[extropy-chat] Failure of low-fat diet

[Hal Finney]

Another example of the failure of modern medicine to improve is discussed
in today's New York Times:
http://www.nytimes.com/2006/02/22/business/22leonhardt.html
also reprinted at the bottom of:
http://www.finfacts.com/irelandbusinessnews/publish/article_10004953.shtml

> With all the tools available to modern medicine - the blood tests and
> M.R.I.'s and endoscopes - you might think that misdiagnosis has become
> a rare thing. But you would be wrong. Studies of autopsies have shown
> that doctors seriously misdiagnose fatal illnesses about 20 percent of
> the time. So millions of patients are being treated for the wrong disease.
>
> As shocking as that is, the more astonishing fact may be that the rate
> has not really changed since the 1930's. "No improvement!" was how an
> article in the normally exclamation-free Journal of the American Medical
> Association summarized the situation.

The reason, according to the article, is how incentives are arranged:

> Under the current medical system, doctors, nurses, lab technicians and
> hospital executives are not actually paid to come up with the right
> diagnosis. They are paid to perform tests and to do surgery and to
> dispense drugs.
>
> There is no bonus for curing someone and no penalty for failing, except
> when the mistakes rise to the level of malpractice. So even though doctors
> can have the best intentions, they have little economic incentive to
> spend time double-checking their instincts, and hospitals have little
> incentive to give them the tools to do so.
...
> Joseph Britto, a former intensive-care doctor, likes to compare
> medicine's attitude toward mistakes with the airline industry's. At the
> insistence of pilots, who have the ultimate incentive not to mess up,
> airlines have studied their errors and nearly eliminated crashes.
>
> "Unlike pilots," Dr. Britto said, "doctors don't go down with their
> planes."

Britto has a company that sells software which provides lists of
alternative diagnoses when doctors type in symptoms.  The article
describes a case of a misdiagnosis which was corrected through the use
of the software, but implies that its $750/yr/doctor cost is too high
for widespread use.

I've been enjoying the TV show "House", which is all about medical
misdiagnosis.  Dr. Gregory House is an expert diagnostician, and he and
his team of three good-looking junior doctors struggle every week with
a difficult case.  The shows always have the same structure, namely
that the team comes up with one wrong diagnosis after another, while
the patient gets sicker and sicker, usually because they are treating
him for the wrong thing and basically killing him.  Finally at the end
they guess right and the patient gets better.

It's not a very realistic show, but still entertaining.  Now whenever
friends or family are having a problem getting a reasonable medical
diagnosis, we say, we need House to come figure things out.  I've been
dealing with some screwy test results myself lately, and I finally
found a good doctor, who takes a scientific approach to the problem.
He's treating me for what he thinks might be wrong, not because he thinks
it is necessarily the best treatment for me, but just to test his theory
and confirm the diagnosis.  It's a great feeling to have a "Dr. House"
on my case.

Back to incentives: when not shoring up the foundations of quantum
mechanics or lending transparency to the murk of Middle Eastern politics,
Robin dabbles in health care economics, and one of his first papers on
that subject has a solution for this problem.  "Buy Health, Not Health
Care," <http://hanson.gmu.edu/buyhealth.html>, talks about getting life
insurance companies to pay for health care.  Since they have an economic
incentive to keep you alive, their incentives are aligned with yours.
And they would develop the expertise to know what works and what doesn't.
To use the analogy above, they do "go down with the plane", in the sense
that they pay out a huge amount every time a patient dies.

Something along these lines might go a long way towards fixing this
stubborn problem of 20% fatal misdiagnoses.  I'd be curious to know what
further ideas Robin has on this since his 1994 paper.

Hal