Everyone wants to cut costs. But what if saving my life is expensive?

At the White House's health care town hall on Wednesday, an epilepsy specialist named Orrin Devinsky asked a question that haunts proponents of reform:

"President Obama, if your wife or your daughter became seriously ill, and things were not going well, and the plan physicians told you they were doing everything that reasonably could be done, and you sought out opinions from some medical leaders and major centers, and they said there's another option that you should—should pursue, but it was not covered in the plan, would you potentially sacrifice the health of your family for the greater good of insuring millions? Or would you do everything you possibly could as a father and husband to get the best health care and outcome for your family?"

That question—the health care equivalent of asking Michael Dukakis whether he would still oppose the death penalty if his wife were raped and murdered—lays out a disturbing scenario. Fortunately—for the White House as much as for anyone with a sick spouse or child—it's not very realistic.

The logic behind the question is clear enough. It's easy to talk about cost cutting in macro terms: As much as 30 percent of health care spending goes to unnecessary treatments. Eliminate those, and we're good. But on an individual level, who defines "unnecessary"? Who is going to decide not to order the extra test or the extra treatment? When it's your life or your child's at stake, are you really going to turn down a long-shot treatment just because it's too expensive?

The answer: Of course not. Obama acknowledged as much: "You're absolutely right that, if it's my family member, it's my wife, if it's my children, if it's my grandmother, I always want them to get the very best care."

But there are problems with that hypothetical, say doctors and economists. One is that it's not accurate. "It's a political cartoon," says Peter Bach, a pulmonologist and epidemiologist at Memorial Sloan-Kettering Cancer Center in New York City. "There is no situation in clinical medicine where I have a patient with life-threatening illness and I'm turning around to grab a drug and big government is barring the doors." (One ad, from Conservatives for Patients' Rights, says just that.)

Nor is a doctor ever sitting there with the patient, weighing chances of survival against the cost of a treatment. "The doctor has to be an advocate," says Arthur Caplan, a bioethicist at the University of Pennsylvania. "That's their job at the bedside." No public plan is going to change that.

What needs to change is the quality of information about tests and treatments. Right now, doctors don't know enough about whether a drug or device will be effective. That's one reason treatment tends to be aggressive. Why risk it? the thinking goes. The problem is, many treatments once thought to be effective are now considered unnecessary. Spine surgery, knee arthroscopy, high-dose chemotherapy, bone-marrow transplants, lung volume reduction surgery, prostate cancer screening, estrogen therapy … the list goes on. Not only are they often unnecessary; they can even be harmful.

"Everyone wants to live," says Bach. "But the data suggest they're not doing things that are making people live. They're just doing things."Upgrading health IT and cost-effectiveness analysis would fix that—they would allow doctors across the country to share data about what treatments work and which ones don't. As a result, they'd order fewer unnecessary treatments, and the ones they do order would be more effective.

The second fix would be reforming medical malpractice laws. One reason treatment is so aggressive—and costs so high—is that doctors are afraid of getting sued. Say someone goes to the doctor complaining about chest discomfort. "In France, they're sent home to rest," says Bach. "In the U.S., they get a cardiac CT, a stress test with thalium, an echocardiogram, then a cardiac catheterization no matter what it showed. All of that is against evidence."