I hope you are well

[Benny G]

I hope you are well

 

[Robin]

I just had my HbA1c come in at 48, and the Libre was predicting 42. I have to take into account the fact that the Libre reads low at night when I’m lying on it, so this will skew the average. I also think it reads lower at the low end generally, and higher at the very high end, and I try and stay out of the very high end, so again, this affects the average.

 

[everydayupsanddowns]

Latest hba1c has come in as 50.

I find it curious that my hba1c and libre/xDrip estimates are so wide apart.

Hba1c result is 50
Libre estimate is 40
Xdrip estimate is 36
~~~~~~~~~~~~~~~
Hba1c result is 50
Libre shows a 90 day average of 6.7 which gives an estimated reading of 40
View attachment 12499

Xdrip estimate is 36
View attachment 12498

Any chance of a little anaemia? That can affect HbA1c results.

I also believe some people are ‘high glycators’ in which case they tend to make more glycated haemoglobin at lower BG values.

Having said that, when wearing Enlites my average TIR is often 80-85%ish in range (1% below and the rest above) and my A1c is generally 47-49ish too.

 

[Eddy Edson]

But an HbA1c of 50 mmol/mol is actually consistent with an avg BG of 6.7 - it's within the 95% confidence interval, according to the original study which established the common HbA1c => avg BG translation.

https://www.ncbi.nlm.nih.gov/pubmed/18540046

Ref this table:



Your 50 mmol/mol HbA1c = 6.7%. The 95% confidence interval range for HbA1c = 7% is 6.8 mmol/L to 10.3 mmol/L. So 6.7 mmol/L would certainly be within the 95% CI range for HbA1c = 6.7%.

That's the foundational data from non-diabetic, non-anaemic etc etc subjects.

It peeves me in a nerdy way that these kinds of numbers always get reported without their uncertainty ranges, as if they were hard and fast definite things, which they ain't.

As the authors express things, one in three "normal" subjects will have avg BG differing by > 15 mg/dL (0.8 mmol/L) from the conventional HbA1c <=> avg BG translation. Assuming your Libre is accurate, you differ by more than 15 mg/dL, but so will a bunch of other people.

In a later study https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5714656/ the same group refines things by taking into account individual variability in avg red blood cell age. With that, only one in ten people have avg BG differeing by more than 0.8 mmol/L from the conventional translation.

(I guess that the profession hasn't considered the accuracy improvement clincally relevant enough to justify the added expense and complexity of measuring avg red blood cell age.)

 

[AJLang]

50 mmol = HBA1C of 6.7 (in old money) which equals an approximately average blood glucose of around 8.0/8.1
I frequently find that my Libre readings differ by about 1.0 to 2.0 in comparison to blood testing readings and I trust the blood testing more.

 

[Robin]

I’ve noticed everyone I've encountered in the Health care field always takes HbA1c as the 'Gold standard' as if it’s infallible, but as has been said, it varies from person to person, depending on how many red blood cells you’ve got, and how long your body makes them last.

 

[Schrodinger]

Personally I never trust Libre or xDrip for HB1Ac figures. I record my figures in MySugar and the more figures I stick in the more the Hb1Ac results change, its rather amusing to watch.
But like all bit of software it does say "Estimated" and in Lire it does say "However, the estimated levels may not match your A1c measured in a laboratory".

So what does the software do? At a guess it simply gives you and average of the data to input to the device and while it might be interesting to see if your control is ~improving~ or not by the fact it is going up or down, I still fail to see why people get so obsessed or upset when the figures don't match at the end of the day xDrip is free and Libre software will be developed by the lowest bidder.

Still could just be me but anything that gives me "an estimate" I always take with a pinch of salt.

 

[everydayupsanddowns]

Personally I never trust Libre or xDrip for HB1Ac figures. I record my figures in MySugar and the more figures I stick in the more the Hb1Ac results change, its rather amusing to watch.
But like all bit of software it does say "Estimated" and in Lire it does say "However, the estimated levels may not match your A1c measured in a laboratory".

So what does the software do? At a guess it simply gives you and average of the data to input to the device and while it might be interesting to see if your control is ~improving~ or not by the fact it is going up or down, I still fail to see why people get so obsessed or upset when the figures don't match at the end of the day xDrip is free and Libre software will be developed by the lowest bidder.

Still could just be me but anything that gives me "an estimate" I always take with a pinch of salt.

well there’s no mathematical conversion between estimated (or measured!) average glucose and A1c. Most of the formulas that are exist are based around the best fit of some empiracle paired data sets (those with measures of average capillary glucose and lab-measured A1c across a bunch of values and than working out what conversion formula gives a decent match).

I always find Libre estimates on the low side for me. Usually by 0.5% - 1% of A1c in old money.