The ACO is Key to Health Care Reform: Part One

by Hamilton E. Davis

    In my last column I wrote about the two-front health reform campaign, focusing on the cost containment issue. The second front went away last December when Governor Peter Shumlin decided he would not pursue the funding of health care in Vermont. Cost containment, however, is not just alive, but thriving.
    Rendering the cost of the health care system in Vermont financially sustainable also has two fronts. The first is the purview of the Green Mountain Care Board, the five-member body tasked by the Legislature with regulating and reorganizing the doctor-hospital system in the state; many health care watchers are quite familiar with the Board because it has been regulating hospital budgets in the state for the last four years.
    The second major element in cost containment is called an ACO, which is—what? In many ways, an ACO is the mystery meat of health care reform. It’s fair to say that hardly anyone really understands the ACO idea, a situation that is exacerbated by the fact that many of the players like it that way because the ACO issue subsumes some of the most difficult political and financial tensions inherent in the reform project.
    It may seem weird, but I think that the best way to get at the essence of the ACO conundrum is to head to the nearest lumberyard. Let’s say that you want to build a new one-car garage and you have worked out on the back of an envelope the number of two-by-fours you’ll need. Say, 60 eight-foot two-bys. So, what will it cost you?
    Off to the first lumber yard, the price per two-by is $2.50. Try another one, $4.00. A short drive away, $5.00. One more, $6.00. Back to the back of the envelope. Yard one total is $150. Yard two is $240. Yard Three is $300. Yard Four is $360. Given that all the pieces are cheap softwood, this is a pretty easy decision. Yard one is the best buy, so you order the wood there.
    Now comes the delivery truck and it drops off 200 two-bys. And the driver hands you a bill for $500, which you are obligated to pay.
    Oh, come on. That would never happen. The lumber yard would be out of business in a week.
    Okay, okay. It wouldn’t happen with a lumberyard. And it wouldn’t happen buying a television set, or carrots at the farmer’s market. But it happens every day in the health care delivery system. And until the delivery system is reorganized, health care costs in Vermont, and in the United States, will never be sustainable year over year.
    Is health care financing really that strange and irrational? And if so, why? The first thing to understand is that the evidence for the dynamic I’ve laid out here is overwhelming. There is little to no opposition in the policy community that the financing system has to be changed because it sets up very powerful incentives to do more than patients need. Which is why the ACO machinery was built into Obamacare: it is considered to be the only way to get costs fully under control.
    To look at the “how” of it, we’ll go back to the lumberyard analogy. In a market economy, buying two-by-fours, or t.v. sets, or carrots, you decide what you want; that’s the Demand.
    The lumberyard or the tv manufacturer or the farmer produces stuff you want.  That’s the Supply. The interaction between the seller and the buyer balances the supply and demand, which is the definition of a market. And a real market is awesomely efficient.
    In health care, by contrast, doctors tell you what you need and then they supply it. Your only role is to pay for it. And the more they supply the more you have to pay.
    In the health policy biz, they call this “supplier-induced demand.” The payment mechanism that underpins it is called fee-for-service. If a health care provider does something, he gets paid for it. If he doesn’t, he doesn’t get paid. A particularly insidious characteristic of the dynamic described here is that competition in such a system increases costs, not decreases them. That “unmarket” is the engine that has driven health care costs in the United States from roughly six percent of our total national output in the mid-1960s to nearly 20 percent today. In Vermont it is 20 percent.

