|
Volume 26, Number 4 (1996) reprinted by permission University of Memphis Law Review Health Insurance Coverage for High-Cost Health Care: Reflections on The Rainmaker ROBERT H. JERRY, II*
and his voice is "weak and raspy."6 He suffers from acute myelocytic leukemia,7 and the only mystery is how long he will live.8 We wonder whether he will survive long enough to know whether the Great Benefit Life Insurance Company (Great Benefit Life) will be ordered to pay damages for having denied him health insurance benefits under a policy issued to his parents. Fictional characters come and go, intriguing us for awhile until we pick up our next novel. But Donny Ray stays with us longer because he reminds us of friends or family members, present or past, who were equally misfortuned to suffer in their youth the random affliction of a fatal illness. The randomness is cruel; there, but for the grace of God, could have gone any one of us. If we linger a second longer, we remind ourselves that each of us must go sometime, either in an accident, or by illness, or simply because the body wears out and an essential part of it fails. If a life-threatening illness strikes, perhaps we will be fortunate enough to discover that one of the truly wondrous possibilities of modern medicine is available to save us. If we are less fortunate, we may discover that our health insurance policy does not cover the treatment for one reason or another and that the treatment is so expensive as to be beyond our own financial means. In other words, this part of the story of Donny Ray could be about almost any one of us. The core of the plot in The Rainmaker involves Rudy Baylor's trial of a bad-faith claim against Great Benefit Life.9 For the bad-faith plot to work effectively, the coverage and claims processing issues cannot be close; without egregious insurer conduct, there is no bad-faith plot to develop. Yet the coverage issue in The Rainmaker, even if Great Benefit Life must take a badly wrong and terribly weak position on every question relevant to coverage, takes us to the vortex of some of the most difficult issues facing our nation's health care system. even if the insurer were to reverse its denial of coverage and thereby en- able Donny Ray to receive it. In late September, some thirteen months after Donny Ray learned he was ill, Rudy Baylor visits Donny Ray, now on his deathbed, and observes, "So this is how the uninsured die."19 A brief amount of time passes, and on a Sunday morning, Donny Ray's mother calls Rudy to give him the news that Donny Ray is dead.20 The insurer's denial of coverage is a major contributing factor in, if not the outright cause of, Donny Ray's death because if the BMT had been performed soon after the discovery of his illness, the chances of Donny Ray surviving his illness were fairly estimated at eighty to ninety percent. Without the transplant, Donny Ray had no chance of surviving his illness.21 The essence of the story is simpler still. Donny Ray needed access to a high-cost treatment. Because he could not obtain that access, he died. In that simple equation lie the twin imperatives of our nation's health care system--providing access to adequate health care at an affordable cost. Although the problems of access and affordability22 are profoundly interrelat- ed, their ramifications are perhaps best understood by reflecting on each in turn.23 look shows this group to be both diverse and fluid. Not surprisingly, the poor are disproportionately represented among the uninsureds,27 as are minorities.28 Because most health insurance is provided by employers as a fringe benefit in employee compensation pack- ages, those who are chronically unemployed as well as those who are between jobs make up a significant portion of the uninsured population.29 In addition, many uninsured people are employed, but in low-wage positions where the employer does not offer health insurance.30 Likewise, many young adults who have recently departed their parents' households, where they were covered as dependents on one or both parents' policies, are also unin- sured.31 In short, the uninsured population is much more diverse than most people realize. Lack of insurance is, however, only one facet of the access issue. Many people who obtain insurance learn later that their coverage has significant exclusions or limitations. For example, most health insurance policies have preexisting condition clauses, which provide that insureds have no coverage, usually for a specific period of time after a policy is issued, for illnesses or conditions that predated the policy's effective date.32 Thus, a person who changes jobs and obtains new insurance through a new employer may find that his or her (or his or her dependent's) existing condition is not covered by the new policy. Most policies also have deductibles or coinsurance clauses. The presence of these loss-sharing clauses helps reduce the premium charged for insurance. To the extent this makes insurance more affordable, more people may be able to purchase it, but the deductibles and coinsurance requirements themselves may make it difficult for insureds to obtain health care if the insured cannot afford the out-of-pocket payment. Some policies have lifetime limits which place a cap on the insurer's total payments under the policy; this leaves some insureds without coverage for catastrophic illnesses or injuries once the policy limits are exhausted. The access issue is, however, even more involved than this. Just because an individual lacks insurance or just because his or her affliction falls within a gap in coverage, it does not necessarily follow that the person receives no medical care. An uninsured person may be more likely to forego preventive care, but when that person suffers a very serious medical condition, he or she is likely to go to a health care provider--most probably the emergency room of the nearest hospital--where that person will receive treatment. If a person is too poor to have insurance, then he or she will be too poor to pay the hospital's bill, and, even if the person is not indigent, the expenses of even a short hospital stay are likely to exceed his or her ability to pay. Hospitals and physicians write off some of these bills as uncompensated professional services,33 but the cost of many of these unpaid bills must be shifted to other paying patients, a redistribution that increases the cost of care and hence the cost of insurance for the insured population. As these costs rise, some insureds lose their ability to afford coverage,34 and these people become part of the uninsured population whose health care expenses will, in turn, be shouldered by the remaining insureds (or by taxpayers if the person is or becomes eligible for Medicaid or a similar state program, such as TennCare). Furthermore, the uninsured person who receives care at an emergency room only after a condition has become acute receives one of the most expensive forms of medical care possible.35 The care is also inefficient36 in at least two respects. First, it would have been better to treat the person's condition earlier when it would have been less expensive to do so, rather than delay to a point when more expensive treatments are needed. Second, it is inefficient to use trauma centers to treat ear or sinus infections, even painful ones.37 To add to the problem, an uninsured who is a rational economic actor is unwilling to invest his or her first discretionary dollars in health insurance when he or she knows that free care is available at the emergency room or from other health care providers. This is particularly true of young adults who tend to be healthy and, therefore, do not perceive a need for health insurance; for many such persons, making a monthly car payment may seem to have more utility than paying a health insurance premium. To the extent young, healthy people do not purchase insurance, the insured population tends to be older and, therefore, more prone to use health care services. This makes health insurance relatively more expensive, which makes it even less likely that a young, healthy adult will perceive health