• Some notes on innovation in health care

    • Rightly or wrongly, the U.S. is often thought of as the global health care innovation leader, pioneering the best devices, drugs, and techniques, and home to the top medical schools.
    • Perhaps a less commonly held, but a fairly pervasive, belief is that regulation unduly stifles innovation in the health care sector, and no more so than with respect to health insurance. For obvious reasons, Obamacare has become the leading health care over-regulation exemplar, harming consumers with government’s heavy (far from invisible) hand.
    • Reihan Salam: “[Obamacare] limits innovation by insurers and providers that can help contain costs. […] The big problem with the Obamacare exchanges is that they don’t give insurers the option of offering consumers a wide range of products suited to their needs, and they don’t offer enough flexibility on pricing. […] Obamacare is narrowly constraining the kind of products available on the marketplaces.”
    • Epstein and Hyman: “PPACA’s fundamental design defect was to superimpose additional layers of regulation and subsidies on a system that was already top-heavy with both. These preexisting regulations and subsidies have already misaligned the incentives within the health care system. The next generation of rules will only compound the errors. In our view, the right approach to these problems is to promptly initiate a program of systematic deregulation that will introduce the choice and competition to which PPACA gives, at best, lip service.  […] All else being equal, the greater the level of market freedom, the higher the level of innovation and the wider the range of choices will be. When government regulators seek to place certain arrangements out of bounds, they restrict the scope of this inventive behavior. The fewer remaining options for potential trading partners means that the search for joint gains will be subject to constraints that produce two kinds of large social losses: (i) increased costs of public oversight, and (ii) inferiority of the private responses that are acceptable in light of that oversight.”
    • In a December 2013 NBER working paper, Jeffrey Clemens estimated the effect of insurance on medical innovation, finding that U.S. insurance expansions (including Medicare and Medicaid) “account for 25 percent of recent, worldwide medical-equipment patenting.” The vector for this relationship is the flow of well-insured patients through the hands and laboratories (e.g., hospitals) of skilled practitioners. Largely shielded from price — and in the context of no serious attempts to manage technology — the (insured) U.S. patient population is arguably the best substrate for innovation. The ACA’s coverage expansion is, therefore, a force for accelerating innovation. However, the law also includes some potential brakes: a tax on medical devices, Medicare payment reductions, movements away from fee-for-service, though that could “increase the rewards for innovation of the cost-conscious variety.”
    • In a December 2013 article in The Review of Economics and Statistics, Aamir Hashmi reexamines the relationship between competition and innovation in U.S. manufacturing. Interestingly, he finds a negative correlation. “Schumpeter (1950) avers that perfect competition is not the best market structure and that ‘the large-scale establishment or unit of control’ is ‘the most powerful engine’ of progress. Since then, many theoretical and empirical studies have explored the relationship between market structure and innovation. The Schumpeterian endogenous growth models pioneered by Aghion and Howitt (1992) formalize Schumpeter’s argument. Their original model (see Aghion & Howitt, 1992) predicts a negative monotone relationship between competition and innovation. The reason is that if innovation is driven by the expectation of higher profits, then any increase in competition (that lowers profits) will reduce innovation.” [Links added.]
    • What, then, are the relationships between competition and innovation in health care delivery and health insurance? It’s presumed by Salam, Epstein, and Hyman above to be the case that lighter regulation and more competition in health care will increase innovation. But is that so, or could the relationship be more like the one found by Hashmi in U.S. manufacturing? (I’m asking. I don’t know. Has it been directly studied? If not, could it? How?)
    • If competition and lighter regulation does increase innovation in health care, does it necessarily or only increase innovation that is of benefit to consumers? One can imagine all manner of innovations that are of benefit to insurers or providers, but not necessarily consumers (or all consumers). Medical underwriting is but one example.

    Comments on this post are open until one week from the date of this post. I’m seeking thoughtful, evidence-based responses and/or links to anything of relevance. Submissions will be moderated accordingly.


    • To the point that you (and Adrianna I think?) made some time ago about focusing only on insurance…

      I think that innovation — in the true sense of biomedical/biotech innovation — is more stifled by a highly risk-averse FDA than Obamacare. Just as one example, the lack of reciprocity with the EMEA is odd to say the least. Where Obamacare likely limits innovation is solely in terms of insurance products and the methods of care that those products can drive. I’m thinking primarily of fully capitated payment models like in Medicare Advantage. Interesting idea actually is to look at the type of plans prevalent in the exchanges — my feeling is that they will mostly be EPOs rather than managed care designs like HMOs.

      Lastly, the innovation and competition link. Probably depends on what industry you look at. I haven’t done this analysis, but I imagine if you look at the consumer tech industry (Apple, Samsung, Google to an extent etc.) you’d find that innovation and competition go hand in hand. One thing that differentiates biomedical innovation though is that you have so much of the basic research conducted by nonprofits, universities, and government. It’s also more reliant on basic research than the consumer tech industry (differentiating a product is also much easier in consumer tech – i.e. iphone vs samsung galaxy – than in medical devices or drugs – lipitor vs zocor vs any generic statin on the market).

