Re: False Positives in Personnel Dosimetry?

[Bob Flood <bflood@SLAC.Stanford.EDU>]

The threshold for reporting non-zero doses should be based on the dosimeter
system's LLD - lower limit of detection. The LLD is the dose level at which
there can be confidence that the reading is not part of the background
distribution. Reporting values below the LLD will inevitably result in false
positives.

The LLD is a function of the average background, the standard deviation for
background measurements, and the standard deviation for low dose
occupational dose measurements. Numerically, the LLD calculation determines
how different from the average background an occupational dose has to be to
conclude it is indeed an occupational dose and not part of the normal range
of background measurements.

Note that the average background is not something over which the dosimetry
service or the customer has any influence. And it is extremely important to
make background measurements correctly, lest biases in background data cause
reporting errors. For example, background should be monitored at the
locations where dosimeters are stored when not being worn. If you use
on-site badge racks for this, you need to have controls on all racks and
calculate the average from all of them. If you use "take home" dosimetry,
i.e., no badge racks, and wearers can take their dosimetry home or leave in
the office as they choose, your background measurements should reflect the
diversity of background conditions across the communities where users live
as well as on-site. We conducted a study of background in the San Francisco
bay area versus our on-site background and found the area-wide variations
tripled our LLD compared to using only on-site data.

A vendor's claim of a detection capability is valid only for specific
background and measurment conditions. Whether the 1 mrem threshold is
realistic for your facility can only be determined by measurements at your
facility. A generic claim of 1 mrem would be correct for your facility only
by coincidence.
============================
Bob Flood
Dosimetry Group Leader
Stanford Linear Accelerator Center
bflood@slac.stanford.edu


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