Antimicrobials (drugs that kill or stop the growth of microorganisms including bacteria, thereby treating infections) commonly used to treat patients with infections are becoming less effective over time as bacteria develop resistance to them. Antimicrobial usage itself can lead to development and spread of antimicrobial resistance. Antimicrobial resistance is now a major threat to patient safety. To conserve the effectiveness of antimicrobials the investigator need to develop ways to use them more sensibly healthcare professionals who diagnose and treat infections must be able to access antimicrobial guidelines and test results at the patient bedside. This needs to be provided rapidly and with support to make sure that the decisions on prescribing antimicrobials are the best that can be made.

Prototype software to achieve this has been developed through collaboration between healthcare professionals and biomedical engineers. This prototype software (run on a mobile device) retrieves patient results from various laboratory and clinical databases (securely within the Trust firewall) and displays this to the clinician making the prescribing decision. Furthermore a machine learning algorithm is applied to the data, and similar anonymised historical cases (and the antimicrobials prescribed and the clinical outcomes) are also displayed to the clinician to further inform their decision making. The prototype has been designed for use in intensive care, where the risk of infection is high, but through the research project detailed here, the software will be developed and validated across other areas of hospital patient care. Furthermore there is a key need to engage patients with how decisions are made around antimicrobial prescribing. The investigator propose to adapt the prototype to meet these needs. This system should improve patient safety and help preserve the effectiveness of existing antimicrobials

Inclusion Criteria:

(i) healthcare professionals for evaluation phases: Have read the PIL and consent to participate in the study

(ii) patients for whom the clinician chooses to use the POC DSS as a resource when prescribing antimicrobials: Adult patients > 18 years old Being managed for infection outside of the critical care setting in Imperial College Healthcare NHS Trust Deemed appropriate for management with POC DSS by attending physician Prescribed antimicrobial agents outside of the critical care setting in last 5 days

(iii) Prescriber / healthcare professional for using POC DSS: Trained Healthcare Professional Working within wards under assessment Deemed suitable for recruitment by senior member of their team

Exclusion Criteria:

(i) healthcare professionals: Do not wish to participate in the study Working across wards which is acting as a control ward Deemed no suitable for recruitment by a senior member of their team Non-permanent member of the Trust Information governance training not up-to-date

(ii) patients recruits Critical care patients Paediatric patients < 18 years old Deemed not suitable for management using POC DSS by attending physician On palliative care, end of life pathway Prisoners / young offenders in custody of HM Prison Service Involved in current research or have recently been involved in any research prior to recruitment (last 3 months)

Patients may be asked to participate in the evaluation of the the patient module.

For prescriber recruitment to use the application the following methods of identifying participants will be undertaken:

Phase 1: Participants will be identified based on their roles within the infectious disease team (level and areas of the hospital covered) and will be directly approached by the study PI / clinical infection lead to request their support with this phase of the project.

Phase 2: Participants (physicians) will be identified based on the wards that they work in.

Data sharing will be considered on individual request in accordance with local information governance and research ethics approvals.

Number of participants that consented to participate

Not assessed

Adverse events were not assessed as this was not an observational study.