The theory of ACOs

    While there has been a consensus on the financing dynamic for decades, finding a way to fix it has been fiendishly difficult. Beginning as far back as the early 1970s, the primary impulse has been regulation—just have government order providers to slow down the rate of inflation. Governments, both state and federal, have done their best by simply refusing to pay the full amount of the bills the providers presented for Medicare and Medicaid services. That hasn’t worked: the providers just shifted the shortages to private sector insurance premiums.
     A second impulse was to set rates for specific services; and states tried that in all sorts of ways. Rate setting failed because of the dynamic illustrated by the lumberyard. Fix the unit cost for, say, an MRI, and the number of MRIs goes way up.
    The poster child for this syndrome is Maryland, where state government moved many years ago to pay all hospitals the same amount for the same services. They never controlled volume, though, and the result is that Maryland has one of the highest cost delivery systems in the country.
    It is important to note that this is not some kind of heinous conspiracy. It is simply the way that the American health care system developed in the modern era. Doctors always took responsibility for their patients, but they almost never took responsibility for what other doctors did, and they wanted nothing to do with trying to manage costs. In the 1970s and 1980s, it was common to hear doctors say that it was unethical, even immoral, to even consider costs. The only thing to be considered, the argument ran, was that whatever the patient needed he should get.
    A potential solution to the problem was designed into the Affordable Care Act, Obamacare, in what the designers called an Accountable Care Organization, an ACO in the current argot. It seems to me that the ACO idea is enveloped in a fog of misunderstanding, fear, political maneuvering and general angst.
    In essence, however, it seems like a plausible and even compelling way forward. At the least, there is nothing on the policy horizon that looks anywhere near as good. The idea works like this:
    An ACO in concept is a grouping of doctors and hospitals that can deliver the full continuum of acute care to a cohort of patients for a fixed, per capita price. Some caveats: We are only talking about acute care here. Long term care, home health and other essentially social services, while critical overall, are not essential in the ACO universe. The continuum starts with primary care delivered by doctors and nurse practitioners and ranges up from there to stand-alone specialists and small and medium sized hospitals and finally to the very sophisticated care that gets delivered in tertiary medical centers.
    The care within such a structure needs to be truly integrated. All of the elements within it are focused on a single task—to treat every patient in the most effective way, both in terms of medical quality and its attendant costs. The concept is similar to that of a complex technical manufacturer, like a car or a computer company, where there are thousands of individuals with a wide variety of occupational skills whose output is expressed in a single product that satisfies the customer, and is affordable.
    Those can exist in medicine now, but they require that all the assets be owned by a single entity. The ACO law makes it possible to form an integrated system without full ownership by exempting the ACO from such federal statutes as anti-trust prohibitions. Readers who have got this far may begin to raise objections to the integration idea, and I’ll deal with those later; for now, it is important to see how the concept works in theory.
    Once you have a fully integrated ACO, you can participate in risk contracts, the kind of arrangement that permeates our economy. If you want to get a contractor to build your new garage, you get bids for the work and you sign a contract for the work at a fixed price. If the contractor gets the work done for the price, he makes his profit. If he gets it done for less, he makes more profit. If the work costs him more than he estimated the difference comes out of his own hide. You can also get the work done for a so-called “time and materials” price—the contractor takes no risk at all, in which case you would probably get a pretty expensive garage. In health care, the thing we know with the most painful kind of certainty is that we’re getting ferociously expensive health care in our current time-and-materials system.
    Here is what a fully functional ACO would look like:
    Its membership would have to be able to deliver the full complement of acute medical care to the patients covered by the contract. The whole process starts with primary care. That’s not only true in the medical sense—everybody’s first contact should be with a primary care provider, either a physician or a nurse practitioner. It is also a key element in the way that the Affordable Care Act was written.
    The only way the patients in the contract grouping can be included in an ACO is if they enter through the primary care portal. They thereby become an “attributed life.” A patient can’t be considered a member of the ACO if he or she goes directly to a hospital or a specialist physician. So, the foundation of any ACO has to be primary care doctors.
    Once an ACO has its primary care doctor base, it needs to build in the necessary services all the way up the acute care severity ladder, from free-standing specialists to small hospitals to major medical centers. The ACO doesn’t have to have the fullest possible range of services, but enough to make risk contract workable. For example, if a Vermonter needs a heart transplant, he can’t get it here. He’ll have to go to a national center, say, Pittsburgh. But there are only a few such cases.
    Especially in Vermont, it is critical to understand that you can’t have a fully elaborated ACO, one that can take financial risk, without having the ability to deliver the full range of services. That is critical in Vermont because there are only two sources of tertiary care—UVM and Dartmouth—and they are both in OneCare. In an urban area like Boston, for example, you could have several. There are four medical schools, Harvard, Boston University, Tufts, and UMass-Worcester along with four of the best tertiary centers in the world—Mass General, Brigham and Women’s, Beth Israel and UMass-Worcester, along with a ridiculous volume of lesser facilities. Also primary care doctors on every block along with a large population.
    Vermont is nothing like that. There is only one source of tertiary care, and that belongs to OneCare.
    In any event, that is a look at the ACO component of the ideal world I set out in last column. Tomorrow I’ll look at what is actually happening on the ACO front in Vermont now.