The latest issue of Journal of the American Medical Association publishes two pieces by plaintiffs'-side experts (including the infamous Dr. David Egilman (e.g., Oct. 2007; Dec. 2005; Jul. 2005)) slamming Merck over Vioxx studies. Ross/Hill/Egilman/Krumholz accuse scientists of ghostwriting studies for Merck, but as Merck and Reuters note, Egilman et al. are smearing dozens of scientists without factual basis. As one tells the Washington Post:
Steven H. Ferris, a neuropsychologist at New York University School of Medicine who is listed as the second author, said it is not true. He said he was on a committee that reviewed the clinical records of 195 patients who received diagnoses of Alzheimer's during the study -- a task that took many hours, and for which he was paid.
"I was directly involved in a very substantive way," he said this week. "If an entity like JAMA is going to make an effort to expose a problem, they better get their facts straight. It is harmful if they don't."
(But see Ed Silverman's take.)
The more serious claim in Psaty/Kronmal—made unsuccessfully years ago in the Humeston Vioxx trial—that Merck did not provide complete mortality data (which includes deaths from car accidents and many other fatalities utterly unrelated to drug ingestion) to the FDA in the Alzheimers' study is simply false: the FDA had the complete mortality data (see p. 27 and p. 124), and concluded "The difference between rofecoxib and placebo in on-drug mortality did not reflect any increases in particular types of events to suggest causality in the Alzheimer's studies."
How unbalanced and unfair is the Psaty/Kronmal piece? The entire piece is about Protocols 078 and 091 (the "Alzheimer's trials"), but nowhere does it note that the very same Alzheimer's trials data found that Vioxx was less likely than placebo to result in cardiovascular events. (Indeed, it was the results of Study 078 and 091 that led Merck (and the FDA) to mistakenly believe that the adverse cardiovascular results in VIGOR were the combination of the cardioprotective effects of naproxen and random chance. See Martin Report 58-60.) There was no scientific reason to believe that Vioxx was causing fatal heart attacks (or car accidents) while simultaneously preventing heart attacks. This sort of for-hire hit-piece is appalling enough when presented in a court of law masquerading as scientific evidence; it's utterly shameful that a political piece has the endorsement of a medical journal.