Vioxx Cardiovascular Concealment (Merck, 1999–2004)
Introduction
Vioxx (rofecoxib) was a COX-2 inhibitor painkiller marketed by Merck beginning in May 1999. It was heavily promoted as a safer alternative to traditional NSAIDs, with peak annual sales of $2.5 billion and approximately 80 million patients prescribed worldwide before its withdrawal. On 30 September 2004, Merck voluntarily withdrew Vioxx after the APPROVe clinical trial confirmed that the drug doubled the risk of serious cardiovascular events, including heart attacks and strokes, in patients taking it for more than 18 months.
The central finding of the Vioxx scandal — confirmed by litigation discovery, Senate hearings, and FDA analysis — is that Merck had internal evidence of cardiovascular risk years before the withdrawal, and made decisions designed to delay and minimise public recognition of that risk.
The VIGOR Trial and the Cardiovascular Signal
In 2000, Merck published results of the VIGOR trial (Vioxx Gastrointestinal Outcomes Research) in the New England Journal of Medicine. The trial was designed to show Vioxx''s gastrointestinal advantage over naproxen. It did — but it also showed a five-fold increase in heart attacks in the Vioxx arm versus the naproxen arm.
Merck''s response was to advance the "naproxen hypothesis" — the claim that naproxen was cardioprotective (like aspirin), not that Vioxx was harmful. Internal documents produced in litigation showed that Merck executives had specifically considered the cardiovascular findings and debated how to frame them. An internal email from a senior Merck executive referenced a strategy to "dodge" questions about cardiovascular risk.
The FDA''s own review of VIGOR in 2001 concluded that the cardiovascular data were more consistent with Vioxx increasing risk than with naproxen being protective. The FDA requested label changes. Merck negotiated those changes for more than a year.
APPROVe and Withdrawal
The APPROVe trial (Adenomatous Polyp Prevention on Vioxx), designed to study Vioxx as a cancer preventive, was halted in September 2004 when interim data showed doubled cardiovascular risk beyond 18 months of use. Merck withdrew Vioxx globally on 30 September 2004 — the same day the APPROVe data were made public.
The FDA''s post-market analysis, published in 2005, estimated between 27,000 and 55,000 excess cardiovascular deaths attributable to Vioxx during its time on the market. This estimate has been the subject of methodological debate but has not been superseded by a credible lower estimate.
Litigation and Settlement
Multidistrict Litigation 1657 consolidated thousands of cases against Merck in federal court. In November 2007, Merck agreed to a $4.85 billion settlement covering approximately 47,000 personal injury claimants. The settlement was notable for requiring individual case-by-case evaluation rather than a class settlement, a structure Merck negotiated to limit precedent.
Separate criminal proceedings resulted in Merck pleading guilty to a misdemeanour charge of illegally marketing Vioxx for uses not approved by the FDA, with an additional $321 million criminal fine in 2011.
Senate Hearings and Internal Documents
The Senate Finance Committee conducted hearings in 2004–2005 examining Merck''s conduct and the FDA''s regulatory response. Testimony included internal Merck documents showing executives had knowledge of the cardiovascular signal and made marketing decisions inconsistent with transparent risk disclosure. FDA reviewer David Graham testified that the agency had failed to act adequately on the safety signal.
Verdict
Confirmed. Internal documents, litigation discovery, Senate testimony, and the FDA''s own analysis confirm that Merck had evidence of cardiovascular risk from Vioxx before the drug''s withdrawal and made decisions that delayed public recognition of that risk for at least four years. The $4.85 billion settlement and criminal guilty plea are matters of public record. The estimated 27,000–55,000 excess deaths represent one of the largest drug-safety failures in FDA history.
Evidence Filters11
VIGOR trial (2000): five-fold increase in heart attacks
SupportingStrongThe Vioxx Gastrointestinal Outcomes Research trial published in 2000 found a five-fold higher rate of myocardial infarction in Vioxx patients versus naproxen. This was the primary cardiovascular signal Merck chose to explain via the naproxen hypothesis rather than Vioxx harm.
Internal Merck emails referencing strategy to "dodge" CV questions
SupportingStrongDocuments produced in MDL 1657 litigation included internal Merck communications discussing how to handle the cardiovascular data from VIGOR and how to respond to physician questions about heart attack risk without triggering regulatory alarm.
APPROVe trial (2004): doubled cardiovascular risk confirmed
SupportingStrongThe APPROVe cancer-prevention trial was halted in September 2004 after interim data confirmed a doubling of serious cardiovascular events in Vioxx patients versus placebo beyond 18 months of use. Merck withdrew the drug the same day.
FDA estimated 27,000–55,000 excess cardiovascular deaths
SupportingStrongFDA analyst David Graham estimated in 2004 Senate testimony that Vioxx caused between 27,000 and 55,000 excess cardiovascular deaths during the five years it was on the market. This estimate has been debated methodologically but not superseded by a credible lower figure.
