Your INR is not a physical property like weight or volume that has a reference material (e.g. a calibrated weight) that can serve as a reference standard. It's been awhile, but I read a few articles on how they determined that home testing machines worked. IIRC the reference test method was a blood draw and analysis by wet chemistry. This is a reference method.
INR difference between lab and my Coagucheck
- Thread starter Eva
- Start date
Help Support Valve Replacement Forums:
Your INR is not a physical property like weight or volume that has a reference material (e.g. a calibrated weight) that can serve as a reference standard. It's been awhile, but I read a few articles on how they determined that home testing machines worked. IIRC the reference test method was a blood draw and analysis by wet chemistry. This is a reference method.
The way that at least one meter passed FDA approval was by comparing its test results to the results from the XS. Because the XS was used as the standard of comparison, wouldn't THIS mean that the XS really is the reference standard?
And as far as lab results go, a method called 'tilt table' is considered to be a standard method -- but it depends on the values and accuracy of the reagent used for the tests. And. IIRC, there's no standard way for the reagent makers to determine an exact value for the reagents -- it's a 'best guess' determination.
For myself, I am more confident in the care taken by Roche to determine a value for each batch of strips that may effect the lives of many thousands than a lab that is trusting the quality of its reagent supplier.
And as far as lab results go, a method called 'tilt table' is considered to be a standard method -- but it depends on the values and accuracy of the reagent used for the tests. And. IIRC, there's no standard way for the reagent makers to determine an exact value for the reagents -- it's a 'best guess' determination.
For myself, I am more confident in the care taken by Roche to determine a value for each batch of strips that may effect the lives of many thousands than a lab that is trusting the quality of its reagent supplier.
pellicle
Professional Dingbat, Guru and Merkintologist
amusingly still:
https://onlinelibrary.wiley.com/doi/full/10.1111/ijlh.13411
Accuracy
Accuracy is defined as closeness of agreement between a measurement and the true value. No standard exists for noncalibrated coagulation screening tests (PT, APTT and TT), so there is no true value. Accuracy is typically performed comparing new instrumentation or reagent systems to an existing or predicate method.
A degree of standardization for INR has been achieved, so verification of the accuracy of a PT/INR may be best achieved by testing suitable EQA materials or through evaluation comparability with an established method. If the PT/INR test system is not validated, this will require assigning an International Sensitivity Index (ISI)/calibrating the INR which is beyond the scope of this document and readers should refer to WHO guidance.23 INR calibrants are commercially available for certain test systems and there are other commercial sources that provide frozen plasma of varying assigned INRs, which are designed to assist local verification of ISI and MNPT.
I keep saying its not like measuring a piece of steel
ATHENS1964
Well-known member
- Joined
- Oct 19, 2019
- Messages
- 459
Today I did an inr test in a lab to check my device. Usually the device gives a smaller result than the lab by about 10% Today the opposite happens lab 3.04 device 2.8. I should note that I am taking antibiotics for a dental problem I had,
the laboratory is always the same where I do the test
Do you think this is the reason?
the laboratory is always the same where I do the test
Do you think this is the reason?
Attachments
Last edited:
ATHENS1964
Well-known member
- Joined
- Oct 19, 2019
- Messages
- 459
It is certainly not a big difference, I say this because it is the first time that the lab gives a value higher than Coagucheck, I expected it to give about 2.4
I agree.
pellicle
Professional Dingbat, Guru and Merkintologist
this may shock you, but not all labs are 100% on the ball with INR work and not all labs are to the standard of scientific research reference.
I've been less interested in keeping this data because even though every time I've got a lab test of bloods I always ask for INR and then do an INR comparison its never different to this.
The differences of Lab - Coaguchek in that data
0.2 |
-0.1 |
0.2 |
-0.2 |
-0.1 |
0.0 |
I've written about this before:
My PCP has its own lab tech. I told him that I self-test and he showed me a practically new XS, still in its box, that a patient gave to him because it was 'inaccurate.' He (and the lab tech) apparently expected an exact match of values. They didn't realize the variabilities in testing -- or that values within 20% of each other should be considered to be valid.
One of these days, I'll get up the nerve to ask for the meter -- and I may show them that the results on my meter and the 'inaccurate' one are close to each other.
My PCP has its own lab tech. I told him that I self-test and he showed me a practically new XS, still in its box, that a patient gave to him because it was 'inaccurate.' He (and the lab tech) apparently expected an exact match of values. They didn't realize the variabilities in testing -- or that values within 20% of each other should be considered to be valid.