insurance to be a sensible investment. The issue of health care's affordability is no less intransigent. The statistics quickly become dated, but the trends are unmistakable and widely documented: as a percentage of gross domestic product, total spending on health care has increased significantly;38 health care costs are growing at about twice the general rate of inflation;39 per-family spending for health care as a percentage of total family income has increased significantly;40 and spending on health care is consuming an ever- increasing percentage of the federal budget.41 Of course, increased spending on health care is not necessarily cause for alarm if the increased expenditures simply reflect the desire of a more affluent society to spend more on health care than on other goods and services. In other words, spending a lot on something is not necessarily bad if this reflects a conscious, voluntary allocation of resources, in the same sense that if people like to play golf often, people will spend a disproportionate amount of their wealth on golf.42 Receiving medical care is not, of course, the same thing as playing golf. People do not aspire to receive more health care in the same way they aspire to play more golf, own a new car, purchase a cellular phone, etc. But when health care is needed, people usually want the very best care possible. Depending on the circumstances, this may mean receiving large quantities of care, the attention of the highest quality (and most expensive) specialists, elaborate tests, etc.43 In other words, health care is a peculiar kind of consumable; when a parent's child is ill, a parent is likely to insist on the very best care available, regardless of cost. When faced with a life-or-death illness or injury, few people make medical care decisions motivated primarily--or even significantly--by cost. Indeed, many, and perhaps most, people prefer extensive medical care when faced with life-threatening illnesses, even if much of the care only extends life briefly with no chance of altering an inevitable outcome. All these motivations are understandable, but it means that the incentives to consume medical care differ from the reasons one has to purchase cars, movie tickets, and other consumer goods and services.44 Yet, even if some portion of increased expenditures for health care reflects an increase that is "desired" by most people, much of the increased expenditure occurs because health care costs more. In other words, cost containment is a problem in health care, and it translates directly into reduced affordability of health care services. The affordability problem has many dimensions. Few would question that large segments of the current system have excessively high administrative costs.45 We praise the effectiveness of new technologies, but their use increases the cost of medical care. Although the magnitude of the effect is vigorously debated, many people believe that the legal system encourages enough unnecessary medical care to affect the cost of health care paid by everyone.46 Unnecessary duplication of expensive medical technologies also increases costs. To illustrate, it is doubtful that each of two adjacent hospitals in most urban centers needs an open-heart surgery capability. It is doubtful that every county in rural areas of the Central Plains needs a hospital, but the controversy over this issue is easy to imagine. The resident of a county with an under-utilized hospital that should be closed does not want the ambulance to have to travel an extra fifty miles to reach the neighboring county's hospital after his or her heart attack. But if an under-utilized hospital is kept open and consolidation is spurned, then the cost of each service must go up in order to pay for the excess overhead. The extent to which the health care industry is sufficiently competitive is much discussed; some believe that certain parts of the industry, such as drug companies, earn excess profits, while others view high drug prices as necessary to cover the high costs of bringing new drugs to the market. Each of these cost factors is complex, and the extent to which each contributes to the problem of affordability is much debated. Perhaps the most important factor increasing the cost of medical care is the one most deeply entrenched. The federal tax code allows employers to deduct health insurance provided as a fringe benefit to employees, and the benefit is excluded from the employee's income. This has greatly increased the amount of health insurance in force, which has greatly increased the demand for health care: to the extent an insured's decision to consume health care is motivated by cost considerations, an insured who has eighty percent of his or her health care bill covered by insurance will continue to consume health care until an additional one dollar of health care services is worth less than twenty cents to the insured. This translates into an enormous increase in demand for health care services: total health care expenditures increase, and to the extent demand rises faster than supply, prices must increase. When to this increased demand for services is added the increased demand generated by government-funded health care programs (i.e., Medicare and Medicaid, or Medicaid substitutes in some states), a simple supply-demand curve predicts significant increases in price, which means health care's affordability is reduced. Moreover, even if every bit of health care services currently consumed were voluntarily and willingly purchased in a perfectly functioning market, the percentage of federal and state governmental budgets devoted to health care would be a serious concern.47 Neither the federal nor state government can continue to increase spending for health care without either increasing taxes (a politically unpopular and often impossible option), increasing the size of budget deficits (which creates another diverse set of problems), or cutting expenditures for other government programs (an extraordinarily difficult and perhaps impossible strategy). Thus, even if the health care system currently operated with perfect efficiency, governments would still be searching for ways to reduce the cost of, and hence expenditures for, health care.48 The problems of access and affordability are, of course, interrelated: when the price of health care increases, more people find health care unaffordable, which further reduces access. When access is provided for those for whom cost is otherwise an insurmountable barrier, systemwide expenditures for health care inevitably increase. These additional expenses must be paid by someone, but there are few alternatives. Physicians and hospitals might provide the care without charge; indeed, the amount of uncompensated care provided by health care professionals is large. But it is unrealistic to expect health care providers to provide the uninsured population with all their health care needs for free. Two alternatives remain. First, higher prices can be charged to those who already pay for health care, in effect making the group of people who receive care in today's system subsidize the care of those who lack access. To the extent the higher prices are paid through an insurance mechanism, the additional insurance payouts must be funded through higher premiums charged to the insured population, which further constricts the affordability of insurance. Second, the government could provide the subsidy rather than impose the burden only on those who receive health care. In other words, those who lack access might be assisted through either publicly provided or publicly compensated health care. This would spread the cost of the subsidy across the tax base, with whatever distributive consequences result from federal or state tax policies. This approach, however, adds to tax burdens at a time when the prevailing mood in our country is that the governmental role (and the concomitant tax burden) should be decreased, not expanded. Sadly, no alternative resonates much excitement, or even hope. Once the generosity of health care providers is exhausted, the cost of providing health care services to