From his perch at Politico, Dan Diamond has launched what promises to be a terrific new health-policy podcast, Pulse Check, with an interview with CMS’s acting administrator, Andy Slavitt.

The interview is refreshingly candid: among other things, Slavitt confesses that the Obama administration has a lot of work to do to win back the hearts and minds of physicians who’ve become disenchanted with health reform. Slavitt also talks openly about the challenges of balancing his devotion to his family with his public service.

One thing in particular caught my ear. Slavitt said that he expects frustration over drug prices to grow. “If we don’t address this problem in the next four or five years, and prescription drug costs grow as they’ve been growing, this will be one of the top three issues for my successor, if not the No. 1 issue.”

For reasons that Austin explored here, I think that’s probably right (although anger over high deductibles will also rank high on the list). Spending on prescription drugs amount to only 10% of overall health-care spending, but that’s the spending that’s most salient to patients. As pressure builds for a solution, policymakers will be on the lookout for bold new approaches.

As if on cue, Amy Kapczynski and Aaron Kesselheim have a new paper in Health Affairs sketching out one such approach.

[A] little-known law, codified at 28 U.S.C. section 1498, … gives the government the right to use patented inventions without permission, while paying the patent holder “reasonable and entire compensation.” In the case of pharmaceuticals, a patent gives a company the right to prevent others from making, selling, using, or importing a covered medicine. The “government use” provision is a form of governmental immunity from patent claims: Under it, patent holders can demand royalties but cannot stop the government from producing the medicine or allowing others (in this case, generic manufacturers) to produce or import the medicine. …

[W]e assess the potential for the federal government to invoke section 1498 to make important new high-cost therapies widely available to patients who need them. We argue that such an approach need not undermine incentives to innovate and could result in net gains from an economic as well as a health perspective. Most prominently, the approach should be considered whenever there is evidence of excessive pricing as measured against the risk-adjusted costs of research and development, and whenever substantial population-level benefits could be realized.

The authors liken their proposal to the power of eminent domain, only for drugs. It’s both provocative and, to my eyes, more promising than other aggressive approaches that are under discussion. (For the curious, Rachel Sachs has succinctly explained why using march-in rights won’t do the trick.) Even the prospect that the government might use eminent domain, as the authors note, might be enough to force drug manufacturers to the negotiating table.

I do have one concern. Kapczynski and Kesselheim are right that, even if the federal government made a habit of infringing patents, drug companies would still retain incentives to innovate. After all, they’d be fully compensated for their risk-adjusted investments in R&D, at least assuming the courts get the valuation right.

But might the use of eminent domain affect which drugs go into development? Cures will become relatively less attractive if drug manufacturers know that the government will probably prevent them from earning outsize profits. At the same time, drugs to treat chronic conditions will become more appealing, even if their health benefits are modest, since eminent domain won’t be used to assure their availability.

I have no idea if this unfortunate incentive effect would be large, but it’s something to consider. Read the whole paper if you want to see where frustration over drug prices might one day take us.

@nicholas_bagley