Constant Dexcom Errors

Hope someone can help me with this problem!

I’m caregiver for my 86 y.o. Type 1 mom. Ever since Medicare allowed her to have a Dexcom 5 years ago, she has experienced very frequent sensor errors, especially in the last few days of a sensor (both G6 and previously G5).

What happens is: The Dexcom graph is cruising along steadily, but suddenly plunges (often even into Urgent Low territory), then we lose data for anywhere from 30 minutes to 3 hours. If we do a finger stick, however, the number is right in line with the previous graph, as is the Dexcom number when the data finally returns. (I.e., the plunging graph is a false low.)

This happens to such an extent that we almost always have to take the sensor off a couple days early. It wakes us up in the middle of the night with regularity, causes us to treat unnecessarily out of an abundance of caution if we can’t stick her finger right away, etc. And now that Mom has an Omnipod 5, we are concerned that the constant bad data from Dexcom is fouling up her Omnipod algorithm.

Although Dexcom is very good about replacing sensors when we take them off early, they’ve never been able to give us any guidance on how to avoid this problem. It seem plausible that, as they say, the sudden plunge (and it’s always a plunge, never a spike) is the result of the sensor periodically having trouble getting data from Mom’s interstitial fluid, resulting in a lower and lower reading until there is no data at all, but that does not help us avoid the problem.

We’ve tried overpatches and putting the sensor in different locations, and probably other things that I’m forgetting, too.

This all being said, sleep deprived though we all are, of course it is better to have the Dexcom than not, and we don’t want to ditch it. We just want it to work as advertised!

Does this sound familiar to anyone? Has anyone got a fix?

(P.S., This forum was so incredibly helpful to us ~three years ago, when we were having pod problems. My mother has never once had to go to the hospital with a number we couldn’t get down ever since we implemented the fix y’all suggested!)


This is an issue that some of us have. I call it the 7 day wonder. Day 7, 8, 9 and 10 the report takes a dive over 15 20 minutes and drops out.

This is not true for all users. Some do a restart for 20 or even 30 days use. But not me, I get better results if the sensor is on my inner upper arm. The officially required abdomen sites are the most troublesome.

My last 2 sensors have actually lasted 10 days.

I wonder if the problem is caused by our immune systems attacking the sensor lead, making it start falling. That’s a wild you know what guess.

1 Like

Thank you, Luis3, for your response! We’ll definitely try the inside of her upper arm.

I’ve wondered about the immune system, too, and even asked my mother’s endo – who doesn’t acknowledge that this is a sensor problem. She keeps telling us that these are actual lows and trying to change all our settings. So patients seem to be on their own here!


Oh, Luis3, PS. Do you have an Omnipod 5 by any chance? Any idea whether this problem corrupts the Omni 5 algorithm?

I’ve uninstalled/reinstalled my mother’s phone app a couple times, since I understand that is the best (or only) way to clear out previous data and start afresh. But that’s a real pain in the neck, and besides, soon we just have more bad Dexcom data.

The reason the abdomen is often preferred as a Dexcom site is the relative amount of “cushion” there. As people get older there’s frequently less tissue that’s suitable for these sensors.

So the best idea is to place the sensors where there’s the best subcutaneous fatty tissues. Adequate hydration can help, as well. And maybe switch the sensors at eight days routinely rather then try to go the distance.


No I’m using a Tandem. What happens is during the loss of data C-IQ stops reverting to the set Basal rate which may be too high for me especially at night. This could cause a real low that isn’t reported.

1 Like

Considering that the sensors work well for several days and then going wacko I don’t think hydration is the problem.

In fact when I’ve been dehydrated exercising or working in the Texas heat I get a low that is false but without the 30-180 minute loss of data.

1 Like

Thank you, MBW! My mom does have at least some fatty tissue there on her abdomen. But as with Luis3’s inner arm suggestion, we’ll look for the place with the most fat and give it a try.

We have pretty much resigned ourselves to changing the sensors early, as long as Dexcom is so forgiving about replacing them. But sometimes we have the problem intermittently as early as Day 4 or 5, and we’re reluctant to change it quite that early, unless we really have to.

Hydration is another issue. It is a challenge to get Mom to drink; like many old people, she doesn’t want to have to get up to use the toilet. But I guess we’ll just have to redouble our efforts!


