Effect of BPA on Anchor Antibiotic Continuity in the ED: Randomized Controlled Trial

The objective is to compare the timeliness of anchor antibiotic administration in the emergency department (ED) after initial dosing with and without a Best Practice Alert in Epic (BPA) implemented to remind physicians to re-order the antibiotic. We hypothesize that post-BPA implementation, physicians will have a higher rate of ordering subsequent doses of antibiotics on-time and with the correct dosages compared to pre-BPA implementation.

Antibiotics administration can sometimes be delayed in the emergency department (ED) environment, where antibiotics are frequently ordered in electronic medical record systems as a one-time dose and the second dose is delayed or missed. This has been described in the literature, a 2017 article in Critical Care Medicine noted in a single academic center that 33% of sepsis cases had 2nd-dose-antibiotic delays greater than 25% of the recommend interval and this was associated with significantly increased hospital mortality (OR 1.61) and mechanical ventilation (OR 2.44). An Epic Best Practice Advisory (BPA) was created by the Mass General Brigham eCare Decision Support team to address this patient safety concern. The BPA alerts providers in the ED or ED Observation when the next dose of broad-spectrum antibiotic dose is due for a patient if no future administration has been ordered, based on patient characteristics (such as renal function) and antibiotic selection. The objective of this study is to analyze the impact of this BPA on our performance in terms of on-time antibiotics through a randomized controlled trial. Patients whose medical record number (MRN) are odd will be assigned to the treatment group, while those whose MRNs are even will be assigned to the control group. The BPA will only be displayed to the patient's providers for the treatment group. We will then compare the performance of on-time antibiotics between the groups and the resulting impact on patient outcomes, such as overall hospital length of stay, admission to the intensive care unit (ICU), and in-hospital mortality.

Patients whose medical record numbers (MRN) are odd will be assigned to the treatment group, while those whose MRNs are even will be assigned to the control group. The intervention, which is the Best Practice Alert in Epic, will only be displayed to the patient's care providers for the treatment group.

Patients will be masked to which group they are assigned. Care Providers will not be masked as they will be shown a Best Practice Alert (BPA) only for patients in the treatment arm. The investigators will not be masked.

Care providers taking care of these patients will receive a Best Practice Alert (BPA) in the electronic medical record (EMR) one hour before an antibiotic expires with no subsequent doses ordered. The BPA will prompt the care provider to re-order the antibiotic and give information on recommended dosage and frequency based on indication and patient characteristics.

The Best Practice Alert described in the Experimental Arm will not be active for patients in this arm. Care providers will proceed with usual care.

This is a customized Best Practice Alert created at Mass General Brigham in the institutional Epic electronic medical record system.

The length of delay between time of expected antibiotic re-administration and the actual time of antibiotic re-administration

Inclusion Criteria: All adult patients (>= 18 years old) who are seen in a Mass General Brigham Emergency Department who had a specific anchor antibiotic administered Exclusion Criteria: None

Given HIPAA concerns and lack of scientific need, we do not plan on sharing individual participant data (IPD).