the uninsured can be paid either with higher prices charged to insureds or with government funds provided through the tax base, or some combination of the two. Moreover, whatever mechanism is used to increase access to health care services, the increased demand for health care leads to higher prices, which has the ironic effect of compounding the access problem.49 employers exert pressure on insurers to keep premiums low. Individual consumers often act likewise. To compete, insurers must try to accommodate this desire, but it necessarily follows, in a market where costs of services are increasing, that stable premiums are likely to be accompanied by either reductions in coverage or at least constraints on expansion of coverage. In The Rainmaker, Great Benefit Life responds to these pressures in the most extreme manner imaginable by adhering to a corporate policy of initially denying coverage in virtually all cases and citing highly dubious and sometimes overtly frivolous grounds. The actual reasons Great Benefit Life denied coverage to Donny Ray were different: the proposed treatment was expensive; neither Donny Ray nor his parents were expected to challenge the denial of coverage; and Great Benefit Life anticipated profit from its behavior. Yet as the reader reacts with anger to the consequences of Great Benefit Life's miserably callous conduct, he or she should not lose sight of the fact that the grounds asserted by Great Benefit Life for denying coverage--the definition of "insured," the preexisting condition clause, the misrepresentation defense, and the experimental treatment exclusion--are important, even essential tools in many other settings where insurers have legitimate reasons for limiting their obligations to their insureds. One of the first reasons asserted by Great Benefit Life for denying coverage to Donny Ray Black is that he, as an adult, was not a covered dependent under his parents' policy.52 Of course, insurance does not extend without limitation to cover anyone with any kind of relationship to a named insured. For example, a common provision in health insurance policies defines persons eligible for coverage as the named insured, the named insured's spouse, and "unmarried dependent children (including stepchildren and legally adopted children) under 19 years of age."53 No one seriously questions the right of insurers to place such limitations in their policies. Indeed, it is important that the persons entitled to assert coverage under an insurance contract be clearly designated and that the insurer have a certain basis for calculating an appropriate premium for a highly material fact--the number of people covered by a policy. The relationship of this named insured and the dependent that gives rise to coverage under standard policy language does not last forever, and no one controverts the proposition that a person who ceases to fit the definition of dependent or to meet a minimum age requirement is not entitled to coverage under his or her parents' policy. The health insurance policy in The Rainmaker does not, however, contain a provision terminating the coverage for Donny Ray when he reaches a certain age.54 We are to assume, apparently, that the policy's definition of "insured" covered the named insured and any "dependent children." Donny Ray and his brother undoubtedly fell within the classification "dependent children" at the time the policy was issued; both boys were seventeen and lived at home. Because Donny Ray never left home, he never lost his status as a "dependent" even though he had reached the age of majority by the time his leukemia was diagnosed.55 There was simply no support for Great Benefit Life's position that Donny Ray was not within the definition of persons entitled to coverage under the policy. Another reason Great Benefit Life offered in support of its denial of coverage was that Donny Ray Black had a preexisting condition,56 which caused him to lose coverage under the preexisting condition clause, a provision found in almost all health insurance policies.57 Although the precise language of the clause varies from policy to policy, its effect is to exclude coverage for sickness or illness commencing before the effective date of coverage.58 Insurers insert preexisting condition clauses in policies to help control adverse selection.59 Obviously, if an insured could obtain coverage for a current illness or condition by purchasing insurance, he or she would wait until the illness or condition occurred to obtain insurance. Thus, insureds would pay nothing (and insurers would collect nothing) for the periods during which insureds are healthy. If insurers sought to solve this problem simply by excluding coverage for illnesses suffered by the insured on the first day of coverage, insureds would have an incentive to conceal their illnesses, claiming they arose after the policy's effective date. Thus, preexisting condition clauses have a broader function than simple exclusions for illnesses active on the first day of coverage; preexisting condition clauses combat the fraud and concealment that would occur in the absence of such clauses.60 By the same analysis, the presence of a preexisting condition clause, if it is known and understood, encourages consumers to purchase insurance when they are healthy. Thus, it is fair to assert that preexisting condition clauses eliminate coverage for illnesses or conditions that motivate the first acquisition of insurance. Through this response to adverse selection, the preexisting condition clause plays an important role in reducing the cost of coverage. With respect to a consumer's initial purchase of health insurance, the cost-reducing benefits of the preexisting condition clause are apparent. The clause is problematic, however, when applied to a person who, already having health insurance, switches employers (and hence insurers or insurance policies) and then confronts a preexisting condition clause once again. At best, this situation will create a gap in coverage, thereby exacerbating the problem of access to health care. At worst, the clause constrains the movement of employees in labor markets; employees with health conditions (or employees who have dependents with health conditions) may not be able to afford to improve their employment status if doing so means they must forfeit coverage for preexisting medical conditions. No single insurer can be expected to voluntarily incur the adverse selection costs generated by eliminating the preexisting condition clause; indeed, such an insurer would attract a disproportionate number of bad risks (i.e., people who having already suffered an illness or condition are now motivated to purchase insurance) and would have to raise premiums above the level of its competitors. Because the preexisting condition clause, and the gap in coverage it creates, is unlikely to disappear given the normal operation of ordinary market forces, governmental regulation of the clause is appropriate--specifically, regulation that would prohibit such clauses in health insurance policies, except for persons who are acquiring health insurance for the first time.61 If there ever was a case where the insurer's reliance on the preexisting condition clause should be rejected, The Rainmaker presents those facts. Donny Ray's leukemia was diagnosed more than four years after the Great Benefit Life policy was issued.62 Even if Donny Ray had symptoms before the diagnosis that should have put him on notice that he had a serious ailment, these symptoms would not have arisen at or near the time of the policy's issuance. In short, there is no adverse selection in The Rainmaker; indeed, Donny Ray's mother observed that the family had never even sought benefits until Donny Ray's illness.63 Moreover, the decided cases stand squarely in Donny Ray's favor. Courts have generally held that preexisting condition clauses apply only to conditions about which the insured was aware on or before the policy's effective date. Thus, an illness that the insured did not recognize as such, or that had not manifested symptoms sufficient to inform the insured about the presence of