Libre 2 support

[richyb]

I just been speaking with libre support about inaccuracy. They keep telling me to allow 10-15 after bg tests. Yet according to there own website it says 2.4 minutes. I told them this lot's of times. You can't have both. They are sending me a reader and sensor. They previously sent a neo meter. The man had no explanation why there own website say's 2.4 minutes. Most times it is accurate, but not always.

 

[In Response]

My understanding is that interstitial fluid is 10 to 15 minutes behind blood readings.
With Libre 2, Abbott have updated the algorithm to extrapolate the current trend line to predict the current reading. Most of the time, this is great and the 2.4 minute delay is probably based upon this. However, if the trend changes in the 10 to 15 minutes, the reading Libre reports will be further out. For example, if you treated a hypo, Libre 2 will not know about the treatment and will extrapolate the falling trend and give you a lower reading until it notices that the curve has changed direction. You may momentarily notice a reading which is higher or lower than your levels actually reach.

Regarding inaccuracies, it has also been noticed that
- sensors can be less accurate during the first 24 to 48 hours after insertion. This is why many of us insert the sensor a day or two before activating.
- sensors are most accurate between 4 and 8. The readings become less reliable outside this range which is why the advice is to always test before correcting a high or low.

 

[StewM]

My understanding is that interstitial fluid is 10 to 15 minutes behind blood readings.
With Libre 2, Abbott have updated the algorithm to extrapolate the current trend line to predict the current reading. Most of the time, this is great and the 2.4 minute delay is probably based upon this. However, if the trend changes in the 10 to 15 minutes, the reading Libre reports will be further out. For example, if you treated a hypo, Libre 2 will not know about the treatment and will extrapolate the falling trend and give you a lower reading until it notices that the curve has changed direction. You may momentarily notice a reading which is higher or lower than your levels actually reach.

Regarding inaccuracies, it has also been noticed that
- sensors can be less accurate during the first 24 to 48 hours after insertion. This is why many of us insert the sensor a day or two before activating.
- sensors are most accurate between 4 and 8. The readings become less reliable outside this range which is why the advice is to always test before correcting a high or low.

A study conducted by Abbott also found it's also more inaccurate in the last 24 hours of its lifespan. I believe this was a study carried out on Libre 1s but I've found this issue consistently occurring on Libre 2s as well (so much so I usually alter my alarm settings on the final day).

The effects you've highlighted in your first paragraph have meant that, I find, the Libre 2 can actually be 'slower' at noticing change than the Libre 1. Though, in the data I've got this isn't consistently true, it's just a frequent occurrence.

 

[richyb]

My understanding is that interstitial fluid is 10 to 15 minutes behind blood readings.
With Libre 2, Abbott have updated the algorithm to extrapolate the current trend line to predict the current reading. Most of the time, this is great and the 2.4 minute delay is probably based upon this. However, if the trend changes in the 10 to 15 minutes, the reading Libre reports will be further out. For example, if you treated a hypo, Libre 2 will not know about the treatment and will extrapolate the falling trend and give you a lower reading until it notices that the curve has changed direction. You may momentarily notice a reading which is higher or lower than your levels actually reach.

Regarding inaccuracies, it has also been noticed that
- sensors can be less accurate during the first 24 to 48 hours after insertion. This is why many of us insert the sensor a day or two before activating.
- sensors are most accurate between 4 and 8. The readings become less reliable outside this range which is why the advice is to always test before correcting a high or low.

What I find strange is the website says 2.4 mins. But support says 10+. It can't both be correct. Surely bg is correct in immediate time. There website say's 2.4 mins from real time. I find the 2.4 to be most correct. Or even less sometimes

 

[richyb]

So the 2 is not accurate for the first 2 day's and mine is not very good now 3 day's to finish. The lib1 was better. I have read a lot are saying this. Might be better to go with 1 and maiomaio

 

[StewM]

So the 2 is not accurate for the first 2 day's and mine is not very good now 3 day's to finish. The lib1 was better. I have read a lot are saying this. Might be better to go with 1 and maiomaio

I have found it common that those who say the Libre 1 is superior have had a rather charmed experience with Libre 1 as many of the problems reported with Libre 2 were also reported with Libre 1.

