It's great to see work being done to highlight an issue but I do wonder what background does the author have? Would recommend gestalt/cleveland as a good grounding, the visualizations is editorial rather than analytical.

Choosing US versus Japan, which Japan has the lowest cost and highest life expectancy in the OECD, it's cherry picking. I'd recommend showing the full distribution of OECD per-capita spending rather than just a single cherry picked comparison.

This also is troubled by McNamara Fallacy, we're looking at metrics that are qunatifiable but ignoring what can't be measured or overlooked, is speed of access being considered, how about innovation incentives, quality and outcomes variation across systems, patient choice and flexibility, in addition to workforce compensation (nurses and physicians in the US earn significantly more). Where are the trade-offs?

Summary Statistics can be dangerous. 254% of medicare is a single ratio summarizing enormous variation across thousands of hospitals and procedures. Median markup of 3.96x inherently hides the distribution, some hopsitals may be higher or lower, why is that?

I think the biggest one to me was the confirmation bias, the $3 trillion gap that represented 'fixable waste' was the conclusion. Every price difference is interpreted as waste rather than investigating the potential cost drivers, was there a null finding framework in place where US spending appears justified or is it all bad?

Overall, glad someone is looking into the data and pulling insights, please don't take this as discouragement just a comment from the peanut gallery.