the ailment, does not fall within the clause.64 This protects insureds from losing benefits on account of unknown preexisting conditions. At the same time, it furthers the purpose of the preexisting condition clause--to protect insurers from adverse selection; an insured who is unaware of an illness cannot be motivated by adverse selection considerations to acquire insurance.65 Great Benefit Life made a feeble attempt to convert the preexisting condition defense into a misrepresentation defense: at the time of the application, Great Benefit Life asserted that no disclosure was made of a doctor's visit which Donny Ray had made to the doctor for treatment for the flu.66 It is, of course, often appropriate for an insurer to deny coverage based on an insured's misrepresentation in an application. Indeed, insurers ask applicants numerous questions about their prior health history for the purpose of gathering information upon which an insurer will rely when determining whether it wishes to issue a policy and assume the risk. Some answers will motivate the insurer to undertake follow-up investigation; in other circumstances, the insurer simply might rely on the accuracy of the information provided by the insurer. If the applicant misrep- resents, either intentionally or negligently, material information that induces the insurer to issue a policy that it would not have otherwise issued or to issue a policy at a premium that would have been adjusted had the insurer known the true facts, then the insurer enjoys a defense to coverage.67 Great Benefit Life's claim that Donny Ray's prior health history was misrepresented is flawed for two reasons. First, Donny Ray's failure to disclose a minor flu condition was not material, and immaterial misrepresentations do not provide a basis for an insurer to void a policy.68 Second, Great Benefit Life did not rely on the misrepresentation. It would have issued the policy anyway; indeed, insurers do not rely on immaterial misrepresentations. The last reason offered by Great Benefit Life for denying coverage to Donny Ray Black is the experimental treatment exclusion.69 In taking this position Great Benefit Life is clearly wrong. The efficacy of BMT treatment for acute leukemia is well established; the survival rates exceed fifty percent if the cancer is caught early enough.70 It may have been overly optimistic for Donny Ray to assert that he would definitely have been cured had the BMT treatment been performed,71 but it cannot be doubted that the BMT would have greatly increased Donny Ray's chances of survival in circumstances where he had no chance for survival without it. Although Great Benefit Life had little or no basis for asserting the experimental treatment exclusion, this facet of the plot invites a more substantial question: what exactly is it that made the BMT treatment nonexperimental? After all, barely half of all people who are identified as well-suited for BMT for acute leukemia in circumstances where the cancer is caught in its early stages and who then receive the treatment are cured. If the cancer is more advanced, the cure rates decline to levels well below 50%.72 One crude, but correct, answer asserts that the line must be drawn somewhere. So, when scientifically validated evidence that a particular regimen reliably and consistently produces a 40% cure rate for a defined group in circumstances where the nontreatment alternative is a 100% certainty of death, we deem, apparently, the treatment to be nonexperimental. In other circumstances, the answer is more obvious; thus, treatment of cancer with laetrile, which study after study has failed to find efficacious to even the slightest degree, is deemed experimental.73 The very nature of medical research requires that new treatments be tested with good science before the treatments are made available to the public for widespread consumption. As much as we might like to make all possibly helpful treatments available to all afflicted people, placing any and all experimental treatments within the scope of health insurance coverage would increase the cost of insurance to unaffordable levels for obvious reasons. Indeed, there are few serious, life-threatening illnesses for which some experimental treatments do not exist; the insured whose only remaining hope for survival (or whose only hope for delaying an imminent, yet inevitable death)74 is an experimental, expensive treatment is likely to request the treatment, even if the value of the treatment is dubious or un- known, as long as someone else pays for it.75 Thus, almost all health insurance policies contain exclusionary language eliminating coverage for experimental treatments, and the rationale for the exclusion is cost-containment. Policies have variously defined "experimental or investigative" treatment and sometimes have not defined the policy terms at all.76 Common formulations have included criteria such as local or national professional standards, scientific standards, or standards created by independent organizations or entities.77 In some cases, courts have refused to enforce the exclusion on the grounds that the definition was inadequate or lacking in sufficient detail and therefore ambiguous.78 Not surprisingly, when insurers have found a treatment to be experimental and, therefore, outside coverage by applying standards not found in the insurance policy, courts have refused to apply the exclusion.79 In one recent case, the court, noting the absence of guidelines or criteria in the coverage at issue, articulated its own non-exclusive list of factors to determine whether a procedure is experimental.80 A troublesome aspect of the case law on the experimental treatment exclusion is that different courts have reached different results on virtually identical facts. A particularly prominent example of this conflict among courts involves the refusal of some insurers to provide coverage for bone marrow transplants to treat certain kinds of cancers.81 Insureds have prevailed in most cases,82 but have lost in some.83 This means that a potentially life-or-death coverage determination--because, absent insurance coverage, few insureds have the personal resources to pay for the expenses of the treatment--turns on the jurisdiction in which the insured resides.84 Moreover, the cases are difficult at another level: how "experimental" does a treatment need to be before it is appropriate to deny coverage? Many medical experts believe that BMT treatment for some kinds of cancer, such as breast cancer, is not experimental, but other medical experts disagree.85 Should the benefit of the doubt on these issues always cut in favor of coverage? "Who" decides whether a treatment is experimental? How many experts must share the view that the treatment has progressed from experimental to conventional in order for the exclusion to be rendered inapplicable? And when experts disagree, how does a court select the "correct" expert opinion? The numerous cases contesting insurers' determinations that BMT treatment for breast cancer is experimental and, therefore, outside coverage have prompted some insurers to revise the exclusionary language. Like Great Benefit Life in The Rainmaker,86 some insurers have attempted to make the experimental exclusion more precise by actually listing the particular procedures that are not covered.87 Thus, if a bone marrow transplant is listed as an excluded procedure, there is no room for the court to find the policy ambiguous and to proceed to interpret the policy in a way that provides coverage.88 That insurers would attempt to draft clearer exclusions in the face of judicial repudiation of insurers' interpretations of more general exclusions should be expected; that has been the normal course of things for many decades. But in this context, the drafting of unambiguous treatment-specific exclusions carries with it the risk that BMT might be excluded as a treatment altogether, which then might deprive leukemia patients--for whom the treatment, though expensive, is not experimental--from relief from what is otherwise a certainly fatal disease.89 With so much at stake, it is not surprising