I was at a social gathering Friday and met an experienced nurse from a local bone marrow transplant unit. She mentioned to me that an FDA panel had met and decided to recommend that Darvon (propoxyphene) be banned from the US market.
"Man, we use Darvon a lot," she said. "It's one of the few drugs we give for our patients that doesn't mask fever. Everyone in our unit gets Darvon. If we suppressed fever, we'd be much slower at recognizing infection in our unit."
The FDA panel made its decision based on some interesting testimony from Sidney Wolfe, MD from Public Citizen. His data were based largely on the Drug Abuse Warning Network (DAWN) that collects data from emergency rooms and medical examiners concerning drugs which have been determined to be related to emergency room visits or deaths. In that testimony, he claimed that 503 deaths were "related" to propoxyphene even though the data "do not imply causality but rather that the death was related to the drug." Note that only 25 of those deaths were classified as single drug. I wonder, was propoxyphene the causal agent in the other deaths? How many of these deaths was the drug actually prescribed to the deceased? Should death from suicide be classified the same as death from accidental poisoning?
Aron J. Hall, D.V.M., M.S.P.H., of the Centers for Disease Control and Prevention, Atlanta, and colleagues conducted a study to determine the risk characteristics and other factors associated with persons dying of unintentional pharmaceutical overdose in West Virginia in 2006. In their study, opioid analgesics were the most prevalent class of drugs, contributing to 93.2 percent of deaths; of these, only 44.4 percent included evidence of prescription documentation for all of the contributory opioids.
Now, compare this to the deaths from acetominophen (Tylenol) in whom more than 56,000 emergency room visits and an estimated 458 deaths from acute liver failure occurred in 2003. Should such similar data lead to acetominophen's removal from the US market, too?
Used properly, statistics can be powerful allies in supporting arguments during public policy decisions. But when those numbers are not couched beside their limitations or other confounding factors, they can misrepresent as well.
The implications of stopping a drug that has been available in the American medical pharmacopoeia for over fifty years should be carefully scrutinized. I can recall when isoproterenol (Isuprel) was discontinued due to the lack of a market - even though it's purely chronotropic (heart rate increasing) cardiac effects were vital to our electrophysiology studies. We quickly learned that our EP studies were markedly adversely affected. Fortunately, after significant outcry from our profession, the drug was reinstated.
As the nurse I met made apparent, some old drugs like Darvon may still have niche areas with distinct advantages and cost savings. These niche areas should be appreciated and recognized before wholesale withdrawal of any drug is mandated. Putting shoddy statistics in perspective might help assure proper public policy decisions are reached.