MDL 1657 settlement: $4.85 billion (2007)
SupportingStrongMerck settled approximately 47,000 personal injury claims in federal multidistrict litigation for $4.85 billion — at the time one of the largest drug liability settlements in US history.
Criminal misdemeanour guilty plea and $321M fine (2011)
SupportingStrongMerck pleaded guilty to a federal misdemeanour for illegally marketing Vioxx for uses not approved by the FDA and paid $321 million in criminal fines. The guilty plea is a matter of public court record.
Merck's naproxen hypothesis was not scientifically validated
DebunkingStrongMerck's explanation that the VIGOR cardiovascular differential was caused by naproxen's aspirin-like protection was not supported by the available pharmacological evidence. The FDA's own 2001 review concluded the data more strongly supported Vioxx harm than naproxen protection.
Vioxx did provide genuine GI safety benefit
DebunkingThe VIGOR trial's primary endpoint — reduced gastrointestinal bleeding versus naproxen — was real. Vioxx did reduce GI events compared with traditional NSAIDs, which was a legitimate clinical benefit. This does not excuse the cardiovascular risk concealment but is relevant context.
Rebuttal
Real GI benefit does not justify concealment of cardiovascular harm. Patients and physicians were not given balanced information needed to make informed prescribing decisions.
Internal Merck Emails Showed Awareness of Cardiovascular Signal
SupportingStrongDocuments produced in litigation revealed that Merck scientists had observed a statistically elevated rate of cardiovascular events in internal studies as early as 2000. Emails showed senior researchers discussing how to present the data in the VIGOR trial publication in ways that minimized the cardiac signal relative to gastrointestinal benefits, a framing the New England Journal of Medicine later characterized as selective reporting.
NEJM Issued Expression of Concern, Not Retraction
NeutralIn December 2005, the New England Journal of Medicine issued a formal Expression of Concern about the original 2000 VIGOR trial publication, stating that three additional myocardial infarctions had been omitted from the data submitted to the journal. The journal stopped short of retraction, concluding the errors resulted from a data-submission cutoff rather than outright fraud — a distinction critics found insufficient.
Show 1 more evidence point
FDA's Own Review Found No Evidence of Active Fraud
DebunkingAn internal FDA analysis completed in 2004 concluded that while the agency's approval process could have been more rigorous, available evidence did not support a finding that Merck had deliberately concealed safety data from regulators. The FDA's safety officer David Graham disagreed publicly, arguing the agency had been too deferential to Merck's submissions, creating an institutional rather than criminal failure.
Evidence Cited by Believers7
VIGOR trial (2000): five-fold increase in heart attacks
SupportingStrongThe Vioxx Gastrointestinal Outcomes Research trial published in 2000 found a five-fold higher rate of myocardial infarction in Vioxx patients versus naproxen. This was the primary cardiovascular signal Merck chose to explain via the naproxen hypothesis rather than Vioxx harm.
Internal Merck emails referencing strategy to "dodge" CV questions
SupportingStrongDocuments produced in MDL 1657 litigation included internal Merck communications discussing how to handle the cardiovascular data from VIGOR and how to respond to physician questions about heart attack risk without triggering regulatory alarm.
APPROVe trial (2004): doubled cardiovascular risk confirmed
SupportingStrongThe APPROVe cancer-prevention trial was halted in September 2004 after interim data confirmed a doubling of serious cardiovascular events in Vioxx patients versus placebo beyond 18 months of use. Merck withdrew the drug the same day.
FDA estimated 27,000–55,000 excess cardiovascular deaths
SupportingStrongFDA analyst David Graham estimated in 2004 Senate testimony that Vioxx caused between 27,000 and 55,000 excess cardiovascular deaths during the five years it was on the market. This estimate has been debated methodologically but not superseded by a credible lower figure.
MDL 1657 settlement: $4.85 billion (2007)
SupportingStrongMerck settled approximately 47,000 personal injury claims in federal multidistrict litigation for $4.85 billion — at the time one of the largest drug liability settlements in US history.
Criminal misdemeanour guilty plea and $321M fine (2011)
SupportingStrongMerck pleaded guilty to a federal misdemeanour for illegally marketing Vioxx for uses not approved by the FDA and paid $321 million in criminal fines. The guilty plea is a matter of public court record.
Internal Merck Emails Showed Awareness of Cardiovascular Signal
SupportingStrongDocuments produced in litigation revealed that Merck scientists had observed a statistically elevated rate of cardiovascular events in internal studies as early as 2000. Emails showed senior researchers discussing how to present the data in the VIGOR trial publication in ways that minimized the cardiac signal relative to gastrointestinal benefits, a framing the New England Journal of Medicine later characterized as selective reporting.
Counter-Evidence3
Merck's naproxen hypothesis was not scientifically validated
DebunkingStrongMerck's explanation that the VIGOR cardiovascular differential was caused by naproxen's aspirin-like protection was not supported by the available pharmacological evidence. The FDA's own 2001 review concluded the data more strongly supported Vioxx harm than naproxen protection.
