Related Experiment Video
Updated: Aug 26, 2026

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
When treatment becomes risk: polypharmacy and inappropriate prescribing in cognitive impairment in 2379 consecutive
Xavier Morató1,2, Adrian Hinojosa-Calleja1, Sara Jofresa1
1Ace Alzheimer Center Barcelona, Universitat Internacional de Catalunya, Barcelona, Spain.
Background:
Polypharmacy, potentially inappropriate medications (PIMs), and drugs associated with drug-induced cognitive impairment (DICI) are highly prevalent in older adults and may affect cognitive decline.
Methods:
This observational retrospective study included 2379 patients evaluated at Ace Alzheimer Center Barcelona. Medication data were extracted from electronic health records using a large language model (LLM). Polypharmacy (≥5), hyperpolypharmacy (≥10), PIMs, and DICI drugs were analyzed across cognitive stages, Alzheimer's disease biomarkers, ApoE subtype, and estimated glomerular filtration rate.
Results:
Among patients with cognitive impairment (Clinical Dementia Rating [CDR] ≥ 0.5), 67% presented polypharmacy and 21% hyperpolypharmacy. Higher odds of polypharmacy were observed across the CDR scale, while higher education showed a protective effect. Altered renal function was found in 20.8% of patients with polypharmacy. Indeed, significant differences were found between amyloid beta-positive and negative (Aβ+/Aβ-) subjects and between the whole population and subjects participating in clinical trials.
Discussion:
Polypharmacy is extremely common in patients attending memory clinics. LLMs could facilitate systematic structured medication reviews.
Related Concept Videos
Drug Dosing: Geriatric Patients
Drug Therapy
Antianxiety Medications
Cognitive Enhancers: Cholinesterase Inhibitors and NMDA Receptor Antagonists
Alzheimer's Disease: Treatment
Pharmacodynamics in Geriatric Patients: Effects of Age
Psychosis: Goals of Pharmacotherapy
