RE: Quality hospitals

["Prevette, Steven S" <Steven_S_Prevette@rl.gov>]

 > I see very few opportunities for us to do that any better by deeper
analysis of our data streams.

One thing to consider is that whether or not you control operational
data, many times, it "controls" you if you let it.  Knowing if ER
admissions are stable and predictable, or are unstable (and if so, under
what conditions) could be very useful for determining staffing levels.

One significant, but unconventional, result of statistical analysis came
from trending radioactive exposure to personnel at one of our projects
(and hospitals have to track radioactive dose also, if you have X ray
machines or isotope treatments).  Suddenly one month the dose for
several work groups dropped below the Lower Control Limit on the control
charts.  I called my point of contact at the project and he did not have
any idea as to why the dose had dropped.  They were still doing the same
work, hadn't taken any new precautions, etc.  After some digging, it was
found out that the laboratory that was doing the dose measurement had
recently adjusted its "correction factors" for estimating person-dose,
and someone had slipped a decimal point.  The doses reported were coming
out 1/10th of what they should have been.  The control charts detected
the change.  No one else prior to that point had.  These incorrect doses
had been entered into the workers' medical records and had been
officially reported out with no one noticing the error.  The use of
statistical analysis (using SPC) made the error very visible.


Steve Prevette
Occupational Safety and Health
Fluor Hanford, A Fluor Government Group Project
ASQ Certified Quality Engineer
steven_s_prevette@rl.gov
509-373-9371