that clear drafting only shifts the legal context in which the exclusion's validity is contested. In recent cases, insureds have argued, with some success, that explicit exclusions of coverage for high-dose chemotherapy treatment accompanied by BMT for most cancers violates the Americans with Disabilities Act.90 Insurers that continue to find themselves on the losing end of the coverage battle must surely be wondering what they must do to eliminate coverage for experimental treatments and to secure their cost-containment objectives. Perhaps our society can reach a consensus that all health insurance policies should cover bone marrow transplants as a treatment for acute leukemia in young adults. What then of bone marrow transplants for older persons, where the cure rate is lower? What then of bone marrow transplants and high-dose chemotherapy for all breast cancer patients? And what about the uninsureds--should access be provided to them? As more coverage is mandated, the cost of insurance must increase; as the cost increases, both access and affordability decline. But to the extent coverage increases, the demand for covered care increases, and this increased demand causes further increases in the price of health care, thereby making health care less affordable. The less affordable health care becomes, the more each of us needs insurance, which in turn costs more because health care costs increase, and so on. Once the potential of the private-market insurance mechanism to provide health services at affordable costs is exhausted, the government can be asked to provide these services. But unless we are prepared to shrink other governmental programs, such as defense, entitlements, social welfare, education, prisons, or highways, the amount of governmental resources that can be devoted to health care is finite, and this means that there will be insufficient funds to meet all the health care needs of every- one in our country. In turn, this means that choices must be made about what kinds of health care services will be provided to whom. Stated otherwise, we must concede that some people will not receive care for some medical conditions, and then we must decide who those people are. Many of these choices are extremely hard because lives hang in the balance. Where, for example, does Donny Ray Black fit in the calculus? Actually, his case is easy; the treatment Great Benefit Life denied him is accepted as standard therapy, and insurers routinely provide coverage in circumstances like his. But The Rainmaker, in presenting us with an easy coverage question, invites us to think about the much more difficult questions--questions that take us to the heart of the fundamental issues confronting our current health care system--where we must make choices about how health care resources should be allocated. More importantly, Donny Ray's story makes clear that the allocative choices, far from being avoidable, are being made right now in our current health care system, where health care is rationed and allocated, where access to excellent care is provided to some and denied entirely to others, where people just like the fictional Donny Ray Black prepare to bid farewell to their futures on account of lack of access to health care that could preserve their existence. The access and affordability imperatives that confronted Donny Ray Black are the same ones confronting our health care system today. At the least, we can hope that we will find answers more easily for our nation's health care problems than Donny Ray was able to find for his. If we fail in our effort, we may discover all too soon that Donny Ray is, indeed, one of us. Great Benefit Life made a feeble and ultimately unsuccessful attempt to establish that Donny Ray had lost his status as a dependent insured. See infra text accompanying notes 52-55. 32. This clause was asserted by Great Benefit Life as a reason for denying coverage to Donny Ray. See text accompanying notes 56-58. 33. For discussion of the frequency of charity care in hospitals, see Edward C. Norton & Douglas O. Staiger, How Hospital Ownership Affects Access to Care for the Uninsured, 25 RAND J. OF ECON. 171, 173-74 (1994); R.G. Frank & D.S. Salkever, The Supply of Charity Services by Nonprofit Hospitals: Motives and Market Structure, 22 RAND J. OF ECON. 430-45 (1991). 34. See Haggerty, supra note 30, at 33 (noting that with continued erosion of employer-funded health insurance in California and nationally, the number of uninsureds keeps increasing). 35. "The cost of [emergency department] visits is almost double that of other substitutable forms of ambulatory care . . . . When [emergency department] visits substitute for physician office visits, which seems to be a pattern among poorer patients, and assuming there are no differences in health outcomes, excess costs may be generated." Scott A. Optenberg et al., Emergency Care Episodes: An Economic Profile, J. OF AMBULATORY CARE MGMT., Jan. 1995, at 1. 36. As used here, efficiency refers "to the relationship between the aggregate benefits of a situation and the aggregate costs of the situation." A. Mitchell Polinsky, AN INTRODUCTION TO LAW AND ECONOMICS 7 (2d ed. 1989). If a situation is inefficient, the situation can be changed so that someone will be benefitted without hurting anyone else. Id. at 7 n.4. 37. "More than half of the 89.8 million emergency department visits made in 1992 were for non-urgent care." Laurence C. Baker & Linda Schuurman Baker, Excess Cost of Emergency Department Visits for Nonurgent Care, 13 HEALTH AFFAIRS 162 (1994) (citing L.F. McCaig, National Hospital Ambulatory Medical Care Survey: 1992 Emergency Dept. Summary, Advance Data from Vital Health Statistics of the Centers for Disease Control and Prevention, Nat'l Center for Health Statistics (Mar. 2, 1994)). 38. In 1992, health care accounted for 13.6% of U.S. gross domestic product (GDP), a significant increase from the 1980 figure of just over 9%. The percentage of GDP devoted to health care in the United States is greater than that of any other developed country. Germany and Japan, for example, had percentages of 8.7% and 6.9%, respectively. This translated to per capita expenditures in 1992 in the United States of $ 3,086, compared to $ 1,376 in Japan. Despite the additional expenditures in the United States, the infant mortality rate in Japan is half that of the United States, and life expectancy at birth in the United States is 4.4 years less for U.S. males than Japanese males and 3.9 years less for U.S. females than Japanese females. See CHART BOOK, supra note 27, at 30, Fig. 16. 39. From 1980 to 1994, the medical care inflation rate averaged 7.7% annually, compared with 4.3% for all items in the consumer price index. CHART BOOK, supra note 27, at 32, Fig. 18. 40. As a share of current consumption (defined as total expenditures minus gifts outside the family, cash contributions, personal insurance, and pensions), out-of-pocket health care expenditures peaked at 6.7% in 1960-1961, declined to 5.4% in 1972-1973, rose to 5.7% in 1988-1989, and rose to 6.9% in 1993. The decline through the late 1960s and early 1970s was presumably due to the Medicare program. Geoffrey D. Paulin & Wolf D. Weber, The Effects of Health Insurance on Consumer Spending, MONTHLY LAB. REV., Mar. 1995, at 34, 34. 41. Federal health spending is forecast to increase from 16.1% of total federal expenditures in 1992 to 23.6% in 1998. The federal government is responsible for Medicare and one-half of the cost of caring for the indigent under the Medicaid program. In state government budgets, Medicaid expenditures are rising more rapidly than any other state expenditure. FELDSTEIN, supra note 24, at 17. 42. Indeed, the aging of the U.S. population has major implications for health care expenditures. As a greater portion of the U.S. population consists of elderly persons, the population as a whole requires more medical services. Id. at 21. 