Following up on Michael's point, we have new evidence of the positive difference Vioxx was making on healthcare outcomes. Dr. Gurkipal Singh, previously a harsh critic of the FDA and Vioxx, released a new study to the American College of Rheumatology annual meeting, finding a 21% increase in the gastrointestinal events Vioxx was designed to avoid:

In 2004, the last year in which all three cox-2 drugs were still on the market, the incidence of serious ulcerations hit an all-time low of 357 cases per 100,000 NSAID prescriptions filled.

But a year later -- after the Vioxx/Bextra withdrawals and attendant controversy -- those levels had already climbed to 434/100,000, a 21 percent increase, the study found.

Experts said they aren't shocked by the findings.

"With decreasing use of gastroprotective therapies, it comes as no surprise to see a substantial increase in NSAID-related GI adverse events," said Dr. Mark Fendrick, a professor of internal medicine at the University of Michigan School of Medicine. He has followed the cox-2 saga closely, co-authoring a commentary on the drugs' risks and benefits for The Lancet earlier this year.

According to Fendrick, it's unclear whether the cardiovascular health gains achieved by removing Vioxx and Bextra from the market have now been outweighed by a surge in serious gastrointestinal events.

"What's always bothered me as a general internist is that every study that you ever read on this either looks at the GI side or the cardiovascular side -- never both," he said. A study that compared both aspects in the same population might provide real guidance, he said.

"Then I might be able to say 'Oh, if we saved 1,000 lives from heart disease by increasing GI adverse events by X percent, then maybe we are doing the right thing,'" Fendrick said. So far, no such study has been done, he added.