We have had some problems lately where the Omnipod 5 can’t find the Dexcom transmitter and thus reverts to the manual settings (technically, called “limited mode”), but I’m not sure whether that synch’s up with Dexcom data issues. (My mother has been with my sister for the last month, and my sister is too harried to be really specific.)

1 Like

This is an issue with the Tandem pump as well. My phone can be in the other room still getting data from the Dexcom, while the pump want’s to be very close to the sensor. Line of sight between the pump and sensor is what is recommended.

1 Like

Yes, we do keep it within line of sight. I think this is probably a function of the Dexcom data being weak-ish. But I won’t be able to say more definitively until my mother and her devices are back with me again…

1 Like

Some people seem to have a real problem with sensors going the full 10 days. I am not one of those (I restart and get 20-30 days). One of the theories is people with less body fat seem to have more of a problem. But that isn’t a given as it can really vary, but it is one we have taken note of a little more often here on the forum. The plunge happens because it can’t read the interstitial fluid right. But the why of that is the unknown factor.

Watch for placement, some areas seem to definitely be better than others and that can vary per person too. I love my arms, but make sure you don’t place it to where you will get a compression low. I sleep on my side and the inside of the arm gets crushed and I will get compression lows. In case you don’t know what a compression low is, it’s when pressure is applied against the sensor cutting off good circulation, you will get a false low reading. That also makes me think that as we get older our circulation has a tendency to get poorer and that might be helping the sensor to go wacky easier with your mother. And with/or as Luis3 has said maybe some kind of immune response over time by individuals reacting to the cannula?

For the endo, that’s ridiculous. They really should know better. It’s not an uncommon problem. But I had an endo once that said I couldn’t be a type 1 and never tested me so… they are out there. I suggest you document a finger stick with the low reading with dates and times so you can present that to them. That way they can see the reading and then the actual Bg level.

Unfortunately the Dexcom is not perfect. I’ve had words with it many times. But I certainly still want it. But our tech can drive us a little nuts. When things go wrong and I can’t figure it out. I start documenting to try to find the pattern. You might not be able to figure it out though. I do find when a sensor is going it starts giving me major false drops. That’s one way I know my restarted sensor is failing. A couple of those and I switch it out right away. I like my sleep and too many things can already disrupt it.


I went from sensors that worked really well for the last few months, to sensors which aren’t reading correctly. Today I was outside doing yard work and my sensor was reading 97. I was feeling off so came inside and did a fingerstick and found out I was at 63. My last 3 sensors have been giving me incorrect readings. I am not a happy camper. I sure hope that the G7 has been improved.


You’re lucky to get so much time from a sensor. But at least my mother is very well insured, and Dexcom does replace the ones that need to be pulled early.

I will keep the compression lows in mind.

Re the endo, Mom also had one who insisted she had to be Type 2, because she didn’t develop diabetes until late in life. Never bothered to test her for antibodies, even after multiple hospital trips.


We’ve had that on occasion, so I’m looking forward to the G7, too. But in my mother’s case, we have to wait for both Medicare and Omnipod to catch up…


Caregiver Daughter, I am really impressed with the care you are providing for your mother. Some of us, who are older, are questioning what kind of care we will get as we age. Your mom is very fortunate to have your help.

Type 1
Dx 1959


Thank you, but my sister and I split the responsibilities of caring for our Mom. (But I’m the “technical one”, so device management falls to me!) I have no children myself, so I also worry about needing that kind of help as I age.

Re my mother, it’s sad that she was not able to get a Dexcom until ~5 years ago when Medicare approved them. I believe that much of her current cognitive decline is due to the many severe hypoglycemic events she had over the years without feeling them until it was too late, or the times the pod didn’t work (and once she ended up in a coma with a BG of 882). None of those things would have happened if she’d had a CGM.

And we only learned the fix for the pod not working here in this forum. (Short answer: Give at least a small bolus immediately after putting on a new pod.)

1 Like

I used Dexcom years ago, and had recurring issues like you describe. I eventually switched my sensor site to my upper arm instead of my stomach with amazing results. It was much more reliable that way. I’m not sure if that site is allowed by the company or not, as I now wear a different brand.

Thanks, HighHopes. Which brand do you use now, and presuming you think it’s better, what do you like about it?

We’ve tried the upper arm, but not the inside of it, so we’ll have to try that.

1 Like

Until recently, I was having great success with placement on the upper arm towards the inner arm. Using the stomach just didn’t work well for me at all.

1 Like