Study: Number of adverse events involves med errors

Facility used CPOE system and on-site pharmacists

A teaching hospital further improved its existing safety mechanisms after a study of critically ill patients found a significant number of adverse events involving medications.

Researchers at the health care institution wanted to study the incidence and nature of serious medical errors in critical care settings. They conducted a one-year observational study, between July 2002 and June 2003, of medical intensive care unit (MICU) and coronary care unit (CCU) patients during nine, three-week periods.

The researchers looked at a total of 391 patients with 420 unit admissions during 1,490 patient days. They found 120 adverse events in 79 patients (20.2%), including 66 (55%) nonpreventable and 54 (45%) preventable adverse events. The researchers also found 223 serious errors.

Among the adverse events, 13% were life-threatening or fatal. Among the serious errors, 11% were potentially life-threatening.

Medications were involved in a large proportion of the adverse events, the researchers say. Among the incidents, 56 (47%) were due to adverse drug events (ADEs), including 19 preventable ADEs and 37 nonpreventable ADEs. Medications also were responsible for 78% of the most serious errors. Wrong dosage was the most common medication error.

The medication categories most frequently associated with errors were cardiovascular drugs (24%), anticoagulants (20%), and anti-infective agents (13%). Anti-infectives and cardiovascular drugs are probably the most common medications on the unit, says Jeffrey M. Rothschild, MD, MPH, assistant professor of medicine at Brigham and Women’s Hospital/Harvard Medical School in Boston, and lead author of the study. "Anticoagulants are not, but they are more difficult to use. We find, for instance, that insulin and heparin are high-risk and are particularly associated with problems."

Going into the study, the researchers didn’t know what their adverse event rates would be, Rothschild says. "They turned out to be somewhat comparable to other studies. We thought we would do better on the medication side because we have good, existing safety mechanisms."

These mechanisms included computerized physician order entry (CPOE) and on-site pharmacists. "It shows that the incidence is probably much higher in most other institutions [that don’t have these mechanisms]," he says. "Any of the medication errors that we found were despite the presence of the pharmacists."

Investigators have studied the value of having pharmacists participate in patient rounds in the intensive care unit (ICU), Rothschild says. "[The literature has] shown that it is equivalent to computerized order entry in efficacy. In places that don’t have CPOE, it is a great interim step. We happen to do both in some of our units."

At his institution, the clinical pharmacists are teaching house staff about proper prescribing. Rothschild adds that "the pharmacists pick up a lot of things before the orders are even entered. They are great assets if they can join physicians on rounds in the ICU setting."

The pharmacists can’t be everywhere all the time, however. Their participation in patient rounds or even availability on the unit usually is Monday through Friday, 7 a.m. to 4 p.m., Rothschild reports. "Things will obviously only be caught a third of the week because of the 24/7 nature of critical care."

Since the study, the institution added more safety interventions, including a new web-based incident reporting system, introduction of smart intravenous infusion pumps, and bar-coded medication administration.

"Our system is getting better," Rothschild says. These interventions are now set up to catch errors that might have slipped through at administration under the previous system.

For more information about the study, see the August 2005 issue of Critical Care Medicine. The study was sponsored by the Agency for Healthcare Research and Quality in Rockville, MD.