Dear Mr Minett,
True, but I think we are talking about slightly different things.
In my letter I only described a rational way to estimate the
uncertainty of measurements in a particular conglomerate of
instruments. The coverage factor is used to modify the probability
that the best estimate is within the coverage limits. The best
estimate itself is the mean of the observations and a weighted mean is
preferable if there are different numbers of observations carried out
by the different instruments. How you use the estimated uncertainty is
another cup of tea and it fits into the Westgard rules as well as any
other estimate. You should realize that the original idea was to set
the alarm limit to +/-2 SD but this will increase the number of false
alarms beyond what most laboratories accept and 2.3 or 2.5 SD are
frequently advised and should be included in your power functions. The
power functions is a balance between false rejects and false accepts -
as usual.
All these aspects have been nicely described by Callum Fraser in
Dundee and Per Hyltoft Petersen in Odense (Denmark) - just across the
North Sea.
Enjoy the weekend!
Anders Kallner
------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/
|