Prescription Drug Coverage and Medicare

Robert D. Reischauer

Mr. Chairman and Members of the Subcommittee, I appreciate this opportunity to discuss with you some of the issues raised by proposals to provide Medicare participants with greater protection from large out-of-pocket prescription drug expenditures. My statement addresses three questions:

  • Why have serious legislative proposals to provide Medicare participants with some type of prescription drug coverage surfaced now, one decade after Congress voted overwhelmingly to repeal the modest drug protection provided by the Medicare Catastrophic Coverage Act of 1988?
  • Which of the various broad approaches to providing the elderly and disabled prescription drug coverage makes the most sense in the current context? and
  • How should a new prescription drug benefit be structured?

Why now?

Three considerations have stimulated renewed interest in providing prescription drug coverage to Medicare participants:

Before World War II, few prescription drugs were available and their therapeutic contribution to health care was limited. The development of new and more powerful antibiotics and antidepressants in the late 1940s and 1950s laid the groundwork for a pharmaceutical revolution. When Medicare was enacted in 1965, this revolution had not yet come to fruition, and the role of pharmaceuticals in health care remained relatively limited. Since 1965, however, there has been an explosion of new drug therapies—more powerful drugs for bacterial infections, immunosuppressant drugs for organ transplants, antidepressants with fewer side effects, vaccines to protect against measles, mumps, rubella, diphtheria, hepatitis B and other diseases, chemotherapies to fight cancer, clot busting and blood thinning drugs, along with pharmaceutical interventions for such chronic conditions as high cholesterol, irregular heartbeats, elevated blood pressure, asthma, and arthritis. The past decades are likely to be just the overture to what lies ahead as new gene therapies and biotechnological applications move from the laboratory to the marketplace.

As the ability of drug therapies to improve health has grown, so too have the total costs of such treatments. When Medicare was enacted, pharmaceutical expenditures constituted 10.6 percent ($3.7 billion) of personal health care expenditures.

From the mid 1960s through the early 1980s, the contribution of drugs to the total health care bill fell fairly steadily, reaching a low of 5.3 percent ($15.0 billion) of personal health care expenditures by 1982. This ratio then began to rise. When the Medicare Catastrophic Care Act was repealed, drug expenditures amounted to 6 percent ($32.9 billion) of personal health care expenditures. In 1999, a decade later, the fraction is expected to be 9.3 percent ($100.6 billion) and it is projected to grow to 12.6 percent ($243.4 billion) by 2008. Many believe these projections are conservative.

Employer-sponsored health insurance policies, as well as privately purchased individual policies, have recognized the increasing importance of drug therapies to overall health care and have expanded coverage and reduced coinsurance for out- patient pharmaceuticals over the past several decades. About 95 percent of employer-sponsored plans now provide some drug coverage and many individual policies offer such protection as well. However, Medicare, with a few exceptions, does not cover the costs of out-patient prescription drugs. This constitutes a serious inadequacy in the program, one that makes no more sense than offering a health insurance policy that does not cover diagnostic imaging such as X-ray, CT, MRI, sonogram, or PET scans.

Most Medicare participants have managed to cope with the program’s failure to provide broad coverage by obtaining some form of out-patient prescription drugs coverage through supplemental policies. The 12 percent of participants who are dually eligible for Medicaid have the most extensive protection and face no, or very little, out-of-pocket exposure. Of the one-third of participants who are covered by a supplemental policy provided by their former employer, roughly nine in ten receive more or less adequate drug coverage through these policies—coverage that is similar to that which they enjoyed when they were active workers.

Roughly one in four of those with Medigap policies—those who purchase one of the three standard Medigap policies that offers prescription drug coverage (policy types H, I, and J) or a nonstandard (pre-1992) policy with such coverage—receive drug benefits that are rather limited and which bear a fairly stiff price. The standard H and I Medigap policies pay 50 percent of prescription drug costs above a $250 deductible up to a maximum $1,250; the J policy has a maximum benefit of $3,000. In a 1998 survey of a sample of metropolitan areas, Consumer Reports found that the median annual premium faced by a 65 year old for the standard Medigap I plan was $1,201 higher than that for the standard F plan—a pretty steep differential considering that the only additional benefit the I plan offers besides limited drug coverage is payment of the Part B deductible ($100). Such differentials reflect the fact that, in a voluntary system like Medigap, less healthy participants are attracted to plans that offer prescription drug coverage.

