April 11, 2015
Ever wonder why ISO or CLSI glucose standards use primarily one set of limits rather than an error grid? Here’s my explanation.
With an error grid – especially a glucose error grid – there are multiple sets of limits. Data inside of the innermost limit implies no harm to patients and data outside of the outermost limit implies serious injury or death. And of course there are limits in between the inner and outermost limits which range in harm to patients. Although the limits are provided without percentages of data that should be in any region, it is implied that there should be no results in the outermost limits.
With ISO or CLSI, the use of one primary set of limits (corresponding to the innermost limits of an error grid) relieves these standard organizations from having to even mention a case where serious injury or death may occur. And this is probably because these groups are dominated by regulatory affairs people from industry.
March 19, 2015
For no particular reason, I searched for Dr. Getzenberg in Google. To recall about previous entries on this blog, search for EPCA-2 on this blog. (there is a search form on the top right of this blog). I found two rather different entries in Google.
One deals with the seventh retraction for articles written by Dr. Getzenberg
Another talks about awards distinction and how he is a senior leader in oncology and urology.
March 16, 2015
There is a new article in Clinical Chemistry about a complicated (to me) analysis of quality targets for A1c when it would seem that a simple error grid – prepared by surveying clinicians would fit the bill.
Thus, this paper has problems. They are:
- The total error model is limited to average bias and imprecision. Error from interferences, user error, or other sources is not included. It is unfortunate to call this “total” error, since there is nothing total about it.
- A pass fail system is mentioned, which is dichotomous and unlike an error grid which allows for varying degrees of error with respect to severity of harm to patients.
- A hierarchy of possible goals are mentioned. This comes from a 1999 conference. But there is really only one way to set patient goals (listed near the top of the 1999 conference): namely; a survey of clinician opinions.
- Discussed in the Clinical Chemistry paper is the use of biological variation based goals for quality targets. Someone needs to explain to me how this could ever be useful.
- The analysis is based on proficiency survey materials, which due to the absence of patient interferences (see #1) is a subset of total error.
- From I could tell from their NICE reference (#11) in the paper, the authors have inferred that total allowable error should be 0.46% but this did not come from surveying clinicians.
- I’m on-board with six sigma in its original use at Motorola. But I don’t see its usefulness in laboratory medicine compared to an error grid.
March 12, 2015
I’ve written before that total error means error from any source not just analytical error. Thus, if a clinician makes an incorrect treatment decision because the test result is wrong due to user error, it is little consolation to know that the analytical system was ok.
All of this applies to SMBG (self-monitoring blood glucose) where the treating “clinician” and user are the patient.
A Letter in Clinical Chemistry (subscription required) shows that whereas 9 out 10 glucose meters met performance standards when the tests were performed by expert users, only 6 out of 10 meters met standards when the tests were performed by routine users.
Of interest as well is that the authors cite as performance standards both the ISO 2013 standard and the suggested FDA draft performance standard from 2014.
February 6, 2015
I had occasion recently to see a final draft of CLSI EP19 – which is a framework for using CLSI evaluation documents. I may review this when it is officially released but here are three comments.
- There is a cause and effect diagram in EP19 listing assay attributes (precision, interferences, and so on) and the CLSI documents that are used to evaluate these attributes. I published (1) a diagram in 1992 (attributes only) and later adapted my diagram to include the associated CLSI documents and this diagram appeared in a 2002 publication (2). In 2005, I proposed to CLSI that the diagram appear in all CLSI evaluation standards – it is in EP10, EP18, and EP21 although it is not in more recent documents. Now I know that CLSI is a consensus organization whereby documents are a collaborative effort and my diagram has been modified further but there should be a citation to my prior work and there isn’t.
- In the clinical performance section, there is no mention of error grids (EP27). In fact, a search of EP19 shows that EP27 is never mentioned. This is most strange. After all, error grids are used to determine if an assay is good enough which is the whole point of an evaluation! Error grids are part of the FDA CLIA waiver recommended guideline and fundamental in glucose meter evaluations. I don’t understand how in years of document development of EP19, EP27 has received zero mention. I did check the list of CLSI publications on their website to make sure that EP27 is still for sale.
- There is mention that assay claims should clear – that’s it! no more details are given. Sadly, there was an entire document about uniformity of claims (EP11) that was killed by CLSI management after one manufacturer threatened to quit.
- Krouwer JS Estimating Total Analytical Error and Its Sources: Techniques to Improve Method Evaluation. 1192, Arch Pathol Lab Med., 116, 726-731.
- Krouwer JS Setting Performance Goals and Evaluating Total Analytical Error for Diagnostic Assays. Clin. Chem., 48: 919-927 (2002).
February 4, 2015
I have previously commented that many CLSI evaluation standards at some point ask the question “is the assay performance good enough” and answer that question with “it’s up to the lab director.”
The problem is that lab directors are not clinicians and do not treat patients. Note that most lab directors are either PhD clinical chemists or pathologists and although pathologists are MDs, they are not clinicians because they do not treat patients.
Of course, lab directors do have a great deal of knowledge about assay performance but in my experience – especially in working on CLSI standards – lab directors tend to focus on analytical errors whereas only total error is of importance to clinicians and the source of errors that contribute to total error is a combination of analytical, pre- and post-analytical error.
So how should the “is it good enough” question be answered? An example appeared recently in the literature (1) where clinicians were surveyed as to what size glucose meter errors would start to cause problems for diabetics under several scenarios. The results provided limits for a glucose meter error grid. Note that there was no attempt to identify error limit sources – the limits simply reflect the observed error, regardless of its source.
- Klonoff DC, Lias C, Vigersky R, et al The surveillance error grid. J Diabetes Sci Technol. 2014;8:658-672.