Friday, July 17, 2009

Health Care Reform

The problem is that the fundamental issue with health care still isn't being addressed or even debated. Certainly the fundamental issue is cost, and while privatization and/or nationalization are two ways by which to deal with cost, neither gets to the primal element of cost which is the nature of health care as a consumable good. I am no health care economist, but my impression and understanding is that health care is a sort of "super-necessity" good, or to put it another way, health care is very inelastic. This in and of itself wouldn't be problematic, as lots of goods are inelastic, but it becomes problematic when combined with the (ir?)rational subjective valuation of health care in this country. The "problem" is that our nation, as a collective and as individuals, puts (near) infinite value on continued life. In our valuation system, the marginal utility of an additional time unit (a minute, day, week, year, etc), when you are facing imminent death, of life is infinite (and perhaps not unjustly so). However, this skews what we are willing to pay for treatment, even when the expected benefit of treatment is very low. Say, for example, that you have terminal cancer (you will die with probability .99 in the next week), and that there is a new drug that with probability .05 will extend your life by some time unit (whatever unit you think is important). What would one be willing to pay for this drug? Well the expected benefit is .05*infinity = infinity (if we place infinite value on extend life), so this person would be willing to pay any amount of money for a 5% chance at an additional day, week, month, whatever....

The above is obviously a bit of hyperbole, but I think it illustrates a mindset that is prevalent in the health care industry - "test and treat" - at any cost (if the patient has insurance of course). A 30 minute doctors visit probably has an actual cost (doctors salary, overhead, materials, etc) of say $300. At this visit and examination, the doctor (if she is good) probably has a fairly reasonable diagnosis (say %80) of what the problem is. However, since this is AMERICA, 80% isn't good enough. So, in addition to running a lab test (another $300) to confirm the doctor's diagnosis (say to 95%, and this isn't necessarily a bad thing), the doctor will also run a battery of tests (another $2100) to make sure she has ruled out all other possibilities, and can diagnose the condition with 99.9% confidence. Thus, "80% good" health care costs $300, with a marginal cost of $3.75 per marginal point; "95% good" health care costs $600, with a marginal cost of $20 per marginal point; and "99.9%" costs $2700 with a marginal cost of $428.57 cost per point. Obviously I've just made all these numbers up, but they illustrate my point and my guess is that they are pretty representative of actual health care costs.

I have numerous friends in the health care industry and I know they conduct all the tests to get to the 99.9% level, even though the marginal cost is ridiculous. This is largely what drives the cost of health care to such crazy levels in this country, and this is driven by "our" subjective valuation of life (and also to some extent by medical malpractice litigation which is inversely related to the probably of the doctor's success). Thus, the debate in this country over health care needs to be a realistic one about what level of care we are willing to pay for. Yes, we have the best health care in the world, but does it really make sense to pay for it? Perhaps this is what you are suggesting in your plan, allowing insurance companies to offer packages at the "80% level" (i.e. you can see a doctor but get no tests), the "95% level" (i.e. you get the doctor and one lab test) or the "99.9% level" (doctor and all the lab tests they want), where the 95% package costs twice as much as the 80% package and the 99.9% package costs 9 times as much.