The overwhelming majority of Medicare participants who are enrolled in a Medicare+Choice plan (M+C)—some 16 percent of all participants today—are provided with some prescription drug coverage, although this protection is often quite limited. Less than 3 percent of Medicare participants get help with their drug expenses through programs operated by the Departments of Defense and Veterans Affairs or one of the drug assistance programs 14 states have established for their low-income elderly and disabled.

Though this patchwork response to Medicare’s inadequate benefit package has functioned tolerably well for many participants in the past, it is inequitable and is starting to erode. There is little or no relationship between access to prescription drug coverage and the need for such insurance or the ability of participants to pay out-of-pocket for their drugs. For those lacking employer-sponsored retiree coverage or Medicaid eligibility, costs and availability can vary significantly. In some areas of the country there are no M+C plans through which a participant can obtain drug coverage. In other regions, participants must pay steep supplemental premiums to obtain M+C drug coverage. In still others, prescription drug coverage is part of the M+C plans’ basic or no-cost benefit packages. Similarly, premiums for Medigap policies that provide drug coverage vary more than five fold depending on the purchaser’s place of residence.

The consequences of this situation are not just financial. Those lacking coverage or having inadequate coverage are more likely to forgo filling prescriptions written by their physicians or to skimp on recommended dosages. Such behavior not only undermines the effectiveness of the medical care these patients receive but, in some cases, also result in complications that require more costly treatment later on.

While estimates suggest that about 65 percent of participants had some form of drug coverage in 1995, the figure is almost certainly lower today and likely to fall further in the future.

The coverage that remains is also likely to be less comprehensive and more expensive for beneficiaries. In response to rising costs and the 1992 Financial Accounting Standards Board statement (No.106) which required that the unfunded liability of retiree health plans be reported on corporate balance sheets, fewer firms will adopt retiree health benefits in the future and more of those that already offer such benefits will drop their coverage, raise the premiums they impose on retirees, or scale back the generosity of their benefits. Between 1994 and 1998, the fraction of employers offering health benefits to their Medicare-eligible retirees fell by one-quarter. The full effect of this retrenchment has yet to be felt because, for the most part, the cutbacks apply to workers who will retire in the future.

The restraints the Balanced Budget Act of 1997 (BBA97) imposed on payments to M+C plans and market pressures are causing these plans to scale back the generosity of their drug benefits. The fraction of M+C plans that provided drug coverage in their basic packages stabilized in 1998 after rising rapidly from 32 percent of all plans in 1993 to 68 percent in 1997.

All indications are that the fraction has now begun to fall. In addition, more plans are imposing caps on drug coverage and these caps are becoming more restrictive. In 2000, some 32 percent of plans will impose caps of $500 or less, up from 21 percent in 1999, and 82 percent of plans will set their maximum benefit at $2,000 or less. Copays will also rise—8 percent on average for generic drugs and 21 percent for brand-name pharmaceuticals. In short, while access to some drug coverage through M+C plans does not appear to be changing significantly, the generosity of the drug benefits offered by these plans is shrinking markedly and there is every reason to expect that this trend will continue.

Furthermore, after a period of fairly modest growth, Medigap premiums have begun to grow again at a rate faster than that of the incomes of the retired population. The premiums charged by the three plan types that provide limited drug coverage, which average around $2,000 a year, are already high relative to the incomes of Medicare participants who lack employer-sponsored retiree coverage. A few more years of increases along the lines of those of the past three years—increases in the 7 percent to 12 percent range—will make this source of limited drug coverage unaffordable to many.

