Dexcom - What do you consider accurate?

I’ve researched the Dexcom (and all the currently available CGM’s) and many people find the CGM to be accurate, but no one really states what they consider to be accurate. I’m currently trying the Freestyle Libre (I was able to get a sample from my doctor) and the readings seem to be within 20-30 points of my meter on average. While I wasn’t expecting the numbers to be exact, I was expecting them to be closer. For those of you who use Dexcom, what is the average difference between your meter and CGM readings? How great a difference do you consider to be accurate?

I think that’s pretty good. There are times when it is spot on and times when the dexcom is more inaccurate than 20-30, especially during sleep or when BG is rising or falling. For me being within 20-30 off wouldn’t make much of a difference for dosing, but it might for others.

It depends where you are.

If you are 65, a 30 point difference is huge. If you are 180, it’s not as big of a deal.

1 Like

Like what Eddie2 said. If I’m in range, I expect numbers within 10. If I get over 200, can easily be off 30.

1 Like

For me dex can be very innacurate, so I never bolus or treat off dex unless I have 2–3 recent sticks which confirm dex. Even a 20-30 point mistake in a bolus can cause a severe drop or a spike. That said it is often dead on for long periods, the trends are usually right and I need it to have any quality of life with this disease such as it is. It has saved me many times and I can keep my bg as low as possible more safely. The first few hours of a sensor are usually a bit crazy.

I spent about a month wearing both the Dexcom G5 and Libre, checking both against a Contour Next meter. I don’t have my data with me right now but can tell you generally the Libre was slightly closer to fingersticks on average. Both were close enough for me to rely on and both are the least accurate when BG is rising or falling. I suspect this is due to the lag time for interstitial fluid. Dexcom calibration is not fool proof and, if done perfectly every time, would possibly make the Dex more accurate than the Libre (but that 100% perfect calibration is easier said than done).

In my view no CGM will be spot on when BG is seriously out of range or rapidly rising/falling. In order to get the alarms, and to avoid the 12 hour Libre warmup, I am presently using Dexcom G5.

1 Like