I am not seeing much reason for me to switch to G7 from G6. I use my arm for the G6 and the words about the overpatch sound like it would be a mess to apply. I don’t really want to buy other patches from amazon that would be another ongoing hassle. To remember to buy when I should not even need an extra patch.
And the the accuracy. One would think that the next version should be more accurate, not less.
Sheesh. I have no problems at all with the G6. When I rarely do a fingerstick the two numbers are always close.
Dexcom plans to phase out G6.
https://www.dexcom.com/faqs/dexcom-g6-discontinuation-and-g7-upgrading
They have rebranded the g6 already and say that it’s not for dosing but it’s the same thing.
I did a short 6 day trial of the g7 and I wore both sensors,and they were pretty much the same. Of course the g6 results are hidden for 2 hours, so I can’t see that data.
I haven’t switched because tandem didn’t support the g7 until just recently. I’m going to get a new pump and I’ll wait till then before I switch for real.
A lot of people complained about the g6 too back when it came out.
It has always worked for me and I only use the back of my arm, so this might be why you get different results.
The g7 gives u a grace period that is supposed to be for presoaking. I forget how long that is though.
I’m expecting someone will hack the g7 to get more life, but not heard yet. https://www.dexcom.com/en-EE/dexcom-one-shop/dexcom-one-starter-kit/stp-do-101
When Dexcomm G7 was fairly new, Dan Heller conducted a 1 month experiment wearing both G6 and G7 sensors. See his post at Q&A on CGM accuracy (follow-up on the G6 vs. G7) – He has links to his original article, which I highly recommend, and to a followup article.
At that time, he discovered G6 (both apps and receiver) smoothed the data (reduced reported spikes in either direction), while G7 apparently does not. So G7 may show significant spikes, then soon after results revert to the original range.
Looking into this further, there seems to be some evidence that these spikes are at least sometimes actually correct readings – the IG (interstial glocose) level really did briefly spike. Regardless, the G6 data smoothing does not lead the user to make inappropriate corrections, while G7 reporting these spikes can lead to unnecessary corrections, resulting in worse overall glycemic management.
The author concluded G6’s data smoothing promoted better overall control and time in range. So I decided to continue using G6 until it’s taken off the market. My supplier had me fill out a form asserting I needed G6 due to pump compatibility.
I encourage people check to out that post and at least the first article. I continue to hope Dexcomm brings back G6 data smoothing to G7.
4 Likes
@Jim10, thanks for mentioning @argv’s posts. @argv’s posts helped me to overcome my initial assertion that the G6 is more accurate. I am happier with the G7 than I was with the G6.
2 Likes
I also went from G7 back to the G6. The last G7 sensor I used was actually very good. So I’m hoping G7 inaccuracy was a quality control or other problem and the G7s the accuracy is good once the G6s are discontinued.
G6 smoothing has been quite good for me. With the smoothing, the BG input to dosing is good and helps prevent poor decisions on insulin doses.
I noticed that the G6 was doing run time averaging across 3 samples from watching what my G6 was showing me and the different values it was recording as history. What I noticed is that the “current” blood glucose that the G6 shows is NOT the value that (5 minutes later) shows up in the history that the app displays. This meant that sometimes I (or my pump) made treatment decisions based on what the G6 was telling me, but then the history erased the value that triggered the decision. Note that this smoothing erased both the (possibly real) spikes in the interstical glucose that you mention and compression lows (which I frequently see in the range of 20 to 40 mg/dl). So while smoothing as implemented by G6 has value in providing a less “noisy” history of blood glucose values, it confounds trying to understand why my pump did something.
What I observed about the “current” and historical G6 blood glucose values is: the “current” value displayed by the G6 is the average of the current and 5 minute ago instantaneous values, while the value recorded in the history (the “smoothed” value) is the average of the (never shown) instantaneous values for the listed time and the times that immediately precede and follow it. Phrased differently, the “current” value is a two point running average while the recorded history is a three point running average.
