PROJECT SUMMARY Drug-drug interactions (DDIs) must be regarded not only as a subject for pharmacoepidemiologists and clinical pharmacologists, but as an important issue for public health, particularly in older adults confronted with multiple chronic conditions. Known DDIs are responsible for 13% of adverse drug events and 4.8% of hospital admissions in older adults. Even these high figures may understate the true impact of DDIs because they include only the effects of known interactions. Many clinically important DDIs take years to discover. Given the widespread and growing use of multiple medications by older adults, there is tremendous potential for DDIs to occur in this population, especially among those treated with high risk therapies like warfarin or direct oral anticoagulants. Based on known metabolic pathways, these anticoagulants may interact with many commonly- prescribed medications. DDIs involving anticoagulants can cause over- or under-anticoagulation, the sequelae of which can be immediately life-threatening and have long-term consequences. The broad objective of this project is to produce clinically-actionable and biologically-relevant knowledge about which drugs interact with anticoagulants to cause serious bleeding and/or thromboembolism, as well as the time-course of such interactions and the subgroups most susceptible to these DDIs. We will achieve this objective by taking a translational science approach to DDIs in which we 1) perform high-throughput simulation of potential DDIs based on pharmacologic knowledge; 2) perform high-throughput screening of healthcare data; and then 3) confirm (or refute) and elucidate selected high-probability DDIs in an independent population by conducting in-depth pharmacoepidemiologic studies. This project will produce clinically-actionable knowledge about which drugs interact with anticoagulants to cause serious bleeding and/or thromboembolism, and generalizable biological knowledge about the drugs and pathways involved in these interactions.