Plaintiffs’ Experts’ “Cherry-Picking” of Scientific Studies Not Permitted

On February 20, 2015, New Jersey Superior Court Judge Nelson C. Johnson issued a decision in the New Jersey Accutane Multicounty Litigation that, if not reversed on appeal, will likely result in the dismissal of several thousand cases pending against Hoffmann-LaRoche. The court granted the pharmaceutical company’s motion to bar plaintiffs’ experts’ testimony regarding Accutane, and its alleged propensity to cause Irritable Bowl Disease (“IBD”), including Crohn’s Disease (“CD”). The court made its determination after conducting an exhaustive analysis of the scientific literature and the testimony of both sides’ experts. The decision is noteworthy for the court’s painstaking assessment of the body of scientific literature on medical causation. Defendant’s counsel clearly did a superb job in educating Judge Johnson on the science.

Accutane was approved by the Food and Drug Administration in 1982 to treat a severe form of acne, which can cause disfigurement on the face and back of those affected. Crohn’s Disease is a serious chronic inflammation of the gastrointestinal tract that can cause gastrointestinal bleeding, among other symptoms, and carries an increased risk of cancer. Most scientific literature has concluded that the cause of CD and IBD remains unknown.

To establish a causal connection between Accutane and IBD and CD, plaintiffs presented the court with two expert witnesses, Arthur A. Kornbluth, M.D., and David Madigan, Ph. D, both of whom are accomplished leaders in their professions, with highly impressive credentials. Dr. Kornbluth is a Clinical Professor of Medicine at the Icahn School of Medicine at Mount Sinai University. He specializes in gastroenterology and has published over one hundred peer-reviewed articles, abstracts and textbook chapters. Dr. Madigan is a Professor of Statistics at Columbia University and Executive Vice President of Arts & Sciences and Dean of Faculty. He has published over one hundred-fifty technical papers and served as an investigator for monitoring the safety of FDA-regulated medical products.

To rebut the plaintiffs’ experts, defendants relied upon the testimony of two experts, Steven N. Goodman, M.D., M.H.S. Ph. D, and Maria Oliva-Hempker, M.D, who are also exceptionally learned and accomplished professionals. Dr. Goodman practices in the field of epidemiology. He is a Professor and Associate Dean for Clinical Research at Stanford University and has received numerous awards in his area of expertise. Dr. Oliva-Hempker is a Professor of IBD and Chief of Division of Pediatric Gastroenterology & Nutrition at the Johns Hopkins University School of Medicine. She has been published in over seventy treatises and medical texts and maintains a substantial presence in educating the public on IBD. Thus, from the standpoint of professional credentials, the two sides were well-represented and seemingly evenly matched.

Of particular significance to the court in making its analysis was The Reference Manual on Scientific Evidence (3d Ed.), issued by the Federal Judicial Center. The state court looked to the Reference Manual for guidance on the reasonableness standards of the scientific community and its evaluation of the epidemiological studies. The use of federal guidance by a New Jersey state court is noteworthy.

In determining that plaintiffs’ experts’ was unreliable Judge Johnson criticized plaintiffs’ experts for “cherry-picking the evidence.”  “The examination of the ‘lines of evidence’ by Plaintiffs’ experts was highly selective, looking no further than they wanted to — cherry-picking the evidence — in order to find support for their conclusion-driven testimony in support of a hypothesis made of disparate pieces, all at the bottom of the medical evidence history.”

The court also caught Dr. Kornbluth taking a position in a published article that was the opposite of what he argued in court. During the pre-publication peer review process, Dr. Kornbluth refused to associate Isotretinoin as a cause of IBD, although in court he had no reservations arguing that Isotretinoin caused IBD. In order to define the prodromal period of CD, plaintiffs’ relied on a single study with only 76 subjects and ignored studies with much larger patient groups. The court found that, “the unsound nature of plaintiffs’ methodology in relying upon such a study becomes readily apparent when compared with the population studies, involving hundreds of thousands of subjects presented to the court. Plaintiffs’ rationalization for ignoring the other studies is assertedly their failure to account for a lengthy prodrome of CD. Yet here again, plaintiffs’ experts engage in their finely-tuned selectivity of the evidence by disregarding eight of nine prodromal studies.” The court found further that plaintiffs’ experts demonstrated the talent of “the self-validating expert, who uses scientific terminology to present unsubstantiated personal beliefs.”

In one instance, plaintiffs’ experts interpreted the results of a study they relied on contrary to the authors’ own stated conclusions. The study’s authors concluded that the risk for IBD following Isotretinoin exposure had no statistical significance, but the experts argued to the contrary. The court commented, “It is one thing to stand alone in the world of science, advancing a hypothesis that others do not accept. It is quite another thing to advance a hypothesis that can only be supported by disregarding valid scientific research.” Even more ineffective was the plaintiffs’ reliance on a second study with a particularly small patient population to define the prodromal period of CD. The court asked rhetorically, “with regard to these two studies, one wonders whether Dr. Kornbluth or Dr. Medigan would cite them as meaningful any place but a courtroom.”  The court found it unlikely that Dr. Kornbluth or Dr. Medigan, who took such pride in their professional accomplishments, would rely on these studies as defining treatises. Excoriating plaintiffs’ counsel and their experts, the court stated that, “the reliance upon these two studies is fatal and reveals the lengths to which legal counsel and their experts are willing to contort the facts and torture the logic associated with plaintiffs’ hypothesis.”

Several observations that may be of use in other cases.

Adverse Event Reports. It is noteworthy that the court did not credit plaintiffs’ reliance on Adverse Event Reports.  Plaintiffs frequently made reference to these anecdotal reports filed with the FDA. But the FDA itself notes that there is no certainty that the reported circumstance are actually attributable to the drug the patient was taking. Causal assessments are difficult to make because these reports are, according to a scholarly text cited by the court, “usually based on suspicion, and may be preliminary, ambiguous, doubtful or wrong.” Moreover, there is concern of potential abuse of these reports, especially by attorneys. “The legal profession is a bulwark of our society, yet the courts should never underestimate the resourcefulness of some attorneys.” Out of 2,214 cases of IBD resulting from Isotretinoin, attorneys reported 87.8% cases, as compared to 6% by physicians and 5.1% by consumers.

Animal Studies. The court also criticized plaintiffs’ reliance on animal studies. Again, referencing the Reference Manual, the court concluded that animal studies have two major disadvantages. The first is the obvious variation in responses among different species of animals. “The second difficulty with inferring human causation from animal studies is that the high dose customarily used in animal studies requires consideration of the dose-response relationship and whether a threshold no-effect does exist.” The problem with the dog study in this case was that even though dogs are mammals, the dogs in the experiment were euthanized after testing, so there was no evidence of whether the dogs’ intestines suffered any permanent harm. Equally important, dogs do not suffer from IBD.

The Reference Manual on Scientific Evidence (3d. Ed.) should never be overlooked by the practitioner preparing a Daubert motion. As evidenced by this case, the Reference Manual has persuasive power in both state and federal courts.

It is defense counsel’s job to teach the science to the presiding judge. Considering the case load of most judges, this is not an easy task. In this case, Hoffmann-LaRoche was fortunate to have a judge willing to make this herculean effort.

Image courtesy of Flickr by Audrey