4 Responses

  1. mcityrk
    mcityrk June 26, 2014 at 9:13 am | | Reply

    This looks to me like an updating of the Clarke grid with the addition of color to signify A, B, C, D, E regions and a gradient of each color across each region going from most useful to least useful accuracy.

  2. Mary Dexter
    Mary Dexter June 26, 2014 at 10:53 am | | Reply

    Left side reads “Measured bg”

  3. John Walsh, PA, CDTC
    John Walsh, PA, CDTC August 12, 2014 at 3:29 pm | | Reply

    Like previous error grids, this one again places people on insulin at high risk, especially for hypoglycemia. The reason error grids fail is that opinion, even from experts, does not match logic.

    If you look at the top of the error grid at the left edge of the middle green color, a glucose meter reading of 600 on the left would actually be a real of 400 at the bottom for the YSI lab equivalent value. This supposedly “safe” area in green would give an excess dose of insulin on a bolus calculator or sliding scale sufficient to lower your glucose 200 more mg/dl than you need. If your target glucose happens to be 100 mg/dl, guess where you end up.

    Logic tells you that an error of more than 30 or 40 mg/dl is the maximum error you would prefer in today’s meters and eventually CGMs. Some of today’s meters already have an accuracy of +/-10%. With these meters, a glucose reading of 400 mg/dl will almost always be within 40 mg/dl of your true glucose. With this meter and a target of 100 mg/dl, you would end up between 60 to 140 mg/dl after an insulin correction dose. This degree of error is quite safe for insulin dosing, at least for values of 300 mg/dl or less.

    Error grids should be ensconced where they belong into diabetes museums! They are simply no longer needed and have luckily been ignored by the FDA for some time for logical reasons.

    Happy dosing to all!

Leave a Reply