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

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. Key Points · GPT-4 has demonstrated potential in retaining complex concepts crucial for developing type 2 diabetes personalized treatment plans. This study evaluated the utility of GPT-4 in enhancing treatment plans for actual patients with diabetes. · GPT-4 accurately discerned personalized treatment goals but was suboptimal in identifying diabetologist-developed treatment plans that could be further improved and personalized. Even so, 40% of GPT-4’s reasoning and explanations were considered helpful when evaluated by blinded expert academic diabetologists. · These findings suggest the potential of broad benefit of GPT-4 in enhancing treatment plans to less expert physicians, such as those practicing in primary care settings.