Bookmarks
Optimizing antimicrobial use in multi-morbid patients through intelligent clinical decision support
Safe People
Organisation name
Imperial College London
Organisation sector
Academic Institute
Applicant name(s)
William Bolton
Funders/ Sponsors
Alison Holmes
DEA accredited researcher?
Unknown
Sub-licence arrangements (if any)?
No
Safe Projects
Project ID
NIBDAPC_2022_0016
Lay summary
Antibiotics are drugs that treat bacterial infections; however, the overuse of antibiotics is driving antimicrobial resistance (AMR) (which is when a bacterial infection is difficult to treat with an antibiotic). AMR is a global challenge that promises to have significant negative effects on health and society. One way to address AMR is to only use antibiotics to treat bacterial infections instead of viral infections, as infections caused by virus do not improve with antibiotics. This can be done through artificial intelligence (AI) where software used by computers mimic aspects of human intelligence. This is a powerful technology that enables us to understand data and make predictions using computers. AI is increasingly being used within medicine and has great potential to provide meaningful benefit with regards to infections and antibiotics. Despite a strong association being shown between other medical conditions and different infection-related risks and outcomes, to date limited AI research has focused on antibiotic use in patients with more than one long-term health condition. This project will use health data to understand the use of antibiotics in patients with more than one long-term health condition and predict patient outcomes and the most appropriate antibiotic treatment through using AI. Ultimately such technology will be incorporated into clinical decision support systems (CDSSs) to provide information to healthcare professionals so they can make good clinical decisions on antibiotic use. November 2023 Update Artificial intelligence (AI) technology to understand patients’ historical medical conditions has been developed. It has been shown to be good at predicting patient death and is able to find historical patient cases that are similar to any patient of interest. Healthcare professionals can use this to learn about previous clinical scenarios and make appropriate clinical decisions. Work on using this technology to learn how to improve antibiotic use, prevent resistance and improve patient outcomes is ongoing.
Public benefit statement
Multi-morbidity and antimicrobial resistance (AMR) are significant challenges to healthcare. Co-morbid conditions put individuals at a high risk of developing an infection with a greater chance of poor outcomes, including increased length of hospital stay, chance of readmission and risk of death. These patients also often fail to be treated appropriately due to polypharmacy and a lack of evidence. Our work will focus on researching intelligent clinical decision support systems (CDSS) to provide clinicians with the information they need to help support antimicrobial decision making in this complex patient population.
Request category type
Public Health Research
Other approval committees
Project start date
17/11/2022
Latest approval date
03/10/2022
Safe Data
Dataset(s) name
ICHT COVID-19 Research Dataset
Data sensitivity level
De-Personalised
Common Law Duty of Confidentiality
Not applicable
National data opt-out applied?
Not applicable
Request frequency
One-off
Release/Access date
17/11/2022
Safe Setting
Access type
TRE