Background: The VA has pledged to reduce overuse of antibiotics, and is a partner in the National Action Plan to combat antibiotic-resistant bacteria. Overuse of antibiotics can lead to Clostridium difficile infections among the individuals receiving antibiotics, as well as antibiotic resistance that can complicate treatment of future infections for all patients. Chronic antibiotic suppression (CAS) is defined as indefinite use of oral antibiotics following initial guideline recommended intravenous and oral antibiotics. There is evidence that CAS may be inappropriately used and overused. Thus, there is a critical need to examine how CAS is prescribed, and who benefits the most, and who is harmed, from CAS therapy. This knowledge can be used to develop an intervention to de-implement and optimize CAS use. In this study, we will examine CAS treatment for prosthetic joint infections (PJIs). PJIs occur after hip, knee or shoulder replacement surgery and are associated with a high clinical and economic impact. In the VHA system, over 10,000 hip, knee or shoulder replacement surgeries are performed each year and this number, as well as the number of associated PJIs, continues to increase. Treatment for PJI after hip, knee or shoulder replacement surgery is important to the VA because osteoarthritis, the most common reason for that surgery, is the fifth most common condition among Veterans. Our pilot data and other studies have found that many PJI patients are prescribed CAS outside of guideline recommendations, potentially due to physician fear of a relapse. This inappropriate use can lead to poor patient outcomes (e.g., C. difficile infection, antibiotic resistance) without any benefit. Research Plan: The specific aims of this study are: (1) Define patient, infection, and treatment characteristics that identify patients who would benefit from de-implementation of CAS; (2) Define current provider CAS prescribing processes and evaluate barriers and facilitators to changing those practices through interviews with health care providers; and (3) Create a decision model to conduct an economic evaluation of CAS use and different interventions to de-implement inappropriate CAS use. We expect to document significant variations in CAS prescribing and identify cost-effective interventions to de-implement CAS or reduce its duration. We propose a 3-year mixed methods study to inform efforts for CAS de-implementation. Aim 1 will use a Corporate Data Warehouse (CDW) database of ~7,500 PJI patients treated at all VA hospitals in a retrospective cohort comparative effectiveness study to determine specific characteristics that identify those PJI patients who will benefit from CAS de-implementation. These characteristics may include young age, infection with a non-virulent pathogen, initial receipt of appropriate antibiotics for PJI caused by Staphylococcus aureus, or receipt of 2-stage exchange surgery. In Aim 2, we will visit 8 diverse hospitals, qualitatively explor...