|
In designing a drug benefit for Medicare, policymakers must make four fundamental decisions:
The answers to each of those questions will have a major impact on the program's cost to the federal government and to enrollees. (This chapter and the next describe, in general, how various approaches to those choices affect costs. Specific cost estimates for particular proposals are discussed in Chapter 4.)
In addition, most of those decisions will involve trade-offs among the various goals for a Medicare prescription drug benefit. For example, features designed to encourage wide enrollment--such as extensive coverage, low deductibles and copayments, broad eligibility, and subsidies for low-income enrollees--will also make the program more expensive for the federal government (and possibly for state and local governments). But features designed to control costs--such as narrow coverage, high cost sharing by enrollees, and limited eligibility--may mean that fewer Medicare beneficiaries receive the benefits of prescription drug coverage.
Any program in which the federal government subsidizes a large share of drug costs for many enrollees will require substantial federal outlays. Medicare beneficiaries are expected to spend an average of nearly $2,500 apiece on outpatient prescription drugs in 2003. Access to better drug coverage would undoubtedly stimulate further spending. However, decisions about how to administer the drug benefit can exacerbate or alleviate the federal cost burden. Consequently, those decisions are especially important.
All of the recent proposals for a Medicare drug program would rely on private entities to administer the benefit, but there are significant differences in the functions envisioned for those entities. Opportunities exist to manage drug use and prices prudently, but pharmacy benefit managers need both the incentives and the tools to do so. The Congressional Budget Office has concluded that among the designs for a Medicare drug program proposed in the past few years, those with certain administrative features offer the greatest opportunity for constraining federal costs and total spending on outpatient prescription drugs. Although not without shortcomings, those designs have three main features: they allow benefit managers to employ the full array of tools now used to administer private-sector drug plans, they force benefit managers to compete among themselves for enrollees' business, and they make managers assume financial risk for delivering benefits.
The Structure of the Coverage
The single biggest determinant of the cost of a Medicare drug benefit is how the coverage is designed. Choices about the structure of the coverage include:
The possible variations on those choices are numerous and could produce benefits unlike anything available in the private sector today. Of the four proposals discussed in Chapter 4, the Clinton Administration's plan and H.R. 4680 both include a capped benefit, then a "hole" (a level of spending at which there would be no drug coverage), and finally a stop-loss provision, beyond which the benefit would pay all drug costs (see Figure 3). Plans currently available through employers, by contrast, almost never include such holes in coverage. The larger the range of spending encompassed by the hole, the less costly the program would be--but also the less coverage the benefit would provide.
The structure of the coverage also indirectly affects the cost of the drug program through the out-of-pocket spending it requires of enrollees. Benefits with low cost-sharing rates would encourage enrollees who were newly shielded from paying the full costs of their drugs to use a greater number of--and more expensive--prescriptions. Conversely, higher cost-sharing rates would induce less new drug spending.
Eligibility for Enrollment
A fundamental design choice concerns whether the prescription drug benefit would be available to all Medicare beneficiaries or only some (on the basis of such criteria as income, wealth, or current drug coverage). The cost of the drug program could be reduced by limiting eligibility. However, most recent proposals for Medicare drug coverage would allow all Medicare beneficiaries to enroll, although they would provide much higher subsidies to low-income beneficiaries.
A second choice for policymakers is whether to make enrollment in the benefit voluntary and, if so, with what restrictions. A voluntary drug program that did not limit when and how often an eligible person could enroll would encourage beneficiaries to sign up only when they expected to incur high drug costs and to opt out again when they expected little need for prescription drugs. All of the proposals examined in Chapter 4 would provide a voluntary drug benefit, but they would restrict that choice by giving people only one opportunity to enroll without penalty (when they first became eligible), or by imposing a surcharge on people who delayed enrollment, to reflect the higher costs that such enrollees typically entail. (When the drug program began, however, all Medicare beneficiaries, regardless of age, would have a chance to sign up for the coverage.)
