Daubert is an important precedent requiring courts to act as gatekeepers to exclude unreliable scientific evidence. Of course, to have that effect, judges must actually provide some set of standards.
The plaintiff in Plunkett v. Merck wishes to use the expert testimony of Dr. Winston Gandy, Jr. The court found that Dr. Gandy's expert report, which, though nineteen pages, had only two and a half pages of analysis, was "wholly conclusive, rather than explanatory." Moreover, his "deposition testimony is littered with circular reasoning and instances where he is unable to answer certain questions regarding the literature and studies he said he had read." The report and deposition testimony of Dr. Thomas Baldwin suffered from similar problems. Nevertheless, Drs. Gandy and Baldwin will be allowed to testify to the jury their opinions that Vioxx caused the death of Dicky Irvin (on which, see Nov. 17). Dr. Lucchesi (Sep. 20) will also be allowed to testify on a wide range of topics.
The court correctly states that its obligation is to review methodology, rather than conclusions. But if scientists are allowed to meet the methodology hurdle by merely asserting that they've based their opinion on the relevant scientific studies, and then jumping to a conclusion, there's next to no bar to junk science whatsoever; experts-for-hire can baldly make whatever claims they want so long as they jump through the large hoop provided by the court.
Merck did benefit from some aspects of the court's ruling: one plaintiff's expert was excluded in part; Merck will be allowed to introduce the April 2005 FDA memo showing that Vioxx is no worse than other NSAIDs; the court found that "there is significant, scientifically reliable information supporting Merck's contention that naproxen is sufficiently cardioprotective to explain the VIGOR test results"; and Merck will be allowed to argue that it needed FDA approval to make the major change to the label that plaintiffs insist should have been made. While Judge Fallon is being lenient in letting in evidence and punting decision-making to the jury, he at least is not following the New Jersey court's lead in excluding scientific evidence that exonerates Merck. Meanwhile, any victory the plaintiffs win in the first federal trial is at risk of being overturned by the Fifth Circuit Court of Appeals because of the use of conclusory expert opinions.
The court's Daubert opinion is available at the November 18 entry on the AEI Liability Project Documents in the News page.