None listed
Conditions
Interventions
Intervention1: To compare the quality of responses generated by seven large language models to standardized pre-anesthetic patient concerns.: A standardized set of pre-anesthetic patient questions h
Sponsors
Government Institute of Medical Sciences
Eligibility
Inclusion criteria
Inclusion criteria: A standardized set of pre-anesthetic patient questions have been developed based on common clinical practice and literature. These Identical prompts will be entered into each LLM and the responses will be collected verbatim without modification and stored in a Google survey form. This survey form will include questions on the accuracy, readability, and completeness of the LLM generated answers. The responses will be recorded on the Likert s scale, in which respondents had to answer on the scale of one to five, with one being the worst response and five being the best-accepted response . This survey form will be sent to two experienced anesthetists, for validation of the survey questions. The final version of the Google survey form will be then sent to 25 anesthetists, chosen on an arbitrary basis. Responses with greater than 3 will be taken as appropriate.
Exclusion criteria
Exclusion criteria: other Large learning models
Design outcomes
Primary
| Measure | Time frame |
|---|---|
| A comparative understanding of strengths and limitations of current LLMs in pre-anesthetic patient education.Timepoint: baseline | — |
Secondary
| Measure | Time frame |
|---|---|
| Identification of ethical risks associated with AI-generated perioperative information.Timepoint: baseline; Evidence-based recommendations for clinicians & institution regarding responsible use of LLMs.Timepoint: baseline | — |
Countries
India
Contacts
Public ContactNazia Nazir
Government Institute of Medical Sciences
Outcome results
None listed