A new study from the University of Liverpool has found that artificial intelligence (AI) can help make better antibiotic prescribing choices in the treatment of urinary tract infections (UTIs).
Researchers used an AI algorithm that “combines human judgement and data” to pick “the most appropriate antibiotic for each patient”, with the aim of reducing “unnecessary use of strong antibiotics and [slowing] down resistance”.
When given simulations using emergency department data, the AI chose an antibiotic that “correctly targeted the urinary pathogen” in 83.6% of cases, while human prescribers were correct in 79.5% of cases.
Read more: Pharmacy First: Report reveals ‘no change’ to UTI antibiotic resistance
The algorithm also picked “more correctly targeted” World Health Organization (WHO) Access category antibiotics in 75.6% of cases, as opposed to just 11.9% by human prescribers, the study said.
Access category antibiotics are those with fewer side effects, lower costs and lower potential to lead to antimicrobial resistance (AMR).
Researchers added that the AI chose the most appropriate oral antibiotic treatment in 69% of cases, while humans were right 22.6% of the time - and the AI was 52.6% less likely than humans to prescribe intravenous antibiotics.
The study concluded that the AI algorithm “could improve antibiotic prescribing decisions by combining human judgement with data-driven probability predictions”.
Innovation needed
Medical microbiology consultant at the University of Liverpool Dr Alexander Howard said AMR is “one of the top global public health and development threats”.
“In an era where antimicrobial resistance continues to increase, innovative solutions to facilitate precision use of antimicrobials are required - our utility-based system may present such a solution,” he added.
“Further research is now needed across a range of global settings to ensure the results apply more widely,” he said, “however, this study shows that using AI alongside doctors’ expertise could improve antibiotic prescribing, help fight resistance and make treatments safer and more convenient for patients”.
Read more: Boots and Oxford University to test UTI trial recruitment
It comes after national surveillance data in November found “no change in resistance” to an antibiotic used for UTIs following its introduction to the Pharmacy First service.
The UK Health Security Agency (UKHSA) report also found that primary care accounted for almost 80% of all antibiotics prescribed in 2024, of which just 4% came from Pharmacy First.
And it follows a recent survey of medical professionals revealing that knowledge of AMR was the “highest” amongst doctors and pharmacists, but the “lowest amongst” pharmacy technicians.
