|
Again I ask, deaths as opposed to what?
IMO, the term "deaths" is a lousy way to express the stochastic
effect of exposure to any harmful agent
(i.e. where the exposure affects the probability for occurrence of the
effect).
Sooner or later, everybody dies. The only question is when
and how does death occur.
On the question of when, effects could be expressed in
terms of changed life-expectancy. However, loss of a few hours or days from
predicted life expectancies of ~ 70 years would have a much lower semantic
impact. For example, in the USA about 2 million people die each year. If some
given harmful exposure were to add 20 deaths/yr. to this figure, the equivalent
effect could be expressed as loss of about 10 hours in life-expectancy.
I think it makes a big difference on which way you want to look at
it.
On the question of how, suppose the effect we want to
avoid is cancer death. It should be noted that increase in the incidence in any
given cause of death will necessarily decrease incidence of other causes. Given
that we want to minimize occurrence of cancer deaths, perhaps we should stop
research on methods of reducing heart disease, the leading cause of death.
Obviously, if we were to succeed in preventing all heart disease, the incidence
of cancer deaths would dramatically increase. Conversely, if heart related death
increases, cancer death would decrease.
Considering all of this, I am not sure what the best way of quantitatively
expressing the effect of harmful exposures should be. I am sure that how
it is expressed has a strong impact on the perception of the problem. Anyone
have any good ideas on the subject???
|