Everyone needs to find their happy place with sensors. Some don’t get accurate readings and some are right on.
I used the G4 before the G6 and I never dosed off it. Never! It just wasn’t accurate enough. And with my correction being 1:100, I couldn’t take a chance.
Now with the G6 and it taking to my pump, I never test before dosing. I trust it and feel confident that if it’s off, the pump will take care of it. I have had no lows that are not explainable but I do once I awhile go higher due to length the pump shuts off, but again, easily fixed. I have come a long way in just letting it go. The only times I test now are during warmup periods if I am eating. Otherwise, I let it do it’s thing.
I am also very fortunate that I am a “usual” person with diabetes. I follow most of the norms that are set out there. So the standards usually work well for me. Of course there are always a few things that are special to my treatment plan but for the most part I do things like the systems say.
But everything with diabetes is an experiment. You just need to keep checking to discover what works and what doesn’t work for you in this moment. I love my CGM but who knows, maybe next year there will be a new company with something I like better. Fingers crossed!
Well said. I have had exactly the same experience and I have done exactly the same thing when talking to Dexcom rep when I had a product failure.
I use a Dexcom G6 with multiple daily injections (no pump; Novolog and Lantus). I check the Dexcom with a fingerstick every night to calibrate the dexcom; I write down the correction. Typically the Dexcom reading is lower than the fingerstick result. The corrections are definitely not random. It’s not uncommon to recalibrate the Dexcom upward by 50. I do use the Dexcom readings to calculate my Novolog doses. I really rely on the Dexcom between meals to see if my hunger pangs are real or not. 
People forget that’s it’s not measuring the same thing as a blood test meter and lags 15-20 minutes behind. The only way to really check is to ensure u have a constant straight line on Dex for around 15 minutes then make sure u wash hands and do a finger prick test and for me it’s usually very close. Real test is how accurate the estimated A1c is on Dex to your official A1c and for the past two years mine has been virtually spot on. Remember finger prick tests can vary as well. If I’m shooting down or up fast I do a finger test as I know the Dex has a lag time
Re: A1C: although Dexcom has been spot on with me finger prick tests (excepting compression lows and measurement during rapid changes) the A1C predicted by Dexcom is usually about 0.6% high.
Most recent A1C was 6.3% while Dexcom predicted 6.9% based upon 30 and 90 days.
You may be unaware that good alignment between CGM numbers and A1c does not occur for everyone. Like many things, it produces a bell-shaped curve. My A1c, for instance, is consistently 0.5% above my CGM prediction.
Let’s start that my range for Dexcom is low 80 , high 180.
I only bolus Dos. Usually 3-4 units never more then 5 oh Navolog. I have been using Dexcom sense the G4 days.
Remember there WILL BE A DIFFERENCE between a dexcom reading and a Glucometer reading.
Both can be correct. I usually don’t cal. the dexcom. If it’s with in 10mg/DL of the Glucometer . I have found over the years that it’s best to follow the arrow on the dexcom as it will tell you which was you blood sugar is trending.
If you cal. The dexcom off a Glucometer reading., cal. Twice and the reading will trend for awhile. It will take some trail and error to figure our how much insulin to take. Before meal. With in my body , even though I took insulin my BS will go up rapidly but then thake several hours to stabilize to a normal level.
Good luck with your ordeal , I’ve been plat the T2 game for about 20 years. Just about the time I think I have if figured out , it changes. Remember if you haven’t already done so , down load CLARITY it will graph out your readings for you.
The never having to do a finger stick is the most ridiculous thing for a lot of us. I once put on a sensor before bed and woke up to an alarm around 8 am and a reading of 185…dosed and then it dawned on me, new sensor, better do a finger check. Sure enough 110…sigh, now fully awake and to the kitchen to eat something.
In my defense I was half awake.
I just have a sneaky feeling that Dexcom pushed the no finger sticks needed to get it by Medicare to cut down the costs of what Medicare has to cover to get it approved. But even Dexcom says when in doubt pull your meter out, but you don’t supposedly need a meter?
But if I let a new sensor go without calibration it is usually reading way under what I am at, 20 or 40 points. I wait until I have an even line, i always do a comparison test at around 95-105. I have waited 24 hours to see what it will do. Since I use it to bolus I do not want to be off 10 or 20 points at those levels. Once I get it calibrated within my 5 points, it stays pretty accurate. And I am then rarely doing finger checks.
And when you do that test to see if it will fall in line? It really screws with your TIR lol…I once had a reading of 50-59 for about 12 hours when I was really in the 90’s. And then I realized what it was doing to my “very low” readings…nope not putting up with that!!!
Dexcom, in fact all CGMs are sensitive to where you put them, how hydrated you are, and if they adequately “soak in” during the warm up 2 hours.
Some things to try: 1) different insertion site. My thighs work best.
2) drink more water
3) Insert the new sensor the night before you remove your old sensor with transmitter. This lets the sensor fiber soak.
#3. Yes! Agree!!!
In my case soaking doesn’t matter enough, I still have to calibrate them, and I drink a ton of water!