Physically Significant Activity

["MacLellan, Jay A" <jay.a.maclellan@pnl.gov>]

After looking at the ABHP Exam Q's & A's in the May Newsletter,  I see no
reference to a statistical "no-man's land" in Dr. Skrabble's solutions.   The
PSA is just an extension of Currie's work. 

Using Currie's approach, the decision level is set at a point on the blank
distribution above which results would be observed with less than a specified
probability (often but not always 5%).  If our sample result is greater than the
decision level we say there is something there, or there is less than the stated
probability that it is just due to random fluctuations.  

The MDA is used to define the sensitivity of the process.  It says how much
activity must be in the sample to have less than a specified probability of
getting a result less than the decision level (a false negative).  

The decision level and minimum detectable activity (a poorly chosen term that
causes much confusion) have different uses.  The MDA defines the sensitivity of
the process.  The decision level is the yardstick we use to evaluate a single
sample.  There is you "no-man's land" between the two, and no additional terms
are necessary.  We just need to use the terms we have properly.

Currie stated the importance of establishing the decision level in the count
domain, prior to correcting results for counting efficiency and chemical yield.
The PSA would translate the decision level for the sample to activity.

Jay A. MacLellan, CHP
Manager, Radiation Records
Radiation and Health Technology
Pacific Northwest National Laboratory
Phone:  509-376-7247
Fax:      509-376-2906
Email:   jay.a.maclellan@pnl.gov
As always the opinions are my own.  Don't blame the government, my employer, or
anyone else.

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