Estimating High-Order, Directional Drug Interactions: Study That Mines Effects on Myopathy Shows the Possibilities

Estimating High-Order, Directional Drug Interactions: Study That Mines Effects on Myopathy Shows the Possibilities

October 2018

Many people in the U.S. take three or more prescription drugs, but very few studies have explored the relationships between high-order drug combinations. The authors set out to make a general tool for quickly and accurately estimating drug-drug interactions, or DDIs, that are directional (a patient already on a given drug adds a new one) and high order (three drugs or more are involved). To accomplish this, they investigated a novel pharmacovigilance problem: mining high-order directional DDI effects on myopathy, using the FDA Adverse Event Reporting System (FAERS) database.

They studied frequent drug combinations from the FAERS database, estimated the risk of myopathy associated with adding new drugs to an existing medication, and created a tree-structured graph to visualize the findings for easy interpretation. The study confirmed that, as previously reported, myopathy is associated with HMG-CoA reductase inhibitors (“statins”) such as rosuvastatin, fluvastatin, simvastatin and atorvastatin—when patients take those drugs alone, when they are newly prescribed at the same time as certain others, and when a patient on, for instance, the immunosuppressant cyclosporine adds simvastatin. The authors also observed other new, previously unidentified but mechanistically plausible associations with myopathy: for instance, the DDI between pamidronate (for bone loss) and the antibiotic levofloxacin.

“This tool can potentially be used to estimate risk for high-order, directional DDIs quickly and accurately, for all different kinds of drug combinations,” say Li Shen, PhD, the study’s senior author. “However, it is dependent on the availability of the data in the FAERS or other electronic health record (EHR) databases. Specifically, the database should contain enough adverse drug events where the corresponding medications contain the studied drug combinations.”


Danai Chasioti, Xiaohui Yao, Pengyue Zhang, Samuel Lerner, Sara K. Quinney, Xia Ning, Lang Li, Li Shen

Share this Content

About Us

To understand health and disease today, we need new thinking and novel science —the kind  we create when multiple disciplines work together from the ground up. That is why this department has put forward a bold vision in population-health science: a single academic home for biostatistics, epidemiology and informatics. LEARN MORE ABOUT US

© 2021 Trustees of the University of Pennsylvania. All rights reserved. | Disclaimer

Follow Us