It might be worth digging into exactly what metrics RAND used to define healthy. For example, if I have a bad knee towards the end of my life, if I have European style care I get put in a queue with a small but significant probability I will die before I get the treatment. In the US of A I get it immediately at great expense regardless of how long I will actually use it. <p>I don't know if knee pain made their metrics, but if not it would be a significant expense with zero gain. The article mentions a negligible effect on survivability but even small changes can muck up results when death is involved. For example, if two countries have exactly the same care but one elects not to expensively treat a terminal disease. Those who are treated in the first country may die anyway and if they don't will probably live out their longer lives at a reduced health status. Since they were equal otherwise, the first will have lower health relative to the second even though they have spent more, an apparent scandal.<p>One solution would be to set health status for the prematurely dead to be as though they were very sickly but alive. But how sickly, and how long do these un-dead individuals get to stick around? You're fudging either way, so I'm suspicious of what they actually did.<p>Maybe I should go find the article... or get back to work... /emote is acting conflicted