One of these days, I'll get up the nerve to ask for the meter -- and I may show them that the results on my meter and the 'inaccurate' one are close to each other.
I'll be testing again at the lab soon and will post the results.
I contacted Roche. I was first impressed, but it turned out to be a waste of time. I was asked too many questions and responded and responded. Then, she (Roche rep) asked me if I was taking antibiotics or sick or taking new medication at the time I tested my INR. I answered I was not and I added that even if I were, the medication/antibiotic or being sick would affect the results of the lab too not just my machine. Then, she responded that maybe I shouldn’t be self-testing!!!
She got me so upset that was her best conclusion after all the wasted time! Maybe she was reading the questions from a computer but not well trained to answer with her brain!
I calmly told her I have been self testing since 2008 and never had this issue before. Then, she ended the conversation by asking for my email to send me a letter naming their recognized distributors if I wanted to buy a new machine!! They won’t give support if the machine is not from their recognized distributor!
Here’s the letter.

Attachments
I agree you are a competent tester, Eva, and think the licencing arrangements in the US are the biggest reason for their unhelpful response - they may well have a regulatory problem if they are seen to help a user of their products not going through the "normal" US procedures.
pellicle
Professional Dingbat, Guru and Merkintologist
I think this sums up the entire global position on this. Its at best patchwork and poorly thought out. Its like the blind applying policy (for which the codification was written before electronics existed) and is certainly not uniform even within a country.
Imagine if this "framework" was applied to blood glucose testing and diabetics?
Unfortunately she didn’t email the other page! I wanted to contact Roche right away, but was distracted with more urgent things then (taking sister to hospital). As soon as I get hold of it, I’ll post it here.
FWIW - the few times I called Roche (including about an XS Pro), they were helpful.
Years ago, when I contacted support for the InRatio meters, they told me to 'trust the labs.'
Coagusense (the manufacturers of the Coag-Sense meters) were of little help the last few times I called them. They were willing to go through the steps taken to make a test (probably from a script), but were unresponsive (or worse) when I told them that the lab results were higher - one tech rep told me to add to my meter's value to match the lab's value. (Like, for example, if I have a clock that's always one hour behind -- they would have told me to just add one hour so I can get a correct time). What good is a meter if it's not accurate or if you have to regularly make an adjustment so it can match the labs?
Roche has provided the most supportive customer service I've encountered recently.
It's a shame that they have to respond to what may be a computer generated list of support options.
Plus, FWIW, these meters are designed to maintain their accuracy - this is the reason that new code chips are shipped with each batch of strips. There should be no reason that a meter - wherever you've bought it - would somehow be inaccurate (especially because QC should catch any problems with meter or strip).
Years ago, when I contacted support for the InRatio meters, they told me to 'trust the labs.'
Coagusense (the manufacturers of the Coag-Sense meters) were of little help the last few times I called them. They were willing to go through the steps taken to make a test (probably from a script), but were unresponsive (or worse) when I told them that the lab results were higher - one tech rep told me to add to my meter's value to match the lab's value. (Like, for example, if I have a clock that's always one hour behind -- they would have told me to just add one hour so I can get a correct time). What good is a meter if it's not accurate or if you have to regularly make an adjustment so it can match the labs?
Roche has provided the most supportive customer service I've encountered recently.
It's a shame that they have to respond to what may be a computer generated list of support options.
Plus, FWIW, these meters are designed to maintain their accuracy - this is the reason that new code chips are shipped with each batch of strips. There should be no reason that a meter - wherever you've bought it - would somehow be inaccurate (especially because QC should catch any problems with meter or strip).
Last edited:
Hello,
I was doing research and came upon your post that describes the same situation that I am in now:
I too have a Coagucheck XS and a INRange/Vantus meter.
- The XS came out 2.7 and the INrange/Vantus 2.9
You can consider both to be accurate. INR testing isn't an exact science - a variation of 20% is considered to be accurate.
Have you tried using the same strips (and the same chip) in both meters? I'll bet that the results are even closer.
Those numbers are just fine -- but I don't understand why you would WANT to check using two meters for each test. Either meter should be fine - in both cases, you know that your result is in range and not to worry about your changing your dose.
A result of a blood draw by a good lab will be close, but also within the 20% acceptable range.
Have you tried using the same strips (and the same chip) in both meters? I'll bet that the results are even closer.
Those numbers are just fine -- but I don't understand why you would WANT to check using two meters for each test. Either meter should be fine - in both cases, you know that your result is in range and not to worry about your changing your dose.
A result of a blood draw by a good lab will be close, but also within the 20% acceptable range.
