudence, Benkwith and Sutherland, as well as the other Zicam cases, also illustrate the following, sometimes critical, lessons:

1. The qualifi cations inquiry is rheostatic, not binary. A witness may have suffi cient credentials to qualify as an expert, but lack the specifi c background and expertise necessary to reliably apply the scientifi c method and meet the level of rigor characteristically employed in the specifi c fi eld of inquiry. A loose “fi t” between the expert’s background and the applicable methodology may undermine the reliability of his opinions. Specifi cally, a clinician seeking to testify to general causation in a toxic tort case may be excluded if he fails to properly understand and apply the principles and methods of toxicology.

2. The general causation inquiry must be properly framed to have real evidentiary meaning and “fi t.” The question is not whether the product contains an ingredient which may cause the endpoint at issue in the abstract; it is whether use of the product, as applied by the plaintiff, is capable of causing the endpoint. Only if the type of exposure experienced by the plaintiff is shown to cause the harm can the plaintiff’s exposure properly be ruled in as a potential cause of the plaintiff’s harm – the prerequisite foundation for any subsequent analysis of specifi c causation.

3. On a practical level, this series of opinions reaffi rms the value of searching deposition exploration of the timing, predicates and nuances of the expert’s opinions. The favorable outcomes would not have been possible without compelling testimony revealing that Dr. Jafek had reached his conclusion fi rst, based on unsupported extrapolations, analogies, and assumptions, and then performed result-oriented testing thereafter, in the course of litigation. Similarly, a detailed record demonstrating the resort to several unscientifi c analytical shortcuts was a key factor in justifying the expert’s conclusion under Rule 702.