Field Tips: Part 4

Field tip #16: The software for most modern meters can display data in several different ways. Modal day is an x/y graph that displays the time of day left-to-right, and blood glucose levels top-to-bottom, for almost any number of days you choose. I find this type of graph to be the most useful. Trend graphs are a bit like the old connect-the-dots game. They can be great for noticing sudden changes that may point to problems like expired insulin or a urinary tract infection, but are less useful in studying routine situations. Some software will also compare weekdays to weekends. This feature can help you see if your therapy is well suited for the change of eating habits and activities that often takes place on weekends. And, of course, most meters still include a digital version of the traditional logbook, with numbers shown in rows and columns.

Field tip #17: Check out the “n.” On most graphs, a simple legend can be found at the bottom, with “n=” followed by a number (n=3, n=17, n=48). This “n” stands for “number,” and it tells you how many blood glucose readings were used to build the graph. The higher the number, the more you can rely on what the graph is showing you.


Field tip #18: Just for fun, as soon as you get your next HbA1c test result, compare it to your meter’s average. HbA1c can be converted to estimated average glucose, or eAG, using this formula: 28.7 x HbA1C — 46.7 = eAG. Use the maximum number of days the meter will let you include in the average. HbA1c represents approximately a three-month average of your blood glucose levels, but it is biased toward your most recent six weeks. Unless you’ve seen some major changes in blood glucose control lately, your meter’s average and your eAG should be in the same ballpark. If not, you have something to investigate.

Field tip #19: Try it for yourself. Log your blood glucose readings on either paper or a software app for one week, then ask yourself: Has your awareness of your blood glucose changed?

Field tip #20: A word of warning: It’s important not to jump to any conclusions based on a sample of one, or even two or three. You need to be able to look at several similar incidents; the only way to spot a trend is to look at data over time.