Many older adults deal with polypharmacy, taking five or more medications simultaneously, which raises the risk of harmful drug interactions.
Sorting through these medications to reduce unnecessary ones, known as deprescribing, can be complicated and time-consuming for primary care practitioners, especially with limited staff. This challenge has spurred a search for efficient tools to manage polypharmacy.
A recent study by researchers at the Mass General Brigham MESH Incubator suggests that ChatGPT, an artificial intelligence (AI) chatbot, could be a valuable tool in managing polypharmacy and deprescribing. Published in the Journal of Medical Systems, this study marks the first application of AI models in medication management.
To test its usefulness, the researchers presented ChatGPT with various clinical scenarios involving elderly patients on multiple medications. Each scenario differed in the patient’s history of cardiovascular disease (CVD) and level of impairment in daily activities.
ChatGPT consistently recommends deprescribing medications for patients without a history of CVD when asked yes or no questions about reducing prescribed drugs. However, it became more cautious when CVD was a factor, often suggesting to keep the medication regimen unchanged. Interestingly, the severity of impairment in daily activities didn’t seem to sway its decision-making.
The researchers also observed that ChatGPT tended to prioritise deprescribing pain medications over other types of drugs like statins or antihypertensives. ChatGPT’s responses also varied across different chat sessions when presented with the same scenario, indicating a potential inconsistency in the trends it was trained on.
With more than 40% of older adults experiencing polypharmacy and an increasing number of specialists involved in their care, primary care providers face a growing challenge in managing medications. An effective AI tool like ChatGPT could offer valuable support in this aspect of care.
“Our study provides the first use case of ChatGPT as a clinical support tool for medication management,” said Dr Marc Succi, Associate Chair of Innovation and Commercialization at Mass General Brigham Radiology and Executive Director of the MESH Incubator.
“While caution should be taken to increase accuracy of such models, AI-assisted polypharmacy management could help alleviate the increasing burden on general practitioners. Further research with specifically trained AI tools may significantly enhance the care of aging patients,” Dr Succi said.
Ritchelle is a Content Producer for Healthcare Channel, Australia’s premier resource of information for healthcare.
