Using Big Data to Prevent Cardiac Arrhythmias

Researchers at the UA and Banner Health will create alerts to identify patients at risk for developing heart rhythm disorders and sudden cardiac death.
"Real-time precision-guided prescribing will save lives and prevent drug-induced arrhythmias for patients in all of the University's clinical partners, including Banner Health."

(May 8, 2019) As many as two-thirds of elderly patients admitted to acute-care settings are exposed to multiple medications known to interfere with the ability of the heart to "recharge" between beats, which could lead to a life-threatening form of cardiac arrhythmia.

Although a rare health occurrence, more than 150 common medications - including azithromycin and ondansetron - can contribute to prolonged heart-recharging intervals, which may lead to serious complications and/or longer hospital stays.

To address this problem, University of Arizona researchers will implement alerts embedded in patients' electronic health records to estimate risk and to assist health-care providers in mitigating sudden cardiac death. Funded by a two-year, $766,000 grant from the Agency for Healthcare Research and Quality, the project uses health information technology to support maximum safety and effectiveness of medicines for patients who are more likely to be affected. The team will work within 28 Banner Health facilities throughout Arizona, California, Colorado, Nebraska, Nevada, and Wyoming.

"Real-time precision-guided prescribing will save lives and prevent drug-induced arrhythmias for patients in all of the University's clinical partners, including Banner Health," said C. William Heise, MD, assistant professor, Division of Clinical Data Analytics and Decision Support at the UA College of Medicine - Phoenix.

The project will use electronic health records and prescribing systems to identify patients at high risk of harm from medications and immediately send electronic messages to advise doctors of the risk and how to prevent the development of potentially life-threatening heart rhythm disorders and sudden death.

The grant also will fund educational programs to inform health-care providers how to best utilize and respond to the computer-generated advisories.

"We are excited to have this opportunity to work closely with Banner Health data scientists and to evaluate a medication safety program within the entire heath-care system," said Daniel Malone, RPh, PhD, FAMCP, scientific leader for the grant, who is professor at the UA College of Pharmacy with a joint appointment in a UA Mel and Enid Zuckerman College of Public Health.

The project builds on data that already is captured in patients' electronic medical records, including data that is imported automatically from the hospital's electrocardiogram equipment.

"In addition to making this alert as specific and precise as possible, we will conduct usability testing to ensure it works in the provider's workflow," said Corneliu Antonescu, MD, MBI, FAAP, clinical assistant professor, Department of Biomedical Informatics, at the UA College of Medicine - Phoenix and a physician informatist with Banner Health.

The research team also includes other UA faculty members, including Raymond Woosley, MD, PhD, co-director of the Division of Clinical Data Analytics and Decision Support at the UA College of Medicine - Phoenix; Tyler Gallo, PharmD, assistant professor, pharmacy practice and science, from the UA College of Pharmacy and UA College of Medicine - Phoenix; and Sheila Gephart, PhD, RN, associate professor with the UA College of Nursing. Assisting in the project is Jim Tisdale, PharmD, professor of pharmacy practice with Purdue University College of Pharmacy.

The project is supported by grant No. R18HS026662 from the Agency for Healthcare Research and Quality. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Agency for Healthcare Research and Quality.

Pictured: Dan Malone, RPh, PhD, FAMCP, scientific leader for the grant.