Meter Variance

Status
Not open for further replies.
Sienne and Gabby (GA) said:
There is a 20% variance between tests due to measurement error inherent in the meter and/or strips.
Click to expand...
That is not how I read the 20% variance. The 20% variance is the variance of the handheld meter reading to the real (laboratory machine) BG reading. The handheld meter reading is made up of different component including: meter error, strip to strip variance, absolute strip error due to how the strip measure BG, variation between lots of strips. If you youse the same meter and same lot of strips the error between reading is much less than 20%
 
Larry and Kitties said:
That is not how I read the 20% variance. The 20% variance is the variance of the handheld meter reading to the real (laboratory machine) BG reading. The handheld meter reading is made up of different component including: meter error, strip to strip variance, absolute strip error due to how the strip measure BG, variation between lots of strips. If you youse the same meter and same lot of strips the error between reading is much less than 20%
Click to expand...
I think that is correct. You also may see a slight variance if you use the same meter with a new strip testing the same drop of blood.
 
Larry and Kitties said:
That is not how I read the 20% variance. The 20% variance is the variance of the handheld meter reading to the real (laboratory machine) BG reading. The handheld meter reading is made up of different component including: meter error, strip to strip variance, absolute strip error due to how the strip measure BG, variation between lots of strips. If you youse the same meter and same lot of strips the error between reading is much less than 20%
Click to expand...
That’s interesting. I was reading it wrong before.

I’m curious what the expected variance is on same meter test to test though. I guess each would be different.
 
Larry and Kitties said:
That is not how I read the 20% variance. The 20% variance is the variance of the handheld meter reading to the real (laboratory machine) BG reading. The handheld meter reading is made up of different component including: meter error, strip to strip variance, absolute strip error due to how the strip measure BG, variation between lots of strips. If you youse the same meter and same lot of strips the error between reading is much less than 20%
Click to expand...
That’s interesting. So if I popped open two lots of strips and mixed them together I might get a larger sampling of variance.

I just have seen in several posts how the Relion Classic Prime runs “high” and wondering just how high. I would like to ultimately get his tests under 100 overall and he has a tendency to have several tests between 90 and 120 so it makes it difficult to gauge if he really is staying under 100. The TR guidelines state I could have say a 110 every now and then but he tends to have frequent PS tests in that range. I am trying to determine just how much meter variance plays a role in his case. If this meter truly does run high he could very well be under 100 for many of these tests.
 
Yes different meter/strip systems would be expected to have different accuracy/variances in the different parameters that make up the system.
Also, there is debate on how to measure accuracy:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2769957/
Also, some meters are significantly more accurate than others
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5505415/

AbbottTheCat said:
That’s interesting. So if I popped open two lots of strips and mixed them together I might get a larger sampling of variance.
Click to expand...
Not really since if you do not know which BG reading was associated with which lot.
 
Thanks for these links…most of it over my head! Although I have a Contour Next and it appears to have done ok…I think.
 
Amy - the articles that were provided, especially the first one, are old. Every so often, Consumer Reports reviews glucometers. What I think is the take away is that no matter how accurate a meter is, there is alway measurement error. If you were to take a sample of blood and test it 10 times with the same meter, you would not get exactly the same result with each test. The same would occur if you had 5 meters and repeated testing 10 times. There would be variability in your results. The same is true if you took your temperature. The reading is not exactly the same -- it may be close but it's not likely to be identical.

With any form of measurement, there is a standard error of measurement that's attributable to the equipment. It's often why you see a +/- (and some percent) attached to a measure so you have some sense of the accuracy or variability.
 
Sienne and Gabby (GA) said:
Amy - the articles that were provided, especially the first one, are old. Every so often, Consumer Reports reviews glucometers. What I think is the take away is that no matter how accurate a meter is, there is alway measurement error. If you were to take a sample of blood and test it 10 times with the same meter, you would not get exactly the same result with each test. The same would occur if you had 5 meters and repeated testing 10 times. There would be variability in your results. The same is true if you took your temperature. The reading is not exactly the same -- it may be close but it's not likely to be identical.

With any form of measurement, there is a standard error of measurement that's attributable to the equipment. It's often why you see a +/- (and some percent) attached to a measure so you have some sense of the accuracy or variability.
Click to expand...
Makes sense. Thanks!
 
Status
Not open for further replies.
Back
Top