Libre (in)accuracy on HBA1C

jellyfishfeet

Member
Messages
17
Type of diabetes
LADA
Hi!

Now I'm not talking about trying to get my finger prick to match my scan, as they are measuring different things that respond at different times etc etc

BUT the difference between my Libreview HBA1C and the hospital blood test has been consistently very different, latest being 32 vs 44.

The higher number is the blood test obviously! This equates to approximately 1mmol/l difference in all my readings on average.

Has anyone else made this comparison?
 

jellyfishfeet

Member
Messages
17
Type of diabetes
LADA
Thank you!
Regarding the 24-48 hours issue, I always apply it 24 hours before activating.
My sugars are within the 4-8 range for 90% of the time.
Therefore these standard errors should be minimised?
Agree with not being factory man, but for me I am happy with management of my highs/lows and wish to bring my average HBA1C down. I am currently using Glimp to see if this will help me, but was wondering what others comparisons are to see whether it is common.
 

Antje77

Oracle
Retired Moderator
Messages
19,284
Type of diabetes
LADA
Treatment type
Insulin
Yes, the difference between fingerpricks and readings nicely match the difference between predicted and real hba1c for me.
 
  • Like
Reactions: jellyfishfeet
M

Member496333

Guest
Worth remembering that HbA1c itself is an indirect measurement and does not always truly represent the real, mean average blood glucose. It's a good estimate in the main and is generally a helpful diagnostic tool but it shouldn't be considered as gospel. In some individuals it can be wildly inaccurate. I believe the primary variables are anaemia and shorter/longer than average red blood cell lifespan.. For these people there is an alternative test called Fructosamine, but I know nothing about it.

Of course none of the above can discount any other inaccuracies in Libre and/or finger prick. There are always errors introduced somewhere in all testing, but if you're experiencing HbA1c results that you're convinced don't represent your average glucose, then it's conceivable that you may fit into the above mentioned category.
 

barrym

Well-Known Member
Messages
800
Type of diabetes
LADA
Treatment type
Insulin
Can't understand why anyone bothers with HbA1c now that we have CGMs. With HbA1c you can on theory get s good reading whilst having lots of hypos and corresponding hypers. It is just a clever weighted average. It was ok when it was all we had, but now you can get a time in target which is much more useful, IMHO anyway.
 

becca59

Well-Known Member
Messages
2,856
Type of diabetes
Type 1
Treatment type
Insulin
@barrym Well if you have been self funding Libre. Lowering your HBA1C whilst on it, is part of the criteria for NHS funding.
 

barrym

Well-Known Member
Messages
800
Type of diabetes
LADA
Treatment type
Insulin
You are totally right. My view is very myopic as one that is "fortunate" enough to be T1 and testing 15-25 times a day and injecting 8-10. My life is obsessed with managing this **** condition and assume everyone else is the same. Stupid, as even this morning I was having coffee with s friend who is pre-diabetic and going in for his annual HbA1c this week. More than adequate s test for his condition to check he's still on the wagon.

I should take a step back before replying.

A genuine apology.
 

jellyfishfeet

Member
Messages
17
Type of diabetes
LADA
My interest in hba1c is that it is used to assess risk of complications, assuming of course it reflects a generally steady glucose level rather than an average of extreme highs and lows.
 

CearaRed1073

Member
Messages
24
Type of diabetes
Type 1
Treatment type
Pump
Dislikes
onions, cheeseburgers, diabetes, colors of red, yellow, and orange, houston (too big and WAY too much traffic, not nearly enough public transportation)
My interest in hba1c is that it is used to assess risk of complications, assuming of course it reflects a generally steady glucose level rather than an average of extreme highs and lows.

which is PRECISELY what mine is, a 90-day range of everything from my odd 400+ as well as my 40-whatever.
on the other hand, i've been type-1 since October 1978 and the only complication i've had is cataract surgery in 2008, three years before my mother had it. my last three HbA1c readings were 6.7 on 29 August, 6.7 on 22 april, and 7.0 on 14 january.
 
  • Like
Reactions: ChrissiStar
M

Member496333

Guest
...assuming of course it reflects a generally steady glucose level rather than an average of extreme highs and lows.

