HbA1c's relationship to bg readings

[Snodger]

I've just been playing with this very useful calculator thingie:
http://www.diabetes.co.uk/hba1c-to-bloo ... erter.html

Can anyone at diabetes.co.uk help, where did you get the info from to make the calculator? i.e. there must be some scientific ratio that explains how hba1c translates to daily bg averages and so on - can you point me in the direction of where to look for the references?

 

[phoenix]

I've answered this before, so just had to cut and paste
I don't know the formula used for the one on this site.

Here's the formula adapted from the ADA one. As far as I know this is the only formula based on recent worldwide evidence.
(though you're supposed to use it the other way round from Hb A1c to estimated average glucose level)
(average glucose in mmol/l x18) + 46.7
...................28.7
Sorry about the extra dots, it's the only way I could get the 28.7 into the 'right' place!
http://professional.diabetes.org/GlucoseCalculator.aspx

 

[Snodger]

brilliant, thanks phoenix. I knew the answer would be on here somewhere but wasn't sure what search terms would bring it up.

I've just been doing some reading for my studies and was looking at old research pre-HbA1c test, wanted to try and compare the bg readings they were using with modern stuff.

Found one study from 1981 looking at what made some diabetics have 'good control' compared with others, and the 'good control' group had an hba1c of 10%. Interesting. Times do change.

 

[sugar2]

Hi,

I was at the diabetes clinic last week, adn they had a conversion chart on teh wall....I remembered that 48 on the new scale, was a 7 on the old, according the DUK site...so I looked at it, to conform that an "under 48" would a good figure, only to fond that they had an old 7, as a 54..

Wish I could remember the name of the company who produced it..might have been useful! I do not know who is right/wrong, or whether statistically this is within experimental error.

 

[Snodger]

that's odd, DUK site at the moment says 48 is 6.5% in 'old money' - http://www.diabetes.org.uk/hba1c - maybe they had it wrong and corrected it?

 

[phoenix]

I've just played around with the converter on this site. It rounds up or down to the nearest whole percentage when you try to convert mmol/mol to DCCT HbA1c % It's more accurate the other way round from HbA1c %.
The formulas are: (I cut and pasted found them from one of the 'official docs but I forgot to link)

IFCC HbA1c unit (mmol/mol) = 10.93 x DCCT/NGSP unit (%) – 23.50
DCCT/NGSP HbA1c unit (%) = 0.09148 x IFCC unit (mmol/mol) + 2.152
therefore
48mmol/mol= 6.54304%

This seems to be a good chart
http://baspath.co.uk/Hba1c_table.pdf