1 Like
I have used G6 sensors for several years and (other than frequent compression lows) found them to quite good. I always checked them against my blood glucose meter (at first readings and every day or two thereafter), but only calibrated them (following Dexcom’s advice) in the unusual cases where the difference exceeded 30%. As a result I never calibrated most of my G6 sensors.
Recently I went a month using G7 sensors. Just in case I would have found something wrong with the G7, I switched before I finished my supply of G6 sensors. Again, G7 sensors generally work well. I was hoping that they fixed the compression low problem, but Dexcom doesn’t seem to have done anything about that.
After that month I switched back to my last box of G6 sensors, then I’m going to switch back to the G7. The first G6 that I installed required me to enter two calibration values before it would even show me the first reading after warmup (which was precisely the average of the two calibration values). It then insisted that I enter additional calibration values every 24 hours after that first pair. I’ve never seen that behavior before.
Yesterday I installed the second sensor in that last box. Again, it wouldn’t display the first sensor value after warmup until I had entered two calibration values. Then 12 hours later it wanted an additional calibration value.
I can think of three possibilities here and am wondering if anyone else has seen this behavior?
- Dexcom changed the way all G6 sensors operate and it just coincidentally lined up with my switch from G7 back to G6
- Dexcom produced a batch of crappy G6 sensors and rather than throw them away they are requiring initial and frequent calibration
- Dexcom noticed that I switched from G7 back to G6 and is trying to encourage me to stick with the G7 sensors
G6s require calibration when the sensor code isn’t entered. Perhaps you accidentally skipped the code or something happened that meant that the code wasn’t recorded. Several years ago, if I started a G6 on my phone and used the phone camera to enter the code it regularly failed to record and I had to do the calibrations. I think they may have fixed that particular problem, but I’ve never used the camera to enter a code again so I’m not certain.
@Tnyc The Dexcom iphone app reading the calibration code using the camera has been working for me since 2019 except for one time. Quitting the app and restarting it fixed the problem.
I was using the libre when y’all were suffering through this. Was it an Android vs iPhone app problem?
That’s a really good thought. I have always entered the code myself rather than scanning it, and I’ve continued doing that. But …
When I started using the G7 sensors my pump battery life dropped from 8 days to 1 even though I was using the pump firmware update that “fixed” the problem. When talking to Tandem they suggested that I un-pair and re-pair the pump every time that I replace CGM sensors. That fixed the pump battery life problem.
Now I’ve switched back to the G6 sensors, but I’m still un-pairing and re-pairing the pump. And I’m (as before) entering the CGM codes on the phone but not the pump. I may need to enter the code in both places now. I’ll have to try that when I switch to my last G6 in another week.
For the G7 yes. For the G6 no. Only enter the code in one device and then have the other device join in, but the pump might be the better device to enter the code.
Do you unpair by turning the Bluetooth off on the pump or by just unpairing the pump on your phone’s Bluetooth list?
(Don’t know if the unpairing method makes a difference, but I want to know how to save my battery. Just this morning my pump unexpectedly turned off from lack of power at 6:30am. When I’d gone to sleep the night before it had been low, maybe 20% but I’m used to that being more than enough to last a night. From this morning to tonight it went from 100% to 65% which is a lot for about 15 hours.)
To unpair: on the pump select Options → Device Settings → Bluetooth Settings → Unpair Device. On the phone, open the Tandem app then choose Settings → Paired Pump → Unpair pump.
Once unpaired, from the same final portions of the menus select pair on the pump and then on the phone. The pump will supply a 6 digit code you’ll have to enter on the phone. The app will then reload your data from the pump and you’re again good to go.
1 Like
I´ve had no use for an overpatch at all. I use my arms for the G7 and it sticks better than the G6 did plus I get no reactions to the adhesive. My reactions to the G6 was bad so this is a great relief.
And I second @Helmut about accuracy and being happier with the G7 than G6.
One tip for all the Tandem-pump users out there is to put your pump in sleep mode the first 24 hours with a new G7 sensor. The spikes from a brand new G7 sensor would sometimes trigger an unwarranted autobolus. Not a big issue, but this way I do not need to adjust my pump settings that work well the rest of the time.