In broad terms, we must ... invest in better information on the effectiveness of treatment options so it might guide the decisions of patients, health care providers, and public and private insurers. Our nation spends over $2 trillion on health care, yet we know far too little about the comparative effectiveness of alternative treatments. Such information is a public good, which has not---and will not---be [sic] spontaneously produced by the private market.
Except: the private market has spontaneously produced comparative-effectiveness information, as I documented in this study released one month prior to the MedPAC report. I also explain why the argument for government provision of public goods is shaky.
Ironically, MedPAC disables the "copy" function in the reports it posts online, so I had to type out the above excerpt by hand. Why is that ironic? It's an example of the sort of innovation that enables markets to overcome the public-good problem. (But since my tax dollars paid for that report, it seems to be an innovation that maybe MedPAC could abandon.)