CGM and fingerstick use

Please help me with a little survey. I found that once I was used to my Dexcom CGM I didn’t need to do as many finger sticks as I had before I started using my Dexcom. I know they say always do a finger stick test before dosing any insulin, but that didn’t seem necessary. I went from 5-6 finger sticks per day to more like 2-3. What’s your experience been?


That has been our experience also. We went from 10 finger sticks a day to around 4. The Dexcom has been so accurate. In fact, tonight I tested my son with a finger stick and the Omnipod gave me a reading of 108. Dexcom said he was 250 so I went ahead and did another fingers stick which gave a reading of 251. Did 2 more finger sticks just to be totally sure and they both gave readings in the 250’s. I’m so glad we have the Dex or I would have sent him to bed without a correction thinking he was in target. Makes me wonder how many other times I had an inaccurate reading and dosed him.

Yes, I too have seen a significant drop in the number of finger sticks that I’m doing. Very similar to what you’re doing in fact. The challenge I have is if I’m rising quickly- that’s when I’ll see a discrepancy in stick vs Dexcom readings. (And the opposite as well when my glucose is dropping quickly.) However, when I’m stable the readings are pretty spot on. And I love the arrows!

Yes, for me there's definitely a significant reduction of postprandial finger sticks because I rely on the CGM's trend. However, I'm not sure that it drastically reduced the total number of tests for me. I often use the trends as an indication to check my actual blood glucose early, which helped me tremendously to avoid hypos (and also detecting highs early). Personally I very rarely dose insulin just based on the CGM reading though.

Related, from Gary Scheiner's recent newsletter article "The Coverages They Are A Changin’":

Again, this is not something I ever thought I would write. However, times have changed. The latest continuous glucose monitors (the Dexcom G4, specifically) perform at a high enough level that between-meal fingersticks may not be necessary. As long as the CGM has been matching recent calibrations closely and the glucose level is relatively steady (“lag time” still has an effect on accuracy), glucose values generated by the CGM can often be taken at face value.

thanks for the re post grom Gary S. When using the Dex 4, it is so accurate that truly wasteful to test between meals. I only test to calibrate ( 2x/day) and if have meal. Previously tested 15 times/day!.

I assume you have the Dexcom G4, because the Seven Plus was so erratic that I ended up using MORE strips - just checking suspicious readings... The G4 is considerably more accurate and I need fewer finger sticks, but I'm not sure I use fewer than before CGM; I always check before correcting either a low or high reading, before meals, exercise, bedtime and with required calibrations, I'm probably just breaking even. Luckily, my insurance allows for a generous supply of strips but if I had to ration their use, I would feel fairly safe relying more on G4 readings.

I stopped doing non-essential (to me) finger sticks when I got the 7+. That continues with the G4. I now do two a day (when the CGM asks for it only). I used to take between eight and 12 tests a day prior to getting a CGM. I know some folks out there have a terrible time with erratic numbers, etc. I've not had the same experience. I dose off the G4 all the time.


New to the glad to read your responses concerning accuracy! I read that lots of users get extended use out of the sensors. Does that seem to effect your readings?

My number of fingersticks has also gone down. If the line on the G4 is pretty smooth, or the peak is reasonable, I don't test. If I'm showing an abnormally high reading, I do a fingerstick, which usually tests out pretty close, but I correct off the fingerstick, not the CGM. Same thing if it shows I'm headed low. The G4 is really a life-saver for me, because it means I can catch these incidents before they get too bad, and since I don't feel them, I need the help.

kennedy does two to three a day now... if she is not widely out of range, then we do not test for meals, even... her hypo % this past week was 1% !!!

Less fingersticks here, although I will be the first to admit it is a slightly bad habit to get into. When I had the Freestyle Navigator, that thing was so accurate after the first day that I got into a very bad habit of almost never doing fingersticks. The Dex Seven+ was a lot more inaccurate (not horrible but did fluctuate and was off by a lot in some cases) and I went back to more fingersticks. The G4 is definitely much more accurate and definitely has made me drop a few sticks during the day. But 3x/day at least it is a good idea to do the fingersticks.

My .02

I have been using the new G4 for a couple of months now. It's accuracy has really changed my sticks. With the old DexCom, I was still sticking my finger about 6 times a day. I shouldn't complain, it was a big improvement over relying on finger sticks alone before the CGM.