The increasing importance of drug therapies to modern medicine and the erosion of access to affordable drug coverage are not the only reasons why there is growing interest in establishing some new mechanism to provide drug coverage for Medicare participants. Such coverage is also an essential component of the leading approach for restructuring Medicare to meet the fiscal challenges that await it in the 21st Century. Premium support—or competitive defined benefit—proposals would encourage competition both among M+C plans and between these plans and traditional fee-for-service Medicare. For such competition to function effectively, the standard benefit package that all M+C plans and the traditional Medicare offered would have to be sufficiently comprehensive so that few participants felt the need for supplemental policies. In other words, the benefit package would have to include some drug coverage. If this were not the case, dual insurance coverage, which is complex, confusing to participants and providers, inequitable, and costly, would persist. Adverse selection would continue to be a problem and the task of adjusting payments to plans for differential risk would be made more difficult.

The approaches

A large number approaches have been suggested to help Medicare participants pay for prescription drugs. These include:

  • allowing Medicare participants an above-the-line deduction from taxable income for prescription drug expenditures that exceed some threshold amount,
  • providing income tax credits to offset large out-of-pocket drug expenditures of Medicare participants,
  • giving states matching or block grants so that they can establish or expand targeted drug assistance programs for low-income Medicare participants,
  • offering Medicare participants separate prescription drug insurance policies through a FEHBP-like structure of competing plans,
  • establishing a stop-loss arrangement financed by government and the private sector to fully cover the pharmaceutical costs associated with chronic or devastating diseases that exceed a threshold,
  • mandating that manufacturers provide drugs at discounted price to retail pharmacies for sale to Medicare participants who lack prescription drug coverage,
  • requiring that all Medigap policies provide prescription drug coverage, and
  • adding prescription drug coverage to Medicare either as part of the mandatory benefit package or as an optional benefit.

While any of these approaches would reduce the burden that drug expenditures now impose on some Medicare participants and would begin to level the playing field between those who have and those who do not have prescription drug coverage, most would provide only a partial and temporary solution to the underlying problem. Moreover, most of these approaches would make the current system even more complex than it already is.

Prescription drugs are an integral and important component of modern health care and, therefore, should be incorporated into the basic Medicare health insurance package. To adopt some other approach will serve only to delay the inevitable. If stopgap measures that rely on tax expenditures or state grant programs are adopted now because they seem to be more affordable, undesirable inequities will be perpetuated and the eventual integration of drug coverage into Medicare may be made more difficult.

The structure of a prescription drug benefit

Policy makers wishing to design a workable system to assist Medicare participants with their prescription drug expenditures must address a such questions as:

  • Should the benefit provide insurance or assistance?
  • Should program eligibility be targeted, that is, limited to those with low incomes?
  • Should subsidies be provided only to those with low incomes or to all participants?
  • Should the benefit be mandatory or optional?

There are no right or wrong answers to these questions. Policy, budgetary, administrative, and philosophical considerations must come into play when answering them.

Insurance or assistance? From both the policy and budgetary perspectives, a prescription drug benefit should be designed to provide insurance protection—security against the possibility that needed pharmaceuticals will impose a financial burden that is large relative to the participant’s resources. All but the poorest participants, many of whom are eligible for Medicaid, should have the financial capacity to budget for a moderate level of out-of-pocket prescription drug expenditures each year.

Political considerations, however, seem to rule out designs that benefit only the minority of participants who incur catastrophic drug expenditures. The lack of popular appeal for catastrophic drug insurance was brought home most forcefully by the fate of the Medicare Catastrophic Care Act of 1988 (MCC). That legislation set a relatively high deductible—one that would be exceeded by only 16.8 percent of participants each year.

When fully phased in, the benefit would have paid 80 percent of approved drug costs above the deductible. The MCC was ignominiously repealed only 17 months after enactment because both the drug and the other catastrophic protections were concentrated on a small number of beneficiaries while the financing was spread broadly across all participants, many of whom concluded that they would receive no benefits over and above those they were already receiving through their employer’s retiree health plan.

