Pleus Stefan, Flacke Frank, Sieber Jochen, Haug Cornelia, Freckmann Guido
1 Institut für Diabetes-Technologie Forschungs- und Entwicklungsgesellschaft mbH an der Universität Ulm, Ulm, Germany.
2 Sanofi, Industriepark Höchst, Frankfurt am Main, Germany.
J Diabetes Sci Technol. 2017 Nov;11(6):1226-1230. doi: 10.1177/1932296817707292. Epub 2017 Apr 26.
Graphical presentation of blood glucose monitoring systems' (BGMSs) accuracy typically includes difference plots (DPs). Recently, 3 new approaches were presented: radar plots (RPs), rectangle target plots (RTPs), and surveillance error grids (SEGs). BGMS data were modeled based on 3 scenarios that can be encountered in real life to highlight strengths and limitations of these approaches. Detailed assessment of BGMS data may be easier in plots with individual data points (DPs, RPs, SEGs), whereas RTPs may facilitate display of large amounts of data or comparison of BGMS. SEGs have the advantage of assessing clinical risk. The selection of a specific type depends mostly on the kind of information sought (eg, accuracy in specific concentration intervals, lot-to-lot variability, clinical risk) as there is no "absolute best" approach.
血糖监测系统(BGMS)准确性的图形表示通常包括差异图(DP)。最近,提出了三种新方法:雷达图(RP)、矩形目标图(RTP)和监测误差网格(SEG)。基于现实生活中可能遇到的三种情况对BGMS数据进行建模,以突出这些方法的优点和局限性。在具有单个数据点的图(DP、RP、SEG)中,对BGMS数据进行详细评估可能更容易,而RTP可能有助于大量数据的显示或BGMS的比较。SEG具有评估临床风险的优势。特定类型的选择主要取决于所寻求的信息类型(例如,特定浓度区间的准确性、批次间变异性、临床风险),因为不存在“绝对最佳”的方法。