Reducing Antimicrobial Overuse Through Targeted Therapy for Patients With Community-Acquired Pneumonia

Community-acquired pneumonia (CAP) is a leading cause of hospitalization and inpatient antimicrobial use in the United States. However, diagnostic uncertainty at the time of initial treatment and following negative cultures is associated with prolonged exposure to broad-spectrum antimicrobials. We propose a large multicenter cluster randomized controlled trial to test two approaches to reducing the use of broad-spectrum antibiotics in adult patients with CAP a) routine use of rapid diagnostic testing at the time of admission and b) pharmacist -led de-escalation after 48 hours for clinically stable patients with negative cultures.

When a patient is admitted with a diagnosis of pneumonia, it will trigger the admission order set and if the physician is in a hospital randomized to the rapid diagnostic testing arm, a CDSS-based alert will be generated in real time, and the form will append orders for viral and UAT testing. For physicians at a hospital randomized to the control condition, ordering will proceed as usual (standard-of-care). A second CDSS algorithm will identify study patients who have negative culture results (blood and/or sputum) for greater than 48 hours and generate a list for the antimicrobial stewardship pharmacist, who will be a member of the study team. The alerts will be audited by the clinical pharmacist daily on weekdays at a centralized location. In clinically stable patients from hospitals randomized to the de-escalation arm, the pharmacist will communicate their recommendations for de-escalation to the clinical providers via a phone call, Epic chat, or page. The primary outcome will be the duration of exposure to broad-spectrum antimicrobial therapy defined by the number of days of antibiotic therapy from initiation to discontinuation. Secondary outcomes will include detection of influenza/RSV, de-escalation and re-escalation to broad-spectrum antibiotics after de-escalation, total antibiotic duration, in-hospital mortality, ICU transfer after admission, healthcare-associated CDI and acute kidney injury after 48 hours.