It also sounds like a number pulled from a hat.
So I went looking.
It's from a Tufts Center for the Study of Drug Development study published in 2014 [0]. Which is, incidentally, supported financially by the pharmaceutical industry [1]. Even the published pdf is pretty lacking on actual data [2]. Because it's a PR piece. This other study, from the Office of Health Economics, 2012, reported the cost as $1.5bn [3]. (OHE is funded by research grants and consultancy fees from - you guessed it - the British pharma industry). Did the cost magically increase by $1bn in the two years between the studies, or is there something fishy going on here?
One point in the Harvard study abstract was,
> The most important factor that allows manufacturers to set high drug prices is market exclusivity, protected by monopoly rights awarded upon Food and Drug Administration approval and by patents.
And that's true! But that exclusivity is an artifact of medicine-for-profit, instead of medicine-for-health.
It's interesting that all of their recommendations for lowering prices fail to take into account (at least in the abstract, as I don't have access to the full article) the single biggest difference between the only market where per capital spending is $850 and the 19 markets where per capita spending is, on average, $400: for profit healthcare and medicine.
Keep in mind that even after that astronomic cost for drug approval (the figure for which is iffy in providence), for-profit pharmaceutical companies still spend more on marketing than they do on research.
0. http://csdd.tufts.edu/news/complete_story/pr_tufts_csdd_2014...
1. http://csdd.tufts.edu/about/corporate_sponsorship
2. http://csdd.tufts.edu/files/uploads/Tufts_CSDD_briefing_on_R...
3. https://www.ohe.org/news/overview-ohe-study-cost-drug-develo...