When Gov. Peter Shumlin launched his single payer health care initiative in 2011, the effort promised to be the policy equivalent of a moon mission, one of the most far-reaching and complex reform projects ever undertaken at the state level. It would be very hard, yet still doable. And critically important, given that the health care system both in Vermont and in the U.S. is simply not financially sustainable.
Last Friday, on the project’s third birthday, Shumlin disclosed on VPR’s Vermont Edition that he was making a midcourse correction, putting off legislative consideration of the financing of the project from early 2015 to at least 2016. We are going to get it right, he said, rather than get it on a specific schedule.
The governor’s comments came at a time when the political atmosphere surrounding the project has soured dramatically. Vermont Health Connect, the federally-financed exchange, has been a mess since it began operations last October. The Legislature has been increasingly skeptical, urging that the administration introduce its “plan” for reform, particularly the financing piece.
The Republican opposition, ineffectual in electoral terms, has been hammering away at the whole idea, claiming that the governor has a secret plan and is trying to hoodwink the public. The left, meanwhile, has been speculating about whether Shumlin was really committed to single payer, and whether he was preparing to pull the plug.
There was very little public notice of Shumlin’s move over the weekend, but the reaction in the health policy community and the Statehouse is likely to run along similar lines. Many legislators, including Democrats, will be unhappy that they won’t get to see a plan in the next few weeks. The left will think that Shumlin is in fact pulling the plug; the Republicans will say we told you so. One particularly shrewd observer of the reform scene said Monday that the “left, right and center” were all unhappy with the governor’s announcement.
Well, the left, right and center are just wrong. The governor did what he absolutely had to do. If he had persisted in forcing out a financing plan in the absence of a detailed implementation plan and assurances on cost containment, the whole effort would have very likely cratered. Now, ironically, despite the financing delay, Shumlin has put single payer back on track.
The governor had to act because the schedule was simply too cramped by at least a year. Possibly more. That conclusion rests on the two fundamental propositions that support any reform effort.
The first is that you can’t reform the health care delivery system unless you can control costs in the system. Health care inflation that substantially exceeds the public’s ability to pay will destroy any delivery system, just the way it is destroying the one we have now.
The second is that you can’t control costs without reorganizing the system, changing both the way doctors and hospitals relate to one another. They have to begin to cooperate rather than competing with one another. And they have to change the way they get paid. That task is the essence of any reform.
Both of these propositions are built into the very fabric of Act 48, the legal underpinning for the Vermont reform effort. The key language is contained in a series of criteria that have to be met before any single payer plan can be implemented. These criteria, known as the Mullin triggers, named after Rutland Republican Sen. Kevin Mullin, add up to a requirement that the administration demonstrate that it can control costs before it implements any plan.
The problem Shumlin faced was that there is no way to demonstrate by early in 2015 that you can control the costs in the system – or even to know if cost control would be feasible. If that’s true, and no one has made a straight-faced claim that it isn’t, then asking the Legislature to vote for a huge financing bill makes no sense at all.
Act 48 doesn’t actually say that evidence for cost control has to precede a vote on financing; it says only that the Mullin triggers have to be pulled before a reform system is actually installed. But in practical political terms, the likelihood of the Legislature placing the financial integrity of the state at risk without such assurances is close to zero.
Certainly House Speaker Shap Smith believes that, and no heavy weight bill is going to pass without his enthusiastic support.
The Shumlin team debated how to handle this dilemma for several weeks, before the governor pivoted last Friday.
What made it so hard is that Shumlin badly wants to get the federal waivers he needs to install a Vermont-specific program before President Obama leaves office in 2017.
Obama’s health care bureaucrats are particularly fond of Vermont, given the lack of political and public enthusiasm for Obamacare in the rest of the country. Here, the governor is a full-throated supporter and the Legislature is completely dominated by Democrats. We wouldn’t have Act 48 if it weren't.
In January of 2017, however, Obama will leave office, and there is no way to tell how something as idiosyncratic as the Vermont initiative will fare with a new administration. It obviously would be helpful to Shumlin if the new President is a Democrat, but even then the reform atmosphere might not be as favorable. A Republican administration would almost certainly be hostile.
So Shumlin hung on to the schedule he announced last year – with a financial plan floated this year and then a vote on the final plan in 2015 – as long as he could. As of 1 p.m. Friday, that schedule is gone.
Given the toxic atmosphere for reform over the last six months and the fact that the governor had to pull back on his announced schedule, why would anybody argue that the program is now back on track, that publicly financed health care is still possible for Vermont?
