Free Style Libre Accuracy Readings

There are many threads devoted to the Dexcom systems, but very few mentions of the FSLibre system. I was on the G4 and the G5 Dex and found little success. So, I moved over to the Libre.

I have found that although when I am properly hydrated, my Libre is pretty accurate, but sometimes it can be quite off the mark, either reading 20 points below what my finger stick reading is or sometimes 20 points higher. I have pretty much learned to ignore direction arrows that don’t make sense or don’t feel right, but sometimes they do help.

That all being said, I would NEVER make any treatment decisions based upon the Libre reading. I would not bolus; I would not correct.

You may wonder why I even bother employing it. But the fact is that it does give me a sense of an approximation of where my BGs are going, so long as I am aware of how a particular sensor has been reading (lower or higher or mostly accurate). There is no way to calibrate the Libre, so I just wait for it to “catch up” with any potential lag time (it usually does within about 15 minutes if it is falling behind (exercise is a main reason, or perhaps sleeping on top of it, or even sharp upwards or dropping BGs)

Please share your experiences with your use of this system. why you chose it, how you like it; I am dying to know, as I can perhaps learn from this.


I use the Libre and I hardly ever do finger sticks. Like you say, the Libre is pretty accurate, but at times the reading can differ 20 points from a finger stick. That’s considered within the margin of error though. In general I’m quite satisfied with the Libre. I must admit that before using the Libre I was pretty skeptical of making treatment decisions based upon sensor readings, but I have no other options if I want to use a sensor. The deal over here is either 400-450 test strips per 90 days or the Libre and 50 test strips per 90 days. More test strips are covered only if you meet strict conditions. I’m sure I don’t qualify and at the moment I don’t want to purchase strips out of pocket, so for the time being I have to accept that I can’t use finger sticks more often. If circumstances change, I might consider buying more strips.


I totally understand. I test so frequently that I do pay OOP. its my treat to myself. But I am fortunate enough that I can do that. Medicare pays for all of my other D stuff 100%.

1 Like

I use the Libre because my insurance covers it. Before they covered the Libre, I paid out of pocket for Dexcom.

The Libre often reads 1.5 to 2.0 mmol/L lower than my actual blood sugar. But it is off by this amount very consistently. So I don’t view it as much of a problem in daily life, but it does make statistics the Libre generates like time in range nearly useless. If I have to make a big correction or treat a major low, then I will test to confirm. But I test once or twice most days. Some days I don’t test at all. On a roller-coaster day I might test six or eight times.





Libre user here… I have varying low readings with mine. The first day is a complete sh*tshow which is expected but it can still be 20-40 points always lower for me the rest of the 14 days. That’s what my insurance covers and its pretty cheap. I am able to cut short my highs and my A1C has improved to 5.5 with VLCD/Dr B eating approach. It’s good enough for me now.

I am doing an experiment to really hydrate well before my new sensor and I’m seeing a little bit of improvement. I tend to overdue my caffeine and alcohol and water is definitely something I don’t drink enough of in the hot climate I’m in. If anything it’s forced me to line up the water glasses every day!



I have emphasized this on another thread; continuous hydration is key to good accuracy (among other things.)I have tested this out and found it to be consistently true. On days that I drink a lot of water, my Libre is almost spot on, give or take +/- 5points. On days that I am not well hydrated, I can see a 20+/- point difference from my finger sticks.

congratulations on your A1c !!! 5.5 is awesome. and thanks for sharing :heart: :smile:


I tried the 10 day Libre for a while about 2 years ago and found it to be quite accurate for me, by CGM standards anyway. If I recall correctly I had more than 90% of Libre readings within 20% of Contour Next fingersticks. It was actually more accurate than the G5 in my short test. I wore both for 30 days.

After that I saved some Libre sensors for backup CGM use and went back to Dexcom. When I (much later) tried an expired Libre sensor, it would not work. Tried another, same thing. The sensors did not last past their expiration.

Thanks for the reminders about hydration…it definitely is important!


Interesting about hydration. I’ll have to try drinking more water and see if it helps with accuracy. The Libre reading low is really the only thing I dislike about it compared to Dexcom. (And I do sometimes like alarms at night, but that’s what hte MiaoMiao is for.)


Before my consistent hydrationing, my Libre was typically about 20 points lower than my finger sticks, but I got use to it and would just calculate in that factor whenever I was reading the Libre. Now I can be a little more all over the map. sometimes, like this morning, it reads spot on. this being said, how do I know when I am reading low or am spot on? this is why I do not bolus off of it nor make ANY treatment decisions based upon its reading. frustrated.


