Inaccuracy at Night

Elizabeth, it would be interesting to know how the DexCom calibration works. Rick once stated that DexCom’s calibration algorithm is proprietary. DexCom makes an effort to conceal the raw sensor data. If I had the raw sensor data I could reverse-engineer how it works. With any calibration algorithm the most accurate readings will be around the calibration point. If you calibrate first at 80 and later at 200 I don’t know whether you give up accuracy at 80 (= one-point calibration) or keep the accuracy at 80 and gain better accuracy at 200 (= two-point or proprietary calibration). After 8 months of using the DexCom 7+ my BG variability has come down so much that I stay below 130 for more than 90% of the time. Thus I don’t really need to care how accurate the DexCom is above 130. If my BG is drifting higher and I don’t expect that it will come back into range by itself then I correct. One thing I noticed: With a fairly new sensor high readings read too high. During the 2nd week of usage high readings read too low. I did not thoroughly study this effect. I put this into the category “Worth studying”, not into “I know something”.

Interesting…I wish Dexcom could give us more information, so we could figure out the best way to get good results throughout higher and lower ranges. It sounds like we’re all kind of winging it, which we shouldn’t have to do. Although if you’re staying <130, that’s fantastic. You’ve definitely made it work for you!

I’ve noticed the same thing, by the way, that this past week (my 2nd week), the sensor was continually lower than actual…in fact it was lower not just when my bg was higher, but also when it was in normal range. That’s why I finally switched the sensor out.

I’m doing fine during the day, but my problem again is that because of gastroparesis, I never know when my dinner’s going to spike. Could happen an hour after eating, or 6 hours, and there are many times I’ll shoot into the 200s overnight for no discernible reason. That’s why I finally started on CGM…But if it’s not going to be accurate for me, then there’s hardly any point. I’ll do some experimenting with this new sensor, and see if I can get more accurate results. Really hope so, because I’m loving seeing the trending during the day!

There is a good chance that I have gastroparesis too. I heard about this condition only a couple of months ago. What I observed was that I would eat a carb and fat loaded dinner at 7pm and my BG would be kind of OK until 2am. Then it would skyrocket and enormous amounts of insulin would have almost no impact. It was not a steady climb, more like an explosion at 2am. When a sensor has earned my trust I typically treat without confirming with a finger prick. During a normal night, if my BG drifts upwards slowly, I just bolus 1 unit when the 120 alarm goes off. If it goes off again after 1 hour I bolus again. When I know that I was bad and my BG takes off I bolus up to 6 units every 2 hours. I notice that this detailed plan gives the impression that I bolus a lot at night. That is not the case. Most nights my BG stays below 120 and I sleep through the night without any alarm going off.

Interesting…I don’t know if mine is a climb or an explosion, just know that I’d wake in the morning in the 2 and 300s a couple of times a week. It got to the point where I’d set an alarm at 2AM, and when I found myself in the 200s I’d bolus. The CGM does work now in that it’ll wake me when it thinks I’m at 160 (although I’m actually at least 220.) So I’m able to catch those couple of nights a week, but sleep through the other nights. What I really want, of course, is for it to warn me when I’m rising, before I get up that high. I’ve lowered my high limit to 140, so that should help. If not, I might lower it again to 120, although it would be nice not to be woken up with a high alarm when my bg actually is only 120 or less!

We’ll see…So far I haven’t been above 100 all day. Will try to eat a higher carb lunch to see if I can get into the 140s or 150s, to set a higher calibration point.