A new technology could be a game-changer for senior living and nursing home residents who have allergic reactions to antibiotics.
JAMA Network Open published a research report on Wednesday, noting a newly developed algorithm demonstrated high sensitivity in detecting antibiotic allergic reactions in patients receiving antibiotic prophylaxis for cardiac implantable electronic device (CIED) procedures.
For the study, researchers developed and tested a set of electronic health record variables that could indicate allergic reactions to the antibiotics typically used to prevent surgical-site infections. They tested Veterans Affairs patients who had undergone CIED procedures and received peri-procedural antibiotic prophylaxis.
Researchers used entries in the VA hospitals’ Allergy Reaction Tracking system and keyword detection in clinical notes to create an algorithm that pinpoints the probability of allergic reactions and patients’ general sensitivity to antibiotics, identifying antibiotic allergic-type reactions with an estimated probability of 30% or more.
For nursing home providers, the study’s findings show the potential of using technology to improve resident care, and how implementing such technologies can keep residents safe, observers say.
“These findings highlight … the need for innovative strategies to provide audit and feedback to clinicians about patient harm caused by unnecessary antibiotic exposures,” study authors wrote.
This algorithm is just one of the latest instances of scientific research and technology coming together to predict, and potentially prevent, adverse reactions and diseases.
For example, scientists from the University of New South Wales Sydney, in collaboration with researchers from Boston University, recently developed an AI-based tool they say can predict the probability of Parkinson’s disease with up to 96% accuracy and up to 15 years before a clinical diagnosis, based on an analysis of chemicals in the blood. Researchers plan to make the tool, which is easy to use and generates results in 10 minutes on a conventional laptop, publicly available to other researchers or clinicians who would like to use machine learning for disease diagnosis.
