by Hamilton E. Davis

Note to my readers: the following text is a model showing how OneCare Vermont, the state’s Accountable Care Organization, will pay primary care doctors who participate in the expanded reform effort that debuted at the first of this year. The purpose of the model is to underpin two articles that are not yet written—one on testimony before the Green Mountain Care Board last week by a group of primary care doctors, the other on the issue of how the Green Mountain Care Board is managing the reform effort. I am putting up the model by itself because some of the questions it addresses are already under discussion by the House Health Care Committee. The two fully elaborated articles will be forthcoming.

The Assumptions

   The following is a model showing the potential payments to a primary physician who attributes his or her patients to the ACO, OneCare Vermont. The size of the panel is taken from the research done by Dr. Deb Richter in support of her proposal for universal primary care—1200 patients. The actual panel sizes could range from 600 to 700 on the low end to 2,000 on the high end. The revenue in the Richter plan to support that panel would be $500 per patient, or a total of $600,000 per practice per year. The Richter number does not include payments for mental health and substance abuse treatment. Nearly all the numbers in the model are available from public sources; I have indicated where I do not have a hard number.
   A second assumption is that the model practitioner would attribute half the panel—600 patients—to OneCare. That percentage could vary widely. Those patients would be arrayed across three payers—Medicare, Medicaid and commercial. The spread could be highly variable based on payer-mix in the provider’s service area; and it could vary also by specialty—one practice could have large numbers of Medicare patients, whereas a primary care pediatrician probably would have no Medicare patients. The benefit package was designed by OneCare as part of its budget submission to the Green Mountain Care Board; the Board approved the budget in December. The model numbers would apply to the 2018 calendar year (with one exception noted below).
   A third assumption is that a primary care doc working full time would take home $180,000 per year, That number is a rough median from Web data on primary care salaries in Vermont. The $180,000 number would probably be high for a rural doc with a lot of Medicaid patients; some hospital-based providers might be higher. The remaining $420,000 to a single-doc practice would pay for infrastructure—nurses, clerks, other expenses.

The Compensation Plan

   Step One: OneCare will pay an administrative fee of $3.25 per member per month to primary care doctors attributing patients to the ACO. In the model that is 600 patients. The $3.25 rolls up to $39 per patient per year; for our model practice the total per year would be $23,400. The purpose of that is to defray the cost of reporting necessary data to OneCare.
   Step Two: OneCare will pay primary care doctors a premium of $15 per member per month for each medically complex patient in those practices. That would roll up to $180 per patient per year. OneCare will determine the eligibility of a given patient using an algorithm designed by the Johns Hopkins academic medical center in Baltimore. There is no way to determine how many such patients a given doctor would have. But national data shows that the medically complex proportion of the Medicare and Medicaid patients of an average primary care practice runs to about 16 percent. For Blue Cross patients, the percentage would be just three percent.
   The payer mix for a given practice can vary widely, of course, but the generally accepted numbers show that about half of acute care is paid for by federal and state governments (Medicare and Medicaid) while the other half is paid for mostly by commercial insurers.
   To get a model figure for the complexity payment, I’ve cut the 600 total attributed patients to 300 (using the 50 percent government parameter). Then, based the national averages, I’ve taken 16 percent of the 300, which is 48 patients. At $180 per patient, our model physician would get an additional $8,640. The private insurance cohort of 300, mostly Blue Cross, would yield just three percent, or nine patients, At $180 per year, Blue Cross patients would yield $1,620. The total complexity of care component would amount to $10,260.
   Step Three: OneCare will operate a Value Based Incentive Fund that will pay a premium based the on the quality of the care delivered by the primary care physician. There is no good way yet to estimate what this program might produce for a primary care doctor. OneCare will use the experience obtained in 2017, but that won’t be known until mid summer of 2018. Based on some conversations with people in the industry I have used a placeholder of $8,000. It is a very rough estimate, but considered conservative.
   Step Four: OneCare will also pay a premium of $10 per member per month, or $120 per year, for those primary care docs who are designated by their complex patients as the “Lead” physician in the management of their care across the continuum of that care. This number is likely to be small for a given doc; there is no way yet to estimate what the experience will be in the system. For a panel of 1200, with 600 of those attributed to the ACO, I’ve used a guess/place holder of 20, which would be an additional $2,400 per year to the physician.
   Step five: One of the most financially important components to membership in OneCare is the federal government program to pay primary care doctors in such organizations five percent more for the care they deliver to Medicare patients. The model uses the Richter proposal. If we cut the panel to 600 and then assume that 200 of those are Medicare and multiply by five percent of that we get $5,000. That benefit would not be available until 2020. A positive wrinkle for primary care docs in this category is that Medicare will pay the five percent bump for all the Medicare patients in OneCare participating practice, even those whose lives are not attributed. An example of such an increment would be Medicare patients enrolled in the Medicare Advantage program run by United Health Care; care for those patients is paid for by United, but Medicare would still pay the five percent bump. For the model, that would double the bump category to $10,000. I have limited this category to $5,000, however, because it won’t take effect in 2018; another reason is to render the whole exercise more conservative.

