This study will be performed at Denver Health Medical Center, a 477-bed urban, safety-net, acute-care hospital located in Denver, Colorado. The adult ED is staffed by board-certified emergency physicians at all times. These physicians supervise the care of all patients being managed by resident physicians, nurse practitioners, physician assistants, and medical students. Approximately 430 patients are admitted to the hospital from the ED each year with community-acquired pneumonia (CAP) and severe sepsis (SS).
Adherence to CPGs will be measured at the level of the attending emergency physician. All employed, attending emergency physicians working clinically in the adult ED at Denver Health Medical Center at the start of the study will be included.
The investigators will use a step-wedge design to randomize physicians into clusters. Randomization of physicians into clusters and randomization of clusters to intervention timing will occur one week prior to delivery of the intervention to cluster one. he intervention will consist of monthly audit and feedback on adherence to CPGs for CAP and SS. Once a cluster enters its first intervention month, all physicians in that cluster will receive an email detailing their adherence to both CAP and SS CPG for every month since the start of the study. Adherence to the entire CPG as well as each component of the CPG will be provided. In addition, physicians will be shown the median prevalence of adherence for all physicians as well as where their adherence ranks among their physician group. Lastly, in a separate secure email, physicians will be given patient identifiers (i.e., name, MRN, date of visit) for each patient that received non-adherent care and will be told which component of care was not adherent to the respective CPG so that they can review the case themselves, if desired.