• Should I say more about the Ryan-Rivlin plan?

    Don Taylor and I have kicked Howard Gleckman’s post back and forth for a day, with me feeling like I’ve got nothing to add, and him thinking he might comment. So far, neither of us has done anything. I’ve got a few minutes, so I’m jumping in.

    Gleckman points out that the Ryan-Rivlin plan for “voucherizing” Medicare bears a striking resemblance to the ACA’s health care insurance exchanges.

    For vouchers to work, insurance companies would have to sell coverage at an affordable price to all, regardless of health status. Seniors would need a way to shop for insurance. To keep premiums reasonably priced, consumers would have to be required, or at least very strongly nudged, to buy coverage before they got sick. Finally, since premiums would still be expensive for older buyers, the government would have to provide seniors with a significant subsidy to make the product affordable.

    As it happens, the first three elements are exactly the model of the ACA. The law includes insurance exchanges, a requirement that private insurers make coverage available to everyone regardless of health status, and the obligation that everyone have at least basic coverage. It even includes subsidies for some low-income buyers. Additional premium support for seniors would be the final piece of the puzzle.

    From a cost-control perspective, there’s a lot to like about the idea, provided it is augmented with a competitive bidding system to set voucher levels. Otherwise, in terms of cost control, what distinguishes it from Medicare Advantage, which is not a model of cost control we should want to broaden (more on this by Coulam, Feldman, and Dowd)? Also, FFS Medicare could participate in that bidding system and, in some rural areas, would outbid private plans due to the inability of private plans to establish any meaningful leverage through selective contracting in such areas.

    In a post in January I attempted to explain how competitive bidding really protects beneficiaries, not just taxpayers. In doing so, I contrasted it with “premium support” in general, which does not include all the protections of competitive bidding.

    One reason premium support frightens people is that no such proposal (or none I’ve seen) includes any plausible means to protect voucher levels from being adjusted upwards arbitrarily by Congress, just as they’ve raised Medicare Advantage payments in the past. This type of cost control is no more believable than that included in the ACA. It depends on the commitment of future Congresses.

    The second reason some fear a voucher program is that it could leave beneficiaries with far greater cost risk than they have today. They’d be on the hook for every dollar increase in coverage above voucher levels, even for a standardized, constant level of benefits. That’s very different than current Medicare. Since people are averse to change, especially at their expense, they’re not keen on this idea. […]

    Then there’s competitive bidding, which is different. The version I’ve written about

    could save 8% or about $50 billion per year (based on a 2010 Medicare cost estimate) through a competitive pricing system in which all plans, fee-for-service [FFS, the public option] included, offer bids for a standardized set of benefits and the government pays all plans based on the lowest of these cost estimates.

    That is, voucher levels are set by the market and at the lowest offered price. They’re large enough so that every beneficiary could enroll in some plan (for standard Medicare benefits) at no additional out-of-pocket cost. If a beneficiary prefers FFS Medicare, that’s an option, though perhaps at a higher premium. If a beneficiary wants additional coverage for enhanced benefits, the market could offer that, but at an additional cost to the beneficiary, not the government. Meanwhile, voucher levels are not set by a political process, but by the market. Thus, both of the concerns about premiums support plans I described above are addressed. Voucher levels are both sufficient to purchase coverage and are not set directly by Congress. Moreover, taxpayers pay no more than is necessary to provide the basic set of benefits.

    It is important to note that neither competitive bidding, nor any voucher program, can solve all our health care system problems. I made this point in a Kaiser Health News column last September.

    Competitive bidding can save money, but it is not a panacea for all of Medicare’s ills, and it doesn’t address every issue associated with the program. It cannot tell us what the standard, required set of benefits upon which plans bid ought to be — only that we need consensus on such a benefit. It would mean that the beneficiary cost of fee-for-service coverage, as well as private plan coverage, would vary across markets, a feature some might consider inequitable. It also cannot, by itself, change the growth rate of health care costs. For that, further reforms to how fee-for-service and Advantage plans pay for care would be required, as well as changes system-wide, well beyond Medicare. However, bidding would ensure that taxpayers get the best value per dollar within the framework of the program’s hybrid public and private structure.

    Should I say more about the Ryan-Rivlin plan? Perhaps you’ll agree, I’ve said plenty already.

