Version: 1.0.0 | Published: 3 Nov 2026 | Updated: 0 days ago
Documentation
Description:
This project has a specific focus in managing the increasing burden from infections caused by bacteria that are resistant to antibiotics (or antimicrobial resistance, AMR).
In this study, the following topics are explored:
a) How advances in artificial intelligence can help to preserve the use of certain antibiotics that are needed to treat AMR infections.
b) How often antibiotics are prescribed without knowing if there was an AMR infection, and how this relates to patient outcomes.
c) How we can identify the type of patients for future trials that are needed to develop new antibiotics to treat AMR infection.
Antimicrobial resistance (AMR) is a global threat of unprecedented scale and is estimated to render many infections untreatable with current antibiotic drugs. Infection caused by AMR bacteria is clinically indistinguishable from the same bacterial species that is drug susceptible. AMR is detected in the laboratory and not in the clinic.
In earlier work, electronic healthcare records (EHRs) were used and emergency admissions for microbiologically confirmed bloodstream and urine infections to develop artificial intelligence (AI) and machine-learning algorithms (MLAs) to anticipate AMR. This approach showed potential to reduce broad-spectrum antibiotic initiation by 40% (the national target for reduction is 2%) when permitted to have the same error rate as current clinical practice.
The challenge is that many patients either survive or die from infection regardless of "effective" antibiotic use, making it difficult to assess the impact of antibiotic choice on outcomes.
Patient factors, such as age and other diseases, play a dominant role in clinical outcomes of infection, and many patients survive AMR infection (after given an antibiotic the organism is resistant to), and yet other patients die from infection despite having received an antibiotic that showed laboratory effectiveness. Between these extremes, there is an unidentified cohort of patients whereby the selection of effective antibiotic therapy can play a meaningful and observable result in survival and other key outcomes. The identification and characterisation of these patients can be used for future clinical trial design intended to show real-world effectiveness of novel antibiotic candidates.
Coverage
Spatial:
West Midlands
Provenance
Origin
Purposes:
Research cohort
Sources:
EPR
Temporal
Accrual Periodicity:
Static
Start Date:
01 January 2015
End Date:
31 December 2025
Time Lag:
Other
Accessibility
Access
Access Rights:
Information Governance and Ethics - West Midlands Secure Data Environment (https://westmidlandssde.nhs.uk/information-governance-and-ethics)
Access Service:
Data Request Process - West Midlands Secure Data Environment
(https://westmidlandssde.nhs.uk/research/data-request)
Access Request Cost:
Please email wmsde@uhb.nhs.uk
Delivery Lead Time:
2-6 months
Jurisdictions:
GB
Usage
Resource Creators:
This publication uses data from WMSDE, an ethically approved NHS Secure Data Environment (NRES Reference 24/YH/0022)
Format and Standards
Vocabulary Encoding Schemes:
- ICD10
- SNOMED CT
Conforms To:
LOCAL
Languages:
en
Formats:
CSV