TheraP ridicules a strawman and does it nicely. She defends the virtue of uniformity over diversity when it comes to health care and paying for it.

The fact is that not everyone has uniform health care values or needs, whether viewed at a moment in time or over a lifetime or over generations. Another fact: Health care is not monolithic, treatments and diagnoses are often heuristic and sometimes conflicting. And except in a non-evolving autocracy, having options is a fundamental principle of both nature and nurture. It's easy to misunderstand "degrees of freedom" and then play with words, harder to build an effective system which accounts for real degrees of freedom, whether in natural or cultural reality.

As we all know, personal responsibility does play a role in health, and therefore consequently in health care needs. And as we've seen on Wall St. recently "moral hazard" is a fact of the human condition. We won't make good progress if we only castigate a strawman notion or fence with windmills. If you want to argue that health care is a right rather than a privilege, you must also argue with equal eloquence for individual responsibility.

The USA is, and should remain, neither an anarcho-libertarian free market nor a communistic autocracy, when it comes to health care. The public interest in proper health care is not zero, nor should the individual get it for free.

TheraP's post demonstrates how not to take the straight line to a destination. To expedite, I suggest finding a better frame next time.

There are two key 2-D frames to consider from the demand side: Need and access, and cost and payment. There are similar frames from the supply side (doctors, hospitals, technology, ...).

A third frame is that of transition, if the "system" is to be transformed from the mucked up status quo, what are the transition costs (dollar and otherwise)? Economic and human dislocations need to be planned for and paid for.

A rational solution will not ignore any of these.

Part 1 mentioned natural and cultural realities. 

I think the "sanctity of life" issue is not discussed enough. We mostly hear it from anti-social Fundamentalists who want to dictate the behavior of women of childbearing ages. But it applies to end of life scenarios, too.

Something like 1% of patients need over $150K medical care while the vast majority don't need anything more than a few $100s/yr. I've seen stats presented at TPM showing that huge fraction of health care cost is spent in the last months of life.  Here are two illustrative scenarios:

Spend a lot of money at decent odds of success failing to save a life which could on to be very productive, say, a 20 year old accident victim in good health; and, spend a lot of money keeping a bed-ridden octogenarian from dying so quickly, aka decent odds of death.

Both fit the simple criteria of $$ and end-of-life. The values in the two scenarios should be very clear, and very clearly different. While the contrast is clear, the frame is of course simple to the point of being simplistic. For example:  It does not illustrate the possible value to medical science or engineering of making the effort in one or perhaps both cases.  But it is accurately illustrative of value conflicts.

Until the culture shifts, both are technically legit expenses of money in service to health care. And with individuals spending their own earned income, few would complain more than perhaps an eye-roll's worth. But with second or third party involvement in payment for or delivery of services, the question of allocation of finite resources applies. At some point someone other than the individual may have to say, "pull the plug".

We don't need to make it a blanket 'either or' issue to see the problem. But in reality, resource allocation does often come down to that, whether we choose rationally, or wisely,  or not.  If we don't choose, then we choose to bankrupt ourselves on health care extravaganzas not even imaginable 100 years ago.

It would be interesting to see a more detailed breakdown of recent actual cost distributions over the "space" my comment defines. I have no idea where to find such, much less create my own graphs out of some raw data which might not be readily available.