I'd imagine that computing the cost of health care is an incredibly complex undertaking. Without being able to review the methodology used, this graph is next to useless. For example, here are just a few questions that come to mind (and I'm sure more would arise if the methodology was properly peer reviewed):
1. Are there no data points between the United States and Switzerland? Are there no data points above the United States and below Mexico? If there are, that would make the graph misleading and they have to specify this limitation if they're intellectually honest. If there aren't, they have to demonstrate that they've gone through the exhaustive list of countries and there are no other "interesting" data points.
2. Is the cost adjusted for cost of living at a given country? If not, that would make the graph extremely misleading. If yes, what adjustment strategy was used? Can we see the ratio between cost of health care and median yearly salary, for example?
3. What does "average life expectancy at birth mean"? Does it account for countries that have a significantly lower birth rate than the United States? I would guess not, which could significantly affect the perception of the numbers. How does one normalize for something like that?
4. How was the currency conversion rate computed? Currency prices fluctuate throughout the year, did they account for that? How much does that affect the numbers? Could be a lot, could be a little, but I need to know whether this was taken into account.
5. What does "universal health coverage" mean? Soviet Union had universal health coverage and no medication, surgery performed by under-qualified residents, and no post-surgery care, unless you know someone or bribe the doctor, of course.
6. Surely there are countries without universal health care other than the U.S. and Mexico - how do they stack up?
7. I'm willing to bet a country like Poland has a lot more homeless people that don't factor into the life expectancy numbers than the U.S. That begs a more general question of whether the same methodology was used for computing the cost or life expectancy for each country. If they simply took official numbers from each country, it's almost certain that they were computed differently. Was this accounted for? If not, I need to know. If yes, how were the numbers normalized?
One could probably come up with dozens of questions like these. Of course any analysis of a problem that complex can be called into question, that doesn't mean every analysis is useless. But before we can seriously discuss this graph, or base any policy decisions on it, they at least have to provide the methodology. It doesn't look like this graph is intellectually honest to me, and the burden of proof is on the author.