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

Meter average VS HBA1c

E

ebony321

Well-Known Member
Messages
1,299
Dislikes
Tomatoes, Rude people, Bees!
Hi,

Just wondering how accurate is the meter average compared to your hba1c result?

I've never looked to see if my 60/90 day average is close to my hba1c.

Just wondering if anyone has taken note of this and how accurate the meter average has been?

spanks :)
 
It's not the same thing. Although you can use your meter average to arrive at an estimated HbA1c using the formula:

HbA1c = Average BG (90 day) + 2.52 / 1.583
 
Why isn't it the same thing?

How is that formula worked out?

Am i going to get confused again? :lol:
 
There is not, as far as I am aware, any way to make a direct and accurate comparison between meter readings and A1c.

There are approximations but that is all they can be.

In the mid ranges around 5-7% I have found some similarity, an average of 6 mmol gives a near to 6% A1c or maybe a bit less. (you will have to do convertions into other units).

All a meter can really tell us is an approximate reading with a not very accurate calibration. It is more a way of noting changes rather than an absolute measurement.

An A1c on the other hand is very accurate and repeatable at most labs. I suppose it is whay the Docs. prefer it!

H
 
Average readings should come pretty close once run through the above formula to your HbA1c... I use a modified version of Patch's above formula on my excel spreadsheets that applies a weight to a set of average over the past 90 days as this gives a more accurate measure (as reading from the last 7 - 30 days are the most important)... but the above is normally within half a percent...

so for my last HbA1c my weighted average said 5.17% and the formula above over 90 days of readings (around 6 - 10) per day came out at 5.35% so not to far off my 5.2% test.
 
I see,

I'm going to keep a note of what my 90 day average, and 60 day average is, a week before my hba1c and see if it comes close.

I understand the readings we get from a meter are slightly out form what they actually are but that was my question how accurate it can actually be.

Interesting your formula almost got it spot on really!

I'm too impatient to wait every 3 months for a result :lol:
 
the blood test gives an average blood glucose over between 2.5 and 3 months. It is reckoned to be more accurate than the meters - which can vary greatly, but are good for an approximation. If you use 2 different meters you will often get 2 different results. I find I can differing results from different hands too.

Cheers
Ailz
 
I can only speak from my experience but I think as long as you test regularly then using the above formula you are going to be reasonably close to your HbA1c... If you aren't then that would suggest a problem to me... i.e. you have a period of time where you aren't testing and you blood sugars are doing something odd?!! The one that springs straight to mind is that a lot of people test during the day but not many people test all to often at night! night-time is for instance 1/3 - 1/2 of your day where your sugars may be doing something you don't expect.

I am the first to admit that I am OCD / Anal about my resting regime and normally test at least once during the night (a five month old daughter helps here?!) this probably isn't for everyone but it may be worth picking a weekend night and setting your alarm for every 2 hours to see what those sugars are doing if you get a vastly different result
 
Convert HbA1c to Average Blood Sugar Level use link
http://www.diabetes.co.uk/hba1c-to-blood-sugar-level-converter.html

What is the HbA1C? read this link below
http://medweb.bham.ac.uk/easdec/prevention/what_is_the_hba1c.htm


in the blood stream are the red blood cells, which are made of a molecule, haemoglobin. Glucose sticks to the haemoglobin to make a 'glycosylated haemoglobin' molecule, called haemoglobin A1C or HbA1C. The more glucose in the blood, the more haemoglobin A1C or HbA1C will be present in the blood.

Red cells live for 8 -12 weeks before they are replaced. By measuring the HbA1C it can tell you how high your blood glucose has been on average over the last 8-12 weeks.
Click to expand...
 
Pneu said:
I can only speak from my experience but I think as long as you test regularly then using the above formula you are going to be reasonably close to your HbA1c... If you aren't then that would suggest a problem to me... i.e. you have a period of time where you aren't testing and you blood sugars are doing something odd?!! The one that springs straight to mind is that a lot of people test during the day but not many people test all to often at night! night-time is for instance 1/3 - 1/2 of your day where your sugars may be doing something you don't expect.

