RESUMO
AIMS/HYPOTHESIS: Continuous glucose monitoring (CGM) improves glycaemic outcomes in the outpatient setting; however, there are limited data regarding CGM accuracy in hospital. METHODS: We conducted a prospective, observational study comparing CGM data from blinded Dexcom G6 Pro sensors with reference point of care and laboratory glucose measurements during participants' hospitalisations. Key accuracy metrics included the proportion of CGM values within ±20% of reference glucose values >5.6 mmol/l or within ±1.1 mmol/l of reference glucose values ≤5.6 mmol/l (%20/20), the mean and median absolute relative difference between CGM and reference value (MARD and median ARD, respectively) and Clarke error grid analysis (CEGA). A retrospective calibration scheme was used to determine whether calibration improved sensor accuracy. Multivariable regression models and subgroup analyses were used to determine the impact of clinical characteristics on accuracy assessments. RESULTS: A total of 326 adults hospitalised on 19 medical or surgical non-intensive care hospital floors were enrolled, providing 6648 matched glucose pairs. The %20/20 was 59.5%, the MARD was 19.2% and the median ARD was 16.8%. CEGA showed that 98.2% of values were in zone A (clinically accurate) and zone B (benign). Subgroups with lower accuracy metrics included those with severe anaemia, renal dysfunction and oedema. Application of a once-daily morning calibration schedule improved accuracy (MARD 11.4%). CONCLUSIONS/INTERPRETATION: The CGM accuracy when used in hospital may be lower than that reported in the outpatient setting, but this may be improved with appropriate patient selection and daily calibration. Further research is needed to understand the role of CGM in inpatient settings.
Assuntos
Diabetes Mellitus , Saúde Digital , Equidade em Saúde , Humanos , Diabetes Mellitus/terapiaRESUMO
Continuous glucose monitors (CGM) provide patients and clinicians with valuable insights about glycemic control that aid in diabetes management. The advent of large language models (LLMs), such as GPT-4, has enabled real-time text generation and summarization of medical data. Further, recent advancements have enabled the integration of data analysis features in chatbots, such that raw data can be uploaded and analyzed when prompted. Studying both the accuracy and suitability of LLM-derived data analysis performed on medical time series data, such as CGM data, is an important area of research. The objective of this study was to assess the strengths and limitations of using an LLM to analyze raw CGM data and produce summaries of 14 days of data for patients with type 1 diabetes. This study used simulated CGM data from 10 different cases. We first evaluated the ability of GPT-4 to compute quantitative metrics specific to diabetes found in an Ambulatory Glucose Profile (AGP). Then, using two independent clinician graders, we evaluated the accuracy, completeness, safety, and suitability of qualitative descriptions produced by GPT-4 across five different CGM analysis tasks. We demonstrated that GPT-4 performs well across measures of accuracy, completeness, and safety when producing summaries of CGM data across all tasks. These results highlight the capabilities of using an LLM to produce accurate and safe narrative summaries of medical time series data. We highlight several limitations of the work, including concerns related to how GPT-4 may misprioritize highlighting instances of hypoglycemia and hyperglycemia. Our work serves as a preliminary study on how generative language models can be integrated into diabetes care through CGM analysis, and more broadly, the potential to leverage LLMs for streamlined medical time series analysis.