The non-partisan Tax Policy Center, after reviewing the two candidates health care plans, found the McCain plan would throw 20 million people currently insured by private employers into the dysfunctional individual insurance market and add just 1 to 5 million new people to the ranks of the insured.

Yet the McCain plan would cost nearly as much as the Obama plan -- $1.3 trillion versus $1.6 trillion over ten years -- even though Obama's plan would insure 34 million people who are currently uninsured, according to an analysis in this morning's Wall Street Journal.

Another review of the candidates' plans in Health Affairs pointed to a little noted aspect of McCain's plan: it would allow people to buy insurance across state lines. In theory, that should increase competition and lower costs. However, it would also undermine state rules that require insurance companies to cover specific services like breast cancer screening, smoking cessation counseling and other preventive measures.

The Tax Policy Center, which is a joint project of the Urban Institute and the Brookings Institution, did not evaluate either plan's ability to hold down the rate of increase in health care spending. But I attended a forum yesterday sponsored by the New America Foundation where notable panelists offered sobering views on that subject.

Everyone wanted to limit waste. Julie Barnes of NAF said a combination of payment reforms (scrapping fee for service for bundled payments; using comparative effectiveness studies to drive medical decision making; establish medical homes -- health care reform jargon for paying a professional to coordinate the care of people with chronic disease) could reduce health care spending growth to the general rate of inflation within ten years.

But I was surprised to hear Henry Aaron of Brookings, who a few years ago wrote a short but eloquent book on the need to ration care that really doesn't add much to health care outcomes, dismiss the potential of such steps. Eliminating waste won't hold down costs because it "isn't very much and isn't a great deal of what is driving the rise in health care costs," he said.

The real driver of the rising cost of health care is what he called "high cost-low benefit" care. "What do we do about those very costly interventions that help some people some of the time? How do we say that some people won't have access to those kinds of interventions? That's the hardest question for a democracy to face," he said.

Aaron then gave the example of a 85-year-old woman who was dying of lung disease. Put on a ventilator, she could last for perhaps a month. Do nothing except make her final days more comfortable, and she would last less than a week. Will we ever have a comparative effectiveness agency willing to declare that the cost of that intervention simply isn't worth it? The woman in his example, when told about her options, said, "put me on the ventilator."

"These decisions have to be made by medical professionals and patients at the bedside -- not by some commission," Aaron said.

Other societies are able to say no in such circumstances. They set payment guidelines. Or they set universal budgets where hospital authorities and physicians, who must live within those budgets, do the best they can for the most people possible.

I, for one, believe that we can actually improve care by eliminating some of the very expensive but marginally useful medical expenditures. It will enable the system to deploy its limited resources into interventions that can be much more effective.

One simple example: Does it make sense to spend $20 billion on cholesterol lowering pills, many of whom are being used by people who are only slightly at risk of heart attack? Wouldn't it make more sense to take, say, $5 billion of that money that gets spent on people at low risk and instead spend it on screening, counseling and other proven interventions for people at serious risk of heart disease, like those who are obese, have untreated hypertension and are pre-diabetic?

The general public has no clue about the choices implied by its personal health care decisions. It still doesn't know that despite spending more than any country on earth, we have worse outcomes. It is still possible for a politician to stand up in public and say we have the best health care system in the world (as the president did earlier this year) because most of his audience had no information on which to refute such a ridiculous statement.

But health care reformers are similarly handicapped. There is a small cottage industry documenting the waste in the system. The Dartmouth Atlas of Health measures it. Books have been written on the topic. But reformers still don't have a language for answering Aaron's questions: Yes, but didn't at least some of those patients benefit from those expensive treatments? And even if the cost of the intervention is $500,000 per quality adjusted life year (QALY), who are you to decide that the patient didn't deserve it?

We need a frank discussion about how much better off the entire population would be if a) resources were considered scarce; and b) those scarce resources were deployed to interventions that gave the most bang for the buck. So if giving statins to people at low risk of heart disease costs $150,000 per QALY, but hypertension screening and aspirin therapy costs $10,000 or less per QALY, someone has to say, we're going to the latter, but not the former.

That's how you bend the cost-curve down. It's not through comparative effectiveness information. It's through comparative cost-effectiveness information. And it's through having the popular understanding that more people will benefit from using that information to drive medical decision making. Societies that have high quality, affordable health care systems that cover everyone and have the best outcomes understand that when it comes to health, "we" is more important than "me."

Henry Aaron was quite eloquent yesterday in defending the rights of individual patients and their physicians to make those decisions. He reminded this observer than we still are a society that puts me first, and as long as that's the case, health care cost control is impossible.