As Congress grapples with how to provide health care to the nation's uninsured, it should be remembered that Maryland has been successfully providing hospital services to the uninsured for 30 years. Amidst all of the talk about health care reform, it may be instructive to look at Maryland's 30 year experiment in setting reimbursement rates for hospitals, and interesting to ask what would happen if Congress expanded the experiment to allow Maryland to set reimbursement rates for physicians.
A. Hospital Rate Setting Waiver
To begin, one must first understand how Maryland sets reimbursement rates for hospitals. That process started in the 1970s, with a Congressionally granted "waiver" from the Medicare Program. The waiver encouraged states to experiment with setting hospital rates, and generally provides that Medicare will pay hospitals the rates set by a state, instead of the rates set by Medicare, provided that two criteria are met.
The first criterion is that the rates set by a state must generally apply to all payors, Medicare, Medicaid, commercial insurers and the uninsured.
The second criterion is that the rates charged to Medicare for inpatient services in the rate regulated state cannot grow faster than Medicare rates are growing elsewhere in the country. The key to this criterion is what it is not. The criterion does not require Medicare to pay less in the regulated state. On the contrary, Medicare is willing to pay more in the regulated state, provided that whatever Medicare pays in the regulated state does not grow faster than Medicare rates are growing in the rest of the nation.
In fact, Medicare has paid, and is paying, higher rates for hospital services in Maryland than elsewhere, but those rates have not grown faster than Medicare hospital charges elsewhere from 1980 to the present.
Why is Medicare willing to pay this extra amount? The answer is access. Medicare pays more so that Maryland's hospitals take care of Maryland's uninsured patients. In other words, 30 years ago, Congress reformed health care by allowing states to experiment with setting hospital rates. Moreover, while other states pursued this experiment for a time, Maryland is now the only state making use of the federal waiver.
B. The HSCRC
How did Maryland do it? Maryland established a quasi-independent commission, the Health Services Cost Review Commission (HSCRC), that sets the rates that Maryland's 47 general acute care hospitals may charge. Each hospital charges each patient on an ˆ la carte basis for the services provided to that patient, and the rates of each hospital are unique to that hospital. However, those rates are generally the same for all payors, Medicaid, Medicare, commercial insurers and the uninsured. There is no negotiation of charges between any payor and any hospital.
Originally, the HSCRC set rates for each hospital at a level that allowed each hospital to cover its costs, including its costs of providing services to the uninsured. Further, during the entire 30 year history of the program, if a hospital treats more uninsured patients in the current year than in the prior year, the hospital is entitled to increase its rates to cover that cost.
This system actually encourages Maryland's hospitals to treat the uninsured. More specifically, each paying patient is charged a little more than it costs to provide services to that patient so that the hospital may also provide services to those who do not pay, namely the uninsured.
C. Case Rates
Over time, the HSCRC has evolved from a "cost" Commission to a "revenue" Commission. Today, each hospital is assigned an average case rate target, for example, $10,000. That means the hospital may charge patients on an ˆ la carte basis, provided that at the end of the year, the hospital has not charged all patients more than $10,000 times the number of the hospital's admissions.
During a year, if a hospital appears to be headed over its average case rate target, it will lower its prices to all payors. Similarly, if the hospital is coming in under its target, it may raise its prices to all payors.
If a hospital's patient mix is sicker one year to the next, the hospital's case rate target increases a little bit. Similarly, recognizing that some costs are fixed, if a hospital's admissions increase, the hospital's average case rate target decreases, since the hospital can afford to charge additional patients a little less.
Once case rates are established for each hospital, the key annual event becomes how much those rates change from year to year. For example, last year the HSCRC only allowed a 1.27% increase in each hospital's case rate target.
D. Setting Doctors' Rates?
Armed with this background, could an all payor system be applied to doctors, and how might it be applied?
First, it must be recognized that, while there are only 47 general acute care hospitals in Maryland, there are 25,000 doctors in Maryland. While one can imagine setting an average encounter rate target for doctors, that target could not practically be set for each doctor, but it could be set for different types of doctors.