Most employer-sponsored plans impose either no deductibles or relatively modest ones for drugs and, therefore, provide some assistance to a majority of participants. It seems likely that any successful drug plan for Medicare will have to follow this practice. This political reality, however, should not be used as a reason for denying true catastrophic protection to the small fraction of participants who face extraordinary drug expenses. In other words, drug benefits should not be capped as they are in most M+C plans, in the H, I, and J Medigap policies, and in the president’s drug proposal. Any new drug benefit should pick up all drug expenditures above some high level; ideally, one catastrophic cap should apply to the out-of-pocket expenditures for all covered services. Within any fixed budget, the resources needed to provide true catastrophic protection could be obtained by reducing the level of assistance to those faced with small and modest expenditures. Participants should accept a program whose assistance became more generous as the burden of drug expenditures rose.

Targeted or universal eligibility? Given the reality of limited resources and the fact that most moderate- and upper-income participants have either prescription drug coverage or access to affordable coverage, some have advocated restricting any new drug benefit to those with low incomes. This could be accomplished by providing states with grants to establish or expand state pharmacy assistance programs or by creating a new drug benefit within Medicaid that serves QMB, SLMB, and QI-1 beneficiaries. Such an approach, however, would be inconsistent with the overarching philosophy of social insurance which holds that, while financing can be income related, eligibility for social benefits should not be. There is no more logic to means-testing drug benefits than to means-testing home health, laboratory, or physicians services. Moreover, as pharmaceutical costs rise and private coverage continues to erode, the need for assistance to help pay for large drug expenditures will creep up the income distribution, eventually encompassing many middle-income retirees. For these reasons, any program to provide prescription drug coverage to the aged and disabled should be universal.

Broadly or narrowly based subsidies? Logically, subsidies in any universal prescription drug program should be restricted to those who do not have the resources to pay the full cost of this protection. The high option in the Breaux/Thomas restructuring proposal, which provided drug coverage as well as broader catastrophic protection, followed this precept. Unfortunately, it is not practical to target subsidies only to low-income beneficiaries as long as participation in a new program is voluntary. In such circumstances, those with middle and upper incomes who expect to incur high drug costs will find the program most attractive and the unsubsidized premiums will be driven up by adverse selection. A broad subsidy equal to the cost associated with adverse selection is probably the minimum needed to make a voluntary program workable.

Mandatory or optional participation? If given the choice, any rational Medicare participant should want to have some modest level of prescription drug coverage if it were available at an actuarially fair price. Judging from the similarity in the drug coverage provided by different employer-sponsored plans, it is likely that the vast majority of Medicare participants would be comfortable with the same modest drug plan, especially if those who wanted more extensive coverage were free to purchase supplemental policies. Mandating participation in such a fair plan, therefore, would not constitute a significant infringement on personal freedom. Nevertheless, as the MCC experience proved, requiring participation in a drug plan that is not heavily subsidized is politically infeasible as long as many enjoy heavily subsidized coverage through a former employer’s plan and others who are healthy and lack coverage are myopic and fail to perceive their long-run interests. Making participation voluntary, however, introduces the possibility of adverse risk selection. To overcome this hazard, policymakers can either provide broad subsidies that are sufficiently generous to make the coverage attractive even to healthy, non-risk adverse individuals or they can restrict enrollment in ways that encourage participation. For example, beneficiaries could be allowed to enroll only when they become initially eligible for Medicare benefits or when their employer-sponsored supplemental plan is no longer available to them. The president’s proposal, quite wisely, relies on both the carrot and the stick to ensure the broad participation that is necessary to ensure stability.


It will be no simple task to design and implement a workable and politically acceptable program to provide Medicare participants with adequate, affordable prescription drug coverage. Nevertheless, the need for such protection is great and will only grow in the future. Furthermore, adoption of a more up-to-date standard benefit package—one that covers prescription drugs, provides an out-of-pocket expenditure cap, and rationalizes copayments—is a necessary first step along the road to making Medicare a more efficient and effective program and allowing it to cope with the demographic and cost pressures that it will face in the 21st Century.