This research aims to compare the accuracy of ChatGPT, an advanced large language model, with that of clinical pharmacists in identifying PIMs among older adult cancer patients. Geriatric oncology, potentially inappropriate medications (PIMs), ChatGPT, artificial intelligence in medication review
Conditions
Interventions
ChatGPT is a natural language processing (NLP) model developed by OpenAI, designed to process text and generate human-like responses using extensive data derived from large-scale internet sources. Sin
Experimental Other,No Intervention Other
ChatGPT,Standard practice
Sponsors
None listed
Eligibility
Sex/Gender
All
Age
65 Years to No maximum
Inclusion criteria
Inclusion criteria: Participants will be enrolled according to the following criteria 1. Patients diagnosed with any type of cancer. 2. Age over 65 years at the time of data collection. 3. Availability of complete clinical and medication data in the Ramathibodi Hospital electronic medical record (EMR) system
Exclusion criteria
Exclusion criteria: 1. Patients with medications that cannot be defined, documented, or classified within the electronic medical record system.
Design outcomes
Primary
| Measure | Time frame |
|---|---|
| Accuracy 12 months identifying number of potentially inappropriate medications (PIMs) | — |
Secondary
| Measure | Time frame |
|---|---|
| time required minute time required for medication review per patient | — |
Countries
Thailand
Contacts
Public ContactPhichai Chansriwong
Ramathibodi Hospital
Outcome results
None listed