Prescribing antibiotics frequently poses problems in practice, since patients don't always receive the right dosage of the right antibiotic for the right period of time. This promotes the emergence and spread of antibiotic resistance. The investigators of this trial aim to develop a system designed to help doctors to use antibiotics more appropriately. Under COMPASS (COMPuterized Antibiotic Stewardship Study), doctors in three Swiss hospitals will receive tips on the use of antibiotics that are integrated directly into electronic health record and will also be given regular feedback on their use of antibiotics. Parallel to this, data on the antimicrobial prescription practices of a control group which is not using the system will be collected.

Inappropriate use of antimicrobials favours the spread and emergence of antimicrobial resistance and other adverse patient outcomes. Antimicrobial stewardship (AMS) programs aim to promote the appropriate use of antimicrobials. Most AMS interventions are based on manual, personalized peer review of antibiotic prescriptions by specialists and are therefore time and resource intensive. Informatics based, computerized approaches to AMS are a promising way to "automatize" AMS, but there have been only few randomized controlled trials analysing their effectiveness in the hospital setting.

The primary research question of this study is whether a multi-modal, computerized antibiotic stewardship intervention (I) reduces overall antibiotic exposure (O) in adult patients hospitalized in acute-care wards of secondary and tertiary care centers (P) compared to no such intervention ("standard-of- care") (C) over a one year time period (T) (the letters refer to the corresponding constituents of the PICOT framework).

The primary objective of the study is to use the methodological rigor of a parallel group, cluster-randomized, controlled superiority trial in three Swiss hospitals to answer the primary research question. Secondary objectives are to assess the impact of the intervention on quality of antibiotic use, patient, microbiologic and economic outcomes.

The primary outcome will be the difference in overall systemic antibiotic use measured in days of therapy (DOT) per admission based on administration data recorded in the electronic health record (EHR) over the whole intervention period. Secondary outcomes will include qualitative and quantitative antimicrobial use indicators (including non-HIV antivirals and antifungals), economic outcomes and key clinical and microbiologic indicators and patient safety indicators such as changes in readmission rates, need for intensive care and mortality.

The study hypothesis is that the multimodal intervention is superior to standard-of-care regarding the primary outcome, i.e. that the intervention leads to a statistically significant reduction in overall antibiotic use expressed as days of therapy per admission compared to no such intervention ("standard-of-care" antibiotic stewardship).

Inclusion Criteria:

CLUSTER (WARD) LEVEL

• Acute-care wards with at least 150 admissions/year
• Use of a computerized physician order entry system (CPOE)

PHYSICIAN LEVEL * All physicians involved in antibiotic prescribing decisions in the participating wards

PATIENT LEVEL

* All patients hospitalized in the participating wards

Exclusion Criteria:

CLUSTER (WARD) LEVEL

• Emergency room
• Outpatient clinics
• Overflow wards
• Absence of a matchable wards with regard to specialty and baseline antibiotic use
• Hematopoietic stem cell

PHYSICIAN LEVEL * None

PATIENT LEVEL

* None

Masking for care providers and investigators is unfortunately not feasible. Outcome assessors and data analysts will be blinded to the study arm allocation.