      Just some thoughts.

      • Thanks. What reading would you recommend with respect to biomedical/biotech innovation and stifling thereof?

        • There’s not much in the way of peer-reviewed lit specifically addressing stifling regulations (that I’m aware of, anyway).

          But I’ll take this opportunity to plug one of our papers (by Avik Roy): http://www.manhattan-institute.org/html/fda_05.htm

          He quotes some good numbers from Dimasi et al on probability of approval at various stages of development. Additionally, some good numbers of increase in clinical trial burdens & costs over the years.

          An NBER paper by Philipson & Sun: http://www.nber.org/papers/w13561.pdf?new_window=1

          They estimate social welfare loss from inefficient FDA regulation.

          More generally, there’s a pretty well-documented “Eroom’s Law” (reverse of Moore’s law): http://www.nature.com/nrd/journal/v11/n3/full/nrd3681.html (gated but one of the really interesting sets of graphs is here: http://www.nature.com/nrd/journal/v11/n3/fig_tab/nrd3681_F1.html)

          Of course, it may be that all of the “low-hanging fruit” were picked already and the decline in productivity says more about pharma than it does the FDA. Recent FDA regulatory actions though (dealing with autologous stem cell treatments and genetic testing firms [23andMe]) are pretty nasty, it seems to me.

          Also, if you have the time, check out the book “The Cure in the Code” by Peter Huber. Very relevant to the topic.

    • Just an obvious if speculative theory: US manufacturing can move overseas and maybe US health care cannot – at least not to the same extent. If the Aamir Hashmi study were to include all manufactured products consumed in the US (not just those manufactured in the US), would it show more innovation with competition a la Schumpeter, or would the findings persist?

    • I would also reiterate that the effect of competition on innovation may depend on which industry you look at. Just from casual observation, the food service industry and the farming industry are hot beds of innovation, and they are textbook examples of highly competitive industries. Relating the issue to hospitals, this 1998 paper (http://www.econ.upf.edu/~puig/publicacions/Any1998/RIO1998.pdf) suggest that technical / productive efficiency (as opposed to allocative efficiency) increases with competition in the hospital market. I have not had a chance to read the paper in great detail, but it may be a good starting point.

    • Austin,

      What do you think about the work of Clayton Christensen in “The Innovator’s Prescription” on the role of innovation in health care?

      They seem to think that a certain level of integration is necessary to implement progressive changes, but there is certainly an upper-limit to this vertical integration in terms of anti-trust concerns.

      Would like to know your thoughts.

    • Innovation in payment policy has largely been pioneered by Medicare (think DRGs, RVUs, rural programs, etc.). CMS houses a large research/demonstration staff and supports research in both academia and in private R&D organizations. Rather than investing in R&D themselves, private payers largely adopt successful efforts started at CMS. Delivery system reforms – usually pivoting off these payment reforms – often come from the private sector. But CMS often plays a seminal role in delivery reform, as well.

      Medicare’s position as a dominant payer clearly plays a role here. It’s in the government’s interest to participate in payment/delivery R&D. The question is why the private sector isn’t devoting equal resources to the effort. Is it more cost-effective to let Medicare be the pioneer? If so, doesn’t that show that private payers are largely satisfied with the quality and direction of the fed’s lead? And if they’re not satisfied, what’s stopping them from getting into the game? Is it excessive regulation (which CMS must follow, as well) or something else?

      • To be fair to private insurers, Medicare is a very large organization, run by its central bureaucracy and governed by statue and the regulations that CMS has promulgated. Every time CMS posts a new regulation it has to solicit public comments and then respond to them. This will probably make it slower to adopt some types of delivery system innovations.

        For example, the Wikipedia article on medical homes cites some examples where commercial or non-profit insurers had medical home provider contracts or pilot programs in place between 2002 and 2008. I believe Medicare had to wait for Congressional authority before it could implement medical home/ACO contracting.


        Similar story with bundled payments. CMS expressed interest and Medicare started a demonstration, but they are only now using bundled payments on a wider scale. Geisinger (a nonprofit integrated health system/insurer) had bundled payment contracts with providers starting in 2006.


        Note that I am a single payer fan. I am saying that Medicare may simply not be as able to adopt payment methods that support delivery system reform faster than the private sector.

    • Hi,

      Informative post — thanks. One thing: you wrote, “However, the law also includes some potential breaks”. In context, you clearly meant “brakes.”

    • There are different types of innovation. Technical innovation (drugs, equipment) probably needs a substantial profit motive to thrive. One of the benefits of the excesses of the US system is the relative health of this type of innovation.

      Innovation in delivery mechanisms that can decrease cost (Bundled payments, global payments, etc to replace FFS) requires some types of intervention by the government since its such a large part of the market. The BPCI experiments which are a part of ACA are a good start.