Nascent Payment Reform
The reason is that the effort to shift the way doctors and hospitals relate to one another and how they get paid is well launched and forging steadily ahead. The road ahead is still long, but the reorganization efforts are picking up steam.
The vehicle for that transformation is an accountable care organization (ACO). This model, which was established nationally in the Obamacare legislation, establishes a clear process to shift the relationship among doctors and hospitals in a given region from competition to cooperation. That in turn will provide a platform for changing the way providers are paid from pure fee-for-service, which encourages overuse, to a reimbursement system that sets up incentives for doctors to focus on the care of populations at an affordable cost. The model is the first credible cost containment vehicle in the modern era.
A few ACOs exist now in the U.S., but they haven’t been fully proven yet. The concept is quite far advanced in Vermont, however, and it’s worth looking at more closely, both to get a sense of what real health care reform could look like and to see why the original trajectory of the Shumlin project was too cramped by at least a year.
The most important ACO now operating in Vermont is called OneCare, which has been established mainly through a joint effort by Fletcher Allen Health Care and the Dartmouth-Hitchcock system in Hanover, N.H. Fletcher Allen delivers nearly half of all the acute health care in Vermont, and Dartmouth-Hitchcock provides considerable primary care to Vermonters and most of the specialty care to residents of eastern Vermont.
OneCare also includes all of the 13 community care hospitals in Vermont; the state’s hospitals employ some 65 to 70 percent of the doctors in the state. So OneCare now includes the great majority of the health care assets in the state. There are two other ACOs in Vermont, but they are smaller and not as far advanced. Moreover, additional ACOs may be brought on line.
In 2013, OneCare embarked on a six-year program to rework the Vermont system. The arrangement was paid for and will be overseen by the federal Centers for Medicare and Medicaid Services, the agency responsible for the two health care programs. Only the Medicare population will be included in the first phase, but the Medicaid population will come along behind them. Commercially financed health services will also eventually fall under this umbrella.
During the first three-year period, which will run to 2016, OneCare management will begin to shift the payment for medical services from pure fee-for-service to a hybrid system called shared savings.
Under shared savings, total costs for all patients who get their primary care from a provider in the OneCare network will be tracked by OneCare and the federal government. If OneCare providers actually spend less than predicted for these patients – and if they meet the quality-of-care requirements – the network gets to share in the savings.
That is described in the health care world as one-sided risk, a sort of “heads I win, tails you lose” situation. One-sided risk will prevail for the first three years of the program, in a sort of training wheels period.
In 2016, the training wheels come off and the whole system shifts to two-sided risk, in which the providers can keep some of the savings below the target if there are any, but if they exceed the target some of the overage comes out of their pockets.
That sort of risk – real risk, in other words – has never existed to any significant degree in American medicine. But the consensus in the health policy community is that without that risk sharing, costs in the system cannot be controlled.
The problem that has bedeviled the Shumlin team is that nobody will be able to see how this methodology works by 2015. According to Todd Moore, the CEO of OneCare, his organization and CMS will begin capturing the first data on shared savings performances sometime this coming summer.
The first year’s results therefore would not be available until the early fall of 2015, long after the Legislature had gone home for that year. And even then, you would only have a single data point – nowhere near enough to make any critical judgments about the ability of this system to control costs.
The far more dependable lever – two sided risk – will not produce even early results until at least the early part of 2016.
Vision Meets Reality
These factors drove the need to accommodate the original Shumlin schedule to the real world. No one – so far, at least – has been able to show how it would be possible to meet the criteria for cost containment before 2016, and even then both Shumlin and the Legislature may feel they need even more data than will be available then.
Politics is a blood sport, of course, and Shumlin will have to face cynicism and skepticism from every quarter. But the fact is that he really had no choice at all but to accommodate his schedule to what is taking place on the ground.
It is also important to understand, however, that the midcourse correction doesn’t necessarily mean that Shumlin can’t launch most or conceivably all of his entire final product by 2017.
There are three basic elements to the reform package. One of those is restructuring, which is well underway already. The second is coverage – the extent to which everyone in the state is assured of access to medical care. The third is financing.
It is at least possible that the cost containment structure could be credible enough by 2016 for the Legislature to be willing to pass the financing necessary to carry out the final extension of coverage to everyone in the state in 2017. What is more likely is that the coverage could be extended in phases, although there is no plan for that yet. Which is another reason why Shumlin is now saying he needs more time.
At present, however, the general environment remains quite hostile. The governor may be able to take some comfort in the thought that we have spent some 40 years building the tangled mess that we have now and that we simply can’t get the whole thing rebuilt on the schedule he originally set.