I started with the 10 day Libre and migrated (mandatorily) to the 14 day when the 10 day became unavailable. The first thing I noticed was that two thirds the time the adhesive came off the sensor (but not my arm!) and once the sensor falls off you have to replace it. I now place surgical tape over it and that solved that issue. Then I started getting sensor error messages “reading not available. Try again in 15 minutes”. That time for re-reading varies from 10 minutes to 3 hours. I have found you cannot try sooner, for sometimes it results in total sensor failure and needs to be replaced. I’ve also had a number of the sensors result in the reader telling me it’s time to replace when only 3-10 days have gone by. It’ll never restart, you have to replace the sensor. Abbott has always replaced it at no charge even sending a pre-paid return box for the defective sensor. It’s just a pain in the ass having to call the Philippine Call Center. I’ve got a handle on the accuracy now too. When my Libre readings go low, the lower it goes the more inaccurate it is: it’s reading lower than the finger stick. When it reads high, it gets much higher than the finger stick. The higher the reading, the more inaccurate it is, compared to the finger stick. My trending seems to be pretty good on the Libre. I use Libre for a “relative” idea of where my BG is going and dose myself accordingly. The graph is especially interesting and I’ve reduced my long acting insulin from 21 units once a day in the morning, a year ago, to 13 units now and because I’m pretty “flat lined” at 55-60 mg/dl throughout the night, I will reduce the Lantus another unit or two. This tact seems to be working but I’ll know for sure when I visit my endo next month. I’m guessing my A1c will be below 6 this visit. It would be nice if Abbott would do a re-design to correct these issues.


It certainly would. I’ve spoken to my endo and CDE about this and they’ve told me that they get the same complaints from their other Libre wearing patients.

I have not had the many difficulties that you have experienced, but I can tell you that I definitely don’t need to keep any extra adhesive over my sensor. It is even challenging to get the darn thing off when it is time to put on a new one. And this despite swimming in a pool for 2 hours and running in this summer humid heat.

But, like yourself, Abbott is great at replacing any failed sensors and they come within 3 days in the mail.


Some very interesting points made above.

I run Xdrip+ off the Libre sensor.

I test and calibrate twice a day so I have detailed figures in my diary for the last three years. For blood glucose tests I use the Contour Next which regularly comes out as the most accurate system.

I have not noticed hydration having any effect but then I am rarely dehydrated.

I would say as far as my experience is concerned Libre sensors vary in their accuracy. My current one for instance for the last 10 days has been as stated above about 20 below very consistently. However the previous one varied from the same to 15 below. I have once analysed 8 + weeks of figures (ayear ago) the details are on my blogsite in an article “Medtrum is it a contender in the CGM stakes?”
There were 3 complete 14 day Libre sensor periods these were from

17.08 – 30.08 39 minus readings 4 plus readings
31.08 – 14.09 15 minus readings 3 plus readings
14.09 – 29.08 9 minus readings 26 plus readings

The minus and plus readings were against the Contour Next readings.

The Libre is almost always low - even the 3rd sensor when below say 90 and quite often high when say over 220. However I do not think any CGM reading the interstitial fluid is completely accurate. Going to sleep or driving I want to see my blood sugar a minimum of 90 and preferably rising.

Personally as I want CGM to warn me from going unconscious at night I would always check before I go to sleep as an absolute minimum.

1 Like

For me, the adhesive separates from the sensor so easily, it just pops off when I pull on it. The adhesive that sticks to the arm just kind of rolls off. No pain for me at all.


1 Like

I used the Libre 1 for about 18 months and got accurate readings on some sensors, able to bolus and treat based on the Libre readings. Other sensors would be wildly inaccurate, a complete waste of money. It was infuriating spending my money a roll of the dice. I stopped using it for about 4 months while waiting for a miao miao, which made it more accurate.
Here in Spain, the Libre 2 came out about 5 months ago, more or less. I have found every single one to be super accurate. I was just on holiday for ten days and only checked my BG twice with a finger stick. Once because my sensor said I was at 342 (I was) and the other because I got one of those error messages that said that I needed to check again in ten minutes and was just leaving to go for a run. I am so impressed with the LIbre 2. I am even more impressed with our socialized medicine, which now supplies all T1s with Libre sensors, so that was like getting a 120euro/month raise! Gravy.


I am eagerly awaiting the release of the Libre 2 and I am hoping that my insurance will eventually cover it (Medicare). I have read up on it and it sounds like a great solution to the inaccuracy problems. fingers crossed.

So happy to hear that you have been so satisfied with the accuracy of your new sensor!!!


I’ve got the same problem. They should make the adhesive stick better to the sensor.


I don’t want to count my chickens before they hatch, but so far so good, this has never happened to me in over 2 years. and this is despite all of the sweat and all of the chlorinated water from the pool. Hummm. unfortunate, though. at least Abbott replaces them for free (as they should).

1 Like

So, the main reason I chose the Dexcom system over the Libre system is because I don’t have to keep checking to see what my reading is. It tells me trends and audible by looking and hearing what my diabetes trends are, especially at night during sleep.

I can honestly that since I started using the Dexcom CGM, my A1C has significantly dropped from 11 8 to 7.2, and I have stayed within the 7 range for 9 months. I do want to lower it to below 7. I believe perseverance is the key. All in all, these baby’s are amazing…when they work. I’ve had numerous issues with sensor errors. Thus obtaining replacements can sometimes be a hassle.

Now, I’ve had to select a new distributor because Dexcom no longer submits requsts to my insurance. Additionally, It is necessary to switch testing supplies due to insurance no longer covering the Contour or the One Touch devices. So now, Freestyle is my new er, choice.


Good to hear the Libre2 is better… hoping my insurance covers it very soon as I’d like to see some serious improvements with accuracy.