Total Financial Benefits to Primary Care

Administrative fees    $23,400
Complex care           $10,260
Medicare bump       $5,000 (not available until 2019)
Quality incentive        $8,000 est.
“Lead care”                $2,400 (guess/placeholder)

Total $49,060

Indirect Benefits

In the planning phase of reform, primary care doctors in Vermont have made it clear that what they want most, in addition to fair financial reimbursement, is a reduction in the administrative burdens they must bear in the process of delivering care to their patients. That theme has led OneCare, the Green Mountain Care Board, federal Medicare officials, and Vermont’s Medicaid agency to develop a series of steps to ease the administrative burden:

In the first step of the execution phase of reform, which debuted last February, OneCare and the Department of Vermont Health Access (the state’s Medicaid Agency) negotiated a partial waiver of so-called Prior Authorization requirement in the reimbursement system. “Prior Auth” forces docs to get permission from somebody outside the delivery system to deliver care like X-rays and MRIs. There is a long way to go on Prior Auth—it is still required on drug prescriptions, for example--but the reform apparatus has made a start.
The Vermont health care reform team—The Green Mountain Care Board, the Governor and the Agency of Human Services-- persuaded federal Medicare officials to authorize the expenditure of $7.5 million for the continuation of the so-called Blueprint program in 2018. Blueprint money basically enables primary care doctors to address the social determinants of care. The extra bump was limited to Vermont because the feds are so hopeful that the state can lead the way to capitated financing for Medicare. Those officials believe that costs in the Medicare program cannot be controlled in the current fee-for-service reimbursement system. Federal Blueprint money has been eliminated in the rest of the country.
The U.S. Congress in 2015 passed the so-called MACRA legislation. This law was designed to provide both a carrot and a stick to encourage providers to participate in an alternative payment model like the one now being operated by OneCare Vermont. In essence, the new federal law says that if you join the OneCare, the operator of the alternative payment model, you will receive up to five percent more for the care you delivered to Medicare patients last year—with no quality reporting requirement…that’s the bump. The stick is that if you don’t join, you have to do considerable quality reporting to Medicare and if you don’t measure up your Medicare payments can be cut by up to nine percent. The carrot here is both the five percent increase in reimbursement, plus the elimination in the amount of federal paperwork you have to produce to show that your quality is high.
An important factor in the whole primary care political environment is that the benefits listed above are available not only to independent primary care doctors, but to primary care docs in Federally Qualified Health Centers (FQHCs). FQHCs already get federal subsidies aimed at ensuring the availability of primary care in underserved areas. The FQHCs that join OneCare will continue to receive those federal subsidies, as well as the financial support from OneCare.

Discussion/caveats

An obvious question is to determine how much of the increased revenue to each doctor is needed to support new demands on the practice, and how much would be an increase in income to the doctor. Participating in OneCare will generate some reporting requirements, but some of that work is already being done now. A rule of thumb in the industry is that one should divide the revenue increment in half, which would mean a $24,500 increase in business expense for each doctor and a $24,500 increase in the doctor’s take home pay. How the new payment system will operate in detail in Vermont won’t be known for some time. A preliminary assessment will probably be possible by mid-summer, when the figures from the first 11 months of 2017 are available. A major source of variability will be economies of scale: a five- doctor clinic would be able to structure support services more efficiently than that of a single doc. Moreover, to the extent that the new payment methodology leads to greater participation in the new new model the pay increase to the doctor would rise also.