|
Indemnities and related problems
Verifying the occurrence of an event covered by the indemnity could be a significant problem. How can an indemnity be designed to overcome it? One approach discussed is to use a multistage verification process. The first stage would be a relatively cheap and reliable determination whether the patient has suffered the condition at all. This could be based on a verified diagnosis of diabetes in a medical record, for example. The indemnity awarded at this initial stage should be relatively small. However, patients should have the right to appeal the initial determination.
In the context of medical indemnities, the appeal could be based on the severity of the disease or condition. Essentially, this is a risk-adjusted indemnity. Similar determinations already are used to establish the severity of long-term care needs, although severity-adjusted payments are made to the nursing home, not directly to the patient, as would be done with an indemnity. Must the Indemnity Be Spent on Medical Care? Recall that "pure" indemnities are fixed amounts of money paid to an individual after the occurrence of a well-defined event. After the individual receives the payment, he or she is free to spend it on a repair or not. Should the same principle be followed for Medicare indemnities?
Public health consequences
The list of such instances includes treatment for infectious diseases that have serious public health consequences, and settlements made on behalf of minors and mentally incompetent patients. These examples involve either substantial externalities where the government already uses its coercive power to force individuals to use medical care (such as vaccinations) or situations where the decision-maker is presumed unable to exercise an informed, independent decision regarding medical treatment. What about compelling informed adults to use medical care when the benefits of treatment are purely private? She did not explain, however, why rational, terminally ill adults should be required to use hospice services if they expressly wished not to.
At least three possible arguments can be made for Feigenbaum's position that a minimum level of coverage should be required. All of them require that we lay aside the economist's usual advice that people should be allowed to do as they wish with an insurance settlement. The first argument is that public medical programs, by their very nature, involve an externality that arises because the program transfers money to "recipients" from "donors" who have preferences regarding how the money should be spent. If recipients were free to use the money as they wished, the donors' utility would be lower than if the recipients had to spend some of the money on medical care. Therefore, letting recipients spend the money any way they wish is not optimal for society as a whole. This argument is most obvious with Medicaid, which transfers money from taxpayers to poor and elderly recipients who meet federal and state requirements for subsidized medical assistance. Medicare also is financed by taxpayers who subsidize medical benefits for the elderly and some disabled beneficiaries. Political support for these programs might be reduced substantially if beneficiaries were free to spend their indemnities on nonmedical goods.
The second argument for requiring the indemnity to be spent on medical care is that contracts in which patients eschewed any use of medical care might not be enforceable. Imagine that a terminally ill patient who refused all forms of medical care, including hospice care, and spent her indemnity on other goods, collapses at the door of a hospital. Would she be turned away? It is difficult to imagine that she would be denied access to all medical care. Quite possibly, she would be treated and referred to a hospice, so society would wind up paying twice - once for the indemnity and again for the charity hospice care.
Patients and their money
Can patients be prevented from spending their money on other goods and then seeking charity care? Clark Havighurst has proposed that patients and health plans, such as Medicare, be permitted to write binding ex ante contracts that specify the rights and responsibilities of each party. In this case, the beneficiary could be required to sign a contract stating that he or she will not seek charity care for the condition that was indemnified. I feel uneasy about dismissing Havighurst's position, but I doubt that society would have the stomach to enforce such contracts. Let me explain why. Havighurst placed most of the blame for the lack of contracts that bind consumers to follow economizing health-care choices on the courts. Such antipathy on the part of courts creates a chilling effect on contract drafters. Nevertheless, Havighurst believed that health plans could use three contracting strategies to authorize rational economizing in the provision of health care. First, they could attempt to specify more clearly the particular services or circumstances in which subscribers do not deem the marginal benefits of medical treatment to exceed the marginal cost. Second, in cases where particular decisions are disputed, the dispute could be settled by an independent third party. Third, the plan could take advantage of clinical guidelines to specify when care is not warranted.
I am much less sanguine about the enforceability of such contracts, not because of antipathy of the courts, but because of the public opprobrium that would be heaped on a health plan (including Medicare) that denied medical care after the fact, even if the patient had agreed a priori to such denial. In other words, Havighurst did not foresee the backlash against managed care that was looming on the horizon in the mid-1990s, centering on the fear that patients in managed-care plans would not receive the services they needed when they were sick. In a survey of managed-care enrollees, a majority (55 percent) said they were at least "somewhat worried" that if they were sick, their health plan would be more concerned about saving money than about what was the best medical treatment; only 34 percent of those with traditional health insurance felt this way. Americans' worst fear about managed care is that their plans won't take care of them when they are sick. Like it or not, the public perception of managed care is driven by "outlier" events - patients who died or were severely injured after their health plan denied coverage for a certain treatment. Such extremely negative perceptions would make unacceptably high the penalties for health-care plans that would try to write economizing contracts.
Minimal amount of medical care
Short of discovering a mechanism that would give patients incentive to purchase a minimal amount of medical care, I think that the Medicare indemnity would have to specify a minimum treatment plan for at least some cases, one of them being Feigenbaum's example of terminal illness in which the beneficiary would be required to use hospice care. Another approach to solving this problem would be to require proof of the patient's ability to pay out of pocket before issuing the indemnity. For example, only patients with assets above a certain minimum level might be allowed to "cash in" the indemnity. This approach has two drawbacks. One is political - it would create the impression that only the rich have this privilege; the other is that it makes little sense to allow patients to cash in the indemnity while restricting an equal amount of their other assets.
The third reason for restricting the indemnity to medical-care spending is that it would reduce the potential for fraud and abuse. Of Medicare beneficiaries, 78 percent have at least one chronic medical condition, and half have three or more. These individuals account for almost 90 percent of total Medicare spending. Putting the problem quite simply, if they were indemnified for their chronic conditions but could spend the money any way they wanted, being diagnosed with a chronic condition could trigger large and quasi-permanent annuity payments. The incentives for patients and physicians to make false claims would be overwhelming.
These incentives could be reduced by restricting the indemnity to be spent on medical care. Before leaving this question, I should note that I don't think unrestricted indemnity payments should be outside the bounds of the debate over Medicare reform. In 1996, a demonstration of unrestricted indemnities actually was undertaken at the Medical College of Virginia in Richmond. In this experiment, terminally ill cancer patients were randomly assigned to traditional fee-for-service care or to an indemnity where they received USD 18,000 to spend on anything they wanted. The USD 18,000 figure was selected because it represented the average cost of six months of chemotherapy. If patients randomized to the indemnity group wanted chemotherapy, they had to pay for it from the USD 18,000. Consequently, the findings from this interesting and potentially pathbreaking study were never published.
Physicians and payments
Community physicians in Richmond wanted to be paid to recruit patients for the study; the investigators did not want to pay, so they relied on their faculty colleagues at MCV to recruit patients. Unfortunately, these salaried physicians were more interested in recruiting patients into their own research studies than assisting in this one. Ultimately, only one patient was enrolled in the indemnity plan. The patient used the money to buy a beach house, but demanded more aggressive care when the illness progressed. The insurance plan that covered this patient relented, and the patient got the aggressive care. This sample, albeit of one patient, illustrates the problem of enforcing contracts with patients to forgo medical care - even if patients agree to these contracts a priori. In addition to the recruitment problem, this study was not a trial of pure indemnities because the only medical services that were carved out of the patient's insurance policy were those considered by the investigators to be "futile" - that is, medical care that purported to cure or extend the life of the terminally ill subject. Medical care for all conditions not related to the patient's cancer, as well as palliative care (such as pain relief), were covered. Therefore, this experiment could be considered to be a mixture of an indemnity and a regular medical insurance policy.