by Hamilton E. Davis
The single most important issue in the Vermont health care reform initiative is to protect and build on the University of Vermont Health Network, and particularly its flagship Medical Center Hospital in Burlington. The Medical Center delivers half of all the acute care in the state—that percentage rises to 60 for the Network when you include the smaller hospitals in Berlin and Middlebury. The quality of their care, measured by outsiders, is far superior to the rest of the Vermont hospitals., while its costs are a full third lower than the rest of the country.
And it isn’t just the volumes of care that sets the Medical Center apart from the rest of the facilities in the state. The Medical Center is the only hospital in Vermont that can deliver tertiary care, like open heart procedures, therapies for the most complex cancers, and Level One trauma care for accident victims. Moreover, in concert with the UVM College of Medicine, its doctors perform important medical research, and train medical students, many of whom remain in state. Finally, aside from the flow of medical benefits, the UVM Health Network, which treats more than a million patients a year, is the single most important component of the Vermont economy.
Yet, the Green Mountain Care Board is doing its best to starve the UVM Network of the revenues it needs to sustain its operations. Those depredations also damage the Network’s reputation and its morale. Plus, the Board’s parallel campaign against OneCare Vermont exacerbates the pressures on the UVM Medical Center, which was the intent of UVM opponents over the last seven years. (I’ll assess the OneCare issue in my next article).
What Evidence Supports that Indictment, and Why is it Happening?
Taking the Why first. The Board’s actions for the most part do not derive from mindless animus toward the Burlington hospital. They arise rather from the huge political pressures that attend the reform process itself, which are generated by the way money flows through the system.
About half of all acute care is paid for by federal and state taxpayers in the form of Medicare and Medicaid, the other half by the private sector in the form of insurance premiums and self-insured employers. Medicare and Medicaid, however, don’t pay the full cost of the care they purchase. So, the shortfall gets added to the bills paid by the private sector—the dreaded cost shift.
A big chunk of the cost shift lands on Vermont Blue Cross because its coverage is so widespread, and an equally big chunk of that goes to the UVM Network because it so dominant: The UVM Medical Center is five times as big as the only sizeable community hospital, Rutland, and more than 20 times the size of the other smaller facilities. The 11 smaller community hospitals waste a lot of money, which I showed in my last series (First, second, third, and fourth), but the Board members are petrified of political blowback if they even mention it. (The whole Blue Cross thing is an issue unto itself, which I’ll asses in a later article.)
The same dynamic terrifies the management of the small community hospitals. They know they can’t begin to perform at the level of the UVM Medical Center on either cost or quality, and if they are forced to try, the odds are they will have to go out of business or step down their operations to clinics that focus mainly on emergency medicine and primary care. No more $400,000 to $600,000 salaries for CEOs of 25-bed hospitals or $500,000 to $1,000,000 incomes for surgeons who want to do too-small numbers of complex surgeries in a cornfield. Those pressures are real, but the Board’s job, for which the members are paid handsomely (six figures for part time work for all but the chair, who gets paid as much as the governor) is to both control costs and make sure that essential services remain available. Spending all their regulatory muscle on hacking away at its most important, least expensive and highest quality hospital is reprehensible.
That’s the Why.
Okay, how about the evidence of dismal performance by the Green Mountain Care Board? Start with the general proposition that an academic medical center like the UVM hospital in Burlington needs a margin of 4.0 percent per year. It also has to recover its expenses from delivering care. If the hospital had been allowed to recover its expenses, plus the four percent margin, the hospital would have retained $308 million over the five-year period from 2018 to 2022. What they got was $87 million, a shortfall of $221 million.
A full $108 million of the shortfall stems from the refusal of the Board to allow UVMMC to recover the costs of the inflation that has ravaged the economy over the past couple of years. Neither the medical system, nor any other sector of the national economy, can escape the consequences of inflation rates reaching 7.0 percent in the Covid era, the highest in 40 years. The remaining $113 million of the shortfall arose because the Covid relief funds did not fully cover the losses from the effects of the virus.
