Indemnity payments and marginal costs for Medicare patients

Indemnity advantages

The primary advantage of indemnity payments is that Medicare patients will face the full marginal cost of consuming expensive medical care. The primary drawback is that patients, especially those with low incomes, could be exposed to a substantial risk that the cost of their treatment will greatly exceed the indemnity.

To illustrate this problem, imagine that your automobile has been damaged, and the insurance adjuster presented this report to you: "The average cost of your repair is USD 1,000, but it could cost anywhere between USD 200 and USD 10,000. Here's your check for USD 1,000. Good luck." Auto insurance with this amount of risk to the policyholder wouldn't last long in the marketplace. Why should we expect that medical indemnity insurance, where the "repair" can cost hundreds of thousands of dollars, will fare any better? Before proceeding, we need to distinguish between "unavoidable" and "avoidable" medical expenses. Part of the reason the cost of the auto repair could vary so much is that some policyholders may install unnecessarily expensive parts. The decision to spend USD 10,000 might be avoidable in the sense that a perfectly decent repair could be done for USD 2,000. The difference between USD 10,000 and USD 2,000 represents avoidable expenses. Unfortunately, neither the measures of central tendency nor the measure of risk in reported medical expenditure data distinguish between avoidable and unavoidable expenses. Therefore, reported medical expenditure data will overestimate the actual risk to which patients are exposed, but the size of that overestimate is unknown.

Cost variation

The coefficient of variation of cost is smallest for colorectal cancer and largest for breast cancer. The larger the variance, the more risk the patient faces when he or she accepts an indemnity payment equal to the average cost per case. To determine how much patients would be willing to pay to avoid 100 percent of this risk, I used an estimate of risk aversion from the RAND Health Insurance Experiment (HIE). Patients in the HIE were willing to pay between forty-two and fifty-five cents to avoid each one thousand dollars of risk (in 1982 dollars).

The cost of risk-bearing ranges from a low of USD 33,722 for breast cancer to almost USD 90,000 for colorectal and lung cancers. I don't claim that these calculations are accurate down to the last dollar. Nevertheless, the message is quite clear: Patients who received an indemnity equal to the average cost of treating these common cancers would be exposed to unacceptably high risk that their treatment cost would substantially exceed the indemnity payment. This means that a system of pure indemnity payments is not likely to be viable for these types of cancer. For others where the range of treatment costs is quite large, such as the cancers shown above, pure indemnity insurance would need to be supplemented with a major medical policy of some type. There are several ways to design this mixed indemnity. One is to charge coinsurance on the excess of actual costs minus the indemnity.

Another, recommended by Gianfrancesco (1983), is to require a deductible before the excess charges are covered by a catastrophic insurance policy. Because the patient's ability to bear risk might vary by income, the coinsurance or deductible could be reduced or waived for lower-income patients. A third strategy is to "carve out" certain services that are part of the treatment bundle and continue to reimburse these through traditional health insurance. The carved-out services could be those that comprise a minimum treatment plan for the condition, or those that are especially risky, such as emergency hospitalizations.

An insurance policy with first-dollar indemnity coverage, followed by a deductible and catastrophic coverage after the deductible is met, sounds like the "donut hole" in Medicare Part D drug coverage, but it is not. All of the money in the indemnity has a marginal opportunity cost equal to that of cash, because the patient can use the indemnity to buy nonmedical goods. The deductible that follows the indemnity also has an opportunity cost equal to that of cash. Therefore, all of the indemnity plus the deductible has a full cash value to the consumer. This means that a mixed policy with a USD 10,000 indemnity followed by a USD 5,000 deductible is equivalent to a policy with a USD 15,000 deductible. Patients have strong incentives to use the first USD 15,000 of medical expenses carefully, as they would if they were spending their own money. In contrast, insurance coverage in Medicare Part D after paying a USD 275 deductible does not provide strong incentives for patients to be careful consumers of prescription drugs. A single indemnity payment covering expected lifetime treatment costs would be very risky for patients with chronic conditions. In such cases, the risk could be lessened by paying a series of indemnities in annual installments corresponding to the expected annual costs of maintenance care, possibly supplemented with major medical insurance.

Feigenbaum proposed a different solution to the riskbearing problem that does not rely on mixed indemnities. While interesting, this proposal has several weaknesses. First, it seems to have been borrowed from the analogy with auto repair shops, but there is an important difference between "auto repair" and "people repair." The auto repair shop has a good idea how much it will cost to fix and repaint a bumper - there is not much risk in accepting this job. But physicians may not know how much it will cost to treat a case of cervical cancer - the risk of accepting this job is substantial. Simply transferring that risk from patients to physicians does not eliminate it, although physicians might be able to pool the risk if they treat many patients with the same indemnified diagnosis. A second problem with Feigenbaum's proposal is that it resembles the physician DRGs. Dividing the indemnity among multiple providers is not a costless transaction.

Patients must be able to decide

More important, this proposal does not really turn decisionmaking power over to patients. Some patients might not want to contract with a single provider for their treatment, and others might not want to purchase standard medical treatment at all. The point of an indemnity is to let patients manage their health care with dollars that have other uses on the margin. On the other hand, Feigenbaum's proposal would be useful for patients who want to contract for a standard treatment at a set price. She specified that physicians who accept the indemnity payment in full would be "preferred," which implies that this decision would be voluntary on the part of the physician, rather than mandated by the Medicare program. This means that it could be one option among several available to patients.

1. Medicare Hospital Insurance Coverage plan A
Medicare Hospital Insurance Coverage Typically, people become eligible for Medicare Part A ...

2. Medicare Coverage under Supplemental Part B
Medicare Supplementary Medical Insurance Everyone who is eligible for Medicare Part A also is eligible to participate in the ...

3. Medicare Prescription Drug Coverage
Medicare Prescription Drug Coverage Medicare Part D coverage was enacted in December 2003 as part of the MMA and became operational January 1, 2006. The program...

4. Financing Social Security and Medicare
The OASI, DI, and HI trust funds are all funded from payroll taxes assessed on current workers. Often called "FICA contributions" [for Federal Insurance Contribution Ac...

5. Medicare Advantage contains five different types of health plans
Medicare Advantage Medicare managed care plans have been available since the 1970s, were paid on a capitated basis using the Adjusted Average Per Capita Cost [A...

6. Competition between Medigap and Medicare Advantage
Because access to employer-sponsored retiree coverage is typically determined before people retire, most of the short-run competition in the retiree market is between M...

7. Health Insurance firm Medicaid Managed Care
Managed care plays a significant role in Medicaid. By 2003, approximately one-half of all Medicaid r...