Original Research
Monday, October 14, 2024
01:00 PM–02:30 PM
Abstract
Introduction: Generative AI (artificial intelligence) holds the promise of transforming clinical care. Utilizing algorithms and datasets, AI can analyze intricate medical information to enhance individualized clinical management for chronic conditions, such as type 2 diabetes. This study aims to compare clinical recommendations for type 2 diabetes management generated by two AI platforms — OpenAI's ChatGPT-4 and Google Bard — based on the American Diabetes Association’s Standards of Medical Care – 2021.
Research Question or Hypothesis: We hypothesized both platforms can achieve internal consistency within each platform and external consensus between platforms and with a clinical pharmacist’s decisions.
Study Design: AI-generated responses were reviewed and compared to the clinical pharmacist’s assessment and plan using a qualitative approach.
Methods: A complex clinical case was chosen, involving a patient with co-morbid conditions and complications related to type 2 diabetes. The note was edited before being analyzed by each AI platform to ensure clarity and HIPAA compliance. The case was analyzed by both AI platforms three times. We then compared the AI-generated recommendations with the assessment and plan written by the clinical pharmacist.
Results: The comparison of ChatGPT-4 and Bard revealed both AI platforms consistently emphasized pharmacotherapy and hypoglycemia prevention. ChatGPT-4 focused on insulin therapy reevaluation and glucagon education, while Bard recommended medication reviews and hypoglycemia monitoring. ChatGPT-4 suggested continuous glucose monitoring (CGM), whereas Bard incorporated HbA1c levels and a GLP-1 receptor agonist for better glycemic management. Both platforms agreed on blood pressure and statin therapy for reducing complication risks. The clinical pharmacist’s note aligned with these recommendations.
Conclusion: Both AI platforms demonstrated different degrees of internal consistency in recommendations. Regarding external consensus, both platforms focused on secondary cardiovascular prevention. The alignment of AI-generated recommendations with a clinical pharmacist’s note suggests that AI can potentially augment clinical pharmacy practice in managing type 2 diabetes and complication reduction.
Presenting Author
Clipper Young PharmD, MPHTouro University California
Authors
Shirley Wong PharmD
Touro University California