For example, the average encounter rate for an orthopaedic surgeon might be $500, and the average encounter rate for an internist might be $100. This would mean that each doctor would be able to charge various amounts for various discrete services, provided that by the end of the year all of the doctor's charges averaged below his or her average encounter target.
Initially, the charge for each discrete service could be set at today's average reimbursement for that discrete service, assuming a certain payor mix, and the amount might result in some percentage of Medicare rates, perhaps 110% of Medicare rates.
For example, a particular specialist's average encounter rate might be $150, and the specialist might perform five discrete tasks, priced at $50, $100, $150, $300 and $500. Moreover, during the first nine months of a year, that specialist's case mix might be as follows:
Discrete Reimbursement Number of Revenue
Task Rate Encounters
1 $ 50 400 $ 20,000
2 100 400 40,000
3 150 400 60,000
4 300 200 60,000
5 500 120 60,000
Total 1,520 $240,000
Average Reimbursement ($240,000 ÷ 1520) = $158
Percent Over $150 Target ($8 ÷ $150) = 5%
Accordingly, in the last three months of the year, the specialist's reimbursement for each discrete service would have to be reduced by 15% (3 times 5%) to bring the specialist under the $150 average encounter target by the end of the year.
The amount charged for each discrete service would also vary per doctor, initially and over time, based on that doctor's historic treatment of uninsured patients. Doctors with more than a certain amount of uninsured patients would be allowed to charge their paying customers, meaning patients with Medicare, Medicaid or commercial insurance, more than the 110% of Medicare rates to allow those doctors to net as much as their colleagues who do not provide service to
Over time, such an all payor physician rate system could also be used to drive quality and efficiency outcomes. For example, a doctor's reimbursement per discrete service, and/or his or her average encounter target, might increase if the doctor could show that by managing the doctor's patients' chronic heart disease or diabetes, the doctor kept his or her patients out of the hospital.
On the flip side, if a doctor provides an abnormally high percentage of his or her patients with the most expensive service that that doctor provides, then that doctor would either have to show the appropriateness of those decisions to have his or her average encounter target raised, or he or she would be discouraged from providing such high end services because the doctor's reimbursement for those high end services would necessarily come down as he or she approaches the applicable annual average encounter target.
Similarly, average encounter targets could be lowered as volumes increase. Doctors would still be incenticized to work harder, but the reimbursement they receive per encounter could go down as a doctor crossed certain volume thresholds.
Also, reimbursement could be increased in shortage areas, geographically or by specialty. For example, if there are too few Ob-Gyns in Western Maryland, physician reimbursement in that area could be increased to attract more doctors.
While 47 hospitals can monitor their charges against their case rate targets during a year, and while hospitals are not likely to go over their targets in one year and leave town the next, 25,000 individual doctors pose a different situation. Accordingly, implementing an all payor physician rate system would likely require an all payor billing clearinghouse that would make reimbursement adjustments in real time.
The cost of such a clearinghouse might be borne by Medicare, Medicaid and the commercial payors, each of which would likely have some offsetting administrative savings. On the flip side, sending one bill for all payors would be welcome news for doctors who have been bearing the ever increasing costs associated with their ever mushrooming billing staffs.
Insurers who believe that they can drive a harder bargain than their competitors might not like an all payor system. Similarly, the State Medicaid Program might find it difficult to pay physicians an all payor rate that is higher than existing Medicaid rates, but there could be an off-setting savings if the State did not have to expand Medicaid to cover as much of the working poor.
Initially, doctors who treat the poor and the uninsured would be benefited by an all payor physician rate system. Such a system would, at least initially, be somewhat neutral to doctors who do not treat the poor or uninsured, if initial reimbursement is based on today's average reimbursement. However, over time some doctors and some specialties would likely do better or worse than others.
Not only would an all payor physician rate system address the issue of the uninsured, but it might also be effective in breaking the back of health care inflation. By applying to all payors and focusing on total revenue, the likelihood of impacting health care inflation would be significantly enhanced. Moreover, an all payor physician rate system would exhibit many of the attributes of a single payor system, while at the same time preserving some of the competitive attributes of our existing multi-payor system.