Educate people and they'll want "death panels"
"Death panels" is a terrible, politically motivated terms for it but, as I recently linked healthcare rationing seems to be necessary and seems to generally be done in public healthcare systems.
Whether healthcare rationing needs to be done was the subject of the latest Intelligence Squared US debate.
The results show the closest to a unanimous vote that I recall having ever seen in one of these debates (the video of which is online, but which I haven't yet gotten around to viewing). What were the results (figures are percentages)?:
| In favor of rationing | Opposed to rationing | Undecided | |
| Before the debate: | 43% | 22% | 35% |
| After the debate: | 81% | 12% | 7% |
Shockingly, a system with finite resources can't spend infinite amounts of money. Of course there's also the need to apply many of the same measures to other types of data. Consider Table 1 on p. 58 of this Cato report on the subject of spending on life saving initiatives (I'd link to a journal paper, but the relevant ones seem all gated).
What the Cato report notes is that some measures adopted seem quite affordable (less than $100,000/life in 1995 dollars) whereas at the other extreme there's a case involving an estimated ~$17 trillion in spending per life saved (1995 dollars) - more than the entire US GDP for the most recent year.
Beyond a certain point, you're better off investing in alternative measures - that same Cato report plots in Table 2 a few different estimates of the point beyond which spending that much to save a life actually statistically will increase mortality.