I am the first to admit that I am OCD / Anal about my resting regime and normally test at least once during the night (a five month old daughter helps here?!) this probably isn't for everyone but it may be worth picking a weekend night and setting your alarm for every 2 hours to see what those sugars are doing if you get a vastly different result
Click to expand...

This was sort of my reasoning behind my question. I wouldn't say i am 'OCD' about my testing, but i do test regularly. I do occasionally do 3am tests, but only really if i'm investigating something, i will also test during the night if i wake up, as i usually sleep like a log, im curious if it's something BG wise that has woke me up.

If i use patch's theory and use my 90day average which is 7.9, gives me an answer of 6.6 (rounded up) so we'll see how accurate it is as i will be having my next hba1c on wednesday.

I'd be quite surprised if it was 6.5 though!
 
There are a number of places where you can read how over trusting the HBA1c, especially for T1's, can not be telling the whole story. For example an A1c of 6.0 can be the 'average' of VERY HIGH numbers AND hypos which will even out to a 6.0. The person receiving the 6.0 could be giving themselves a pat on the back when they're still potentially doing a lot of damage to their bodies.

Monthly meter averages tell a better story in my opinion for the T2's. The reason for this is that some people have RBCs (red blood cell) lives that are shorter or longer than others. This will affect A1c in an 'overinflated' or 'underinflated' number where monthly meter averages are really telling the more accurate picture of your control regimen. The docs prefer A1c over meter log books for obvious reasons it seems to me. I agree meters can vary.. And because I know that my A1c isn't an averaging of lots of highs and hypos -- I actually chose my final meter that I use BASED on how close it matched/predicted my next A1c. I'm always within 0.1% on my next Lab A1c using the meter averages from the meter I use now.

For those interested. I've found the Freedom FreeStyle series of meters to be slightly inflated and the Walmart ReliOn Micro and Bayer AccuCheck to me a little 'under inflated'. I use the ReliOn CONFIRM and Bayer Breeze 2 and have found them to be spot on when their Monthly BG averages are compared to my Lab A1c. This also means I have pretty 'normal' Red Blood cell lifespan. If I didn't then such a tight tracking wouldn't be possible. That may be the case for several of you.
 
NewdestinyX said:
There are a number of places where you can read how over trusting the HBA1c, especially for T1's, can not be telling the whole story. For example an A1c of 6.0 can be the 'average' of VERY HIGH numbers AND hypos which will even out to a 6.0. The person receiving the 6.0 could be giving themselves a pat on the back when they're still potentially doing a lot of damage to their bodies.

Monthly meter averages tell a better story in my opinion for the T2's. The reason for this is that some people have RBCs (red blood cell) lives that are shorter or longer than others. This will affect A1c in an 'overinflated' or 'underinflated' number where monthly meter averages are really telling the more accurate picture of your control regimen. The docs prefer A1c over meter log books for obvious reasons it seems to me. I agree meters can vary.. And because I know that my A1c isn't an averaging of lots of highs and hypos -- I actually chose my final meter that I use BASED on how close it matched/predicted my next A1c. I'm always within 0.1% on my next Lab A1c using the meter averages from the meter I use now.

For those interested. I've found the Freedom FreeStyle series of meters to be slightly inflated and the Walmart ReliOn Micro and Bayer AccuCheck to me a little 'under inflated'. I use the ReliOn CONFIRM and Bayer Breeze 2 and have found them to be spot on when their Monthly BG averages are compared to my Lab A1c. This also means I have pretty 'normal' Red Blood cell lifespan. If I didn't then such a tight tracking wouldn't be possible. That may be the case for several of you.
Click to expand...

I understand that HBA1c's aren't always truthful in reflecting control, as it is an average, but i personally know wether mine is reflective or not, because of regular testing. I know my last hba1c of 7.2% wasn't entirely truthful. I think it should of been higher, late 7's possibly, this was due to a fair few hypo's whilst starting insulin pump. So i know if my next one is the exact same, it will be because i'm not dipping low and having high swings, which i know i'm not because i regularly test.

