No home use glucose meter is really accurate . I think theyhave to be within 15% 90 % of the time .I have two meters one is a Braun and the other is a code free meter, I have used both meters at almost the same time to experiment and the Braun is reading 2 to 3mmol higher than the code free meter. not sure which one is correct.
I am self funding that is why I have two meters at the moment, I am using up the Braun strips before I start using the Code free meter as the strips are cheaper.No home use glucose meter is really accurate . I think theyhave to be within 15% 90 % of the time .
Also blood is not homogeneous
If you are self funding the strips, I would just use the meter with the cheapest strips
remember and check the date on the test solution, sometimes they aren't very well dated, also if its been open over a certain amount of time it shouldn't be trusted either xxI am self funding that is why I have two meters at the moment, I am using up the Braun strips before I start using the Code free meter as the strips are cheaper.
I have just remembered I have got some testing fluid for the Braun meter so will test the meter when I get home tonight, that should give me an idea if at least that one is working ok.
I think theyhave to be within 15% 90 % of the time
ThanksA bit better than that now. See https://medicines.necsu.nhs.uk/bloo...mplementation-of-iso-2013-accuracy-standards/
That still means that now and again you'll get two readings from meters which conform which are 2 or 3 mmol/L apart.
- 95% results must fall within ± 0.83 mmol/L of laboratory results at concentrations of less than 5.6 mmol/L
- 95% of results must fall within 15% of laboratory results at concentrations of 5.6 mmol/L or more.
- The 2013 guidelines also now stipulate that 99% of readings must fall within zones A and B of the Consensus Error Grid for type 1 diabetes.
(I seem to remember a report about actual meters in the real world being a lot worse than the original tested ones (because normally nobody does follow up tests and manufacturers make changes (which shouldn't make any difference) after the initial testing). I don't know whether that was satisfactorily resolved; seems obvious that checks ought to be performed on meters that people actually buy.)