Patient Safety and Quality

Examining use of "rescue drugs" in a hospital can reveal previously unreported adverse drug events

When hospitalized patients are given "rescue drugs" to revive them from a dangerous state, it may be due to a previously unreported adverse drug event (ADE), suggests a new study. These events are any incident in which use of a medication, a medical device, or a special nutritional product (for example, infant formula) may have harmed the patient.

The study authors performed 3,572 patient chart audits at a rural hospital to identify patients given "rescue" drugs during a 6-month period. They then looked for a specific diagnosis of ADE or a mention of possible drug reaction, overdose, adverse effect, or signs of toxicity in nursing notes. Using this approach, they found that less than 4 percent of all ADEs involving use of rescue drugs had been reported, comparable to published rates.

Of the chart audits, 1,011 included administration of rescue drugs. In 109 of these cases, or 3 percent of discharges, an ADE was believed to have occurred. The ADE was deemed to have been preventable in about one-quarter (0.8 percent) of these cases. Rescue drugs included vitamin K (reverses the blood-thinning effect of warfarin), naloxon (reverses oversedation of narcotics), sodium polystyrene sulfonate (reverses high serum potassium), flumazenil (reverses the effects of benzodiazepines), diphenhydramine (treatment of allergy), digoxin immune fab (reverses digoxin toxicity), and D50W/oral glucose (reverses the effects of insulin or oral hypoglycemic agents).

The researchers concluded that their approach to surveillance of ADEs to detect underreporting is feasible, but labor-intensive. The study was supported by the Agency for Healthcare Research and Quality (HS14920).

See "Improving self-reporting of adverse drug events in a West Virginia hospital," by Charles P. Schade, M.D., M.P.H., Karen Hannah, M.B.A, Patricia Ruddick, M.S.N., A.P.R.N.-B.C., and others, in the September 2006 American Journal of Medical Quality 21(5), pp. 335-341.

Return to Contents
Proceed to Next Article