The effect on the UVM Medical Center can be seen in one of the most important financial metrics, the Days-Cash-on-Hand available to company management. That figure is critical to the hospital’s ability to raise money in the financial markets, which make lending rates based on recommendations from the three major bond rating agencies. They look like this:
As you can see, the UVM figures are in no-man’s land for both Moodys and Fitch, and they edge into the barely acceptable range only for S&P in 2018 (205) and 2021 (185). No financially competent person can look at that picture and see anything but the edge of a financial cliff. There is no real way to estimate the monetary cost of a reduction in UVMMC’s bond rating, but it will obviously be considerable, especially given the current lineup of new equipment and structural needs, and the need on a somewhat longer reach for a new bed tower.
The consultant, Berkeley Research Group, stated in 2021 that UVMMC needs at least another 60 beds by 2026; and that didn’t take into account any movement of complex care out of small regional hospitals and into the academic medical center, which has to be done. Call it 100 new beds, at a minimum.
I’ve been highly critical of the Board’s management of this issue, and it’s fair to ask whether the Board shouldn’t be rigorous in examining the performance of 60 percent of regulated system. The obvious answer to that is, Yes, but it is also fair to test the Board’s performance in that regard. Have the members looked fairly at the UVMMC performance? Let’s look at that.
A good place to start is the performance of UVMMC and its smaller partners compared to the national spending rates. Everybody brags about that, including the UVM Network. They publish figures from the Dartmouth Health Atlas showing that Vermont has the lowest Medicare costs in the country. Those figures are accurate, but misleading.
Vermont’s favorable cost profile is generated mostly by the UVM health network. If you compare the UVM Network to the national rates, the cost profile looks considerably better; and if you stack the 11 non-UVM hospitals against the country the Vermont community group looks far more ordinary. Those spending profiles are show in the graph below.
A question worth asking is the amount of money Vermonters would save if the community hospitals operated at the level of cost and quality as the UVM Network. You can visualize the potential in the graph below.
The Green bar shows the 2018 Medicare cost performance by the UVM Network compared to performance of the 11 non-UVM facilities in the state. Note the extreme outliers, all over $9,000 per Medicare recipient—Rutland Regional Medical Center, Southwestern Vermont Medical Center in Bennington, Gifford in Randolph. And Grace Cottage in Townshend, which shouldn’t be a full-service hospital at all (in some years, its average daily census is one) and Mt. Ascutney in Windsor, which functions essentially as a rehab hospital for Dartmouth-Hitchcock.
We got at the actual dollar amount of saving by multiplying the population in each hospital service area by the per capita rate achieved at the Medical Center in Burlington. We then determined the cost in each service area by multiplying the actual per capita rate by population of each hospital. We then subtracted the efficient level cost form the actual cost.
The weighted averages capture the savings:
$134 million per year, as measured from the most reliable data set in the country—The Dartmouth Health Atlas.
Another look at the same issue is contained in a state-by-state look at per capita in the Dartmouth Health Atlas. For the year 2018 it shows that Vermont has the lowest costs ($8,010) for its Medicare population in the country: Only a few states out of the 50 are even close: Hawaii ($8,090), Alaska ($8,251), Oregon ($8,614), and New Mexico ($8,905). The weighted average for the entire United States is $10,779. Most of the states range from $9,000 to $10,000, or higher. You can see the whole U.S. layout here
Of course, the academic medical center is expected to operate at a high level, but what has so far gone unremarked is that the UVM Network has demonstrated that it can wring high performance out of its small hospital units as well as the flagship. Take a look at the data. The first graph shows the Network’s Central Vermont Medical Center stacked up against the other four med sized facilities. Note the size of the overspending gap at Rutland, the second biggest hospital in the state, and Southwestern Medical Center in Bennington.
The same kind of spending gap can be seen in the graph below that shows the performance of the UVM Network’s Critical Access Hospital against Vermont other 25-bed hospitals.
There is no evidence that the members of the Green Mountain Care Board have the faintest notion of how their decisions have run counter to the needs of the Vermont hospital system and the need of Vermont for national class health care in the state’s center of population, Chittenden County.
The Board’s management of OneCare Vermont, the statewide Accountable Care Organization has been even worse. We’ll look at that in tomorrow’s article.