Dexcom g5 accuracy off!

#1

Is anybody else’s dexcom g5 accuracy off. It seems that when I bolus and eat food that is slow absorbing it will tell me I’m low. I will test and my meter will say normal. It even goes as low as saying it’s an urgent low. I am so frustrated with the accuracy of it.

#2

All my dexcoms have frequently been very off in both directions, I have g5 now. But it has a lot of accuracy as well and trends are usually right so for now it is the best one for me. I really need cgm, it is sad there is not something better and less expensive and lessirritating- the sites cause some terrible itching, rashes and reactions. Some people seem to have better results. I never bolus off it without 2-3 close fs.

#3

I’ve been having lots of transmitter errors and less accuracy with my G4 for some months now, I think.

#4

I need my cgm also as I have hypoglycemia unawareness. I just wish it was accurate enough to be able to make corrections from it. I find it a little more accurate at nighttime.

2 Likes
#5

It would be nice, but since it is so off and there is no way to tell finger sticks it is then. One bad decision due to an off dex number could be disastrous.

1 Like
#6

That is so true! That is why I can hardly wait for the g6 to be available in Canada.

#7

Now, obviously, you’ve gone through proper calibration techniques…that may have gotten me a while ago. Did you have better luck on the previous model? I’m worried about switching. But, this thing isn’t gonna last forever.

#8

Why are u sure that the G6 will work much better for you than the G5? Maybe it’s an issue with your body. I have zero luck with Medtronic sensors (different sensor tech) but have awesomely accurate numbers from the G5. About 20 months of usage and the thing is so close I hardly test. I’ll calibrate mostly to get rid of the annoying “blood drop” screen on my receiver and I even let that go for hours before finally calibrating. You can’t make assumptions about the G6 based on anyone else’s luck with it. It may be great, or you might still have issues as do many G6 users. It seems to be a very polarizing product.

1 Like
#9

Yeah. Agree.
Everything is not a good fit for everybody.

If the G5 simply is not providing good readings then I would not be holding my breath assuming that the G6 is going to simply fix everything. In fact, I would assume that the data problems from the G5 are going to continue with the G6 until I was able to prove otherwise.

BTW @Dee_Meloche in your other posts, you are talking about the Tandem t:slim X2 along with posts both about the Basal-IQ and Control-IQ algorithms. Obviously those algorithms are ENTIRELY dependent on good data from the Dexcom G6.

1 Like
#10

Yes I realize that the cgm must be spot on in order to make adjustments. I’m not sure it will be spot on but one thing I am sure of is that I’m going to give it a try.

#11

By no means am I saying that.

Certainly I do not require nor expect 100% accuracy. I may not even believe in the concept.

However the data has to be good enough FOR YOU whatever you consider to be acceptable. Given that you are reporting G5 data which is unacceptable, I merely suggest that you consider if an automated algorithm of any sort which depends on this data would be appropriate FOR YOU or not.

#12

Pretty much what Tim already said.

#13

I am going to change the times when I calibrate. Usually I would do it any time during the day when the arrow was going straight. I think first thing in the morning is going to be best and at bedtime considering the arrow is straight. I am going to wash my hands before testing. When I test I’m not going to use the first drop but the next one. I going to see if that affects it. It’s not always inaccurate. I would say about 20% of my readings are off.

1 Like
#14

Excellent idea. I support that as a 1st try. No eating or insulin taking within 4 hours. If you want to check manually an hour before calibrating, just to be sure, then do it. Gotta be sure whats going on with devices.

#15

Yes, the G5 can be notoriously inaccurate. However, my Freestyle BG meter can vary widely also. So, I have a range where I just go by my CGM because it really doesn’t matter. If it says I’m low, I treat the low. If it says I’m something crazy like below 50, I test. If it says I’m high, I bolus accordingly. If it says I’m over 200, I test. This system works for me and helps keep my sanity.

#16

To restate what I have said before.
I have found that with Dexcom, being properly hydrated has made the biggest difference in accuracy for me.
Do a test, drink as much water as you can for a few days straight. It can be a bit annoying with all the bathroom trips, but it will quickly tell you if this is one of the culprits for your bad readings. I takes more than a day to test, but you should start seeing a difference fairly quick if this is it.

Also, I found that on the G5, that Xdrip+ was FAR superior for accuracy over the native dexcom algorithm. I let my pump read the G5, but I only used Xdrip on my phone.
To the point that on most sensors I completely stopped calibrating after the initial 2x calibrations. I just ignored all the prompts to calibrate and let xdrip do it’s thing.

1 Like
#17

I agree!

#18

Yes for sure…as well as “waiting out” the 15 minute delay when you are rising or falling. I have had times when G5 was low or high because it had not yet “noticed” a rising or falling BG. For example after eating or bolussing.

Yes and make sure your hands are clean.

When I started with Dexcom I was using a One Touch meter and had some problems with accuracy. I switched to a Contour Next and found that G5 accuracy improved a lot…because I was calibrating and comparing with a more accurate fingerstick BG. Maybe you can switch meters and see if things improve?

#19

I think I have it figured out! My dexcom numbers have been very accurate. I have been writing down the cgm blood sugar and my glucometer blood sugar to keep track of it.

2 Likes
#20

I also don’t trust ANY meter with a single sample.
I know I do this differently, but I will test the same drop of blood twice.
If there is a big difference, I will test a third time.
If they are really close (usually are,) I use that number.

Remember, no matter how good your meter is, it still has an error factor.

1 Like