Featured Expert Contributor, Judicial Gatekeeping of Expert Evidence
Although often couched in gentile terms, the real concern underlying both Daubert’s core requirement of reliability and the gatekeeping role of district courts more generally is that all too often expert witnesses see their role as hired guns, offering—for a price—whatever opinions are necessary in order for their clients to prevail. The U.S. Court of Appeals for the Fourth Circuit recently issued an extensive decision politely but firmly renouncing such testimony.
In re Lipitor (Atorvastatin Calcium) Marketing, Sales Practices and Products Liability Litigation arose out of multi-district litigation in which the plaintiffs alleged that the cholesterol-lowering drug Lipitor caused them to develop diabetes. After excluding or sharply limiting the testimony of the bellwether plaintiffs’ expert witnesses, the district court granted summary judgment in favor of the defendant, Pfizer, Inc., on the ground that the plaintiffs lacked sufficient evidence of causation.
In affirming the district court’s decision, the Fourth Circuit provided a cogent primer on many recurring issues in litigation involving medical causation. The court began by emphasizing the underlying concern: “due to the difficulty of evaluating their testimony, expert witnesses have the potential to be both powerful and quite misleading. And, given the potential persuasiveness of expert testimony, proffered evidence that has a greater potential to mislead than to enlighten should be excluded” (internal quotation marks omitted). It then went on to systematically demonstrate why the district court was right to exclude the testimony of the three experts whom the plaintiffs tried to defend on appeal.
The first expert—Dr. Nicholas Jewell—was offered to establish a statistical association between Lipitor and diabetes. Dr. Jewell performed a “reanalysis” of the data from several clinical trials. But his methodology was transparently result-oriented—or, as the Fourth Circuit more politely put it, “too tainted with potential bias and error to pass Daubert muster.”
In particular, Dr. Jewell included in his sample set patients who had a single elevated blood glucose reading as well as those who had elevated levels at the beginning of the clinical trial. Including the former information conflicted with the position of the plaintiffs themselves that a single elevated reading is not a reliable indicator of diabetes. And including the latter introduced “confounding variables”—those that may cause a correlation to exist between the independent and dependent variables without causation being present. Specifically, the significant number of patients with elevated glucose levels after treatment with Lipitor may be explained by the fact that a significant number started with elevated levels and may not necessarily be because Lipitor caused them to experience elevated levels.
Another problem with this testimony was that Dr. Jewell is a statistician, not a medical doctor. As such, he “‘lacked the expertise to opine about any implications that single glucose readings might have about the possibility of new-onset diabetes’” (quoting district court). In other words, he was not qualified to opine on the validity of a pivotal assumption of his study and had not identified another reliable source to support that assumption. This holding is important, because it is common for expert witnesses to attempt to bootstrap opinions that are outside their areas of expertise either by obfuscating the distinction between areas of expertise or, as here, purporting to offer opinions within their area of expertise that rely on unsupported assumptions that fall outside that area. It is critical that district courts bar the gate to this hallmark of result-oriented expert testimony.
To make matters worse, Dr. Jewell engaged in a result-oriented analysis of the data. Specifically, after his first pass through the data showed a statistically insignificant association between Lipitor use and diabetes, Dr. Jewell “omitted that result from his initial expert report, and instead reported the results using” a different lens. The district court deemed Dr. Jewell’s choice of methodology to be “results driven,” and the Fourth Circuit agreed, explaining: “Result-driven analysis, or cherry-picking, undermines principles of the scientific method and is a quintessential example of applying methodologies (valid or otherwise) in an unreliable fashion.”
Dr. Jewell’s reanalysis of the data from another Lipitor trial fared no better. Although the study had found no statistically significant difference in diabetes onset between patients who were administered Lipitor and patients who were administered a placebo, Dr. Jewell reanalyzed the data and found that Lipitor use was associated with a significantly increased risk of new-onset diabetes. He came to this conclusion by employing a definition of new-onset diabetes that differed from the one used by the authors of the study.
Again pointing out that Dr. Jewell is not a doctor, much less an expert on diabetes, the district court held that Dr. Jewell lacked the expertise to “second guess” the doctors involved in the study and “to replace the[ir] determination of new-onset diabetes with particular unadjudicated raw data, namely lab values of his choice.” As the Fourth Circuit explained, the district court correctly “stressed the importance of using prespecified diagnostic criteria … to “help guard against bias.” “By contrast, Dr. Jewell … had the benefit of hindsight when he selected his definition.”
The Fourth Circuit acknowledged that “a reanalysis of a clinical trial’s data may sometimes be appropriate.” At the same time, it explained, “such a reanalysis isn’t per se admissible under Daubert. As with any expert undertaking, a district court must review a reanalysis to ensure it meets Daubert’s demands of reliable methodologies reliably applied.”
It concluded that “when the results of a reanalysis are squarely at odds with the conclusions of a published, peer-reviewed study; the methods of the reanalysis are questionable because of the absence of properly adjudicated data; the expert performing the reanalysis lacks expertise in the portion of the study he’s modifying; and that expert offers an unpersuasive rationale for why the original findings are wrong and his correct, then skepticism by the district court is warranted.”
Next on the chopping block was Dr. Sonal Singh, whose opinions were offered to establish general causation—i.e., that Lipitor use is causally related to increased risk of diabetes. Because the plaintiffs in the case were prescribed four different doses of Lipitor—10 mg, 20 mg, 40 mg, and 80 mg—the district court required that Dr. Singh provide a reliable methodology for determining that each of those doses is causally related to increased risk of diabetes.
