Jul 8, 2013
For the first time, experts from a broad spectrum of fields have come together to begin to address a widely recognized and potentially dangerous weakness of drug warning systems which are used now by nearly every pharmacy and hospital in the United States and will soon be used by most physicians.
A research team at the University of Arizona College of Pharmacy, headed by Professor Dan Malone, is leading a national initiative to improve drug-drug interaction (DDI) alert systems.
DDI alert systems warn healthcare professionals of potentially dangerous interactions between two or more drugs prescribed for the same patient. The systems are so sensitive, however, that they send alerts for drug combinations that have a very low probability of causing problems. This results in thousands of alerts being displayed per week at busy healthcare facilities, which leads to “alert fatigue” in prescribers and pharmacists.
To combat alert fatigue, many healthcare workers end up ignoring or overriding alerts. Studies have found up to a 90 percent override rate.
“This is a potentially catastrophic situation for patients taking multiple medications,” says Malone, who is an expert in drug-drug interactions, “and it is not an easy problem to solve. There are so many factors that go into designing, implementing and using these sophisticated systems. We needed to bring together everyone involved: leaders from academia, clinical practice, government agencies, industry and international organizations.”
To help those stakeholders collaborate, Malone and his team hosted an invitational conference of experts in DDI clinical decision support (CDS) systems at the United States Pharmacopeia Convention Headquarters in Rockville, Md., in May. The meeting was one step in the effort, known as the Drug-Drug Interaction Clinical Decision Support Conference Series, to develop an ongoing, structured process to improve the quality of DDI alerting systems.
Participating in the conference were representatives from the Food and Drug Administration; the Office of the National Coordinator for Health Information Technology; the Agency for Healthcare Research and Quality; the RAND Corporation; drug knowledge database vendors Wolters Klewer, Epocrates, Lexicomp and First DataBank; CDS software companies Epic and Cerner; the Pharmacy Quality Alliance and the American Society of Health-System Pharmacists.
“Drug-drug interactions represent an important cause of harm, and because providers in the U.S. are rapidly adopting electronic health records, it will be much easier to prevent them,” says David Bates, professor of medicine at Boston’s Brigham and Women’s Hospital and Harvard Medical School and a participant in the May conference. “But it is critical that we get the settings right—many organizations have displayed too many interactions, and the really serious ones often look the same as some that are not very important in today’s systems. At the conference, we talked about how to approach these issues, which will be central to getting value from electronic records.”
Standards are one of the aims of the Drug-Drug Interaction Clinical Decision Support Conference Series. The goals of the series are to:
1. Develop an ongoing process for DDI evidence integration into clinical decision support systems;
2. Recommend standards for DDI classification for CDS; and
3. Establish basic standards for communicating DDI information within CDS.
Participants appreciated the collaborative nature of the May meeting.
“An old African proverb states: ‘If you want to go fast, go alone. If you want to go far, go together,’” says Raymond Chan of Sentara. "This conference definitely brought together many of the brightest minds in CDS design and utilization, ultimately shooting for shifting the bar up.”
The Drug-Drug Interaction Clinical Decision Support Conference Series will include more working meetings through 2013. Malone’s goal is to have recommendations and guidelines for improving DDI alert systems ready to distribute for comments and feedback by early 2014.
This project is supported in part by a grant from the U.S. Department of Health and Human Services Agency for Healthcare Research and Quality, No. 1R13HS021826-01.