Introducing Rational Risk
I was invited to give a
presentation at the Biomedical Standards eXchange 2013 meeting sponsored by the
Singapore Standards Council. The meeting
was held in conjunction with our International Organization for Standardization
(ISO) meeting for Technical Committee 212.
As opposed to the ISO meeting, the eXchange was well organized, and
provided a lot of interesting information that would likely form the basis of
some future action. [The ISO meeting was
pretty much the opposite.]
I was invited to give a
presentation on the application of Risk Management for the medical
laboratory. The sub-title for my
presentation was “Reducing Risk through
Application of Standards”. For those
interested, the presentation slides will be available at www.POLQM.ca by November 27, if not sooner.
The message
from my presentation was not a traditional Risk Management message, but I think
it was both realistic and appropriate, and therefore an effective message for
medical laboratorians.
I started
with some basics from Frank Knight that Uncertainty is made of two components;
measureable uncertainty (“risk”) and the unmeasurable component (the unknown
unknowns). I argued that in controlled
situations, like manufacturing, one can expect to considerably reduce
uncertainty variables (measureable>>unmeasureable), but when one thinks
of all the variation associated with patient selection, test ordering, sample
collection and transport, the unknown unknowns in medical laboratory testing are
huge and the relationship is likely inverted.
Then I
introduced observations from medical laboratory error investigation which
indicate the large impact and frequency of testing slips. Game and Chaos theories say that all events
have a cause and effect, but often the cause is unrecognized. Causes may be unproportional, meaning that
seemingly trivial causes are associated with major effects (downstreaming).
My main argument
was that while there is a role in using risk management tools to reduce bad
outcomes, at the same time, we have to acknowledge that until many other
control issues come to play in the medical laboratory, one needs to understand
that the goal is to reduce risk of bad events rather than prevention.
We are
more likely to achieve success through the application of outcome oriented risk
management tools (such as severity-occurrence grids) than we would with a
static precision tools (like Failure Mode – Effects Analysis or FMEA).
I demonstrated how to set up an S-O grid,
and how to use one for setting priorities and further how an S-O grid helps in
implementing risk mitigation strategies.
I think
that was a helpful message, and it seemed to be well appreciated.
I also
pointed out a risk strategy flaw in ISO 15189:2012. As good as this standard is it introduces
risk management as a way to reduce error soley associated with patient
safety. That is a good message, but does
not go far enough. Risk in the medical
laboratory has a much broader footprint than just patient safety. We also have risks that impact clinical
staff, laboratory staff, the community, the environment, and the
institution. Setting up risk reduction
without taking these issues into consideration would likely leave the
laboratory with all sorts of liability issues.
One of the
other speakers, coming from the cosmetic industry, also talked about risk, but
from another perspective. He introduced
the following formula:
RISK
= HAZARD X EXPOSURE
His point
was extremely relevant to the microbiology laboratory. Just because E. coli can produce exzymes and
toxins and can cause infection (all examples of inherent hazard), this does not
equate to risk level because exposure effects are moderated by handwashing,
containment, packaging, laboratorization, antimicrobials, etc, etc, etc.
The
formula is critical in recognizing and understanding the differences between risk
perception and risk reality. It is the
key issue that should distinguish the general public from what should be the
knowledgeable public [politicians, regulators, biosafety officers].
All too often, it seems that “knowledgeable
public” see a microbial name and then based on that information alone make all
sorts of inappropriate assumptions and judgements about risk to public, many of
which are just wrong.
Recently I
tried to have a discussion on the subject with one of our Canadian transport
regulators, who was clearly incapable in understanding the concept. Too busy first protecting her right to write
regulation to make sure that the regulations were rational or appropriate.
Write
before being right.
It was
disappointing and sad comment on the ability and mindset of what should have been
a responsible professional. Tragic.
