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

A1cnow

O

Onerunshort

Member
Messages
13
I decided to try out A1CNow at home as an indication of how things are going in between official tests. My latest NHS and private tests have been between 38 and 41. I followed the instructions of A1CNow carefully and got a mildly shocking result of 47. I mulled it over for a few minutes and decided, what the heck, to run the test again about 15 minutes later. Following the same procedure I got a result of 36! I think I might have to forget this experiment and await lab test results in the future.
 

Onerunshort - I use A1cNow regularly, just for my own info. On several occasions I have utilised an A1cNow on the same day as my lab tests, and have never found them to be more than 0.1% adrift on the lab.

I have recently finished a box, so I really should order some more.
 

My A1cNow came out at 33. My lab test came out at 43 a few days later. Now I know my HbA1c is ALWAYS a lot adrift of my own data (regular finger pricks at peak times and also a Libre sensor that all tallied with the A1cNow) So I queried it with my GP who contacted the biochemist consultant at the lab. This consultant suggested I have blood drawn again, sent off to her lab and then forwarded to another lab that uses a different method of testing. Results came back. Normal lab 44. New lab 40. Same vial of blood. My GP has now told me she will send all future HbA1c tests to both labs.

I have no confidence whatsoever in my HbA1c results.
 
Not wishing to be pedantic but as the A1CNow results are whole numbers I don't see how they can be 0.1% out. Even if your lab result was 40.1 against an A1CNow result of 40 that's considerably higher than 0.1%.
 
Onerunshort said:
Not wishing to be pedantic but as the A1CNow results are whole numbers I don't see how they can be 0.1% out. Even if your lab result was 40.1 against an A1CNow result of 40 that's considerably higher than 0.1%.
Click to expand...

A number of the packs I have had, including the box I just finished expressed as DCCT, and therefore as a %, rather than IFCC.

My spreadsheet tracks my A1c in both formats.
 
DCUKMod said:
A number of the packs I have had, including the box I just finished expressed as DCCT, and therefore as a %, rather than IFCC.

My spreadsheet tracks my A1c in both formats.
Click to expand...
OK, sorry I understand. I am just concerned that 2 tests so close to each other could give 2 such different results. If I had a lab test today and it came back at 36 I would have no idea if the A1CNow was 11 points out or completely accurate. Then again that assumes I have complete faith in the lab test result. It does make it difficult to manage having readings of 36,38 and 47 in such a short space of time.
 

I can appreciate how unsettling that would be. I was only reporting back my personal experience.

Like @Bluetit1802 , above, my A1cs never match y average finger prick tests or Libre sensor readings, but for some reason, the A1cNows have always tallied well against my lab results when I have done them on the same day.

I guess you checked against all the usual stuff, and had washed your hands before testing, then re-washed before the second?
 
Yes followed the same procedure. The only difference was I used a different finger. Obviously a less sweet one
 
Did you get the test strips out of the fridge for the correct time?
 
There's always going to be an error in any measurement like this. Finger prick BG tests have a precision of around 15%. Depending on the method used to measure HbA1c, the error can be as high as 10% (note, that's 10% of the measurement value, not 10% difference in the DCCT result). So the 34 & 36 reported by @Onerunshort are, effectively, equal. So are the 40 & 44 by @Bluetit1802.

My suspicious are that A1cNow won't be as accurate as a lab result. A bit of Googling show that to be true:

"Although the A1CNow+ had good sensitivity, its accuracy was insufficient for use as a replacement for laboratory measurements of HbA1c, for glycemic control monitoring in diabetic patients."

...from here
 

Can you tell me where you get this information from please? If the error can be as high as 10% it could make a massive difference to type 2s on diagnosis if they hover around the threshold number of 48. It could mean some (many?) are diagnosed as diabetic, possibly for life, when in fact they were not diabetic at that point and may never be.

I read somewhere, and can't recall where, that the error could be up to 4%. I haven't seen any reference to 10% so would be interested to read about it.
 
Sure, if you Google for “coefficient of variation for HbA1c”, you’ll get a load of peer reviewed papers that have compared labs and home testing kits. One study found that the variation was 10%, most seem to be around 3-5%.

I’m not sure it will affect diagnosis, since nondiabetics have an HbA1c around 35 mmol/mol. Accounting for a max of 10% error it could show as high as 39 mmol/mol, which I don’t think will trigger a diagnosis; nor will the patient have any symptoms of diabetes at that level.
 
My results were 47 and 36 within 20 minutes of each other, not 34 and 36.
 
SamJB said:
Sure, if you Google for “coefficient of variation for HbA1c”, you’ll get a load of peer reviewed papers that have compared labs and home testing kits. One study found that the variation was 10%, most seem to be around 3-5%.
Click to expand...


We are diagnosed with diabetes at 48 and above, not around 35 as you suggest. With HbA1cs of 42 to 47 we are told we have pre-diabetes but are not diagnosed and not put on the register. So a difference of 10% can make all the difference in the world.

I will have a google. Thank you.
 
Onerunshort said:
My results were 47 and 36 within 20 minutes of each other, not 34 and 36.
Click to expand...
Definitely inconsistent then. Although these measurements will never be completely precise, particularly when comparing the results from different labs and different measurements devices.
 
Sorry, I don’t think I explained myself well enough. I’m not saying 35 is a diagnosis of diabetes - I’m saying the opposite! If someone went to the doctors and the doctor suspected diabetes, did an HbA1c test, which came back with a non-diabetic reading of 35+/-10% (I.e a range of 32-39), they wouldn’t be diagnosed with diabetes. Nor would they likely have any symptoms of diabetes.
 

No, but someone with an HbA1c of 48 will be diagnosed with diabetes, and also unlikely to have symptoms. But if there is a 10% error he may well not be diabetic, but will still be entered on the register even though his true level may only be 43/44. Being on the register can, in some cases, have consequences with employment, insurance, and so on.
 
You must log in or register to reply here.
Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn More.…