Abstract Body

Systems for continuous glucose monitoring (CGM) have become an essential tool in the therapy of people with diabetes that provides information to the patent and allows to calculate CGM derived parameters like TiR etc. Accuracy of CGM systems has since become a topic of discussion, promoting the mean absolute relative difference (MARD) as one of the key characteristics of a CGM system. Experiences from practice as well as from structured head-to-head studies, however, have shown that different CGM systems may exhibit considerable variations in clinically relevant accuracy parameters even if they claim similar MARD values. On the one hand, this shows that the MARD alone is inadequate to fully characterize the accuracy of CGM system, which has led to the proposal of several alternatives. On the other hand, it shows that the apparent accuracy of a CGM system can be highly dependent on various factors related to the design and evaluation of the respective performance study. Among these factors are the selection of study participants as well as the characteristics of the comparator measurements (e.g. range and rate of change) and their traceability. This leads to a broad range of reported accuracy and makes comparison of different systems indeed difficult because all aspects of the respective study design have to be taken into consideration which requires extensive background knowledge that cannot be expected from practitioners or users.
Given the importance of CGM in current diabetes therapy, reliable and transparent performance declaration is essential which can be reached by standardized and scientifically reasonable study procedures. In his talk, Dr. Freckmann will provide an overview on past and current efforts, including the work of the IFCC working group on CGM, to achieve this goal of standardization.