would set a useful broader context within which to consider the NRC's regulatory process better.
Hospitals in the United States provide more than $5 billion worth of prescription drugs and drug products per year (Manesse, 1990). In addition, more than 1.5 billion prescriptions are filled each year in outpatient pharmacies, and additional prescriptions are delivered in outpatient medical facilities, including nursing homes and medical clinics. Although American consumers have greatly benefited from prescription drugs, these medications carry both predictable and unpredictable risks. In addition to known side effects and drug interactions, adverse drugs events (ADEs, defined as an injury resulting from medical intervention) can occur because of medication administration errors.
The numbers of ADEs were researched for comparison with byproduct radioactive materials misadministrations. No comprehensive, consistent, reliable, valid, and accurate accounting of ADEs or drug misadventures is available in the United States (Allan and Barker, 1990). Drug misadventures overall have an elusive denominator and a moving numerator because no one agency or reporting mechanism is standard to all hospitals, outpatient facilities, pharmacies, and nursing homes. Manesse (1989) has suggested that 535 ADEs occur each year, or perhaps 1 per 1,000 prescriptions.
Although the JCAHO requires that hospitals have an ongoing drug surveillance program designed to monitor adverse drug events, mandatory reporting to the FDA is required only for investigational drugs and blood products. The FDA's Adverse Drug Reaction Voluntary Reporting Program collects information on adverse drug reactions from pharmaceutical companies and from physicians, pharmacists, other health care personnel, and patients. In 1987, these groups filed 50,000 adverse drug incident reports with the FDA; of these, 12,000 deaths were attributed to adverse drug reactions, and 15,000 cases required hospitalization (Perry and Knapp, 1987). Adverse drug events may be markedly underreported for hospitalized patients (Classen et al., 1991).
In a recent prospective cohort study, Bates et al. (1993) reported the incidence of ADEs and potential ADEs in 4,031 adult patients admitted to nonobstetrical units of two tertiary care hospitals in Boston from February through July 1993. A total of 247 ADEs and 194 potential ADEs were identified, with extrapolated event rates of 6.5 percent ADEs and 5.5 percent potential ADEs per 100 admissions. Of these ADEs, 1 percent were fatal; 12 percent, life threatening; 30 percent, serious; and 57 percent, significant. Chemotherapy ADEs are generally grouped in with other medications in ADE statistics. In the Harvard Medical Practice Study II, ADEs accounted for 19.4 of the total number of disabling adverse events (Leape et al., 1991).