The medical examples are particularly interesting because these are practical problems we face.

A very interesting thing is that there are many preventative measures that can be taken that are less intrusive, though certainly costly.

An example: one thing I’m reading about is that nutrition plays a huge role in activating disease. The book “The China Study” by T. Colin Campbell suggests a plant-based diet (no dairy, no meat) wards off against a variety of illnesses. Another book “Dr. Dean Ornish’s Program for Reversing Heart Disease” suggests that a program of meditation, a well balanced low-fat vegetarian diet, and exercise can actually reverse heart disease.

For an average American, these are huge costs to take. It is socially awkward at times to be a low-fat vegetarian or vegan. But then I remember that the average American gets heart disease. Many people are unwilling to take comprehensive changes.

Your example about crossing the street is very interesting too. Years ago I virtually stopped jaywalking for the very reason of eliminating catastrophic risk. I got made fun of a lot from my friends, but words do not hurt as much as a car collision.

I read somewhere that Nassim Taleb, author of “The Black Swan,” was going to write an entire book about dealing with such catastrophic risk.

I’m looking forward to it.

What if the cost of “insurance” were non-trivial (e.g., handwashing were painful, or if each handwashing significantly diminished your drinkable water supply)? At some point it becomes a better strategy to forego insurance and take the risk. In terms of expected value this depends on the cost of the insurance, the probability of the ‘bad event’ if you do vs. don’t take insurance, and the cost of the ‘bad event.’

I’m particularly interested in the limit where the probability of the bad event is low, but its cost is huge — and insurance is costly too. A practical example: generically, men/women have a ~10% chance of early death from prostate/breast cancer; “insurance” against this could be had (at a non-trivial cost) from preemptive prostatectomy/mastectomy. But this would be curative, and avoid early death (at least from the cancer).

What’s the right way to think about this problem? Lets say genetic counseling tells you you have an increased chance of getting such a cancer. You can have regular tests (incurring financial and emotional cost, not to mention some risk of missing the disease when its still curable) or the pre-emptive ‘ectomy (which carries its own risks and diminished quality of life — but at least you don’t die). What to do?

There are lots of risks we incur that have potentially catastrophic consequences (i.e. death, a seemingly infinite “cost”). For example, we do cross the street even though the gain in getting to the other side pales in comparison with DEATH. What’s the framework for thinking about this?

P.S. No, I’m not a crazy hypochondriac!