Sooo frustrated.

[pellicle]

Did another comparison test this past weds.

Lab Draw- 3.4
InRatio2- 3.8

pretty close stuff, doubt you'd see much better results between 2 different labs from the same draw.

Remember, this isnt meant to be calulated to 2 tenths, thats why there is a range.

Micrometers are for steel ;-)

 

[Protimenow]

The use by labs of hundredths has baffled me, too. This implies that there's a degree of accuracy that doesn't really exist. (I don't think that the reagents that the labs use for calculating the INRs are standardized to that degree of accuracy, so any numbers beyond tenths are a statistical artifact).

Personally, if my range was 2.5 - 3.5, and my InRatio gave me a 3.8, I wouldn't change a thing. (And I don't when my INR looks to be just a bit above range on the InRatio).

(Also, FWIW, when I asked Alere about the difference between lab and InRatio, their answer was three words: Trust the Lab. I would accept the lab's 3.4 and not mess with a thing).

 

[marc_kowal]

My concern is that since I call into Alere with my results, they are always higher then a lab draw (which I personally feel is usually more accurate then a home testing unit). So my cardio will be making adjustments based on a number that is skewed and it could possibly make me go below my range, unless I always adjust my number that I call into Alere using Pem's formula. I just wish that wasn't necessary.

 

[pellicle]

Hi

... a number that is skewed and it could possibly make me go below my range, unless I always adjust my number that I call into Alere using Pem's formula. I just wish that wasn't necessary.

what is your range?

Personally I'd just report the INR from the reading and leave it at that.

Anyway, all numbers are skewed, its not like any system (labs included) is going to give you the sort of certainty in measurements your asking for. I feel rather like when I was helping an engineer mate of mine work on his house. We were trying to cut wood for flooring and he was trying to get less than 1mm tolerance on flooring. The beams expand more than that in various temperatures and humidities for crying out loud.

Also the factor gets less as the number gets less (right?) so if your machine is reporting 2 (and assuming that the function we apply is correct) then your "actual" INR would be 1.8

I seriously doubt that any INR dosing Dr will see a trend of your INR falling from (say) 3 down to 2 and say "well, your within range, so lets see what happens"

What is more likely is that they'll steer you up a wee bit as a result of seeing the trend. You would have to get 2.3 (assuming as I said the correction factor) for your INR to be 2. In effect all it does is compress your actual range.

Whack it in a spread sheet and look at the graphs

I think you are stressing over this for no nett benefit.

 

[Protimenow]

From a statistical standpoint, your InRatio is probably 'reliable.' (The term 'reliable' means that the results are consistent - reproducible. A stopped watch is 'reliable' because it always shows the same time). Assuming that your InRatio 2 always varies from the labs by a set amount, or using a consistent correction formula, you should be able to get some good results from the meter. The adjustment algorithm may be a slightly different formula from the one that Pem originally proposed (and Pellicle put into one of his postings).

FWIW -- I trusted my InRatio meter to be accurate (rather than higher than actual lab values), and wound up with a minor stroke because I maintained my INR a bit lower than I should have. Personally, with a range of 2.5 - 3.5, and an InRatio meter, I would probably be happy with a reported INR between 2.9 (or so) and 3.9 (or so).

And, yes, the higher the INR actually is, the higher the variance from actual INR, for most meters.

 

[pellicle]

From a statistical standpoint, your InRatio is probably 'reliable.' (The term 'reliable' means that the results are consistent - reproducible.

Interesting point, one that I had assumed but never stated the obvious as you now have.

A stopped watch is 'reliable' because it always shows the same time).

and its exactly correct twice a day