When I calibrate the G5 it typically adjusts to about halfway of the finger stick reading I entered. For example if the finger stick reader is 104 and the sensor reading is 90 then the CGM will adjust to 97. But I noticed that if I calibrate to 104 twice, then it will adjust to 104. Is there a problem with adjusting it twice? Why doesn’t the CGM automatically adjust to the finger stick reading we enter?
The G5 typically uses a number of calibrations points in its algorithm. It is not typically only using the last one. This generally provides superior performance as opposed to only using one point.
Doing a double calibration entry is apparently like a calibration reset or something. Not sure exactly. But we had done that on occasion while on the G5 so I completely understand what you mean.
However that was certainly not our standard calibration method. There are downsides and that is not what you want to do as standard practice. Generally, just do the two calibrations per day for the G5 and if it is close enough for you then let it be. Excessive calibrations will cause older calibrations to drop out and then you end up with the algorithm using a less than ideal set of calibrations going forward.
The double calibration method was our last ditch attempt to correct a seriously bad data issue if the sensor did not appear to have completely turned to garbage. The other method would be a session stop/start which clearly removes previous calibration data for that session.
u can double calibrate if you want to force the G5 to match the latest bg readings on your meter. we do it here and it works fine.
Tim35, thanks. So do you work for Dexcom?
Thanks, Dave. I’m torn between calibrating it one or twice. I would not think that a calibration five days ago would have any effect on sensor accuracy so I would not be too concerned about older calibrations dropping off. I think I will try a double calibration and see if accuracy becomes affected.
Just to clarify a double calibration is when you calibrate then calibrate again 20 minutes later?
Well when I calibrate it twice within seconds (one finger stick put in twice) it will reset reading to finger stick reading. That’s what I am going to try unless it will result in the Dexcom becoming inaccurate.
Richard J. Carli
1 Like
I think that’s correct given what I’ve seen with G6. When I calibrate the reading will move closer to the entered calibration value, but continue to move closer to the actual reading within a few additional readings. Here’s an example of what I saw this weekend when calibrating a G6 sensor after startup. At calibration the reading dropped from 26% to 14% above finger stick, but within 3 additional readings matched the finger stick number.
I recall seeing similar behavior with my G5, but that’s anecdotal. I never documented it.
That’s what I did when the sensor went wacko, typically near the end of a session,
It would seem to me start/stop has some advantages.
It restarts readings that should already be pretty good as they are not within the first 24 hours.
The first and next calibration will be the only one in history and that would seem to better than the old calibrations which gradually trend you toward accuracy.
Have you ever tried “cheating” your finger stick number. In other words if your CGM says 60 and your finger stick says 80, you know that entering 80 in your CGM will then show a mid point reading of 70. If, however you “cheat” the CGM and enter a slightly higher bogus fingerstick of let’s say 90, then your CGM will read 75.
I have never seen this discussed in this community, but can’t believe that I am the only person that has played around with that option. If you only add a few points it will take and all will be good. If you add too many points, you will get a notice of a drop to calibrate. I only do this when running near my bottom limit of 60 and need to get the CGM reading accurately as soon as possible.
I used a 10 point “Cheat” just as an example for easy computation. I normally will not add more than 5 points which often gives me a better value long term than a 2 step calibration.
I use this method also. I’ve noticed no change in my accuracy between Dex and my meter. I always have to double calibrate in the morning. I guess compression messes things up for me. After this double calibration things are spot on til the next morning. 
CJ114,
I think we all strive to be creative when it comes to managing T1D. I don’t feel it is cheating at all but also feel putting in the meter reading twice or start/stop gets you there too but thanks for that idea. If one is confident with your meter (I am with Contour Next) I prefer one stick and two calibrations.
1 Like
No problem especially if you think it is off. The problem could be if your finger stick is off. So it makes sense to do 2-3.
1 Like
no, it is when u calibrate 2x moments apart to force the system to show the current bg reading instead of something in between the last cgm reading and a single calibration
1 Like