Assessment of Large Language Models for Enhancing Diabetologist-Developed Personalized Treatment Plans in Complex Type 2 Diabetes

2025/08/20 by Michelle Wang, Madhumita Sushil, Christopher Y.K. Williams +8 · 1 citation

paper · doi:10.2337/cd25-0015

published in Clinical Diabetes 43(4), 545-553 (American Diabetes Association)

crossref issued 2025/08/20 · crossref published 2025/08/20 · crossref published-online 2025/08/20 · crossref created 2025/08/20 · crossref published-print 2025/10/15 · crossref deposited 2026/06/18 · crossref indexed 2026/08/01

Abstract

Treatment decisions for type 2 diabetes are complex and multifactorial. This study evaluated the utility of OpenAI’s generative pre-trained transformer-4 (GPT-4) large language model in enhancing treatment plans for people with diabetes. The authors assessed GPT-4’s ability to extract and critically evaluate diabetologist-developed treatment plans from patients’ clinical notes. GPT-4 accurately discerned personalized treatment goals but performed suboptimally when identifying plans that could be further improved and personalized. However, 40% of GPT-4’s reasoning and explanations were still considered helpful by experienced academic diabetologists, suggesting a potential broad benefit to less expert physicians, such as those in primary care settings.

Citations

• DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines
• Global, regional, and national burden of diabetes from 1990 to 2021, with projections of prevalence to 2050: a systematic analysis for the Global Burden of Disease Study 2021
• Capabilities of GPT-4 on Medical Challenge Problems
• GPT-4 Technical Report
• Precision Medicine in Type 2 Diabetes: Using Individualized Prediction Models to Optimize Selection of Treatment