Re: Analytical lab performan

["ROB FORREST" <ROB_FORREST.RSO#u#MAILROOM@rsomacserver.rso.upenn.edu>]

                      RE>Analytical lab performance on spi         6/15/95
My previous employer performed numerous urine bioassays principally for uranium
and mixed fission products.  The bias in ANSI N13.30 seemed a bit large for
spiked samples so we narrowed the requirements based on vendor passed
performance and discussions with several radiochemists.  The final criteria was
+/- 20% for spiked samples.  Considering the uncertainity in everything else
concerning internal dose, this range seemed to be acceptable.  Depending on the
vendor, they should be able to meet this criteria without too much problem. 
Good Luck.

-Rob Forrest
UPENN


--------------------------------------
Date: 6/14/95 6:46 PM
To: ROB FORREST
From: radsafe@romulus.ehs.uiuc.edu
Received: by rsomacserver.rso.upenn.edu with SMTP;14 Jun 1995 18:44:56 U
Received: from romulus.ehs.uiuc.edu (romulus.ehs.uiuc.edu [128.174.74.24]) by
postoffice.cso.uiuc.edu (8.6.12/8.6.12) with SMTP id RAA56376; Wed, 14 Jun 1995
17:37:02 -0500
Received: by romulus.ehs.uiuc.edu (NX5.67d/NeXT-2.0)
	id AA17052; Wed, 14 Jun 95 17:36:44 -0500
Date: Wed, 14 Jun 95 17:36:44 -0500
Message-Id: <9505148031.AA803169076@ccsmtp.sandia.gov>
Errors-To: melissa@romulus.ehs.uiuc.edu
Reply-To: radsafe@romulus.ehs.uiuc.edu
Originator: radsafe@romulus.ehs.uiuc.edu
Sender: radsafe@romulus.ehs.uiuc.edu
Precedence: bulk
From: "Mallon, James" <JMALLO@saix367.sandia.gov>
To: Multiple recipients of list <radsafe@romulus.ehs.uiuc.edu>
Subject: Analytical lab performance on spike samples
X-Listserver-Version: 6.0 -- UNIX ListServer by Anastasios Kotsikonas
X-Comment:  Radiation Safety Distribution List 

  New Thread!

The internal dosimetry group at Sandia is in the process of developing a 
procedure to document the review of lab results for spiked samples.  I have 
reviewed the ANSI N13.30(draft 1994) guidance for performance criteria on QC 
samples.  The ANSI standard identifies numerical criteria for the avg. bias of
a group of spike samples.  This parameter alone is insufficient to evaluate the

performance of an analytical lab.  Individual sample bias could swing wildly, 
while the average bias is within the -0.25 to +0.50 criteria.

We are thinking of looking at the standard deviation of the bias to provide a 
measure of how the bias is behaving.  So the questions are:

1. Is anyone aware of any formal guidance in this or a related area(e.g. TLD's,

   radiochemistry)?

2. Has anyone adopted 'rule of thumb' guidance for their own use?

Thanks for the help.

Jamie Mallon

Environmental Dimensions Inc.
Phone 505-844-8287
Fax   505-844-0701
Email JMallo@SAIX367.Sandia.gov

P.S. I'm a Scorpio!