Medication Use Leading to Emergency Department Visits for Adverse Drug Events in Older Adults

This study was an analysis of data compiled from the National Electronic Injury Surveillance System-Cooperative Adverse Drug Event Surveillance (NEISS-CADES), National Ambulatory Medical Care Survey (NAMCS), and the National Hospital Ambulatory Medical Care Survey (NHAMCS). It attempted to estimate the number of Emergency Department (ED) visits for adverse drug events involving medications that by consensus have been deemed potentially inappropriate for use in elderly patients (Beers criteria) compared with other medications. By searching the NAMCS and NHAMCS databases, the number of outpatient prescriptions written for Beers criteria as well as other more suitable medications was determined. Data from patients aged 65 years or older seeking ED or outpatient care in 58 non-pediatric hospitals participating in NEISS-CADES from January 1, 2004 to December 31, 2005 were also obtained and were used to estimate ED visits for adverse drug events that were defined as events that the physician attributed to the use of a drug or to a specific drug effect. Visits related to intentional overdose for self harm, therapy withdrawal, drug abuse, adverse events that occurred as a result of treatment in the ED, and follow-up visits for previously diagnosed adverse events were excluded. In 10.5% of all outpatient visits, a prescription was written for a Beers criteria medication. In an additional 9.4% of visits, a prescription was written for an appropriate anticoagulant, antiplatelet, or oral hypoglycemic agent. On the basis of 4492 reported events, the authors estimated that in 2004 and 2005 there were 177,504 ED visits annually for adverse drug events nationwide. Of those, 3.6% were related to medications listed in the Beers criteria, whereas 47.5% of events were related to oral anticoagulants, antiplatelet agents, oral hypoglycemics, and narrow therapeutic index agents not listed in the Beers criteria. Thus, the number of visits for adverse drug events was lower than expected for Beers criteria medications, but significantly higher than expected for certain agents not included in those criteria.