Gene testing

Mark D. Garfinkel garfinkl at
Sun Feb 27 11:26:33 EST 1994

In article <J.Oosting.67.2D6F9FB1 at Farmaco.RULimburg.NL>
J.Oosting at Farmaco.RULimburg.NL (Jan Oosting) writes:

>I see Insurance as a kind of betting. (a 1000:1 on me to get in the hospital)
	I agree that there is a gambling aspect to the insurance business.
The premiums are based on statistical studies of groups, but each of us
lead individual lives that may deviate from the mean.

>No insurance company will take this bet when I am already in the ambulance
>after a heart attack.
	By definition. Insurance, as I said, is a buy-now-use-later contract.

>But I do think that insurance companies have a social function too. The
>prices for insurance should be leveled off to the middle. People with a poor
>medical condition should get better bets than calculated for their risk
>factors and people with good health should pay more than they ought to on
>basis of health statistics
	Insurance is already a form of income redistribution in that within
an actuarily defined subpopulation the number of premium-paying customers
(usually) exceeds the number of claim-receiving customers. Often, but not
always, the claim-recipients obtain value greater than what they paid in
premiums for a given year. Furthermore, there are, to some degree, random
factors at work: People who make insurance claims because of accident,
injury, disease, are in one sense unlucky: something bad's happened to
them. In another sense, though, they win the lottery: their claim benefits
may substantially exceed the premiums they have paid in. What makes the
system fair is premiums being assessed according to a statistically sound
process, that premium-payers participate voluntarily, and that of course
next year it's different what happens to whom.

	I have to ask you: Why should there be a further redistribution
of premium burden so that those who are in the statistically healthier
subpopulation pay more than those who are statistically less-healthy?
There is already a redistribution from those who are lucky (all other
things being equal) to those who are unlucky. Now you propose to eliminate
the "all other things being equal" clause. Why? I think you have to justify
your position with something more than a vague appeal to an insurance
company's "social function."

	Let me put the matter on a personal hypothetical basis: You have
healthy habits, eat right, exercise regularly, etc., and I do not. You're
never sick and I am sickly, and there is a convincing cause-effect
relationship between your good habits & good health, and my bad habits
& poor health. What possible basis could *I* have in saying, "you should
pay disproportionately larger premiums"? The only answer *I* can come up
with is that in the hypothetical case I'm trying to be a thief, that I am
compounding my irresponsible behavior with larceny. I am trying to take,
from you, a subsidy on my health insurance premiums. What possible basis
could *you* have for going along with that scheme? Why should you do so?
In real life, you're proposing that a whole group of healthy people
agree to subsidize another whole group of unhealthy people. In my view,
such subsidies constitute grand theft.

>At the moment DNA testing gives only little detailed information about
>someones health risks because only few diseases can be detected nowadays.
>This disfavours people with these bad genes compared to people with bad
>genes that can not be detected at present
	So? If the bad genes that are detectable are strongly causative
of particular diseases, then genotype is itself a risk factor for *those*
diseases. If I am *known* to lack a genetic factor that predisposes me
towards cancer type-1, my premiums ought to reflect that fact. And the
size of the premium reduction should somehow be proportional to the
frequency of the disease in the population at large and the strength with
which genotype is causative. The fact that we cannot also test if a
genetic factor predisposes towards cancer type-2 should be without effect.
In any case, epidemiologically we know that cigarette smoking is strongly
causative for lung cancer, and so my insurance company would *still* be
much more interested in my behavior than in my genes since the incidence
of lung cancer is so much higher than other forms of cancer.


Mark D. Garfinkel (e-mail: garfinkl at
My views are my own, which is why they're copyright 1994

More information about the Bioforum mailing list