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

Hba1c accuracy

F

Flora123

Well-Known Member
Messages
1,078
Type of diabetes
Type 2
Treatment type
Tablets (oral)
I recently had my annual checks with my GP. At the blood draw at the hospital I asked them to do extras for Medichecks tests. Interestingly the Medichecks Hba1c was 34 (34.01) and the NHS at the same time was 36 (the same as I have been three times previously)!

Funny how it can be different. I’m not bothered, but thought I’d mention just in case anyone gets disheartened at a 1 or 2 point rise. It may not be that accurate. I wonder which one is right though!
 
Flora123 said:
I recently had my annual checks with my GP. At the blood draw at the hospital I asked them to do extras for Medichecks tests. Interestingly the Medichecks Hba1c was 34 (34.01) and the NHS at the same time was 36 (the same as I have been three times previously)!

Funny how it can be different. I’m not bothered, but thought I’d mention just in case anyone gets disheartened at a 1 or 2 point rise. It may not be that accurate. I wonder which one is right though!
Click to expand...
As @Bluetit1802 will confirm, different labs use different methods of analysis with consequent variations in results. Your results only vary by about 5% and are very good anyway so I wouldn't be concerned.
 
No test or measurement will ever be repeatable with absolute accuracy. There are always errors of some magnitude. I would imagine especially so when dealing with biological samples, and definitely from one lab to the next.
 
While the blood was taken at the same time, there will have been differences in how it was stored and transported which may affect the end result, along with when the actual test took place.

In the grand scheme the difference is marginal, and both show excellent control.
 
Flora123 said:
Interestingly the Medichecks Hba1c was 34 (34.01) and the NHS at the same time was 36 (the same as I have been three times previously)!
Click to expand...

I'm assuming that's mmol/mol. A mole is 6.02214076×10 (to the power of 23) particles, typically molecules. That number is known as Avogadro's Number. So, with a difference between 34 mmol/mol and 36 mmol/mol being 2 mmol/mol, although decimal 2 seems like a lot, 2 mmol is not.
 
I’m not worried but was only commenting as there have been a couple of people posting about a slight rise and how disappointed they were.
 
Flora123 said:
I’m not worried but was only commenting as there have been a couple of people posting about a slight rise and how disappointed they were.
Click to expand...

Last year my GP agreed with me that my HbA1c results were significantly higher than any other data suggested so she arranged for my HbA1c to be tested at 2 different labs, which use different machinery. From the same vial of blood the usual lab result was 4mmol/mol higher than the other lab, and was 11mmol/mol higher than an A1cNow test on the same day. I no longer rely on my HbA1c, I prefer to rely on my own finger pricking tests.
 
Bluetit1802 said:
Last year my GP agreed with me that my HbA1c results were significantly higher than any other data suggested so she arranged for my HbA1c to be tested at 2 different labs, which use different machinery. From the same vial of blood the usual lab result was 4mmol/mol higher than the other lab, and was 11mmol/mol higher than an A1cNow test on the same day. I no longer rely on my HbA1c, I prefer to rely on my own finger pricking tests.
Click to expand...

If both labs used the same vial of blood, that itself could be part of the anomoly, as they would not be done at the same time - one sample being older than the other, in terms of transportation one lab to the other?

I agree we have to look on our HbA1cs as a different snapshot (to finger prick testing or Libre trace calculations) and an indicator, rather then the absolute our GPs might suggest.

@Flora123 - You could always go for 35 as the average? Nice result. :)
 
Back
Top