Test meter readings discrepancy.

[Paul520785]

Background - T1 for over 55 years
Because of changes to the required accuracy level OneTouch notified users of the coming withdrawal of the Ultra range of meters.

As a result of this, after testing various makes of meter I decided to use the OneTouch Verio range.
I very quickly started to disbelieve the reported results based on how I felt.

OneTouch have been brilliantly helpful - - - BUT
when using Control solution to verify results on old and new meters (3 old and 3new) the following became obvious.
Each type of meter produced the same results for the respective control solutions which confirmed consistent operation of each type.
The old meters all displayed in the lowest 15% of the permissible range.
The new meters all displayed in the TOP 10% of the permissible range.

For me this translates into a discrepancy to the results which I have become accustomed over about 20 years
The comparison readings obtained by testing the same blood drop on 2 meters shows a position as below
Ref...............New..........Old.............Diff..........Diff as %age of Old
A...................3.4............2.8.............0.6.................21%
B.................10.3............8.4.............1.9.................22%
C.................16.2..........12.2.............4.0.................33%

I am aware that the rules on accuracy of meters have been updated .

I have 2 annoyances -
My brain is struggling to deal with the change in readings and allow me to act appropriately.
People who tell me not to compare old and new meter readings . . . . . ... . .

Has anyone else spotted this ??

 

[tim2000s]

You might want to have a chat with @Oldvatr @Paul520785

 

[Oldvatr]

Background - T1 for over 55 years
Because of changes to the required accuracy level OneTouch notified users of the coming withdrawal of the Ultra range of meters.

As a result of this, after testing various makes of meter I decided to use the OneTouch Verio range.
I very quickly started to disbelieve the reported results based on how I felt.

OneTouch have been brilliantly helpful - - - BUT
when using Control solution to verify results on old and new meters (3 old and 3new) the following became obvious.
Each type of meter produced the same results for the respective control solutions which confirmed consistent operation of each type.
The old meters all displayed in the lowest 15% of the permissible range.
The new meters all displayed in the TOP 10% of the permissible range.

For me this translates into a discrepancy to the results which I have become accustomed over about 20 years
The comparison readings obtained by testing the same blood drop on 2 meters shows a position as below
Ref...............New..........Old.............Diff..........Diff as %age of Old
A...................3.4............2.8.............0.6.................21%
B.................10.3............8.4.............1.9.................22%
C.................16.2..........12.2.............4.0.................33%

I am aware that the rules on accuracy of meters have been updated .

I have 2 annoyances -
My brain is struggling to deal with the change in readings and allow me to act appropriately.
People who tell me not to compare old and new meter readings . . . . . ... . .

Has anyone else spotted this ??

Hi Paul. Yes I am experiencing a similar issue with my meters. I am being forced to change my meter by my CCG and it has given me a similar headache. I have now been running 4 meters in parallel testing for a couple of weeks, and I have shared my journey on another thread entitled Hypoglycemia and Plasma Meters which you might find of interest, It is not quite up to date and I have recently found that the replacement NEO meter is actually (probably) behaving according to the ISO Specification most of the time which is a correction to what I originally reported in that thread. I will be updating the graph soon, but I have run out of strips and have had to delay this.

In summary. the two meters I received this year (2016/17) seem to meet the new ISO requirements, but the two older meters do not. I have no explanation for this at the moment,

I note that you express your error as being (difference / LOWER reading x100), whereas it is more likely that the actual value that a Lab would report will probably be somewhere in between, so this is why I have used an average as my baseline, Also, for actual bgl below 5.2 mmol/l then the error allowed is no longer a straight percentage, but a fixed value of 0.83 mmol/l. so I think your meters are actually closer than shown in your posting above.

One reason why people say don't parallel test is because each meter has an error allowance (per ISO) so two meters can differ by twice that budget worst case, so the ISO allowance of 15% for one meter means that two meters can differ by up to 30% and both could still be meeting the requirement. So at a bgl of 10 mmol/l they can differ by 3 mmol/l and pass. Since we do not know what the true bgl value is when home testing, then the comparison can become quite woolly. So it is difficult to PROVE anything without a lab result (and even the lab can be 2% out compared to a spectrometer value.)