Making medication alerts in electronic medical record systems more useful and usable

A study by Regenstrief Institute and U.S. Department of Veterans Affairs investigators provides the first in-depth look at how health care providers react to medication alerts generated by electronic medical record systems. The researchers plan to use this information to improve the design of medication alerts and diminish the phenomenon known as alert fatigue, where providers can become desensitized and may start unintentionally ignoring some important warnings.

For this study, the researchers observed providers as they treated patients to learn about the strengths and weaknesses of medication alerts.

Medication alerts provide the health care team with computer-generated information on a variety of drug-related issues. Among the most common medication alerts are warnings about patient allergies, drug interactions and duplicate prescriptions. The alerts, critical to patient safety, can be triggered by many factors including the prescription of a new medication or a change in a patient's laboratory test results.

But health care providers may experience alert fatigue and unintentionally overlook important alerts if the electronic medical record system generates too many medication alerts; if alerts do not apply to the patient (for example, warning about a drug the patient has already been taking without problems); or if the alert provides too much extra information. The goal is to develop alerts that aid healthcare providers more effectively and enhance patient safety.

"Prescribers' Interactions With Medication Alerts at the Point of Prescribing: A Multi-Method, In Situ Investigation of the Human-Computer Interaction" appears in the April 2012 issue of the International Journal of Medical Informatics.

"As a human factors research scientist, I am interested in learning how to improve the usability of electronic medical records systems so doctors, nurses and pharmacists can work more effectively. I also am interested in finding ways to make health care delivery safer for patients," said Regenstrief Institute investigator Alissa Russ, Ph.D., a research scientist with the Center of Excellence on Implementing Evidence-Based Practice at the Richard L. Roudebush VA Medical Center in Indianapolis. She is first author of the study and is an adjunct assistant professor of pharmacy practice at Purdue University.

During the study, 320 medication alerts were generated by an electronic medical record system as 30 doctors, nurse practitioners and pharmacists treated 146 patients in a variety of outpatient clinics. The study authors observed and analyzed factors that influenced how health care providers perceive, interpret and respond to alerts. The authors identified nine factors that influence prescribers as they encounter alerts, providing a detailed description of 44 components that contribute to these factors.

The researchers found that prescribers were sometimes unsure why an alert was appearing, and they also determined that alert designs were more pharmacist-oriented than physician- or nurse practitioner-oriented, in spite of the fact that doctors and nurse practitioners were the principal prescribers.

"Too many alerts and overly detailed alerts are a common source of frustration across electronic medical record systems," Dr. Russ said. "Unless we improve medication alerts so they contain information that users need to make decisions, the problem of alert fatigue will grow as EMR systems expand beyond single hospitals and share more data."

"When interface usability, workflow integration, and potential information overload are addressed, electronic medical records have a tremendous advantage over paper records, such as rapid accessibility by multiple providers, improved documentation, and implementation of computerized decision support, such as clinical reminders and medication alerts," said Regenstrief Institute investigator Jason J. Saleem, Ph.D., a research scientist with the Center of Excellence on Implementing Evidence-Based Practice at Roudebush VAMC. Senior author of the study, he is a human factors engineering specializing in the delivery of medicine care and an assistant research professor of electrical engineering and computer science in the Purdue School of Engineering and Technology at Indiana University-Purdue University Indianapolis.