Court Refuses to Scrutinize the Sufficiency of Differential Diagnosis Analysis and Permits Causation Testimony in Absence of Controlled Studies
This past week, an Ohio federal court considered under Daubert the standard by which specific causation experts in drug and device cases are to be scrutinized. While the Court precluded the opinions of certain experts who did not engage in a differential diagnosis analysis before forming specific causation opinions, it stopped short of precluding one expert who engaged in what was essentially an incomplete differential diagnosis analysis, concluding the sufficiency of the analysis goes to the weight of the evidence and not its reliability under Daubert. The Court also permitted specific causation expert testimony even though the expert testified there were no controlled studies supporting general causation.
The products at issue are Novartis’ Aredia® and Zometa®. An MDL was created for litigation about these drugs and whether they cause osteonecrosis of the jaw (“ONJ”). The MDL court denied Novartis’ Daubert motion to preclude plaintiffs’ general causation experts. In re: Aredia and Zometa Prods. Liability Litigation, No. 3:06-md-1760, Docs. ##2763, 2764 (M.D. Tenn. Aug. 9, 2009). Thereafter, the lawsuit by plaintiff Bowles was transferred to federal court in Ohio for trial. In that court, Novartis challenged plaintiff’s specific causation experts.
Several of the proffered specific causation experts were treating dentists and oral surgeons. While the court permitted plaintiff’s treaters to opine as to plaintiff’s condition, it did not permit causation opinions because these treater-experts did not conduct a differential diagnosis (i.e., what are the possible causes, and what was ruled out) as to the cause of plaintiff’s injuries. Similarly, the Court precluded another expert who, while purporting to be familiar with the medical literature related to the drugs at issue, did not remember if he engaged in a differential diagnosis process. In absence of any attempted differential diagnosis analysis, the Court precluded causation opinions of these experts.
There was a non-treating, litigation expert that was challenged by Novartis, but permitted to testify. Novartis argued that the expert’s causation opinion was formed on the basis of an unreliable methodology because the expert had not ruled out certain specified potential causes of the ONJ which are a part of the differential diagnosis. The Court acknowledged that some causes were ruled out by the expert, although others were not. Nevertheless, the Court ruled that the failure to rule out causes goes to the weight, and not the admissibility of the testimony.
Novartis alternatively argued that this same expert should be precluded because he testified at his deposition that a causal relationship between the drugs’ use and ONJ has not been scientifically proved through controlled studies. Novartis argued that since one needs general causation before one can establish specific causation, this expert must be precluded. The Court rejected this argument, reasoning that because the MDL court had already denied a Daubert general causation preclusion motion, this previous ruling that there was a question of fact as to general causation was the law of the case, even though this particular expert had testified that there were no controlled studies establishing causation.
Put into context, general causation must be proved scientifically before opinions can be offered that in a specific case that that plaintiff’s injury was in fact caused by the agent at issue. The expert in Bowles that was permitted to testify had opined that no controlled studies had established causation. In many courts, that would be, and should be, the end of the story. But because the MDL court had previously determined that there was a question of fact on the general causation issue, this Ohio court let the opinions stand.
Having gotten over that hurdle, this same expert testified that there were known causes of the condition that he did not rule out. Not having ruled them out, the question begs to be asked – how can a reliable opinion be offered that the drugs in fact caused the condition? Certainly, in a real-life, non-litigation posture, treating a patient on this basis, would be asking for a medical malpractice claim. Yet this court held that the use of a reliable methodology (differential diagnosis) incompletely does not preclude the testimony, but only goes to how it will be weighed by a jury. If the differential diagnosis methodology was not done completely, then we wonder how it can be reliable and admissible.
It will be interesting to see how far the plaintiff’s bar will be permitted to go in using the ‘differential diagnosis’ process (a clinical tool for practicing physicians) to get around the necessity that both general and specific causation be scientifically proved.