HEALTH-SYSTEM EDITION
PROFESSIONAL PRACTICE

This system for cutting errors uses trained observers

Auburn University, Auburn, Ala., has licensed its system of identifying, measuring, and tracking medication errors to MedAccuracy of Lenexa, Kan. AU MEDS (Auburn Medical Error Detection System) has been set up at eight hospitals, and more than 100 others have expressed interest.

AU MEDS was conceived primarily by Kenneth Barker, Ph.D., R.Ph., director of the Center for Pharmacy Operations & Designs at Auburn and Distinguished Sterling Professor in the department of pharmacy care systems at Auburn's school of pharmacy.

This type of research methodology, used to study the concept of unit dosing, has been in use for about 30 years in studying hospital medication systems, said Barker. He has used this system of direct observation of medication administration in a hospital setting in his research on medication errors. He and colleagues reported in 2002 that a medication error occurs in nearly one in every five medication doses given in hospitals and skilled nursing facilities, with an error defined as a discrepancy between the dose ordered and the dose received.

A large hospital system approached Auburn to ask if such a system could be revised and developed into a version to be used within a hospital for the purpose of continuous quality improvement.

AU MEDS has two parts. One involves having a certified medication observer watch as nurses in a hospital unit administer medications to a patient, said Jim Martin, VP of marketing and sales for MedAccuracy. Observers intervene only if a mistake is being made, he said. After watching about 100 medication administrations per week, observers review patient records for discrepancies between the administered drugs and the original orders. The information is then recorded in AU MEDS software.

In the second part of the system, proprietary software takes the observers' information, tabulates it, and formats it into accuracy reports. These reports show where mistakes are more likely to happen, either in the hospital pharmacy or on the hospital unit, and why the mistakes occurred. They also indicate the categories of drugs most likely to be involved in an error.

DCH Health System, Tuscaloosa, Ala., has used AU MEDS for 10 months, said Tim Martin, Pharm.D., DCH's director of pharmacy. "It took us about 10 months to go through training, initially survey all our patient care areas, and collect an initial round of data," he said.

DCH includes four hospitals in western Alabama with a total of about 1,000 licensed beds. While initial reports show that DCH had a better medication accuracy rate than expected, some units had very high accuracy rates and some did not, Martin said.

At DCH, total staffing for the AU MEDS system is a bit more than one full-time employee, Martin said. The system uses half the time of a pharmacist, half the time of one nurse, and part of another nurse's time. DCH added the equivalent of one-half of a full-time employee to offset the time taken up by AU MEDS.

It takes about a week to train and certify medication observers, MedAccuracy's Martin noted. Typically, more than one observer is certified at each hospital. Depending on the size of the hospital, AU MEDS costs between $30,000 and $100,000, not including the cost of additional employees. Staff retention has not been a problem for the hospitals using AU MEDS, Martin told Drug Topics. If there is staff turnover, MedAccuracy comes back to train and certify a new observer.

Most nurses have been comfortable with the system, Martin added. Nursing units are not compared with each other in terms of how they prevent drug errors. "You have to be careful not to set up competition," he said.

"We think AU MEDS works best as a peer-review process by nurses, because nurses are the ones who administer medications," said Barker. Nurses are a good source of information about clues to causes of medication errors and of suggested interventions and solutions. "We have used pharmacists and nurses and even pharmacy technicians as observers during testing," he added. "They can be equally effective, but the other half of the observer's job is to take that information and go back to the nurses and ask them how to prevent similar errors. That part is easier when the observer is a nurse."

Barker said that an additional feature for AU MEDS would be launched shortly. BenchMeds will allow hospitals to confidentially compare their accuracy rates to a national sample of similar hospitals around the country.

Valerie DeBenedette

The AUTHOR is a healthcare writer based in New York.

Valerie DeBenedette. This system for cutting errors uses trained observers.

Drug Topics

Dec. 8, 2003;147:HSE18.