Abstract
The Evaluation of Processes and Indicators in Infection Control (EPIC) study assesses the relationship between hospital care and rates of central venous catheter-associated primary bacteremia in 54 intensive-care units (ICUs) in the United States and 14 other countries. Using ICU rather than the patient as the primary unit of statistical analysis permits evaluation of factors that vary at the ICU level. The design of EPIC can serve as a template for studies investigating the relationship between process and event rates across health-care institutions.