The court concluded that the data were sufficient to show a relationship at the 80 mg dose, but it found that the data did not reveal a sufficiently strong association between the 10 mg dose and diabetes. It therefore excluded Dr. Singh’s opinion regarding that dosage; and because Dr. Singh admitted that his opinions regarding the 20 mg and 40 mg doses were derivative of his opinion regarding the 10 mg dose, the district court excluded those opinions as well.
The plaintiffs argued on appeal that the district court abused its discretion by scrutinizing the reliability of Dr. Singh’s opinion for each dosage. That gave the Fourth Circuit an opportunity to provide clear guidance about the role of dosage. As the Fourth Circuit bluntly put it, “dose matters” (internal quotation marks and alteration omitted). “[I]n many cases, … substances that might be quite harmful in high doses are innocuous in smaller amounts.” “Pharmaceuticals like Lipitor … lend themselves quite well to dosage analysis. Unlike substances in other toxic tort cases …, pharmaceutical drugs are typically prescribed and consumed in measured and knowable quantities.”
The court allowed that not “every case involving a claim of injury resulting from a pharmaceutical drug will require a dose-by-dose analysis, and an expert witness will not necessarily need to define the precise lower bound of exposure risk. The appropriate level of analysis will depend on the circumstances of the case and the capacity of current scientific methods.”
Having said that, however, it concluded that “where, as here, each plaintiff took one of only several commercially available doses, clinical data exist that enable an expert to perform a causation analysis at each dose, and experts (including plaintiffs’ own) acknowledge that there is some relationship between dosage and harm, the district court doesn’t abuse its discretion in asking the expert to produce a dose-by-dose analysis.” Indeed, in such circumstances, it is hard to see how a district court could fail to require such an analysis without abusing its discretion.
The plaintiffs also challenged the district court’s holding that it was improper for Dr. Singh to apply the well-known Bradford Hill factors to determine causation without first demonstrating the existence of a statistically significant association between Lipitor use and diabetes. Instead of first identifying a statistically significant association, Dr. Singh relied on “trends” in the data that admittedly fell below the normal threshold for statistical significance.
The district court had reasoned that Dr. Singh’s opinion was unreliable because the plaintiffs had “failed to demonstrate that Dr. Singh’s reliance on non-statistically significant ‘trends’ is accepted in his field, that non-statistically significant findings have served as the basis for any epidemiologist’s causation opinion in peer-reviewed literature, or that standards exist for controlling the technique’s operation.” The Fourth Circuit found that conclusion to be “well within the district court’s discretion.”
That left plaintiffs’ specific-causation expert, Dr. Elizabeth Murphy, in the case of bellwether plaintiff Jaunita Hempstead. Her testimony too was excluded because it was transparently result-oriented.
As the Fourth Circuit described it, the district court “expressed skepticism” about Dr. Murphy’s analysis because Dr. Murphy “‘could not identify any organizations or peer-reviewed texts that contain this methodology,’ nor any colleagues who used the methodology to determine the cause of diabetes. Additionally, Dr. Murphy had not personally used the methodology to determine the cause of her own patients’ diabetes and had never diagnosed one of her patients as suffering from statin-induced diabetes.”
Of particular concern was Dr. Murphy’s result-oriented differential diagnosis. Performed properly, a differential diagnosis (or, more aptly, a differential etiology) can be a valid methodology for determining specific causation. It entails “determining the possible causes for the patient’s symptoms and then eliminating each of these potential causes until reaching one that cannot be ruled out or determining which of those that cannot be excluded is the most likely” (internal quotation marks omitted).
But while Dr. Murphy purported to apply this technique, in fact she stacked the deck. The Fourth Circuit explained that although “Dr. Murphy did consider (and purportedly ruled out) several other risk factors, including Ms. Hempstead’s family history, race, body mass index (BMI), and age,” “her analysis of those factors—and, more importantly, her reasons for rejecting them as the likely cause of Ms. Hempstead’s disease—fell short.” As the Fourth Circuit colorfully put it, “[a] rose by another name may smell as sweet—but simply calling an analysis a differential diagnosis doesn’t make it so.”
The district court observed that Dr. Murphy identified several factors for which the risk “greatly exceed[ed] the risk of developing diabetes associated with Lipitor.” Yet she nevertheless concluded that Lipitor use was a substantial cause of Ms. Hempstead’s diabetes. The court concluded that “[t]he powerful evidence that Plaintiff’s many other risk factors can independently cause diabetes and cannot be ruled out … undermine Dr. Murphy’s testimony. The gap between the available scientific evidence and Dr. Murphy’s opinions [is] too great to survive a Rule 702 review.”
Although Dr. Murphy had dispensed with certain risk factors—such as age and BMI—by invoking studies that ostensibly show that Lipitor use increases the risk of diabetes even after adjusting for those factors, the Fourth Circuit called her to task for conflating general and specific causation. As the court explained, “these explanations (even if true) do not accomplish the specific causation expert’s task: accounting for the development of the disease in a particular plaintiff. That Lipitor may cause an increased risk of diabetes notwithstanding certain other risk factors is insufficient to conclude that the drug was a substantial contributing factor in an individual patient. To hold otherwise would obviate the need for any specific causation evidence at all.”
The appeals court explained that the district court “found that Dr. Murphy’s report appeared to dismiss other possible causes in favor of Lipitor in a cursory fashion that appeared closer to an ipse dixit than a reasoned scientific analysis. Dr. Murphy’s conclusions focused almost exclusively on the fact that Ms. Hempstead took the drug and later developed the disease, rather than explaining what led her to believe that it was a substantial contributing factor as compared to other possible causes. Simply put, Daubert requires more.”
As the foregoing summary reflects, the Fourth Circuit’s decision is a treasure trove of helpful insights about inappropriate, result-oriented expert testimony. This decision not only should help rein in such testimony within the Fourth Circuit, but also should serve as a model for judges throughout the country.