Featured Articles

QuarterWatch™ (Quarter 4 and 2009 totals): Reported Patient Deaths Increased by 14% in 2009

QuarterWatch™ is an ISMP program used to monitor adverse drug events (ADEs) in the US. During the most recent period of monitoring, ISMP analyzed 116,174 serious, disabling, and fatal ADEs reported to FDA during 2009. A 2.8-fold increase in reports has been observed since 2000. The volume of reports from health professionals has been relatively stable since 2004, while consumer reports have risen steadily since 2000.

Fatal ADEs

During 2009, FDA received 19,551 reports of patient deaths associated with drug therapy, a 14% increase compared to 2008 and a 3-fold increase over the past decade. At least three factors contributed to the striking increase in reported deaths:

Increased awareness. Reported deaths increased for some drugs because FDA warnings, product recall notices, new medical studies, and media publicity reached more consumers and doctors, who then reported events with these drugs.

Known risks continue unabated. Some high-alert drugs—notably, powerful opioids and acetaminophen/opioid combinations—have been associated with patient deaths for many years, which has not decreased over time. This indicates little or no progress has been made in managing the risks associated with these important but potentially harmful medications.

Company direct-to-consumer contacts. Drug companies are increasingly in contact with patients who use their products, enrolling them in informational and refill reminder programs, directly providing them with drugs, and registering them in special restricted programs to distribute drugs and collect information from patients. When treatment is discontinued because a patient dies, companies may report this as a “fatal” ADE even if no relationship is suspected between the patient’s death and the drug.

Most Frequent Drugs Associated with Reported Deaths

The drugs most frequently identified as suspect in reported patient deaths are shown in Table 1. drug deaths 2009As in prior years, a relatively small number of drugs accounts for a disproportionate share of all ADE deaths in 2009. For example, QuarterWatch™ identified 1,986 drugs involved in ADEs during the 4th quarter of 2009. Of these, 1,604 drugs (80.7%) were not associated with patient deaths. Another 191 drugs (9.6%) had two or fewer patient deaths reported during the quarter.

Signals for Specific Drugs

Rosiglitazone (AVANDIA) is a prime example of where a growing number of reported deaths is likely the result of increased awareness of the drug’s risks. In 2009, this antidiabetic agent accounted for 1,354 reported patient deaths—more than any other prescription drug. In part, we attribute the large number of reports to FDA warnings and publicity about cardiovascular safety risks with rosiglitazone.  

Deferasirox (EXJADE) is an oral drug used to treat chronic iron overload due to repeated blood transfusions (transfusional hemosiderosis). Given that the drug is used in a limited population, 1,320 deaths is unusual but may be explained by two related events. In 2009, the manufacturer (Novartis) analyzed reported ADE data and updated their database, adding deaths from the drug that had occurred years earlier. Secondly, as a result of the analysis, a new boxed warning about the risk of kidney failure, liver failure, and gastrointestinal hemorrhage was added.

Digoxin (DIGITEK brand) was the target of large-scale drug recalls in 2008 and 2009 due to over-strength tablets, leading to a massive increase in reported patient injuries and deaths in both years, as we have previously reported.

FentaNYL (DURAGESIC, FENTORA, ACTIQ, ONSOLIS, SUBLIMAZE), a powerful synthetic narcotic, provides an example of continued problems with an important high-alert drug. In 2009, fentaNYL accounted for 397 reported patient deaths. For years, use of fentaNYL patches has presented an array of drug safety risks, which appear to be unresolved. 

Acetaminophen and acetaminophen-HYDROcodone (VICODIN, HYCET, LORTAB, others) are among the most widely used pain medications. But a modest overdose of acetaminophen can result in severe or fatal liver damage. The acetaminophen-narcotic combination has special hazards if patients develop a tolerance to the narcotic and increase their daily intake, leading to a potentially fatal acetaminophen overdose. Acetaminophen is also frequently identified as a suspect drug in intentional fatal overdoses.

Ibandronate (BONIVA), a once-a-month drug for osteoporosis, provides an example where a company’s consumer outreach program generated numerous reports of patient deaths in which the drug was not necessarily the cause. The company sent monthly refill reminders to hundreds of thousands of mostly-elderly patients. If a postcard was returned “addressee deceased,” the company reported it as a fatal ADE, accounting for 85-90% of the 250 deaths reported in 2009.

Product Quality Concerns

TYLENOL (acetaminophen). Since September 2009, Johnson & Johnson has recalled at least 120 million packages of Tylenol for infants and children after FDA inspections disclosed evidence of bacterial contamination of raw materials. We identified more than 400 reported ADEs for these products during the last quarter of 2009. However, we suspect that, in the majority of reported cases, it is unlikely that a child was actually injured by bacterial contamination of a Tylenol product. These cases were also reported following publicity about the first Tylenol recall. On the other hand, we identified 66 cases involving hospitalization, 1 case of a life threatening condition, 1 case that required intervention to prevent harm, and 4 patient deaths.

DURAGESIC (fentaNYL) Patches. Quality concerns were also reported for Ortho-McNeil’s Duragesic patches, primarily the 100 mcg/hour strength. The main problem appeared to be that the patches did not adhere well or were falling off. A patch that falls off of a patient can be picked up by pets who might eat it or children who may place it on their bodies as a “Band-Aid.” From the information available, no determinations could be made regarding why the patch adherence problem was occurring.


The total of 19,551 reported patient deaths in 2009 is large by any measure, even without adjustment for underreporting. In comparison, for 2007 (the most recent data available), the National Center for Health Statistics reported 17,520 deaths by homicide, 33,185 deaths by suicide, and 42,031 deaths from motor vehicle accidents.

Other risks of serious injury and death in our society, such as crime, airline safety, railway safety, motor vehicle injuries, are monitored carefully and in depth. But as far as we know, the ISMP QuarterWatch™ is the only ongoing assessment of the risks associated with drug therapy. Additionally, our effort is subject to substantial limitations, which we take care to disclose in our full report, found here.

The number of patient deaths associated with drug therapy ought to be declining thanks to closer scrutiny of products before approval and better programs to understand and manage their known risks. That outcome seems unlikely unless we develop better measures of drug risks and learn more about which drugs are implicated and the types of reactions they most frequently cause.