
Artificial intelligence (AI) chatbots, such as ChatGPT, can be used by clinicians as a resource for disease management and education, but researchers of a recent study urge users to consider the “gaps” seen in managing individuals with type 2 diabetes mellitus (T2DM) and metabolic syndrome (MetS).
This study considered three nutrition management-related domains, namely dietary management, nutrition care process (NCP), and menu planning (1,500 kcal). Sixty-three prompts were used in ChatGPT, and two dietitians assessed the chatbot responses’ concordance with existing guidelines.
The two dietitians gave similar assessments for most conditions examined in this study. On the other hand, gaps were identified in the ChatGPT-derived outputs, including weight-loss recommendations, energy deficit, anthropometric assessment, specific nutrients of concern, and the adoption of specific dietary interventions.
There were also gaps in physical activity recommendations noted, which highlighted the chatbot’s limitations in providing holistic lifestyle interventions.
With regard to NCP, ChatGPT provided incomplete examples of diagnostic documentation statements and showed gaps in the monitoring and evaluation step. For the 1,500 kcal 1-day menus, discrepancies were observed in the amounts of carbohydrates, fat, calcium, and vitamin D relative to dietary recommendations.
When it comes to clarity, both dietitians agreed that the output provided by ChatGPT was either good or excellent.
“Although ChatGPT is an increasingly available resource for practitioners, users are encouraged to consider the gaps identified in this study in the dietary management of T2DM and the MetS,” the researchers said.