References; Linked-In Posts; White Papers
AACC Forum: Westgard & ZB re QC Chart Mean
Short Comments about SQC Strategies for Multiple Instruments
There has been a running discussion about the use of means and standard deviations for the same methods on multiple instruments and multiple laboratories. While many of the comments are instructive, it's clear there is a wide range of approaches currently in practice.
Selecting SQC strategies for multiple instruments is a sufficiently difficult problem that the most recent CLSI C24-Ed4 guidance document [essentially the "Best Practices" for QC guideline] did NOT provide an answer to this issue, instead stating that "although significant advances in QC thinking have occurred, there are still important areas that could benefit from additional developments, such as QC strategy design and implementation for laboratories with multiple instruments of the same type performing the same measurement procedures." The C24-Ed4 guidance therefore is limited to the case of a single method where the individual mean and individual SD are the basis for planning an SQC strategy.
C24-Ed4 recommends "Best Practices" for the establishing the mean and SD, but also for defining the quality for intended use. By defining intended use, and observing actual performance, laboratories can optimize the selection of the control rules, number of control measurements, and frequency of QC. C24-Ed4 defines an SQC strategy as "the number of QC materials to measure, the number of QC results and the QC rule to use at each QC event, and the frequency of QC events" and provides a "road map" for planning such SQC strategies. The best practices for multiple instruments should also consider all these factors, thus the issue is actually more complicated than it initially appears.
As an example of "complications," consider a laboratory that sets 2 SD control limits at 92 and 108 based on a pooled mean of 100 and a pooled SD of 4.0, but a particular instrument has an actual mean of 100 and an actual SD of 2.0. The lab thinks it is implementing a 1:2s rule and expects to achieve high error detection (as well as experience high false rejections, by the way). However, the actual control rule being implemented is truly based on the individual mean and individual SD, which define a 1:4s rule (control limit/SD=8/2=4s) and will lead to low error detection. After implementation, the lab may be happy to observe a low number of rejections using pooled means and pooled SDs, without realizing that the SQC strategy may lack the necessary error detection to assure the quality required for intended use, giving the laboratory a false sense of security.
Assessing the error detection of the actual control rule requires knowledge of the rejection characteristics, i.e., power curves that show the probability for rejection as a function of the size of errors occurring. The available power curves, however, may not be applicable for a network product in a regional service because an additional component of variation is expected between instruments and/or between laboratories. Additional research is needed to develop a rigorous QC design approach that can apply to multiple instruments.
We should also recognize that rigorous SQC design even in the simple cases of single instruments, following the existing C24-Ed4 road map, has not yet been widely implemented. Even on a single instrument, the complexity of the theory and calculations often overwhelm the lab. Until laboratories are able to improve their QC design capabilities, I think the safest practice is to use individual means and individual SDs for each method and instrument. That approach adheres to the basic principles of SQC, the "Best Practices" recommended by CLSI, and is supported by existing SQC software.
[An expanded discussion is posted at www.westgard.com/means-for-multiples.htm]
James O. Westgard, Westgard QC, Inc., Madison WI: email@example.com
James O. Westgard
Reply to Group Online Reply to Group via Email View Thread Online
Sent: 12-14-2018 17:28
From: Martin Abrams
Subject: Establishing QC mean on multiple instruments
I work for a health system with 12 labs each have at least two main chemistry analyzers. Currently for establishing new QC lot they run QC over multiple days on all instruments to get about 20+ data points on each system. They then calculate the mean for each instrument but use the same SD for all instruments. So one lab has 3 chemistry systems each with slightly different means. Is this correct practice? Is this what others do in their institution? In previous labs I have worked in we would establish an use one mean for all 3 instruments at that lab. It is OK for each lab to establish their own mean but a single lab should they not be using one mean for all 3 instruments? Just want to get feedback on what others are doing.