Posted: 22 September 2008 at 4:07pm | IP Logged

When our reports are printed out, if a centre has failed for one of the three samples provided, they are automatically failed for all three - the other results are regarded as suspect.

What are the criteria for other schemes? I understand that the way CAP in the USA works is that if you fail one sample but pass the others you still pass for a given distribution.

Opinions/thoughts?

Posted: 13 January 2009 at 6:58pm | IP Logged

UK NEQAS for Blood Coagulation does not 'fail' a participant in a survey on the basis of one test; each test result is assessed independently. For further details please see the following link to our website:

http://www.ukneqasbc.org/content/PageServer.asp?S=529234305& amp;C=1252&ID=52

Please also follow the subsequent link to a detailed explanation of our performance criteria.

Posted: 14 January 2009 at 10:20am | IP Logged

Thanks for the reply. That's broadly similar to how we operate. Do you provide more than one sample for each test? If not, I guess that would explain the difference in how we operate.

Regardless, we're in the process of upgrading our stats programs to show which sample of the three a centre might have failed on - which moves us in line with the CAP schemes in the US.

Posted: 20 January 2009 at 6:32pm | IP Logged

Some tests do have more than one sample in each distribution (e.g. Prothrombin Time/INR where there are normally 2 samples). In some infrequently distributed surveys classed as 'Supplementary Exercises' there may be 3 or more samples covering a range of results (e.g. lupus anticoagulant)

We do indicate in reports for which sample(s) a participant has been classified as 'outwith consensus' (rather than 'failed' or 'poor performance') or 'persistently outwith consensus' where applicable

HTH