Randomized controlled trials and other prospective studies for novel interventions in people with diabetes have traditionally reported HbA1c (A1C) values, an estimate of the mean blood glucose for the previous 3 months. This is based on the Diabetes Control and Complications Trial’s established correlation between A1C and risk of microvascular complications. Change in A1C has been used by regulatory agencies for more than 30 years when evaluating diabetes medications’ efficacy. However, with the increasing development of continuous glucose monitoring (CGM), researchers have been increasingly interested in using CGM devices in prospective randomized clinical trials (RCT) to complement A1C and further assess the effects of therapeutic interventions. The diaTribe Foundation invited CGM experts with experience using CGM in clinical trials, FDA, NIH, ADA, and others to become part of this consensus initiative. People with diabetes were also invited. Participation was voluntary and not remunerated. Consolidated draft consensus recommendations and manuscript were developed. Draft materials were then the focus of a meeting in Barcelona, Spain just prior to the 15th ATTD on April 25, 2022. All members of the writing group attended the consensus conference either in person or virtually. The discussion and outcomes from the meeting were used to further refine and revise the consensus recommendations and subsequent manuscript. The final draft was then approved by all members of the writing group. International consensus recommendations include how to optimize CGM-derived glucose data collection in clinical studies: CGM device selection for RCT use, how to report time in range and other metrics, application of CGM sensors in RCT, classification of hypoglycemia or hyperglycemia in RCT, clinically meaningful differences, and CGM data inclusion in the final analysis set. The consensus statement provides recommendations on how to optimize CGM-derived glucose data collection in clinical studies, including CGM-derived glucose metrics that should be used by all researchers and regulatory agencies.
Read full abstract