Subscribe Subscribe   Find us on Twitter Follow POL on Twitter  



Lying with Statistics - Ernst v. Merck edition

In trials corporate defendants have to fight fair, or even fairer than fair, because of the risk of gigantic sanctions if they are even perceived to have crossed the line, as recently happened to Morgan Stanley (OL May 18). This puts them at a disadvantage in courtrooms where plaintiffs' attorneys suffer from no such constraints. The coroner's report indicated that Roger Ernst died of an arrhythmia rather than a blood clot; Ernst's attorneys didn't disclose that they intended to call the coroner as a witness or that they had any evidence that the coroner's report was anything other than correct and complete; so Merck attorneys assumed that they could rely upon a coroner's report at trial. Instead, the coroner is being called as a surprise witness, and, if allowed to testify Monday (as the trial judge has indicated he will do), Dr. Maria Araneta will clarify her report to speculate that Ernst really did die of a heart attack, though she admits that she has no recollection of her original report. Meanwhile, the jury will remember Merck's opening statements (OL Jul. 15) that relied on that report. And rest assured no judge would allow a corporate defendant to introduce an undisclosed star witness at the last second.

The New York Times uncritically repeats Lanier's sound-bite that he's just interested in getting the truth in front of the jury. Then why wait until the last minute to announce Araneta as a witness, and risk that she won't be allowed to testify or risk that the case would be thrown out on the strength of Araneta's original report? The question answers itself: there's a substantial chance the plaintiffs are hiding something that they didn't want Merck to have the opportunity to discover. This would imply a motive completely the opposite of wanting to get the truth in front of the jury.

As does another tactic: Plaintiffs' lawyer Mark Lanier has repeated to the jury numerous times a statement from the Merck Manual of Medical Information that "Abnormal heart rhythms (arrhythmias) occur in more than 90% of people who have had a heart attack." The press has repeated it uncritically. Dr. David Egilman, who somehow has become a cardiological expert after years of testifying about occupational lung disease, told a jury under oath, according to one account, that this demonstrates that Ernst must have died of a heart attack that was caused by Vioxx. (Merck requested to enter into evidence a decision by a Colorado state judge barring Egilman from testifying because he was "not objective, reliable, or credible"; only the AP notes this, but noone says the result of the motion. In the earlier case, Egilman claimed the defense firm "bought [the] Colorado judge," argued that the law firm illegally obtained the evidence they used against him, and sued in Texas after federal and state prosecutors declined to intervene.)

But the "90%" statistic is irrelevant to Ernst's case; that Lanier is being allowed to prejudice a jury with it demonstrates only how unfair this trial is.

Consider the following thought-exercise: Imagine a scenario where attorney Roscoe W. Chandler declares that African explorer Hugo T. Spaulding must have been attacked by a hyena because Spaulding was attacked by an animal with four legs, and, after all, over 90% of hyenas have four legs. One can immediately see that this is absurd. The question is not what percentage of hyenas have four legs, but what percentage of four-legged animal attacks are by hyenas: Spaulding could have been attacked by an elephant, a rhinocerous, a giraffe, a lion, a tiger, or even a dog.

Yet Lanier and Egilman are telling a jury that 90% of heart attacks involve arrhythmias, rather than what percentage of fatal arrhythmias involve heart attacks. The question why they feel the need to use a misleading statistic, rather than a truthful one, also answers itself. But why is the media going along with it? (Alex Berenson, "Judge Denies Merck Request at Vioxx Trial", NY Times, Jul. 29; Richard Stewart, "Surprise witness OK'd in Vioxx trial", Houston Chronicle, Jul. 29; Alex Berenson, "In Vioxx Trial, Battle Nears Over Coroner's Testimony", NY Times, Jul. 28; Kristen Hays, AP/Miami Herald, Jul. 22; Steven Stycos, "Web trap misfires for workers' advocate", Providence Phoenix, Jan. 16, 2003; see also my Jul. 11 disclaimer). Jim Copland also discussed the case Jul. 18.



Rafael Mangual
Project Manager,
Legal Policy

Manhattan Institute


Published by the Manhattan Institute

The Manhattan Insitute's Center for Legal Policy.