Uncertainty of Measurement in Transfusion Services

[YorkshireExile 23]

This question may be mainly for UK Blood Transfusion labs, but we have just had an ISO 15189 inspection and our inspector says we should be doing Uncertainty of Measurement (UoM) in Blood Transfusion labs, especially with regards to Blood grouping.  Now, I have no idea how UoM relates to Blood Bank procedures, and my inspector failed to convince me, but he said that this is quite common in the UK now and many Transfusion labs are doing this. So for anyone working in a UK Blood Bank who are ISO 15189 accredited, are you doing UoM  calculations for any BB procedure? If you are, can you please explain to me how it is relevant and what are you actually doing? UoM is relevant to Haematology and Chemistry with all their CVs and SDs and Bias calculations etc, but what the heck has it got to do with Blood Bank work!

Edited April 9, 2020 by YorkshireExile

[Malcolm Needs ★ 4,921]

I couldn't agree more with you YorkshireExile.  At a stretch, and I mean, at a stretch, it MAY be relevant to such tests as quantification and titrations, where you are giving a result involving a measured number, ascertained with red cells that may express different numbers of antigens, which may themselves involve protein or carbohydrate substitutions, but that is all.

How on Earth your inspector thought that this was remotely relevant to blood grouping, with all the positive and negative controls used to ensure the antisera are working properly, and the temperature mapping of everything these days, like you, I cannot see the point.  Either he or she was trying to justify their position as an inspector, and/or was going well over the top.

I would be amongst the first to say, very loudly, that Quality in the world of Blood Transfusion was pretty low at one point, but now, there are times when Quality issues actually interfere with the laboratory doing its job, for no good reason, and this seems to me to be one.

END OF RANT.

[John C. Staley 1,193]

I'm not sure I want to admit this, and I have been out of the business for a while but, what on earth is "Uncertainty of Measurement"???  Can anyone tell be in 50 words or less?  I think it's one more reason I'm glad I retired when I did.  

[YorkshireExile 23]

You are lucky you have retired!

Definition is (I think!) - a parameter associated with the result of a measurement, that characterizes the dispersion of the values that could reasonably be attributed to the measurand.

The ISO requirement we are supposed to meet is:

5.5.1.4 Where examinations include a measurement step but do not report a measured quantity value, the laboratory should calculate the uncertainty of the measurement step where it has utility in assessing the reliability of the examination procedure or has influence on the reported result.

 

The explanation is this: To ensure that measurement results are useful and safe in medical practice and to permit meaningful comparison with medical decision limits and previous results of the same kind in the same individual, medical laboratories require estimates for the overall variability in values reported by their measurement procedures."

 

I imagine that all makes you none the wiser!! Sorry - I couldn`t do it in 50 words!

Edited April 10, 2020 by YorkshireExile

[David Saikin 1,439]

ABORhs aren't measurements - are they?  I agree w Malcolm.  Though, I don't know how you would do the calculation for titers (I guess you quantitate Ab levels too (anti-D).  I also think I've seen questions and comments on this topic here in the past.

[DPruden 129]

My argument would be that blood bank testing is qualitative and not quantitative.  We have run into this a little bit in the US as well, they re-organized the federal regulations and starting using chemistry and hematology requirements for blood banking.  Our method comparison requirement in particular has never made sense to me.  Of course LISS, PEG, solid phase, and gel methods give different results, they are designed to!