Who is Responsible for Notifying Patients?
This editorial by Sidney Wolfe, M.D. appeared in the December 2002 edition of Pharmacoepidemiology and Drug Safety.
Schiff and colleagues are to be commended for stepping in and notifying those patients most at risk of cerivastatin (Baycol)-induced rhabdomyolysis after the drug was voluntarily withdrawn from the U.S. because of 416 cases of rhabdomyolysis including 31 deaths associated with its use. Their finding that 40% of the patients notified had muscle-related symptoms consistent with myopathy and that 5/8 symptomatic patients monitored had significantly elevated creatine kinase levels underscores the public health importance of their effort.
However, their study raises an important question concerning the responsibility of the FDA, the pharmaceutical industry, pharmacists and physicians to notify patients when a drug is withdrawn from the market.
Since virtually all drug safety withdrawals are done voluntarily by the companies making the drugs and there are no federal laws or regulations governing this process, FDA’s involvement is at best a passive oversight function as to the completeness and promptness of the product withdrawals. In addition, such drug safety withdrawals are posted on the agency’s web site (FDA.gov) but most patients or physicians are unlikely to regularly use the site.
The Pharmaceutical Industry
The pharmaceutical industry, in the absence of the FDA having mandatory drug recall authority, sets the pace and scope of the safety withdrawal. The pace is quite variable and the scope may or may not involve immediately removing withdrawn drug products from the shelves of retail pharmacies. From a product liability perspective, it might be assumed that the company’s best interests would be served by making sure no patients used the drug once the safety withdrawal was decided on. To accomplish this, the company could fund pharmacists to notify those patients who were using the drug, identified in the computerized pharmacy records now used by most pharmacists in this country. To our knowledge, this rarely if ever happens, confirming the idea that, with few exceptions, the pharmaceutical industry pays out only a small fraction of their profits for a given drug in the form of losses in litigation brought by people injured by the drug or families of those killed by the drug. Thus, companies are apparently willing to take their chances by not funding patient notification.
The Model State Pharmacy Act, of the National Association of Boards of Pharmacy (NABP), states, in sections addressing drug recalls, that “The Pharmacist-in-Charge shall develop and implement a procedure for proper management of Drug recalls which may include, where appropriate, contacting patients to whom the recalled Drug product(s) have been Dispensed.” These procedures are intended to “be adequate to deal with recalls and withdrawals due to: any action initiated at the request of the FDA… any volunteer action by the Manufacturer to remove defective or potentially defective Drugs from the market….” According to a summary of state pharmacy practice acts, 48 of 51 acts contain language about establishing such procedures for recalls and 34 of these states also include language concerning withdrawals of drugs, most of which are for safety reasons. Thus, in most states, pharmacists would not be complying with acceptable standards of practice were they not to notify patients in the event of a drug safety withdrawal. According to the Executive Director of NABP, no action by a State Pharmacy board has been brought for violation of this aspect of the state pharmacy practice act. 3 Unlike the widely-publicized situations in which patients’ pharmacy records were used, in efforts paid for by drug companies, to get them or their physicians to switch to another drug, this use of pharmacy patient records is both ethical and an important public health effort.
According to one of the study authors, when the withdrawal of cerivastatin was announced, he sought information from two large pharmacy chains on their patient notification policies. At OSCO, a supervisory pharmacist told him that the chain was notifying all patients using the drug about the withdrawal but at Walgreens, he was told that the chain thought there was no need to notify patients.
Unlike pharmacists, most physicians do not have easy access to a list of those patients for whom they have prescribed a drug. For those that do, notification should be part of their responsibility to their patients. In a large outpatient clinic such as Cook County, it is more likely that such records can be accessed through the pharmacy database. Because of limited resources, the notification efforts at that medical facility were limited to the 67 patients simultaneously using cerivastatin and gemfibrozil but the more than 10,000 patients who had been prescribed cerivastatin without gemfibrozil were not notified. Since approximately one half of the cases of rhabdomyolysis in users of cerivastatin reported to the FDA were not simultaneously using a fibrate such as gemfibrozil, the notification of these additional patients would have been important.
A further issue raised by the study by Schiff, et al, involves the routine informing and monitoring of all patients on statins concerning early evidence of rhabdomyolyis. Given that 40% of notified patients had symptoms consistent with myopathy and that five out of eight such patients had significant elevations of creatine phosphokinase (CK), routine monitoring for CK should be done for all patients using statins. In addition, FDA should mandate government-approved MedGuides, patient information leaflets handed out at the time a prescription is filled, which inform patients of early symptoms of myopathy. Patients should be instructed, ideally by their physicians and, for double protection, in the MedGuide, that at the first sign of muscle pain, muscle tenderness, muscle weakness, tiredness or darkened urine, to stop taking any statin drugs to reduce the likelihood that further muscle damage might occur. Routine monitoring of CK and contacting those patients with elevated levels should be included in this rhabdomyolysis-prevention strategy.
Remedies for the Lack of Patient Notification
Ultimately, the cost for such notification should be borne by the pharmaceutical company making the dangerous drug. Pharmacists and/or physicians, if the latter have record access, should be reimbursed for doing the company’s work. In addition, the FDA should be given new legislative authority to impose mandatory drug recalls with an enforceable withdrawal schedule governing the level and rapidity of recall and patient notification. Currently, the agency lacks any authority to impose civil monetary penalties on pharmaceutical companies for any violation (or even repeat violations) of FDA laws or regulations and this new statutory authority is also sorely needed along with the authority over recalls.
Why were so many Cook County patients taking Baycol?
About 10 months before the August 2001 withdrawal of cerivastatin, Bayer underbid drug companies that had been previously supplying statins for Cook County. The bid had been won in the fall of 2000 and the switch began in November of that year. It is of interest that the three previous statins which had been used there were first lovastatin, then simvistatin, then pravastatin. These three earlier drugs have all been found in clinical trials to not only lower cholesterol but also to reduce the risk of mortality and/or major coronary events whereas for cerivastatin, the clinical benefits are unproven.
 Schiff GD, Keehr LM, Sai TT, Bult J. High rates of patient unawareness of withdrawn lipid lowering drug combination in a public hospital clinic. Pharmacoepidemiology and Drug Safety 2002 xxxxxxxxx.
 Personal communication, Carmen Catizone, Executive Director, NABP,
 Personal communication, Gordon Schiff, MD,
 Public Citizen Health Research Group petition to the FDA for box warning concerning rhabdomyolysis on all statin drugs.
 Furberg CD, Pitt B. Withdrawal of cerivastatin from the world market. Curr Control Trials Cardiovasc Med. 2001;2(5):205-207.