What I am doing now is calibrating in the morning when BG is very steady. One finger stick and two samples. If they are more than 10 points apart, I take another sample from the same finger stick. I enter the two that are the closest together. Unless something unusual happens, that is it for poking myself with sharp stuff.

If I have an unusual high or low, I will do another stick. What I have found is NOT to calibrate the G4 when BG is moving up or down. There are differences in the interstitial glucose and blood glucose and calibrating when they are moving actually comprimises the accuracy of the G4.

The tips of my fingers are back to normal.


Wow, that's amazing! So you actually only calibrate once a day?

If you rely on the sensor that much, how long do you usually wear it? What are the signs that make you realize that it's no good anymore?

I use the sensor for the 7 days. I hacked it once to extend the sensor because a shipment was delayed. Even then I only extended it a day and a half. I worry about infection and irritation.

There is no difference in the accuracy from beginning to the end of a sensor.

I find the two factors play off each other. When I calibrate the CGM, if the scores are quite close I feel more comfortable using the CGM, rather than take the time to also do a finger stick, for example at 2AM when I’ve been woken up by the high glucose alarm.

I change sensors using the same criteria, when the CGM reading seems too far off the calibration finger stick glucose, it’s time to change. For me this is usually around the two week mark. I also found that by using waterproof tape when I shower, I can extend the life of the sensor adhesive, but I have to be sure to take the tape off as soon as I’m out of the shower. I tried leaving it on to extend the life of the sensor adhesive even longer and it was quite hard on my skin.

I just wanted to thank you for your original post. I've given it a try, i.e. really carefully calibrating once in the morning and then only once again in the evening, and only referring to finger sticks during the day before treating lows or during unusual readings. I'm amazed how well it works. It has changed my daily life quite a bit.

I'm also very surprised by the accuracy and longevity of the G4 sensors. Since I've never had problems with infection or irritation at the sensor site, I'm wearing the sensors as long as possible. Accuracy seems great even after three weeks, which is totally different to the Medtronic Enlite sensors, which basically became unusable after one week (especially in the last half year or so).

Since I'm applying an extra adhesive on top of the G4, I also don't have issues with the sensor falling off and I replace the additional adhesive every week or so. Now I really wonder how long I can push the sensor life. In my opinion this is a really important consideration, since it might make CGM affordable for people without health insurance coverage and drastically improve their quality of life.

Is anybody here using the G4 sensors regularly for three weeks or more? If yes, at which point do you usually decide to change the sensor?

The reason SSD is seeing a discrepancy in glucose test strip readings compared to the Dexcom CGM (continuous glucose monitor) readings when glucose level is changing quickly is because CGM readings lag about 10 to 15 minutes behind real-time. A Wikipedia article describes this situation:
"Anecdotally, some users of the various systems report lag times of up to 10–15 minutes. This lag time is insignificant when blood sugar levels are relatively consistent. However, blood sugar levels, when changing rapidly, may read in the normal range on a CGM system while in reality the patient is already experiencing symptoms of an out-of-range blood glucose value and may require treatment. Patients using CGM are therefore advised to consider both the absolute value of the blood glucose level given by the system as well as any trend in the blood glucose levels. For example, a patient using CGM with a blood glucose of 100 mg/dl on their CGM system might take no action if their blood glucose has been consistent for several readings, while a patient with the same blood glucose level but whose blood glucose has been dropping steeply in a short period of time might be advised to perform a fingerstick test to check for hypoglycemia."

The reason why this happens is aptly described in the article "Continuous Glucose Monitoring"
" Even with good calibration, there can be a 15% to 20% difference between blood glucose readings and sensor readings. This is because of the way glucose moves through the body: After food is eaten, glucose is absorbed through the small intestine. From there it enters the bloodstream. And from the bloodstream, it is moved to the various cells of the body and to the interstitial fluid fluid. Consequently, the glucose level of interstitial fluid rises about 10–15 minutes after blood glucose level rises, and it similarly falls about 10–15 minutes later. This is commonly referred to as the “lag time.”

Because of this lag, it’s common for CGM users to set their hypoglycemia alarm higher than the blood glucose level at which they normally treat for hypoglycemia: If the CGM receiver reads 80 mg/dl, the blood glucose level may actually be below 70 mg/dl. Any symptoms of hypoglycemia should be checked out with a conventional blood glucose meter."