Re: formaldehyde and truth

Dan Shapiro (dan@good.stanford.edu)
Thu, 10 Aug 95 12:22:33 -0700

Roman makes a good point that we are really operating on theory vs fact
in many of these long term exposure problems. This makes it even harder
for us to estimate and compensate for risk... the stake in the ground
is kind of fuzzy. (In order to directly measure low level toxicity we
would have to wait for natural experiments to conduct themselves -
across many people and lots of time. So, we resort to injecting rats
with omigod quantities of suspected toxins, producing arguable
conclusions.)

Here's my cut on a solution: let's come up with a simple set of Standard
Operating Procedures - a table that classifies compounds by "risk level"
and associates reasonable precautions with each category. The
information might be surprising to some. The table would represent
someone's bias, but it would be better than our current situation which
amounts to random guidelines, one per person/class/institution. Don't
chemists use something like this all the time to decide whether to run
their next experiment on the table top or under the hood, or in a P3
isolation chamber?

There is even a way to attach some intuitive meaning to a table of this
form. Define each category of toxin + precautions so as to yield equal
risk, defined as so many chances in a million of dying today. (The
technical term for this is a "micromort" - cute, huh? It comes out of
some theory in Decision Analysis.) We all live in a background
radiation of micromorts ... we absorb a couple extra here for riding a
bicycle to work, a couple there for playing sports, a big spike for
sky-diving. It's like calorie counting; a few examples would give you a
good feel for what you are doing to yourself by eating that
cookie/putting your hands in hypo.

Dan Shapiro