 |
 |
Deluged by all the statistics, I'm somewhat confused by the absence of any reference in this discussion to the fact that the great majority of substances assayed by clinical chemists are heterogeneous in molecular composition, assays of which are therefore method dependent. In these circumstances, there is no correct result, and the concept of 'bias' becomes meaningless. Furthermore the precision of measurement of an analyte concentration generally varies with concentration, a relationship represented by the "precision profile". This too seems to be ignored by discussants. Roger Ekins >I have been following the discussion on >estimation of uncertainty with much interest. >However, I am concerned that the >interchangeability in the use of the terms total >error (TE), total error allowable (TEa) and >measurement uncertainty (U) may be confounding >the understanding and application of these >inter-related concepts. >The concept of measurement uncertainty is meant >to reflect the actual variation of a measured >result, not its potential to be in error. While >it is true that any given QC procedure will >allow for a certain amount of error before >detection, this does not mean that the >uncertainty of a measured result is impacted to >the full extent of that potential error. Let me >illustrate my point with the following example: >CV = 1.895%, bias = 0.1%. Suppose the QC >procedure used gives an SDcrit of 2.584, then >TEa = CV(SDcrit + 1.65) + bias >TEa = 1.895(2.584 + 1.65) + 0.1 = 8.1% >Note that this is total error allowable, not >total error. Total error would be: >TE = (CV x 1.96) + Bias >TE = (1.895 x 1.96) + 0.1 = 3.8% >Uncertainty (U) would be calculated (using a coverage factor of 2) as follows: >U = CV x 2 >U = 1.895 x 2 = 3.8% >Other contributing sources of uncertainty (such >as calibration uncertainty) can be included if >known in the estimation of uncertainty (by >adding in quadrature). Note that bias is not >included as it is assumed that any significant >bias is corrected and if so the uncertainty of >this correction can be included. > >Now, let us say the performance of this assay >changes such that the CV increases to 3.79%, >which increases the TE to (3.79 x 1.96) + 0.1 = >7.5%. The total error allowable (TEa) remains >the same as it is based on a CV obtained during >stable analytical performance. Measurement >uncertainty changes to 3.79 x 2 = 7.58%. > >One of the provisos of measurement uncertainty >and TE is that the variation measured must >reflect the real variation of the measurement >process across all analytical influences, not >just under ideal conditions. On the other hand, >the variation for calculation of TEa should >reflect stable analytical performance, which we >want to use to determine the control procedure >for the assay. > >Cheers >David Parry >Winnipeg, Canada > > > >------ACB discussion List Information-------- >This is an open discussion list for the academic >and clinical community working in clinical >biochemistry. >Please note, archived messages are public and >can be viewed via the internet. Views expressed >are those of the individual and they are >responsible for all message content. >ACB Web Site >http://www.acb.org.uk >Green Laboratories Work >http://www.laboratorymedicine.nhs.uk >List Archives >http://www.jiscmail.ac.uk/lists/ACB-CLIN-CHEM-GEN.html >List Instructions (How to leave etc.) >http://www.jiscmail.ac.uk/ > >This email and/or any documents in this transmission is intended for the >addressee(s) only and may contain legally >privileged or confidential information. Any >unauthorized use, disclosure, distribution, >copying or dissemination is strictly prohibited. >If you receive this transmission in error, >please notify the sender immediately and return >the original. > >Ce courriel et tout document dans cette >transmission est destiné à la personne ou aux >personnes à qui il est adressé. Il peut contenir >des informations privilégiées ou >confidentielles. Toute utilisation, divulgation, >distribution, copie, ou diffusion non autorisée >est strictement défendue. Si vous n'êtes pas le >destinataire de ce message, veuillez en informer >l'expéditeur immédiatement et lui remettre >l'original. > >------ACB discussion List Information-------- >This is an open discussion list for the academic >and clinical community working in clinical >biochemistry. >Please note, archived messages are public and >can be viewed via the internet. Views expressed >are those of the individual and they are >responsible for all message content. >ACB Web Site >http://www.acb.org.uk >Green Laboratories Work >http://www.laboratorymedicine.nhs.uk >List Archives >http://www.jiscmail.ac.uk/lists/ACB-CLIN-CHEM-GEN.html >List Instructions (How to leave etc.) >http://www.jiscmail.ac.uk/ -- With best wishes, Roger Ekins Prof Roger Ekins, PhD DSc FRS Windeyer Institute University College London London W1T 4JF Phone +44 20 7679 9410 Fax +44 20 7679 9407 ------ACB discussion List Information-------- This is an open discussion list for the academic and clinical community working in clinical biochemistry. Please note, archived messages are public and can be viewed via the internet. Views expressed are those of the individual and they are responsible for all message content. ACB Web Site http://www.acb.org.uk Green Laboratories Work http://www.laboratorymedicine.nhs.uk List Archives http://www.jiscmail.ac.uk/lists/ACB-CLIN-CHEM-GEN.html List Instructions (How to leave etc.) http://www.jiscmail.ac.uk/