Vioxx did provide genuine GI safety benefit
DebunkingThe VIGOR trial's primary endpoint — reduced gastrointestinal bleeding versus naproxen — was real. Vioxx did reduce GI events compared with traditional NSAIDs, which was a legitimate clinical benefit. This does not excuse the cardiovascular risk concealment but is relevant context.
Rebuttal
Real GI benefit does not justify concealment of cardiovascular harm. Patients and physicians were not given balanced information needed to make informed prescribing decisions.
FDA's Own Review Found No Evidence of Active Fraud
DebunkingAn internal FDA analysis completed in 2004 concluded that while the agency's approval process could have been more rigorous, available evidence did not support a finding that Merck had deliberately concealed safety data from regulators. The FDA's safety officer David Graham disagreed publicly, arguing the agency had been too deferential to Merck's submissions, creating an institutional rather than criminal failure.
Neutral / Ambiguous1
NEJM Issued Expression of Concern, Not Retraction
NeutralIn December 2005, the New England Journal of Medicine issued a formal Expression of Concern about the original 2000 VIGOR trial publication, stating that three additional myocardial infarctions had been omitted from the data submitted to the journal. The journal stopped short of retraction, concluding the errors resulted from a data-submission cutoff rather than outright fraud — a distinction critics found insufficient.
Timeline
VIGOR published: five-fold MI increase downplayed as naproxen benefit
The VIGOR trial results, published in the NEJM, show five times more heart attacks in Vioxx patients. Merck advances the "naproxen is cardioprotective" hypothesis. The FDA's own review concludes the data more likely reflect Vioxx harm. Label negotiation begins and continues for over a year.
Source →Merck voluntarily withdraws Vioxx from global markets
After the APPROVe trial was halted early due to a statistically significant increase in heart attacks and strokes in patients taking rofecoxib, Merck announced a worldwide withdrawal within 24 hours — the largest drug recall in U.S. history at that time.
APPROVe halted; Merck withdraws Vioxx globally
Interim data from the APPROVe trial confirm doubled cardiovascular risk beyond 18 months. Merck withdraws Vioxx globally on the same day. The simultaneous withdrawal and data release limit the pre-withdrawal regulatory window for action.
Senate hearings: FDA reviewer Graham estimates 27,000–55,000 excess deaths
FDA safety reviewer David Graham testifies before the Senate Finance Committee, estimating 27,000 to 55,000 excess cardiovascular deaths attributable to Vioxx during its market life. Graham states the FDA failed to act adequately on the safety signal.
Source →
Verdict
Internal documents and litigation discovery confirm Merck had evidence of Vioxx cardiovascular risk from the VIGOR trial (2000) and delayed public disclosure for four years. APPROVe (2004) confirmed doubled CV risk; Merck withdrew the drug globally. FDA estimated 27,000–55,000 excess cardiovascular deaths. MDL 1657 settled for $4.85 billion (2007); criminal misdemeanour guilty plea and $321M fine (2011). Senate Finance Committee hearings documented internal decision-making to minimise the cardiovascular signal.
Frequently Asked Questions
Did Merck know about Vioxx's cardiovascular risk before the withdrawal?
Yes. Internal Merck documents produced in MDL 1657 litigation, and the Senate Finance Committee record, confirm that Merck had internal evidence of cardiovascular risk from the VIGOR trial (2000) and chose to explain it as a naproxen benefit rather than a Vioxx harm. The APPROVe trial (2004) confirmed the risk and Merck withdrew the drug the same day the data became available.
How many deaths were caused by Vioxx?
FDA safety reviewer David Graham estimated 27,000 to 55,000 excess cardiovascular deaths attributable to Vioxx during its five years on the market. This estimate has been debated methodologically — some analyses suggest lower numbers — but no credible analysis has concluded that the drug caused zero excess deaths.
What was the naproxen hypothesis and why was it rejected?
Merck argued that the higher heart attack rate in VIGOR's Vioxx arm was because naproxen (the comparator drug) had aspirin-like heart-protective properties — making Vioxx look worse by comparison. The FDA's own 2001 review concluded the available evidence more strongly supported Vioxx causing harm than naproxen providing protection. The APPROVe placebo-controlled trial later confirmed Vioxx harm directly.
What happened to Merck legally?
Merck paid $4.85 billion in November 2007 to settle approximately 47,000 personal injury claims in federal MDL 1657. In 2011, Merck pleaded guilty to a federal misdemeanour for illegally marketing Vioxx for unapproved uses and paid $321 million in criminal fines. No individual Merck executives were criminally charged.
Sources
Show 6 more sources
Further Reading
- articleThe Vioxx legacy: lessons for drug safety — Richard Horton (2004)
- bookOverdosed America: The Broken Promise of American Medicine — John Abramson MD (2004)
- paperAPPROVe trial — Cardiovascular Events Associated with Rofecoxib — Bresalier et al. (2005)
- paperSenate Finance Committee: Merck and the FDA — US Senate Finance Committee (2005)