43. This description does not, of course, fit everyone. Some people simply do not like to go to doctors, regardless of their illness or other circumstances. Some people are satisfied with a minimum amount of high quality care, assuming this degree of intervention resolves the ailment. All things being equal, I would prefer that my physician spare no expense to cure a persistent earache, but I tend to avoid visiting the doctor for such afflictions at all, even though the out-of-pocket cost to me for such services is de minimis. This is because large quantities of time are lost anytime I visit my physician. This is, of course, an access problem: to keep my premiums low, my health maintenance organization (HMO) has an extremely large patient membership, which frequently leads to long waits in the physicians' offices, which is a cost--lost time--that deters me from seeking care (that is, I consider lost time a greater cost than the annoyance of the earache). If the earache goes away on its own, as most of them do, the HMO saves money (thereby permitting lower premiums) by deterring my visit. This outcome is efficient, however, only if these savings exceed the additional costs incurred by patients who delay the receipt of health care for minor afflictions until the conditions become more serious and, hence, much more expensive to treat. 44. The situation is, of course, somewhat more complicated than this. The incentives to consume care to address health care needs differ from the incentives to receive preventive care. In other words, most people want a minimum amount of care (e.g., periodical physical examinations, immunizations, mammograms, EKGs, etc.) in order to reduce the risk of a catastrophic, serious illness. Few people are motivated to receive 10 EKGs a year. The amount of preventive care thought optimal is likely to be affected by education, the physician's recommendations (to the extent supplier-induced demand exists, see infra note 49, an above-optimal amount of preventive care may be ordered), and the time-costs of receiving the care. Yet whether a person has health insurance is, apparently, directly related to the amount of preventive care received. See note 25. 45. This problem is, of course, more complicated than the text allows. For example, the advent of "managed care" and "managed care organizations," such as HMOs and PPOs (preferred provider organizations), is a direct consequence of the cost problem. To the extent many health services rendered are thought to be unnecessary or inappropriate, thereby driving up overall system-wide costs, managed care, through review or intervention, seeks to deter health care providers from prescribing unnecessary or inappropriate treatments. Managed care may also involve a managed care organization negotiating, on behalf of a large group of consumers, with health care providers to receive discounted prices for services. Thus, managed care often requires new or additional administrative mechanisms, which have the effect of increasing administrative costs. If these additional administrative costs result in net savings to the health care system, then the increased administrative costs are desirable. 46. A 1993 study of the Competition Center of the Hudson Institute claimed that legal liability expenses added $450 to the average patient's hospital bill in a large urban hospital in Indiana and that defensive medicine accounted for 3.9% of the hospital's total operating expenditures. On the other hand, a 1994 report of the Office of Technology Assessment labeled earlier estimates of the cost of defense medicine "unreliable" and found that less than 8% of all diagnostic procedures were consciously defensive. MEDICAL MALPRACTICE, INSURANCE INFORMATION INSTITUTE REPORTS (Ruth Gastel ed., 1996). 47. See supra notes 41-42. 48. See FELDSTEIN, supra note 24, at 17-19. 49. In this discussion, it is assumed that access for the uninsured will not be increased by reducing benefits for the insured population--a solution which merely transfers some of one group's access to another group, i.e., the group lacking access. Such a solution would not increase demand, but would merely shift the access problem, not solve it. See Rashid L. Bashshur et al., Beyond the Uninsured: Problems in Access to Care, 32 MEDICAL CARE 409, 410 (1994) (arguing that some recently proposed solutions to the problems of the uninsured reduce benefits to insureds, "thereby inadvertently decreasing access overall"). Merely increasing the supply of health care does not guarantee that prices will fall. The impact of the increased supply of physicians in recent years illustrates the point. Some commentators believe that physicians act in their own self-interest to protect their own incomes, which means that physicians will prescribe additional services to increase their revenue. This theory of physician behavior, called "supplier-induced demand," is controversial, but there is some evidence to support the notion that increasing the supply of physicians also increases demand. The competing theory considers the physician a "perfect agent" for the insured, meaning that the physician will act in the insured's interest. To the extent insurance pays for the insured's treatment, the physician as a perfect agent will prescribe services where the costs outweigh the benefits, thus increasing demand and adding to system-wide costs. See FELDSTEIN, supra note 24, at 35-39; MARK V. PAULY ET AL., PAYING PHYSICIANS: OPTIONS FOR CONTROLLING COST, VOLUME, INTENSITY OF SERVICES 35-45 (1992). 50. The fictitious Great Benefit Life is a commercial insurer, of which there are about 350 that sell health insurance in the United States. See WARREN GREENBERG, COMPETITION, REGULATION AND RATIONING IN HEALTH CARE 48 (1991). The other kinds of private health insurance are Blue Cross/Blue Shield plans, self-funded employer plans, and prepaid plans (such as HMOs). See generally BARRY R. FURROW ET AL., HEALTH LAW section 11-1, at 498-502 (1995). 51. Between 1991 and 1994, the percentage of total employer-provided compensation devoted to health insurance in goods-producing industries rose from 6.9% to 8.1%; in manufacturing, the percentage rose from 7.5 to 8.6; in services, the percentage rose from 5.5% to 6.0%. NATIONAL CENTER FOR HEALTH STATISTICS, U.S. DEPT. OF HEALTH AND HUMAN SERVICES, HEALTH UNITED STATES 1994, Table 122, at 227 (1995). 52. GRISHAM, supra note 1, at 26. 53. Family Hospital Expense and Surgical Policy, Form IM 6187, in ALLIANCE OF AMERICAN INSURERS, THE INSURANCE PROFESSIONALS' POLICY KIT: A COLLECTION OF SAMPLE INSURANCE FORMS 95 (1994-1995). Definitions of "dependent" can, of course, be more elaborate. Consider, for example, the definition of "dependent" in the policy at issue in Kornman v. Blue Cross/Blue Shield of Louisiana, 662 So. 2d 498 (La. Ct. App. 1995), writ denied, 667 So. 2d 1054 (La. 1996): "Dependent" means a Member for whom current fees have been paid and is:Id. at 503. 54. During her initial meeting with Rudy Baylor, Donny Ray's mother commented that she had read the policy several times and had found no language terminating her dependents' coverage upon their reaching a certain age. GRISHAM, supra note 1, at 16. The policy listed two dependents, Donny Ray and his twin brother. Id. at 15. 55. Counsel for Great Benefit Life explored this issue in the deposition of Donny Ray Black, but was unable to elicit any fact tending to show that Donny Ray left home to live elsewhere for any period of time, even a period as short as a week. Id. at 237. 56. Id. at 26. 57. This discussion is substantially based on JERRY 1996, supra note 9, section 103. 58. One typical clause defined a preexisting condition as follows: "Any condition that was diagnosed or treated by a physician within 24 months prior to the effective date of the coverage or produced symptoms within 12 months prior to the effective date of coverage that would have caused an ordinary prudent person to seek medical diagnosis or treatment." See Holub v. Holy Family Soc'y, 518 N.E.2d 419, 420 (Ill. App. Ct. 1987). 59. Adverse selection refers to the phenomenon that results in an insurer's pool of insureds inevitably being composed of a disproportionately high number of less desirable risks. This results from the greater tendency of higher-risk people to seek insurance because they receive a better return on their premium investment than lower-risk people. See JERRY 1996, supra note 9, section 10c[2]. 