Detecting Novel Drug Combinations Associated with Adverse Events (MedCare) – PHRT


Detecting Novel Drug Combinations Associated with Adverse Events (MedCare)

Short Summary

In medicine, there is the underlying principle to apply treatment for the patient’s benefit without risking harm. This is often referred to as the “do no harm” oath. However, in patients with multiple health conditions multiple medications are often prescribed. While done with good intention to treat the individual conditions, it can increase the likelihood of experiencing adverse drug events due to a harmful interaction between two or more medications. Despite the knowledge that the risk of an adverse drug event increases with the number of medications, our understanding of which drug combinations are harmful is limited. Thus, the driving motivation of MedCare is to overcome this limitation by developing screen algorithms that can applied to two different types of healthcare data – adverse event reporting data (i.e., pharmacovigilance) and routinely collected healthcare data (i.e., timeseries data) – to detect and validate adverse drug events associated with harmful drug combinations.


The work proposed in MedCare aims to develop and apply new approaches to identify and evaluate DDIs in large healthcare data.


MedCare will include a collaboration between epidemiology, pharmacology, and data science to tackle this complex problem. It is expected that the results will advance the knowledge of harmful DDIs and improvement of patient safety.


Drug-drug interactions (DDIs) are a leading cause of adverse drug events (ADEs), which are a major cause of preventable harm and mortality. While a small number of DDIs are known at the time of market approval, based on the pharmacokinetics of drug metabolism, the potential for additive effects from multi-drug combinations (MDCs) in patients with high medication use (polypharmacy) remains largely unknown. Moreover, as the number of possible drug combinations is immense, it is not possible to identify new harmful DDIs using traditional methods. Thus, advances in machine learning (ML) are needed.

Data-Intensive Research Project

Prof. Dr. Andrea Burden

ETH Zurich


  • Dr. PD. Stefan Weiler, ETH Zurich


  • Institute of Pharmaceutical Sciences, ETH Zurich
In Progress

Funded by