Impact of a Simple Automated Best Practice Alert (BPA) on Quantity and Quality of In-hospital Antibiotic Use in a Tertiary and Three Secondary Hospitals

NCT07115966 | Active Not Recruiting

• 1 other identifier: 2025-00735
• Study Type: interventional
• Target: 58
• Locations: 1 country
• Sites: 1

Brief Summary

The goal of the stepped-wedge cluster-randomized trial is to assess the impact of an antimicrobial stewardship intervention: a simple, automated Best Practice Alert (BPA) that reminds prescribers to reevaluate antibiotic therapy after 72 hours (or 24 hours for prophylaxis), in accordance with guideline recommendations. The primary hypothesis is that this simple BPA reduces antibiotic use in terms of quantity (amount and duration) and quality (spectrum breadth), measured by days of antibiotic spectrum coverage at the patient level (primary outcome), as well as at both patient and cluster levels using various metrics of antibiotic use. The trial will introduce the BPA in a stepwise manner, with all wards implementing it by the end. It will compare the intervention period to the baseline (pre-intervention) and control periods.

Conditions

Antibiotic Prescriptions; In-Patient Treatment

Keywords

best practice alert; quanlity and quantity of antibiotic use; in-patients

Outcome Measures

Primary Outcomes (1)

• Days of antibiotic spectrum coverage (DASC) per patient admission (PA)

  Overall score of days of antibiotic spectrum coverage per patient admission. The score is composed of the breadth of the bacterial spectrum covered by the administered antibiotic (according to Kakiuchi 2022 - the broader the antibiotic spectrum, the higher the score), summed over the number of days the antibiotic is given. Accordingly, there are no maximum or minimum values

  12 months

Secondary Outcomes (11)

• Days of antibiotic spectrum coverage (DASC) per patient antibiotic day (PAD)

  12 months

• Days of treatment (DOT) per 100 patient days (PD) and per patient admission on ward level

  12 months

• Defined daily doses (DDD) per 100 patient days (PD) and per patient admission (PA)

  12 months

• Antibiotic (AB) days per patient admission (PA)

  12 months

• In hospital mortality

  12 months

Study Arms (2)

Best practice alert (BPA)

EXPERIMENTAL

Intervention: The BPA will appear to the prescribing physician for every patient on the respective ward. Activation of the BPA is ward-based. A total of 58 wards will be stratified according to their focus-surgical, medical, intermediate care, rehabilitation, or pediatric-and grouped based on their baseline antibiotic consumption, measured in days of antibiotic spectrum coverage per patient admission (DASC/PA). This stratification will result in 9 to 10 clusters across 6 groups. The clusters will then be randomized to the timing of BPA activation, and all wards will sequentially switch to the BPA arm every 2 months over a 12-month period. By the end of the trial, after 12 months, all wards will be using the automated BPA.

Behavioral: Computerized decision support by best practice alert (BPA)

Controls - No BPA

NO INTERVENTION

Control: standard patient care with no BPA activated. By the end of the trial, after 12 months, all wards will be using the automated BPA.

Interventions

Computerized decision support by best practice alert (BPA) BEHAVIORAL

The antimicrobial stewardship intervention encourages prescribers by a simple alert to follow guidelines for reviewing antimicrobial prescriptions after a set timeframe for potential de-escalation to targeted therapy or discontinuation of the antibiotics, as recommended by national and international guidelines. An automated simple BPA will trigger after an antibiotic prescription with therapeutic indication (72 hours) or surgical prophylaxis (24 hours, reminding prescribers to reassess treatment for possible de-escalation, adaption to targeted therapy, or cessation. If ignored, the prescription remains unchanged, but the alert will continue until addressed. Prescribers must select reasons for not changing the prescription, such as pending microbiology results. The control group corresponds to the inpatient wards not yet receiving the BPA, where antibiotics are managed according to "standard-of-care".

Best practice alert (BPA)

Eligibility Criteria

• Sex: all
• Healthy Volunteers: No
• Age Groups: Child (0-17), Adult (18-64), Older Adult (65+)

You may not qualify if:

• Ward level:
• Emergency departments
• Outpatient clinics
• Haemato-oncologic stem cell transplant wards, where daily ID visits are performed
• ICU wards, where daily ID visits are performed
• Indvidual patient data for analysis
• Refusal of institutional general consent for further use of patient data

Sponsors & Collaborators

• Insel Gruppe AG, University Hospital Bern: lead

Study Sites (1)

Inselgruppe

Bern, 3007, Switzerland

Location

Study Officials

• Philipp Jent, PD

  Inselgruppe Bern

  STUDY CHAIR

• Nasstasja Wassilew, Dr. med.

  Inselgruppe Bern

  PRINCIPAL INVESTIGATOR

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: RANDOMIZED
• Masking: SINGLE
• Who Masked: OUTCOMES ASSESSOR
• Masking Details: Cluster allocation will also be blinded to the wards and prescribers; however, prescribers may become aware of their cluster assignment when working across different wards in both intervention and control clusters, based on whether or not they receive the BPA.
• Purpose: PREVENTION
• Intervention Model: SEQUENTIAL
• Sponsor Type: OTHER
• Responsible Party: SPONSOR

Model Details: Wards will be cluster-randomized according to specialty and antibiotic use (DASC/PA), based on retrospective data. The BPA will be activated in a stepwise manner for each cluster every two months over a 12-month period. By the end of the study, the BPA will be active in all participating wards.

Study Record Dates

• First Submitted: August 4, 2025
• First Posted: August 11, 2025
• Study Start: August 1, 2025
• Primary Completion (Estimated): July 31, 2026
• Study Completion (Estimated): July 31, 2026
• Last Updated: September 2, 2025

Record last verified: 2025-08

Data Sharing

• IPD Sharing: Will not share

Locations

CH (1)