Which of course is the greatest weakness of the test. It cannot assume a steady glucose level. It literally is a mean average of the extreme highs and lows, and even then it's an indirect measurement with potential for error. It's a useful tool if used alongside actual glucose data, but on its own it's a very weak measure of control. Certainly in early diagnosis of hyperinsulinemia (and inbound type 2) it's next to useless, but that's a different subject for another day.
 

smc4761

Well-Known Member
Messages
1,039
Type of diabetes
Type 1
Treatment type
Insulin
Hi @jellyfishfeet i had my annual check up last week. I had put in a great deal of effort over previous 6 months or so and my libre was advising a reading of 51. I was shocked and disheartened when my Hba1C came back as 60. The doctor reviewed my Libre reading s as did my specialist diabetic nurse and they too were surprised my reading was so high

However i was told not to get too hung up on the Hb1aC figure

libre is not perfect but is generally a good indication of how were managing and i would not be without it
 

LooperCat

Expert
Messages
5,223
Type of diabetes
Type 1
Treatment type
Other
My Libre predicted an HbA1c of 26 last time, compared to a lab result of 35. It consistently reads 2-3mmol below blood - but not consistently. However, the app I use with my MiaoMiao that collects the signal from the sensor is blood calibratable, and that predicted 34.5 - so accurate. I rarely bother scanning with the official app these days, the algorithm is way out for me.

Time in range is much more important imo, and used in conjunction with standard deviation, glucose variability index and the HbA1c, our medics can get a very good impression of our control. I can easily justify my non diabetic HbA1c to my team when they get twitchy that’s it’s so low by showing them my graphs and stats.
 

jellyfishfeet

Member
Messages
17
Type of diabetes
LADA
My Libre predicted an HbA1c of 26 last time, compared to a lab result of 35. It consistently reads 2-3mmol below blood - but not consistently. However, the app I use with my MiaoMiao that collects the signal from the sensor is blood calibratable, and that predicted 34.5 - so accurate. I rarely bother scanning with the official app these days, the algorithm is way out for me.

Time in range is much more important imo, and used in conjunction with standard deviation, glucose variability index and the HbA1c, our medics can get a very good impression of our control. I can easily justify my non diabetic HbA1c to my team when they get twitchy that’s it’s so low by showing them my graphs and stats.

Would you mind sharing which app you use? And do you show them the calibration? Have they made any comment on the libre performance or suggested a dexcom etc? Thanks for your reply - I think we have a similar experience of using the libre!
 

LooperCat

Expert
Messages
5,223
Type of diabetes
Type 1
Treatment type
Other
Would you mind sharing which app you use? And do you show them the calibration? Have they made any comment on the libre performance or suggested a dexcom etc? Thanks for your reply - I think we have a similar experience of using the libre!
I use an app called Spike, which is only for iPhone - similar apps are available for Android phones. You get it from spike-app.com and it needs to be installed onto an iPhone using a computer, as it’s not in the App Store. I show them my data through Nightscout, Spike can set up to send data there.
 
  • Like
Reactions: jellyfishfeet

ChrissiStar

Well-Known Member
Messages
90
Type of diabetes
Type 1
Treatment type
Insulin
I just checked my Libre app for the estimated HbA1c and it’s showing 39, however I am not expecting an HbA1c anywhere near that as I found I was getting a lot of false lows while using it and I had whole nights where the Libre had me under 4, but finger tests had me sitting at 5.4 or so.
 
  • Like
Reactions: jellyfishfeet

slip

Well-Known Member
Messages
3,523
Type of diabetes
Type 1
Treatment type
Insulin
There are other factors that affect the HbA1c as well, anaemia being one of them, there is no way the Libre could take that in to account when 'averaging' your readings.
 

barrym

Well-Known Member
Messages
800
Type of diabetes
LADA
Treatment type
Insulin
@barrym Well if you have been self funding Libre. Lowering your HBA1C whilst on it, is part of the criteria for NHS funding.

Hmm, I did self fund, and am now getting them from NHS but my consultant looks first at my time in range. Her first comments are about below range as I used to spend too much time down there. Then she moves up the range.

At my last meeting with her I was going out the door when she said "Oh, don't you want your HbA1C?"

I suppose they are only human, some like some indicators better than others. HbA1C just seems so coarse.