Pharmacoepidemiologic Approaches to Studying Drug-Drug Interactions

‎1. Course Aim ‎
The course will cover the foundations of pharmacoepidemiologic research on the health ‎effects of drug-drug interactions (DDI) in populations. We will introduce the main sources ‎of bias in drug-drug interaction studies, cover main study designs and techniques to address ‎bias, introduce the concepts for high-throughput data mining for drug-drug interactions, and ‎walk the audience through several case studies. ‎

‎2. Requisites Statement ‎
This is an intermediate level course. Attendees should have basic knowledge of ‎epidemiology or pharmacoepidemiology. The following papers are recommended for ‎preparation for the course, but are not required:‎

Hennessy S et al. Pharmacoepidemiologic methods for studying the health effects of drug-‎drug interactions. Clin Pharmacol Ther 2016;99(1):92-100.‎
Bykov K, Franklin JM, Li H, Gagne JJ. Comparison of self-controlled designs for evaluating ‎outcomes of drug-drug interactions: Simulation study. Epidemiology 2019;30(6):861-866. ‎

‎3. Course Objectives ‎
‎- To describe the role of pharmacoepidemiology in generating real-world evidence on the ‎population health effects of drug interactions
‎- To describe main sources of bias in drug-drug interaction research
‎- To describe the application of a cohort design to study the population health effects of drug ‎interactions
‎- To introduce self-controlled designs and their application to identify and evaluate ‎clinically relevant drug-drug interactions.‎
‎- To present new and advanced topics in drug-drug interaction research, including but not ‎limited to causal inference, study design bias, high-throughput screening, or recent ‎publications.‎
‎ - To facilitate discussion of challenges and solutions in drug-drug interaction research ‎

‎4. Syllabus Outline‎
Introduction, format, and objectives of the course | Presenter: Katsiaryna Bykov, PharmD, ‎ScD (Brigham and Women’s Hospital and Harvard Medical School)

Lecture 1: Introduction to drug-drug interaction research | Presenter: Todd Miano, PharmD, PhD, FCCM ‎‎(University of Pennsylvania)

This lecture will provide an overview and relative merits of available research designs for ‎studying the health effects of drug-drug interactions. Further, it will describe methodologic ‎issues and biases that are unique to or more prominent in studies of drug-drug interactions ‎compared to those examining individual drugs.‎

Case-studies | Design-related decisions for cohort studies on drug-drug interactions | ‎Presenter: Antonios Douros, MD, PhD (McGill University)

This interactive talk will focus on the cohort design for the study of drug-drug interactions. ‎Using examples Dr. Douros’s research, the talk will describe the decision-making process ‎with respect to important aspects of the cohort design (e.g., cohort assembly, exposure ‎definition, choice of comparators) and the minimization of potential biases.‎

Lecture 2 | Self-controlled designs for drug-drug interactions | Presenter: Katsiaryna ‎Bykov, PharmD, ScD (Brigham and Women’s Hospital and Harvard Medical School)

This lecture will provide an overview of self-controlled designs and their role in DDI ‎research. We will discuss two major self-controlled designs: self-controlled case series and ‎case-crossover, along with their pros and cons. The lecture will include empirical examples ‎of how self-controlled designs are used to identify or evaluate drug-drug interactions in ‎pharmacoepidemiology.‎

Lecture 3 | Advanced topics: Time-related bias in drug-drug interaction studies | ‎Presenter: Katsiaryna Bykov, PharmD, ScD (Brigham and Women’s Hospital and Harvard ‎Medical School)
This lecture will provide examples of how time-related biases, including immortal person-‎time bias, can be introduced in DDI studies. We will further discuss strategies to identify and ‎avoid time-related bias in DDI research.‎

‎ Q&A session | Presenters: All faculty