Clarke Error Grid Analysis (CEGA) is an essential tool for checking the clinical accuracy of self-monitoring of blood glucose (SMBG) monitors. Currently, there is no freeware (software) that can be used to generate the error grid.
If we think about statistical analysis of data, the first thing comes to our mind is—which software to use—SPSS, GraphPad, R, or any other? The machine has now taken all the burden of our complex statistical analysis. However, in many resource-limited settings, the costly statistical software package is still not available. Recently, we faced difficulty in the conduct of a CEGA.1 We were searching about it on the internet; however, we could not find a solution which we can adopt. Then, we decided to do it manually. First, we took a graph paper and marked it according to the original CEGA plot.2 Then, we plotted the glucose levels (X-axis for reference method and Y-axis for SMBG) on the graph paper and tabulated the numbers according to different zones (viz., A, B, C, D, and E).
We faced the second challenge while making a digital image of the graph. Simply taking a picture with a smartphone was not of sufficient quality. Furthermore, scanning a large graph paper was not feasible in our setting. Hence, we used Microsoft Excel. We entered the data (reference method in the first column and SMBG in the second column) and inserted a scatter plot. We formatted the X- and Y-axis from 0 to 450. Then, we manually inserted lines to make the grid. Later, we converted it to a portable document format (PDF) and generated a JPG image from the PDF. Figure 1 shows an example of the graph with grid lines with some random data as red dots.
Figure 1.
An example of a Clarke error grid generated with scatterplot and inserted lines in Microsoft Excel.
Anyone facing a similar problem in any resource-limited settings can carry out CEGA manually on graph paper and can generate a manuscript-ready digital image with the help of spreadsheet software. Furthermore, Parkes Error Grid analysis, another method of reporting accuracy of SMBG monitors, can also be conducted with the similar method.3
Footnotes
Declaration of Conflicting Interests: The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The author(s) received no financial support for the research, authorship, and/or publication of this article.
ORCID iD: Himel Mondal 
https://orcid.org/0000-0001-6950-5857
References
- 1. Clarke WL, Cox D, Gonder-Frederick LA, Carter W, Pohl SL. Evaluating clinical accuracy of systems for self-monitoring of blood glucose. Diabetes Care. 1987;10(5):622-628. [DOI] [PubMed] [Google Scholar]
- 2. Clarke WL. The original Clarke error grid analysis. Diabetes Technol Ther. 2005;7(5):776-779. [DOI] [PubMed] [Google Scholar]
- 3. Pfützner A, Klonoff DC, Pardo S, Parkes JL. Technical aspects of the Parkes error grid. J Diabetes Sci Technol. 2013;7(5):1275-1281. [DOI] [PMC free article] [PubMed] [Google Scholar]