When Did Merck Know Vioxx Was Deadly?

Study Finds Danger Signal in Early Merck Data; Merck Says Study Flawed

Medically Reviewed by Louise Chang, MD on November 23, 2009
From the WebMD Archives

Nov. 23, 2009 - Merck should have known Vioxx was deadly years before they pulled the drug from the market, a study of Merck's own data suggests.

The study, published in the current issue of the Archives of Internal Medicine, is by Joseph S. Ross, MD, of Mount Sinai School of Medicine, and colleagues. The study authors were paid consultants to plaintiffs' lawyers in Vioxx lawsuits -- in which much of the Merck data first was revealed.

"By our analyses, the association is clear that by June 2001 -- more than three years before the drug was eventually taken off the market -- the risk could have been known," Ross tells WebMD.

Merck took Vioxx off the market in November 2004, after the "APPROVe" study conclusively demonstrated that Vioxx users had more heart attacks and strokes than patients receiving a placebo.

In their study of data from Merck-sponsored clinical trials, Ross and colleagues found that:

  • By December 2000, data from 21 trials showed that the risk of a heart- or stroke-related adverse event (a cardiovascular thromboembolic event or CBT) or death (from all causes) was twice as high in Vioxx patients -- but the finding was just shy of statistical significance, meaning it could have been a chance finding.
  • By June 2001, pooled data showed Vioxx increased the risk of a CBT adverse event or death by 35% -- a statistically significant finding, meaning it is unlikely to be a chance finding.
  • By April 2002, adding new studies to the pooled data showed Vioxx increased the risk of CBT or death by 39%, a statistically significant finding.
  • By September 2004, adding new studies to the pooled data showed Vioxx increased the risk of CBT or death by 43%, a statistically significant finding.

In a written statement and in an interview with WebMD, Merck says the study is flawed. It maintains that the company acted responsibly both in conducting safety studies and in pulling the $2 billion per year drug from the market as soon as it was aware that the drug increased the risk of heart attack and stroke.

"We fundamentally disagree with their conclusions that there was an actionable signal before September 2004," Doug Watson, PhD, senior director of medical science for Merck Research Laboratories, tells WebMD.

Watson says that when Merck looked at the data, its researchers were much more thorough than Ross and colleagues. Most importantly, Watson says, the company used independent evaluators to determine whether adverse events were truly related to heart attack or stroke, and whether deaths were due to heart attacks or strokes or to non-Vioxx-related causes such as infections or trauma.

"We disagree with their conclusions that there was a finding that should have resulted in earlier action on our part, because we disagree with the methods they use," Watson tells WebMD. "We investigated Vioxx while it was on the market ... based on the most rigorous scientific methods, vetted with the FDA, and we did not see a signal prior to the APPROVe study."

Ross defends his team's methods as the best way to look for a safety signal. He notes, for example, that if a patient on Vioxx died due to an accident, it may very well have been a heart attack that led to the accident. And he points to the fact that the heart/stroke risk of Vioxx is now a widely accepted medical fact -- and that Merck's studies failed to detect this risk until some 20 million people had taken the drug.

Ross says the real message of the study is not to point fingers at Merck. The point, he says, is that routine safety evaluations -- using all clinical data, not just published data -- should be performed on all new drugs.

"What we are saying is this kind of study should be done routinely, particularly when there is a known safety concern about a drug," Ross says. "This safety monitoring is not rocket science. We would be protecting the publics' health because we would be identifying risks earlier."

Federal law enacted since the Vioxx debacle now requires companies to disclose clinical trial data to the public within 12 to 24 months of study completion.

In late September, Merck settled more than 3,000 Vioxx lawsuits in a $4.85 billion settlement. That reversed a company policy of fighting each lawsuit one at a time, even though the company has won most of these suits.

Show Sources


Ross, J.S. Archives of Internal Medicine, Nov. 23, 2009; vol 169: pp 1976-1984.

Woloshin, S. and Schwartz, L.M. Archives of Internal Medicine, Nov. 23, 2009; vol 169: pp 1985-1987.

Hirsch, L.J. Mayo Clinic Proceedings, September 2009; vol 84: pp 811-821.

News release, Merck, Nov. 23, 2009.

Joseph S. Ross, MD, MHS, assistant professor of Geriatrics and adult development, Mount Sinai School of Medicine, N.Y.

Doug Watson, PhD, senior director of medical science, Merck Research Laboratories.

© 2009 WebMD, LLC. All rights reserved. View privacy policy and trust info