It's a good point that meter's aren't always accurate. I've used a fair few meters. But have never compared the averages to my hba1c before. The meter i use now is also a bluetooth remote control for my insulin pump, so finding a meter that reflects my hba1c (unless it already will do) isn't much of an option for me.

Again this is why this site and many other places (i hope) will always encourage T2's to test, because the hba1c relied upon alone doesn't show a good picture.

Doc's like hba1c's as it's a result, it can be recorded and is also dependant on the figures for how well the diabetes clinic does. A percentage of people with diabetes that have hba1c's in target with determine funding, bonuses etc.

For a while now my BG's are stable, not so many swings and not very many highs at all. The pump has been brilliant in enabling me to control my diabetes, so i look foward to a good result. I will certainly post to let peeps know if they are interested wether my meter was on target! i will also put patch's thoery to the test too :) so i shall have 3 results, all should be similar but we'll soon find out :)
 
As everyone has said whether your HbA1c and your averages will be the same as the 'average persons' does depend on frequency of testing, accuracy of meter etc.
There are some other caveats.
The study that produced the most recent formula did not include all ethnic groups 8% were of African descent. 8% were of 'Hispanic' descent but there were only 8 others of non white ethnicity so it may not apply for all ethnicities.
Whether you are 'average' may also depend on whether you're what they call high or low glycator. In a group of people with a similar average some will always have a lower HbA1c than the average, some a higher HbA1c than average possibly because some people glycate more haemoglobin than others.
The authors of the study felt that if 90% of the estimates being within 15% of the regression line this would be a valid measure and this was fulfilled but this also means that 10% of people had HbA1cs that were either below or above that level ... and that's a lot of people.
Lastly the people studied didn't have any other problems and they had stable control. People with some problems such as kidney disease may well have HbA1cs that don't conform to their average levels.
So in the end an estimated average is just that, not a precise figure . It may work for you, but there are a lot of explanations if it doesn't.
http://www.aacc.org/publications/cln/20 ... _1008.aspx
 
this is extremely creepy..

my results came in today.

Patch, your little thoery is spot on.

Using my 90 day average with patch's equasion it gave me a result of 6.6%

today my results are in and my hba1c is 6.6%

:shock: :D
 
I might be able to help here. First of all, the conversion from average bg to HbA1c depends on wether your bg meter is reading whole blood plasma or not. As I understand it, newer meters read whole blood plasma and older ones don't.
If yoiu read whole blood plasma, thern the equation is to add 4.29 to your average bg then divide by 1.98. This is your HbA1c equivalent. If it is not whole blood plasma you then multiply by 1.12 to get HbA1c. Your meter provider can tell you what you're reading.
The bigger problem is in arriving at an accurate average bg . You can't just take an average of all your readings. You need to know a "base" level that your blood glucose is at before it goes up after eating, how long it goes up for as well as the peak and thus, the area under the curve of the blood sugar rise. You need this for each meal type. You also need to know what happoens overnight. Does your bg straight line between the "before bed" reading and the fasting reading, or does it go down and then up, or vice versa? Each of these will give a different overnight average. As this is about a third of each day an error here can have a big affect on your averages. You have to make assumptions about how your sugar levels behave inbetween testing.
Youy also have to realise that our meter readings may only be within 5% accuracy, and a 5% error could represent an HbA1c difference of, say, 0.3 between the lab result and your calc on an actual reading of 6.0
I have a spreadsheet I developed with about 40 calcs behind it that gives me an automatic update on my "expected" HbA1c each time I enter another bg reading. Seems pretty accurate within the constraints above. Last time I predicted 5.6 and got 5.9 from the lab, but that was during a period of big swings which make your assumptions more difficult. I'm currently much more stable and I expect to be much closer to my current prediction of 5.3

Malcolm, T2, diet and exercise only, Diagnosed with HbA1c 8.2, down to 5.9 after 4 months on reduced (not low) carbs and low GI diet.
 
Very interesting post malc,

My meter is a newer model so i assume it will definately read whole blood plasma.

I'm aware the 90 day average will obviously not take into account the falling and rising between tests. Also that meters aren't fully accurate.

That's why i asked how accurate the comparison would be.

