Pop Quiz: How many equations are needed (at minimum) to calculate measurement uncertainty?

Posted by Sten Westgard, MS

In Biochemia Medica, a recent article detailed the Minimum requirements for the estimation of measurement uncertainty:Recommendations of the joint Working group for uncertainty of measurement of the CSMBLM and CCMB.

Can you guess how many?

Turns out there are 19 different equations you need to know.

 

Now of course some of these equations are quite simple and you already use them: mean, SD, etc.

But this is quite a list of mathematics. And when many labs simply take the intermediate SD and call it MU, and take the SD * 2 and call it expanded MU, this is a bit of a difference. You can see the common practice is very different than the committee recommendations.

It's valid and worthwhile work to determine these minimum equations. But one should also bear in mind this is a world where most labs find no value (beyond mandatory compliance) in calculate MU, never use it, never get asked for it, never have seen any patients impacted by it. Do we really want to ask those labs to make more complicated calculations?

P.S. a corrigendum has already been issued on these corrections: http://www.biochemia-medica.com/2018/28/011201