Acceptable Performance vs. Performance Standards

["Sandy Perle" <sandyfl@earthlink.net>]

The issue of performance that is better than existing performance 
standards was posted today. While this issue had been addressed 
previously, from the peripherary, it warrants a few comments:

(1) Standards, such as NVLAP, ANSI N13.11-1993 and 15CFR285 
were developed and only signify that the facility meets the "minimal 
requirements" deemed acceptable for a dosimetry processor. 
These standards should not be assumed to demonstrate 
exceptional, or commendable performance.

(2) As an assessor, I can only issue a deficiency if the facility does 
not meet the appropriate requirements established in 15CFR285 
and Handbook 150 and 150-4. While there may be a significant 
issue, unless it's addressed in one of the above documents, the 
facility is not considered deficient. I can only offer written 
recommendations, to which the facility need not apply. It would be 
prudent for them to evaluate the recommendation, since it is highly 
likely that the next assessor will ask them if they took the time to 
evaluate the recommendations, and what actions if any, did they 
take, and if not, why not.

(3) The fact that a facility has passed the proficiency testing does 
not and should not imply that they can accurately assess dose. It 
is one thing to test to a standard in a known environment, sterile as 
it is, compared to field dosimetry. The assessment of dose in each 
of these environments is quite different.

(4) The most important factor to deem a processor fit is the On-Site 
Assessment, taking into account the limitations of the assessor, 
as I described above.

(5) Passing a standard, at the upper limits, such as the 0.5 
mentioned previously, gives some a false sense of security. I 
wouldn't want to hang my company's financial security in a 
litigation case based on having the NVLAP Certificate and 
Accreditation.

(6) I would be more concerned about having a viable blind testing 
program and inter-comparison program established, and to have 
my program fully documented whereby the Uncertainty of 
Measurement is quite understood, and utilized within the program.

(7) If I am on the stand defending a dose assessment, I would 
rather be standing behind a system whereby I can demonstrate +/- 
10% of known spectra assessments, with +/- 5% better yet. I 
really don't want to say I am NVLAP accredited, and have 
performance quotients approaching the upper quartile of all 
processors, nearing the 0.5 limitations.

(8) I am a proponent of taking all reasonable and economically 
sound steps to improve the accuracy and precision of my systems 
.. without causing detriment to personnel resources. Remember, 
there are diminishing returns, and there is a point where 
performance will just not be improved without major efforts, if then.

------------------------
Sandy Perle
E-Mail: sandyfl@earthlink.net
Personal Website: http://www.geocities.com/capecanaveral/1205

************************************************************************
The RADSAFE Frequently Asked Questions list, archives and subscription
information can be accessed at http://www.ehs.uiuc.edu/~rad/radsafe.html