I'm also interested in how people came up with these equasions to get a likely hba1c from the meter averages but so far nobodys coughed that one up!

Is it a coincidence my result matches patch's theory perfectly? or has he hit gold with his equasions?

hmmmm...
 
I would say, the fact that Patchs equation gave the same result as your HbA1c result was a bit of science and a bit of luck. I think it is "Fairly" accurate as discussed....in a your HbA1C will be aruound 6.6 type of way. The fact that it was bang on is possible, but not a given...if it can narrow it down to say +/- 0.75, then there is a reasonably chance that it is going to be right some of the time.. (I do not claim to know the accuracy of Patchs formula, just using 0.75 as an example)

Another thing I noticed, on a day where I had 2 vials of blood taken for HbA1C analysis, at different labs, the results differed by about 0.3 anyway. Not sure what the accuracy of the HbA1c. I just know, that I always get a better results from teh lab y GP uses, than the one the hospital use.

For me, I tend to look it as...

4 somthing...wow!
5 something great control
6 something good control
7 something OK control
8 something not great control
9 something bad control.

This is for me, as a T1 adult I hasten to add...measures of control inchildren have to be different, because their bodies are changing so much.
 
I've never had the oppurtunity to experience two different results, as they only take one sample for the one lab in the hospital.

There's other factors that can produce a lower result such as anaemia if i remember correctly.

It's creepy that it predicted my result, but i'm quite a non-believer until i have solid evidence where it came from, and patch still hasnt posted how the equasion came about so i'll raise my eyebrow judgingly until it can be explained scientifically.

I think how you've judged the numbers is similar how i'd put them, but there's alot of factors that effect people's abilities to gain certain numbers.

Especially for type 1's when you get into the 5's, making sure that it's not due to hypo's and your hypo awareness is still good. But of course this is still acheivable without hypo's or risking your hypo awareness.

Some people also prefer to run at a higher level too, there was a woman who started her pump same time as me and when we were asked where our targets were, i said i'd like mine between 5-7mmol but she said she prefers hers to be between 7-8mmol.

I think if i got a hba1c result in the 4's i would be very worried that my hypo awareness is at risk, confused how the heck i managed it, and i probably would fall off my chair too :lol:

My ideal has always been around 7% ... Now i see what kind of levels i can acheive a 6.6% with.. i think if i tweak things more i think i can get it a little lower without risking anything.

But for now i'm very happy with my 6.6% and so is my nurse :)
 
This is the formula from Dr Bernstsein's book:

average plasma glucose = (35.6 x Hba1c) - 77.3 then divide by 18 to convert to mmol.

I can't remember how to swap the hba1c over the 'gate' in order to get the hba1c as the answer so I just guess at the hba1c to see what the average comes out as. I find it to be fairly accurate.
 
I might be able to help here. First of all, the conversion from average bg to HbA1c depends on wether your bg meter is reading whole blood plasma or not. As I understand it, newer meters read whole blood plasma and older ones don't
Click to expand...
All meters in the UK now give plasma readings. Accuchek meters still gave whole blood readings until the end of last year when they changed their strips.
http://www.diabetes.org.uk/Guide-to-dia ... se/Meters/

I'm also interested in how people came up with these equasions to get a likely hba1c from the meter averages but so far nobodys coughed that one up!
Click to expand...
Have a look at this graph.
http://care.diabetesjournals.org/conten ... nsion.html
They studied 507 subjects, including 268 patients with type 1 diabetes, 159 with type 2 diabetes, and 80 nondiabetic subjects from 10 international centers. These subjects all underwent at least 2 days with a cgm and did 7 tests a day for at least 3 days a week over 3 months. They had monthly HbA1c tests. The object was to find out if the HbA1c and the averages were linearly related. If they were then it was easy enough to calculate the formula. They found that for most people in the studies the results were within 15% of the line .
http://care.diabetesjournals.org/content/31/8/1473.full
But there were ethnic groups left out, children and pregnant women were not included, nor people with problems that are known to effect HbA1c results. There are as I said in the last post quite a lot of people who aren't 'average'.
 
Back
Top