lower= flatter but yeah, I agree the number is more meaningful over longer periods of time from day to day.
I’m not totally sure I’d agree that jumping around is as bad a staying high all the time? I agree that it’s not ideal but, in terms of complications I would think that any low normal or even “not that high” time would be preferential to being high all the time? OF course, I don’t have a study to back me up.
There have been newer studies that point to that. I lost all my links when my computer crashed but I posted some here. Remember Pete on the other board? He posted a Science Daily one there once.
Agreed, that’s why I said with a ‘reasonable average bg’, or something like that. A sd with an average of 200 isn’t appealing!
The variance (like the mean/average) only starts to become meaningful when you get like 30 samples. Really, that turns out to be 30 days of data since we get big variations during the day and independent big variations from day to day.
I agree that for people that are not testing very much, the data is not meaningful. My SD is based on 18 tests a day so means more than someone testing 4 times a day. My Dexcom tests every 5 minutes so a days worth of SD has 1440 readings in it unless I am away from my receiver (which happens when I am in the shower). Natalie was looking at her CGMS - I don’t know many times hers gets recorded. I know that the Navigator tested every 3 minutes but for reports, it only used the readings every 10 minutes so the Dexcom actually gave me more SD readings than the Navigator.
I don’t have serious lows either so I think in our cases the lower the A1c the better. Above 7 may be necessary for some. Dr. R Bernstein claims that on very low carb you can do better than 6.
There is a nice graph in “Pumping Insulin” which comes from the DCCT study…see Firgure 2.5 in the 4th edition, for example…It shows that complications are lower (even dramatically so) at the same A1C if the variability (SD) is less.
There is also a reference to a study by Dr. Irl Hirsch at UW.