My Freestyle Libre 2 plus has for some time estimated that my A1c was 37mmol/mol, based on 90 day glucose data from the sensor. My actual result from a blood test just taken was 43mmol/mol. I would say the Libre fairly consistently underestimates my blood glucose by perhaps 20%. I tend to forget this and get carried away with brilliant readings, but I don't mind, as long as it records the ups and downs fairly accurately. How accurate is yours?
-
Guest - w'd love to know what you think about the forum! Take the 2025 Survey »
How Freestyle Libre 2+ measured against an actual A1c result.
- Thread starter Rosie9876
- Start Date
- Messages
- 667
- Type of diabetes
- Type 2
- Treatment type
- Insulin
I’ve tracked mine for a couple of years now and it is always 5 points under a blood test.
Always 8 mmol/mol below my measured hba1c for over 8 years.
Oddly, the LibreView website is closer in their estimation, based on the same data.
Oddly, the LibreView website is closer in their estimation, based on the same data.
grantg
Well-Known Member
- Messages
- 1,035
- Location
- scotland
- Type of diabetes
- LADA
- Treatment type
- Other
- Dislikes
- needles, bloods sampling
90day a1c in app? vs glucose report time period perhaps? estimated changes in report if i swap 2 week (default) to say a 60 day report. very odd if same period if different estimations based on same data. libres estimated been a touch lower for the few hba1c's have had too.
some ppl libre reads lower others higher guess depending if constsistantly reading higher/lower that will reflect change. I do wonder however if 'compression lows' have something to do with it when sleeping, even thou no pressure on the sensor if i lay down in bed can see it drop a little bit so to me that would explain part of estimated a1c appearing to read a little lower.
My lab HbA1c was always higher than the Libre 2+, difference anywhere between 1 mmol/mol and 8 mmol/mol!
I’ve recently switched to DexcomONE + so interested to see how that pans out. After 30 days the estimate seems a lot more realistic than the Libre, but will have to see.
Edit to correct unit of measurement
I’ve recently switched to DexcomONE + so interested to see how that pans out. After 30 days the estimate seems a lot more realistic than the Libre, but will have to see.
Edit to correct unit of measurement
Last edited:
grantg
Well-Known Member
- Messages
- 1,035
- Location
- scotland
- Type of diabetes
- LADA
- Treatment type
- Other
- Dislikes
- needles, bloods sampling
1st may: 41 mmol/mol (data spans 90 of 90 days) compared to Hba1c: 46 mmol/mol (6.4%) not too far out 
october: 29mmol/mol (data spans 90 of 90 days) compared to Hba1c: 38 - slightly further out
19/01/2025 06:33 162.0 HbA1c (mmol/mol)
honeymoon period advantage alongside fairly lowish carbs but not keto.
october: 29mmol/mol (data spans 90 of 90 days) compared to Hba1c: 38 - slightly further out
19/01/2025 06:33 162.0 HbA1c (mmol/mol)
honeymoon period advantage alongside fairly lowish carbs but not keto.
Both set on 90 days.
The lower estimation on the LibreLink app seems to correlate with my Libre readings, which are always 1 to 2 mmol/l lower than a fingerprick.
So I have no idea how LibreView arrives at a higher number, which is much closer to my real hba1c but doesn't add up with Libre reading lower than finger pricks.
By the way, I've just had a look at the estimated hba1c in the app, and it looks like the estimated hba1c is still to low, but less than 8 mmol/l out. I haven't has an hba1c test since january but I'm quite sure it hasn't risen by 5 mmol/mol, so maybe they've recently changed the calculations.
