How accurate are test meters?

Thread view:

Enable EMail Alerts

Start New Thread

Thread rating:

Scowcroft - 25 Mar 2006 18:39 GMT

Hi all
I've been reading and learning in this group for some time now. Most of my
questions have been covered by reading the posts of the knowledgeable
members of the group. Thank you everyone for your unwitting help however I
wonder if anyone can help to explain a recent observation.

I use a One Touch Ultra meter and have found it's readings to fit what I
might regard as reasonable given recent meals and exercise. Occasionally
however I've got an "odd" reading - higher than I might expect. On
retesting the result has generally been better but I decided to squander
some strips checking this variability out. I have discovered that testing 3
times in rapid succession can give quite different readings, 6.8, 7.5, 8.3
was one set, done an hour after a meal.

I tested on clean hands, strips within date, test solution used to confirm
everything should be OK etc.

Is it me? Do different fingers give different results!!!? Is my meter
faulty? or are meters prone to wide variation?

Signature

Brian S

Reply to this Message

Pete - 25 Mar 2006 19:21 GMT

>Hi all
>I've been reading and learning in this group for some time now. Most of my
>questions have been covered by reading the posts of the knowledgeable
>members of the group. Thank you everyone for your unwitting help however I
>wonder if anyone can help to explain a recent observation.

>I use a One Touch Ultra meter and have found it's readings to fit what I
>might regard as reasonable given recent meals and exercise. Occasionally
[quoted text clipped - 3 lines]
>times in rapid succession can give quite different readings, 6.8, 7.5, 8.3
>was one set, done an hour after a meal.

>I tested on clean hands, strips within date, test solution used to confirm
>everything should be OK etc.

>Is it me? Do different fingers give different results!!!? Is my meter
>faulty? or are meters prone to wide variation?

I have a OTU and have never found it to be inaccurate to
that degree. There is a test solution provided with the
meter, have you used it?

As another poster pointed out, all those readings are
rising, there is a possibility that it is indicating
correctly.

Reply to this Message

Scowcroft - 25 Mar 2006 19:48 GMT

.....can give quite different readings, 6.8, 7.5, 8.3 was one set, done an
hour after a meal.

There is a test solution provided with the
meter, have you used it?

Yes. It gave a midpoint reading as expected

> As another poster pointed out, all those readings are
> rising, there is a possibility that it is indicating
> correctly.

Actually they don't consistently do that. I am also doing the three tests
within a range of at the most 2 minutes, does blood sugar really change as
rapidly as that?

Has anybody else experimented to find out the range of values a meter gives
with (almost) simultaneous tests?

Brian S

Reply to this Message

Paul - 25 Mar 2006 20:39 GMT

>Actually they don't consistently do that.

Do you mean that you have repeated the test, and the figures do not always
follow a rising pattern?

> I am also doing the three tests
>within a range of at the most 2 minutes, does blood sugar really change as
>rapidly as that?

You'd normally only expect BG's to change that fast if you'd ingested sugars.

>Has anybody else experimented to find out the range of values a meter gives
>with (almost) simultaneous tests?

I think you have to accept that, if your calibration tests sit close to the mid
point, and yet you repeatedly get fluctuating readings when testing real blood,
there must be something wrong with your testing procedure (even if you are
following the instructions - are you using a different finger for each test - if
you were to keep using the same prick site, I'd expect the readings to go up
because water would evapourate from the blood between tests, concentrating the
glucose).

Try doing three consecutive tests on waking, using three different sites - these
should certainly be the same.

Reply to this Message

Scowcroft - 25 Mar 2006 20:50 GMT

>>Actually they don't consistently do that.
>
> Do you mean that you have repeated the test, and the figures do not always
> follow a rising pattern?

Yes. In fact typically, but not always, they go down.

>> I am also doing the three tests
>>within a range of at the most 2 minutes, does blood sugar really change as
>>rapidly as that?
>
> You'd normally only expect BG's to change that fast if you'd ingested
> sugars.

Which I don't

>>Has anybody else experimented to find out the range of values a meter
>>gives
[quoted text clipped - 12 lines]
> the
> glucose).

I'm using different fingers but after washing my hands. I appreciate that
this isn't a carefully controlled experiment nor have I collected that much
data (I don't have a bottomless supply of test strips!!) but I'm pretty
confident that my testing procedures are fine

> Try doing three consecutive tests on waking, using three different sites -
> these
> should certainly be the same.

Soon perhaps, I'm running out of strips!

Brian S

Reply to this Message

Alan S - 26 Mar 2006 10:00 GMT

> I have discovered that testing 3
>times in rapid succession can give quite different readings, 6.8, 7.5, 8.3
>was one set, done an hour after a meal.

As Pete mentioned, they are all rising. How far apart in
time were those tests, and what did you eat before them?

Cheers, Alan, T2, Australia.
d&e, metformin 2x500mg

Signature

Everything in Moderation - Except Laughter.

Reply to this Message

African Bios - 31 Mar 2006 11:28 GMT

> Hi all
> I've been reading and learning in this group for some time now. Most
[quoted text clipped - 17 lines]
> Is it me? Do different fingers give different results!!!? Is my meter
> faulty? or are meters prone to wide variation?

Dear Sir
To partly answer your question have a look at the ff medical journals
The lancet 1999 An independent pilot study into the accuracy and
reliability of home blood glucose monitors M Urdang et al
Diabetes Research and complications 2006
A comparison of blood glucose meters in Australia M Cohen et al

As a rule most meters are designed to be quite accurate within a normal
glucose range ie 4-7mmol/L thier accurancy falls outside the extremes of
this range. Most meters start to fall off in accurancy with glucose
reading below 3.5mmol/L and above 16mmol/L. Scrupulous meter
manufacturers often will simple make thier meter say Hi once the
capillary glucose sample reading is greater than 17mmol/L

HTH

Signature

Remove nyet / block or change dot to .
to reply Thank you

Reply to this Message