• Guest - w'd love to know what you think about the forum! Take the 2025 Survey »

How accurate do people find libre predictions of hba1c?

Boo1979

Boo1979

Well-Known Member
Messages
1,849
Type of diabetes
Other
Treatment type
Tablets (oral)
I’ve been diagnosed T2 for 21years and on a stable low dose of Gliclizide for all of that time apart from a month or 2 on insulin at the beginning. Control has always been good although I have had to gradually reduce carbs as my initial tolerance of around 35-40g per meal has dropped such that 35-40g is closeer to my daily tolerance
I recently invested in 2 month supply of sensors in order to get a better picture of overall pattern of control leading up to next review in the endo clinic, where I plan to ask them to think about tests to follow up a suggestion they made that I may have monogenic rather than T2 diabetes
6 weeks in and the libre is producing pretty flat graphs and reports that my sugars heve been within the target level thatt I set (4-7.8 mmol) for 91% of the time (100% for the last 2 weeks !) with no hypos throughout. It gives a projected hba1c of 39 with overnight and mornig average BG in the 5’s and afternoon and evening average in the 6’s
I’m thinking through how to use that info at my annual review but, being of a sceptical nature, also wonder how accurate the system is.
Any thoughts / experiences?
 
Last edited:
I don’t know about hba1c but I found the Libre very inaccurate compared to finger pricks.
I understand some people find them accurate and some do not. I just have an incompatible body/lifestyle :-)
 
For me, it wasn't particularly accurate at best, but it's tendency to get a bit overexcited in the 4 - 5 range, and read it as 2 - 3, meant that the hba1c prediction was not even close - if memory serves, it was about 10 - 15 under, which is quite a big difference. I guess it also depends on if you think you are likely to be lying on it at all overnight, because the 'compression hypos' it suffers from are extreme, even if it is just squeezed between your arm and your body.

TL;DR - I wouldn't put too much stock in the predicted hba1c value.
 
cheers
Ive tried doing some finger prick tests alongside the libre but find the results difficult to assess - the readings are often identical but at other times can differ by up to 1.4 mmol.
Ive tried different gaps between taking readings with the libre and BG meter (0-30mins) because of the time lag between blood and interstitial fluid glucose - the longer gaps tend to bring out of synch readings closer togather but its unpredictable
I dont think the compression thing is a factor as the libre shows sugars happily pootling about in the 4’s and 5’s overnight and throughout the morning. All very puzzlung but I still think the sensors were worth the investment, but only in the short term
 
Last edited:
My experience with the 9 sensors I've used is that I totally ignore the estimated HbA1c. One sensor gave mine as 27. It was actually 41! :arghh: All the others have been in the mid 30's when I have never dropped below 41 - and that includes one I sent back because it was giving me very silly high readings. Having said that, my HbA1c is always higher than expected possibly due to my haematocrit levels.
 
Well my suspiscion that sensor 3 was reading low was confirmed last night, which was its last hurrah before expiring. According to the sensor my sugars were 3,4-3.9 at 1.00am ish but meter test put them 1mmol higher. I took10g carbs on board ‘ just in case’ so I could sleep without worrying.
I was still soundlyasleep at 8am, but the Libre shows me at beng on a 2.6 at that point wheres 90 mins later meter put me at 5.7.
Just to confuse me furtherr, the new liibre sensor and my meter currently agree Im at 8.5 despite a total carb intake so far today of 11g alongside 2x Gliclazide with nothing eaten for nearly 3 hours - Its certainy messing with my head!
 
Last edited:
My libre readings and predicted HbA1c are always significantly lower than prick tests and NHS HbA1cs.

My sensors are anywhere from 0.5 to 1.5 mmol/l below prick tests (but each sensor is usually very consistent about it)
And my HbA1c is higher than the Libre predicts. I think my last NHS HbA1c was 40 compared with a prediction on the Libre of 33.

Even allowing for the difference between the Libre mmol/l readings and my prick tests, I was expecting an NHS reading around 36, not 40.

But then, NHS HbA1cs have never lined up with my prick tests. In fact the disparity has been so great that I now consider myself a high glycator and think that my HbA1cs are not representative. I just shrug and disregard those figs preferring to go with prick tests, occasional Libre sensors, and focusing on keeping my bgs steady to avoid damaging peaks and troughs. I disregard the numbers and look at the graph lines, aiming for the flatter the better.
 
I find that if I get up and walk to the kitchen and back..Take a BG test straight away..The reading is higher. I do a second test sitting for 4 mins, then the reading is quite a bit lower..I never take my BG test now straight away from even walking to the kitchen.
 
Back
Top