ChatGPT deprescribing decisions varied along activities of daily living status, cardiovascular disease history, and medication type
By Lori Solomon HealthDay Reporter
WEDNESDAY, April 24, 2024 (HealthDay News) — ChatGPT shows promise in polypharmacy management and deprescribing medications among the elderly, according to a study published online April 18 in the Journal of Medical Systems.
Arya Rao, from Harvard Medical School in Boston, and colleagues evaluated ChatGPT performance in polypharmacy management via its binary (yes/no) deprescribing decisions in standardized clinical vignettes.
The researchers found that in yes/no binary deprescribing decisions, ChatGPT universally recommended deprescribing medications regardless of activities of daily living (ADL) status in patients with no overlying cardiovascular disease (CVD) history. Among patients with CVD history, ChatGPT answers varied by technical replicate. The range for total number of medications deprescribed was 2.67 to 3.67 (out of seven) and did not vary with CVD status. However, the number of medications deprescribed increased linearly with severity of ADL impairment. ChatGPT preferentially deprescribed pain medications.
“Our study provides the first use case of ChatGPT as a clinical support tool for medication management,” senior author Marc Succi, M.D., from Medically Engineered Solutions in Healthcare Incubator, Innovation in Operations Research Center in Boston, said in a statement. “While caution should be taken to increase accuracy of such models, artificial intelligence-assisted polypharmacy management could help alleviate the increasing burden on general practitioners. Further research with specifically trained artificial intelligence tools may significantly enhance the care of aging patients.”
Abstract/Full Text (subscription or payment may be required)
Copyright © 2024 HealthDay. All rights reserved.