    • The concern I have with a competitive bidding system is that instead of tinkering with voucher levels (or MA payments) we would tinker with what standard benefits mean. And not simply in a way that is driven by research, but rather driven by holding costs down.

      That is not to say I do not care about controlling costs, but I worry that our cost control would come on the backs of the same people who normally bear the burden of government cuts (poor) and how inequitable that would be.

      Striking a balance between what is affordable for taxpayers and what is equitable is not an easy task. But I worry that the same folks who offer up competitive bidding are generally the same folks who advocate for vouchers in Medicare (that over time would be inadequate to provide decent coverage for many Americans) and block grants for Medicaid (whose designs would clearly mean significant cuts in the program over time). Call me jaded or cynical, but I do not trust their motives.

      • @Justin Anderson – I understand your concern. I’m not one of those folks who offer up competitive bidding in conjunction with block grants for Medicaid. In fact, I can’t see why one would advocate one system for one set of people and not believe it was good for everyone else. In other words, why not have the exchange-like system for everyone. Throw in a public option (FFS Medicare) because it can get lower prices in some places. Set subsidies based on income and health conditions (risk adjust). Allow (or mandate) that everyone participate.

        Will benefits levels erode? Maybe. But I hope they do so based on evidence. Will we get a two-tired system? Yes we will! It exists today.

        I have not yet encountered any other way forward that takes full advantage of every tool at our disposal. Why should taxpayers pay more than they need to for basic benefits (as they do today for MA plans and as they likely will for exchange-based subsidies)? Without bidding, we’re throwing money away. For what?

    • My main fear is that we might exacerbate our present two-tiered system. I can imagine a time when, due to budget pressures, the definition of standard benefits becomes so narrow that a large number of people will not just want, but need, a more comprehensive plan. Given that most folks on Medicare do not have particularly large disposable incomes, that would mean an unacceptably large number of people without adequate insurance. And that has consequences not just for those with substandard insurance, but for providers who, in the absence of payment, would be forced to swallow losses or try to collect from their patients (which is not without cost).

      I’m also struck by what I’ve heard from a grad school friend who is now quite high at CMS with respect to how much cheaper we could do Part D without the private plans. This leads me to wonder if we’re not better off keeping the traditional Medicare model with some tweaking.

      • @Justin Anderson – We can do far, far better on Part D, even keeping the private plans. (I don’t even entertain the notion of jettisoning them. It just isn’t going to happen. Same goes for the rest of the system.)

        We do have to decide who gets access to what benefits. Seeming not to decide is still a decision. For example, right now, oddly, the decision isn’t even geographically consistent. Some have access to lavish benefits at very low cost, some don’t. It depends on where plans participate and what they’re able to negotiate with providers.

    • From conclusion of paper:

      “The frequency of diagnoses reported in claims data are routinely used in methods for risk adjustment in comparative effectiveness research,6​,7 the evaluation of readmissions following hospitalization,8​,9 and in paying insurance plans under the Medicare Advantage program.10​ If diagnosis is not solely an attribute of underlying disease burden, adjustments based on frequency of diagnosis may introduce bias into efforts to compare outcomes, pay for health care, and assess the extent of geographic variation in health care delivery.”

      While body of evidence still maturing, this paper, along with Song et al (NEJM 2010), make a compelling case for, as the investigaors cite: “increased intensity of observation.” Cute. I call it cooking the books, but OK.

      My sense, for the two nickels its worth, is that this phenomena is real–and have talked to Dartmouth folks offline, as well as others who research in this field, and they believe something is astray and data does not square.

      My question: a) how long until we say, “wait a minute,” and b), do the cries of “no fair” from those who are penalized hold water–when some coverage system goes live?

      How do you square that circle, mainly, propose a voucher-ized solution whose underpinnings are based on a shaky RA foundation?


      PS–incdentally, the threads and subjects are engaging the last few days. Engaging.

    • @Austin Frakt- Agreed on the wild inconsistencies in terms of access to benefits and care. Just not sure how we get there from here. My intellectual predisposition, generally speaking, is to blow up everything and begin anew. But my physician fiancee keeps telling me that will not work.

    • @Justin-You are both right.


    • who benefits from the ryan rvlin notion?

    • Oops. Late day reread.

      I find that multitasking on a on handheld detracts from my primitive writing skills. My earlier post above references a paper Austin blogged on earlier in the day:

      My quote is from the paper’s summary/conclusion and is germane to the risk adjustment overtures nested in the post above..