60. See Mogil v. California Physicians Corp., 267 Cal. Rptr. 487 (Ct. App. 1990). 61. As of May 1996, both the U.S. House of Representatives and the U.S. Senate had passed bills that would regulate exclusions for preexisting conditions. The proposed law would generally allow health plans to deny coverage for preexisting conditions for up to 12 months, but would reduce the 12-month period by the amount of time a new employee was covered under prior health plans. Although this proposal has widespread support, prospects for its ultimate enactment are uncertain due to differences between the House and Senate versions on other issues. See Kathy Kristof, A Continuous Chain of Health Insurance Proposals Limit Use of Pre-Existing Condition Clauses That Can Make It Difficult to Switch Jobs or Medical Plans, CHI. TRIB., May 8, 1996; Robert Pear, Health Care Bill Draws Opposition on Key Provisions, N.Y. TIMES, May 26, 1996, at 6, col. 6. This proposal, if enacted, will have the effect of limiting the impact of the exclusion to persons acquiring health insurance for the first time or upgrading to more comprehensive plans. 62. GRISHAM, supra note 1, at 14-15. 63. Id. at 15. 64. See, e.g., State v. Carper, 545 So. 2d 1 (Miss. 1989) (policy excludes only conditions that manifest themselves prior to date of coverage); Holub v. Holy Family Soc'y, 518 N.E.2d 419 (Ill. App. Ct. 1987) (insured who had minor bowel symptoms and who was told by doctor she had "nothing to worry about," but who in fact had rectal cancer, had coverage despite preexisting condition clause). 65. There is, of course, a risk that a person may apply for insurance and understate the significance of symptoms about which he or she, at the time the insurance was sought, actually had great concern. Thus, some courts have held that a condition thought to be minor but found to be more significant after the policy's issuance may trigger the preexisting condition clause. See, e.g., Golden Rule Ins. Co. v. Atallah, 45 F.3d 512 (1st Cir. 1995) (preexisting condition clause does not require insured to suspect a particular diagnosis; if insured experiences symptoms, whatever illness is ultimately determined to have caused those symptoms is a preexisting condition and is excluded from coverage; no exception exists for insured who in good faith obtained an incomplete or incorrect diagnosis and therefore failed to disclose the full extent of the illness before purchasing insurance); Mogil v. California Physicians Corp., 267 Cal. Rptr. 487 (Ct. App. 1990) (cancerous mole was preexisting condition, in circumstances where insured's prior moles had been benign but insured knew that moles needed continued monitoring). But these circumstances did not exist in Donny Ray's situation either. 66. GRISHAM, supra note 1, at 296. 67. For a more complete discussion of the misrepresentation defense, see JERRY 1996, supra note 9, section 102. 68. GRISHAM, supra note 1, at 296. As explained more fully there, even a fraudulent immaterial misrepresentation is unlikely to invalidate the coverage. Id. Even so, there was no indication anywhere in The Rainmaker that Mrs. Black's failure to disclose on the application that Donny Ray had once had the flu, an illness that afflicts virtually everyone at one time or another, constituted a willful attempt to deceive Great Benefit Life. 69. Portions of the following discussion are based on JERRY 1996, supra note 9. 70. See Thomas H. Maugh II, Improving the Odds, L.A. TIMES, Jan. 11, 1996, at B2 (reporting a recent large study at a Seattle cancer research center showing that the five-year survival rate for leukemia patients who received a BMT was 50%, compared to 20% for those who received chemotherapy or radiation alone); Wisconsin Bone-Marrow Transplant Success Improves Slightly as Leukemia Treatment, CANCER WEEKLY, Aug. 17, 1992 (use of BMT to cure leukemia became "modestly more successful during the 1980s, reaching a 57 percent survival rate in cases caught early enough," according to a new study of 7,788 sibling BMTs reported by 185 transplant teams world-wide; survival rates were 36% for intermediate leukemia, and 18% for advanced leukemia; "survival" was defined as no recurrence of leukemia two years after the transplant). 71. GRISHAM, supra note 1, at 103 (Donny Ray states that "the transplant would've saved my life"); id. at 104 (Donny Ray states that if he had received the BMT six months earlier, he would have had a 90% chance of cure); id. at 345 (Dr. Kord testifies that a BMT for Donny Ray would have increased his likelihood of surviving acute leukemia by 80% to 90%). 72. See Experimental Therapies for Treating Leukemia, GENESIS REPORT, Aug. 1, 1993 (reporting that some types of childhood leukemias have cure rates of 95% or higher, but that "cure rates for the most common acute and chronic adult leukemias are 40% and less"; "so few bone marrow transplants are successful in people older than 55, who account for about half of all adult leukemia patients, that transplants are not worth doing"). 73. Although the observation that administering laetrile for cancer is experimental would seem to be beyond serious dispute, this has not prevented the issue of whether the use of laetrile as a treatment for cancer is covered under a health insurance policy from being litigated. See Free v. Travelers Ins. Co., 551 F. Supp. 554 (D. Md. 1982) (holding that insured's use of laetrile for treatment of lymphoma was not an expense necessarily incurred and thus was not within the policy's coverage). See generally Michael G. Walsh, Annotation, Right of Medical Patient to Obtain, or Physician to Prescribe, Laetrile for Treatment of Illness--State Cases, 5 A.L.R.4th 219 (1981). 74. One court described one insured's stake in a particular treatment this way: "Twenty to thirty percent of Stage IV breast cancer patients receiving HDC-ABMT have a long-term survival rate of at least two or three years free of cancer with no need for additional treatment. On the other hand, the Stage IV breast cancer patients who receive conventional chemotherapy have a cancer-free outlook of zero percent." Taylor v. Blue Cross/Blue Shield of Mich., 517 N.W.2d 864, 869 (Mich. Ct. App. 1994). 75. If of unknown or dubious value, a treatment may also be considered not to be "medically necessary" because of the lack of an apparent benefit. Although the subjects are usually treated separately, here it may make sense to view them as raising a common question of "medical appropriateness." For a thoughtful treatment of the entire area, see Mark A. Hall & Gerald F. Anderson, Health Insurers' Assessment of Medical Necessity, 140 U. PA. L. REV. 1637 (1992). 76. See generally Barbara A. Fisfis, Comment, Who Should Rightfully Decide Whether a Medical Treatment Necessarily Incurred Should be Excluded From Coverage Under a Health Insurance Policy Provision Which Excludes From Coverage "Experimental" Medical Treatments?, 31 DUQ. L. REV. 777 (1993); Jennifer Belk, Comment, Undefined Experimental Treatment Exclusions in Health Insurance Contracts: A Proposal for Judicial Response, 66 WASH L. REV. 809 (1991). 77. See, e.g., Fuja v. Benefit Trust Life Ins. Co., 18 F.3d 1405 (7th Cir. 1994); Lubeznik v. Healthchicago, Inc., 644 N.E.2d 777, 780 (Ill. App. Ct. 1994) ("appropriate technology assessment bodies"), appeal denied, 652 N.E.2d 343 (Ill. 1995). 78. See, e.g., Dahl-Eimers v. Mutual of Omaha Life Ins. Co., 986 F.2d 1379 (11th Cir. 1993) ("considered experimental" held ambiguous because policy did not specify who would make determination); Pirozzi v. Blue Cross/Blue Shield of Va., 741 F. Supp. 586, 589-90 (E.D. Va. 1990) (policy failed to define "experimental" or "clinical investigative" and failed to indicate what evidence was required to classify a treatment as experimental); Tepe v. Rocky Mountain Hosp. & Medical Servs., 893 P.2d 1323 (Colo. Ct. App. 1994) (history of plan changes culminating in exclusion was unclear and misleading to insured, who had reasonable expectations of coverage); Taylor v. Blue Cross/Blue Shield of Mich., 517 N.W.2d 864, 869 (Mich. Ct. App. 1994) ("experimental" and "research in nature" are ambiguous). 