The Libre 2 algorithm is different. This does not mean it’s inferior (or superior) and there has not been enough testing yet to determine if there are any trends one way or the other.

The Tests that found out about the Libre’s inaccuracies on the first and last day were carried out on Libre 1s for example. I was merely surmising that has continued with the 2, it may in fact be better with the 2. There is no comprehensive evidence to suggest it’s worse.

In my personal experience it performs better in the first 24 hours than the Libre 1.

 

[richyb]

Well I have had enough now and removed it. I have had about 10 lib 2 and not had a good one. I got far less poor ones with the 1.
They are now sending me a reader and another replacement. If that's no good I will ask to go back to a 1.
A few years ago I had a dex privately also use an x drip system. With that it showed every reading that it took. It put a line through the average to give your result(very accurate). I reckon that is what libre are doing. That could be why the 2 is not as good, as it is only taking the average over a 2.4minute time, so more likely to have errors. I think they shortened the time to get a sales advantage over the others. (It is a commercial business) The dex was wonderfull but too expensive.

 

[richyb]

Just received a lib2 reader + sensor, to try, as well as my phone. But will have to wait a few day's for my present one to expire.

 

[richyb]

I am pleased to say the last 3 sensors have been very good. So perhaps I had a lot of bad sensors. If that is true I suggest the problems occur in manufacture. Each sensor should be good. If not they should be thrown out by their quality control officers

 

[Mungobean]

I have had lots of issues with Libre 1 readings and sensors. Now I have got the sensors prescribed, I am going to change to the Libre 2, primarily so that I will get the hypo alarm facility. I’ll be interested to see how the 2 measures up. I have had so many sensors replaced, I start to see why they are so expensive, because they wouldn’t be making money! I wonder how much a sensor costs Abbott and whether or not they build in the cost of replacing each sensor in the cost of a sensor? I see the system as saving me many finger pricks, and it gives me a good guide as to where my readings are heading.

 

[Deejaypdex]

I have found the accuracy of sensors really does depend on the factory calibration, and the bad ones seem to come in batches (eg if I collect 4 from the pharmacy, they are often from the same batch, and the each batch seems to have similar performance). They all tend to overestimate highs and lows, and will show the biggest difference vs finger sticks when blood glucose levels are changing rapidly (after eating or exercise).

a good time to check the health of your sensor is when you’ve had steady reading for a little while, such as first thing in the morning, or before meals. If the readings at these moments are close enough, then you’ve got a good sensor.

they will never be spot on, but are supposed to be within 10% either way of a finger stick when glucose levels are steady.

hope that helps to set the correct expectations for what ‘working well’ looks like.

In my experience, about 1 in 5 sensors fail or are way off and need to be replaced, which is acceptable to me

 

[bonerp]

I've had dexcom on for a day and finding it FAR more accurate than Libre2. Got both on at the moment and going to keep an eye with finger pricks too.
I have one failure out of every 2 libre2. Dexcom need to get NHS approvals!

 

[h884]

Hi Folks

Your experiences with Libre 2 all sound very familiar. I have found Libre 2 very disappointing. I have lost count of the number of sensors Abbot have replaced.

I self funded dexcom for a while and found it much more accurate & reliable albeit costly. With the new NICE guidelines it maybe the dexcom and other CGMs will be provided by the NHS in England. I live in Scotland so am not sure what will happen here.

I will follow this thread with interest

 

[richyb]

I find it strange that they can produce accurate sensors, but most are not. My latest batch are not very good. Often the reading is 30% or 40% error. Even when I am not rising or falling. But the cost of them are significantly cheaper than the rest, as they last 14days. Strangely I have had an HBA1C blood test done of 46, which is what my libre say's also. But scan results are always higher than bg by 1.5 plus