By Jesse Marx
By Chris Parker
By Jake Rossen
By Jesse Marx
By Michelle LeBow
By Alleen Brown
By Maggie LaMaack
By CP Staff
The final version of the study found that people taking drugs like Levaquin were 1.2 times (or 20 percent) more likely to rupture their Achilles tendons than people not taking such drugs. But if the flaw that the journal editors flagged was indeed present, that rate could be a gross undercount.
"In other words, the weak association of 1.2 could be stronger and point to the drug as a risk factor for Achilles tendon rupture," says DeAnn Lazovich, a professor of epidemiology and community health at the University of Minnesota, "but we cannot know for sure because the observed association is fairly weak."
There was another, more serious problem with the study results.
In research, every study is assigned a "power," which describes the study's ability to compare groups and draw significant conclusions. This is calculated by a number of factors, including sample size. If a study's power is low, then it can't properly measure what it purports to track—its results could be nothing more than random noise. The Ingenix study's power was low.
A second measure of a study's reliability is its confidence interval, a numerical range that represents the trustworthiness of the results. The confidence interval is written as two numbers, with a dash between. If the range between the numbers is small, researchers consider the study results reliable. If the range is broad, the results are less certain.
In the Johnson & Johnson study, the confidence interval was 0.9-1.7. Statistically speaking, that's a gap the size of the Grand Canyon. But more importantly, the range includes the number 1.
"Any test where the confidence interval includes 1.0 is generally considered non-significant," explains Russell Luepker, a Mayo Professor of Public Health at the U, "whether from inadequate power or the lack of a real difference."
The greatest criticisms of the study came directly from Aventis, and they had to do with the study's design, which systematically eliminated the elderly—precisely the population most at risk.
"The elderly population is under-represented," wrote Carla Canabarro, one of the doctors at Aventis. "Regarding age distribution, the Ingenix research database population does not reflect the elderly age group as per the United States 2000 census population."
The study algorithm had also eliminated tendon ruptures in old people at a higher rate than the overall group.
By January 2008, with no substantial response to the petition for a black box warning from the doctors at the Illinois Attorney General's Office, consumer watchdog advocacy group Public Citizen sued the FDA, demanding that it mandate the warning.
"These stronger warnings to doctors and patients could lead to earlier intervention to stop tendon pain from progressing to frank tendon rupture by changing to other antibiotics," the lawsuit read. "Therefore, to protect public safety and prevent needless injury, Public Citizen seeks a declaration that FDA has acted unlawfully by withholding action."
By the end of the year, the FDA required Johnson & Johnson—and all makers of antibiotics like Levaquin—to add a black box warning to its label.
Vicki Schedin was watching television one day when a commercial caught her ear. A lawyer was looking for people who'd taken Levaquin and ruptured their tendons.
"Give us a call," the television voice intoned.
Vicki rushed to tell her husband.
A few days later, Bob phoned the hotline from the TV. He reached a receptionist who was working for Lewis J. Saul, an attorney in Portland, Maine, who had been prescribed Levaquin and suffered tendon problems himself. Saul was suing Johnson & Johnson.
Last fall, Schedin, now 83, was the first plaintiff in a series of lawsuits in the massive case against the drug maker. The jury awarded him $700,000 in compensatory damages plus $1.1 million in punitive damages.
Next week, another trial begins in federal court in Minneapolis before Judge John Tunheim. It is the second of six cases slated to be tried as part of the complex litigation.
The central issue at trial is whether Johnson & Johnson adequately warned doctors about the risks of taking Levaquin. The company argues that the warning was sufficient.
"We want to make clear that ever since it was first approved by the FDA in 1996, the LEVAQUIN label, in addition to the labels for the wider class of fluoroquinolone antibiotics, has included warnings about tendonitis and tendon ruptures," William Foster, a Johnson & Johnson spokesman, told City Pages in an emailed statement. "This risk was well known to prescribers."