79. See, e.g., Kekis v. Blue Cross & Blue Shield of Utica-Watertown, Inc., 815 F. Supp. 571, 578 (N.D.N.Y. 1993) (insurer used wrong definition of "experimental"); Pirozzi v. Blue Cross/Blue Shield, 741 F. Supp. 586, 591 (E.D. Va. 1990) (use of technical evaluation was unacceptable because such criteria were not specified in policy). 80. See Heasley v. Belden & Blake Corp., 2 F.3d 1249, 1262-64 (3d Cir. 1993) (judgment of other insurers and medical bodies; amount of experience with the procedure; demonstrated effectiveness of the procedure). 81. More specifically, the procedure is high-dose chemotherapy with autologous bone marrow transplant, or "HDCT-ABMT." The treatment involves removing part of a patient's bone marrow, giving the patient extremely high doses of chemotherapy, and then replacing the bone marrow. According to a study published in the New England Journal of Medicine, the reasons health insurers most often give for denying coverage for BMTs for breast cancer is that the treatment is considered experimental. See Breast Cancer: Bone-Marrow Transplants, HARVARD HEALTH LETTER, July 1, 1995. For a discussion of this issue and other legal issues relevant to breast cancer, see Nancy A. Wynstra, Breast Cancer: Selected Legal Issues, 74 CANCER SUPPLEMENT 491 (July 1, 1994). 82. See, e.g., Kekis v. Blue Cross & Blue Shield of Utica-Watertown, Inc., 815 F. Supp. 571 (N.D.N.Y. 1993); Adams v. Blue Cross/Blue Shield of Md., Inc., 757 F. Supp. 661 (D. Md. 1991); Kulakowski v. Rochester Hosp. Serv. Corp., 779 F. Supp. 710 (W.D.N.Y. 1991); White v. Caterpillar, Inc., 765 F. Supp. 1418 (W.D. Mo.), aff'd, 985 F.2d 564 (8th Cir. 1991); Pirozzi v. Blue Cross/Blue Shield, 741 F. Supp. 586 (E.D. Va. 1990); Tepe v. Rocky Mountain Hosp. & Medical Servs., 893 P.2d 1323 (Colo. Ct. App. 1994); Lubeznik v. Healthchicago, Inc., 644 N.E.2d 777 (Ill. App. Ct. 1994), appeal denied, 652 N.E.2d 343 (Ill. 1995); Taylor v. Blue Cross/Blue Shield of Mich., 517 N.W.2d 864 (Mich. Ct. App. 1994). 83. See, e.g., Harris v. Mutual of Omaha Cos., 992 F.2d 706 (7th Cir. 1993); Nesseim v. Mail Handlers Benefit Plan, 995 F.2d 804 (8th Cir. 1993); Holder v. Prudential Ins. Co. of Am., 951 F.2d 89 (5th Cir. 1992). For more discussion, see Paul E. Pongrace III, Comment, HDC/ABMT: Experimental Treatment or Cure All? (Ask the Insurance Companies), 2 J. PHARM. & L. 329 (1994). 84. The problem, however, may have deeper roots. A 1994 study reported on the process through which insurers approve the treatment with high-dose chemotherapy and autologous bone marrow transplantation of breast cancer patients in grant-sponsored clinical trials. It concluded that the predetermination process was "arbitrary and capricious," "did not correlate with protocol-based medical decision making," and "was a barrier to obtaining treatment." The study noted "substantial inconsistency in the frequency of approval of coverage both among insurers and between decisions made by some individual insurers, even for patients in the same study protocol." William P. Peters & Mark C. Rogers, Variation in Approval by Insurance Companies of Coverage for Autologous Bone Marrow Transplantation for Breast Cancer, 330 N. ENG. J. MED. 473 (Feb. 17, 1994). 85. The difficulty in determining when experimental treatment becomes customary therapy is starkly illustrated by the dissension over the utility of BMTs, accompanied by high-dose chemotherapy, for the treatment of breast cancer. Breast cancer strikes more than one out of every ten women at some time in their lives. More than 180,000 new cases are diagnosed annually. Each year more than 45,000 women succumb to breast cancer, making it the leading cause of death (as of 1990) among women in the United States. Wynstra, supra note 81, at 491 (citing E.J. Feur, The Lifetime Risk of Developing Breast Cancer, 85 J. NAT'L CANCER INST. 892 (1993)). It is estimated that about one-third of those who die might be good candidates for BMT treatment, which would cost $ 2.25 billion annually. Marilyn Chase, Medical Quandary: Breast-Cancer Patients Seeking New Therapy Face Tough Obstacles, WALL ST. J., Feb. 17, 1993, at A1, A9. Expert opinion on whether the treatment is superior to conventional therapy is divided. See Breast Cancer: Bone-Marrow Transplants, HARVARD HEALTH LETTER, July 1, 1995, at 1 ("The procedure is costly, controversial, and traumatic and is not as successful as some research physicians had hoped."). The mortality rate from the treatment itself is 1% to 5%, but many experts believe that the treatment increases longevity beyond what is possible with conventional therapy. Because the treatment has been in use for barely a decade, long-term follow-up studies have not yet been possible. Id. It has been suggested that BMT is routinely covered for some rare forms of cancer, such as testicular cancer, where there is less evidence of BMT's utility than with breast cancer, but insurers do not contest coverage because the treatment is less frequent and therefore less costly. Chase, supra, at A1. 86. See GRISHAM supra note 1, at 311. 87. See Patricia Anstett, Cancer Patients Battle With Insurers, HOUSTON CHRONICLE, July 5, 1995 (reporting that Blue Cross in Michigan has recently issued "riders" that approve payment for BMTs to treat a half-dozen kinds of leukemia and related blood cancers, but which reject payment for the treatment in all other common cancers; an insured or employer can purchase a policy covering all BMTs, but most employers decline the more expansive coverage because of its high cost). For the text of such a policy, see Nesseim v. Mail Handlers Benefit Plan, 792 F. Supp. 674, 678 (D.S.D. 1992), rev'd, 995 F.2d 804 (8th Cir. 1993). 88. See, e.g., Caudill v. Blue Cross & Blue Shield, 999 F.2d 74 (4th Cir. 1993); Nesseim v. Mail Handlers Benefit Plan, 995 F.2d 804 (8th Cir. 1993). 89. See Joyce Price, Bone Marrow Transplants Gain Favor But Insurance Often Won't Pay, WASH. TIMES, June 20, 1994, at A10 ("there have been isolated cases where insurance providers have denied coverage of HDC/BMT to treat diseases such as leukemia, where its benefits have been proven in clinical trials"); Michelle Slatalla, Leukemia Aid Denied, NEWSDAY, April 28, 1994, at A8 (reporting on a 28-year-old leukemia patient who was denied insurance coverage for a BMT; "experts say the controversy over whether to cover experimental marrow transplants for breast cancer patients may be creating a backlash against the long-accepted practice of covering the procedure for leukemia patients"). In In re Gonzalez, 621 A.2d 94 (N.J. Super. 1992), the issue addressed by the court was how surplus monies in a fund created by public donations to defray the decedent's operative procedure should be disposed. According to the court, the decedent was diagnosed as suffering from acute leukemia in 1987, and the treating physicians recommended a BMT. Id. at 94. The insurer denied coverage on the basis that a BMT was experimental treatment for leukemia. Id. at 95. Gonzalez died before a BMT, which was to have been funded by private donations, was performed. Id. at 94. 90. In Henderson v. Bodine Aluminum, Inc., 70 F.3d 958 (8th Cir. 1995), the court reversed a district court's denial of a preliminary injunction that would order an ERISA health plan to provide coverage for an HDCT-BMT treatment. The insured argued that because the plan covered the treatment for cancers for which it is an accepted treatment, denying the treatment for breast cancer was discrimination based on disability type and that such discrimination is prohibited by the Americans with Disabilities Act, 42 U.S.C. sections 12101-12213 (1990). Henderson, 70 F.3d at 960. The court agreed that the insured's "argument has a sufficient likelihood of success on the merits." Id.; cf. Hilliard v. BellSouth Medical Assistance Plan, No. CIV. A 3:95-CV-793WS, 1995 WL 815238, at *10 (S.D. Miss. 1995) (denying insured's request for preliminary injunction that would order HDCT-BMT for multiple myeloma and distinguishing Henderson). |
