Room For Debate wrestles with the question. Carol Levine recommends redefining what is medically necessary:
Emergency surgery after a hip fracture is clearly medically necessary. The insertion of a feeding tube in a severely demented person with advanced cancer is medically unnecessary and fails to meet ethical standards of beneficence — doing good or at least doing no harm.
Robin Hanson wants to copy the British:
The United Kingdom, where, on average, people live longer than in the U.S., spends only about 9 percent of gross domestic product on medicine, compared with our 18 percent. … My solution: admit we are cost-control wimps, and outsource our treatment evaluation to the U.K. Pass a simple law saying Medicare (and Medicaid) won’t cover treatments considered but not positively appraised by the Britain's national health institute.
Alex Massie doesn't expect "this idea to catch on." Austin Frakt places his faith in the ACA reforms, but admits he's engaging in wishful thinking:
If we can’t just whack spending and expect better outcomes, what can we do? The answer must lie in more nuanced reforms to payment policy, towards a system that rewards the good performance we seek. That is, we need a scalpel, not a hatchet. That’s an obvious conclusion, neither helpful nor controversial.
Slightly less obvious – if you don’t know the research – is that there is no evidence that beneficiaries know how to select good performance even when they’re required to pay for it on their own. When they economize, they cut both helpful, necessary care and wasteful, useless treatment in about equal amounts. To continue the metaphor, beneficiaries don’t know how to use scalpels. This is the fatal flaw in the Republican plan for Medicare.
The Dish 