Besides those two restrictions, another way to reduce beneficiaries' incentives to enroll or drop coverage at will would be to couple the drug benefit with Part B of Medicare (Supplementary Medical Insurance). In that case, beneficiaries could choose either Part B plus drug coverage or no Part B and no drug coverage. Because Medicare currently pays 75 percent of Part B benefits, and because enrollment in Part B without penalty is restricted to a brief period after eligibility begins, virtually all Medicare beneficiaries sign up for Part B. Linking the drug benefit to enrollment in Part B would probably ensure reasonably high participation in the benefit. However, if enrollment in Part B required enrollment in the drug program and the full premium (for Part B plus drugs) rose substantially as a result, some people who would otherwise enroll in Part B might drop that coverage.
The Level and Structure of Federal Subsidies
Two important design choices are how much the federal government would contribute to the cost of drug coverage for Medicare beneficiaries and how such subsidies would be structured. Those design choices have consequences not only for federal costs but also for costs to state and local governments. In addition, the level of subsidy offered to enrollees would have an important effect on people's willingness to take part in the drug program.
Most of the Medicare drug benefits proposed recently in the Congress would provide some level of federal subsidy for all enrollees, but they would still require those enrollees to contribute substantially--through both cost sharing and premiums. To make drug coverage more affordable to low-income Medicare beneficiaries, most proposals would provide a higher federal subsidy for enrollees who met certain eligibility criteria.
Subsidies for All Enrollees
Even if the federal subsidy was small, enrollment would still be fairly high because many people with private or public drug coverage would probably be required to enroll. For example, employment-based health plans would probably require retirees eligible for a Medicare drug benefit to participate in it, just as they now effectively require that retirees participate in Part B. Even employers who offered to pay Medicare's drug premium for retirees would have lower costs so long as that premium was subsidized to any extent (assuming their retirees were not markedly less costly than the average Medicare participant). The more comprehensive Medicare's drug coverage was, the more employers' health care costs for retirees would be reduced, but they would probably take advantage of even a limited drug benefit. Likewise, state Medicaid agencies, even if not required to do so, would choose to enroll people eligible for both Medicare and Medicaid in a Medicare drug program if states' costs under the new program were less than under Medicaid's current drug benefit.
Thus, the people whose participation would be most sensitive to the size of the federal subsidy would be the one-quarter of Medicare beneficiaries without drug coverage from private or public sources. If the drug benefit was not subsidized and enrollment was always open, few of those beneficiaries would be likely to enroll. If, however, they had only one chance to enroll without financial penalty, some of those beneficiaries would probably take part without any federal subsidy, and most would choose to enroll at subsidy rates well below 100 percent of the cost. The reason is that virtually all beneficiaries would expect to receive more in benefits in some years than the premiums they paid. One-time-only enrollment would also increase the extent of participation for any given level of subsidy. However, a drug benefit that was financed mainly by enrollees could be difficult for some people to afford--specifically, people whose income or assets were only slightly too high to qualify for drug coverage through Medicaid or other public programs.
Subsidies for Low-Income Enrollees
Who Would Be Eligible? As noted earlier, some low-income Medicare beneficiaries receive assistance for part or all of their medical costs through the federal/state Medicaid program. Those beneficiaries fall into three categories.
Most proposals for a Medicare drug benefit include some form of low-income subsidy for beneficiaries in all of those categories.
In addition, most recent Medicare drug proposals would assist other low-income Medicare beneficiaries. Proposals typically call for providing subsidies to all enrollees with income below 135 percent of the poverty level (and limited assets), which would cover their premiums and cost sharing for prescription drugs. Enrollees with income between 135 percent and 150 percent of the poverty level would have some or all of their premiums subsidized. Many proposals would extend the subsidies to enrollees with even higher income. (A few proposals would remove limits on assets.)
How Much Would States Pay? A key design choice for low-income subsidies is how much would be funded by the federal government and how much by the states. The federal government now pays 57 percent of Medicaid costs, on average, with the states covering the rest. Some proposals for a Medicare drug benefit would maintain the current federal percentage for dual eligibles and QMBs. However, the federal government would pay the full subsidy for other low-income enrollees (those who met special income and asset limits established for low-income subsidies in the drug program). Other proposals would require the federal government to pay the full cost of the subsidies for all low-income enrollees, including dual eligibles. In many of those cases, states would be required to reimburse the federal government for some share of the amount they saved under the proposal.
A proposal that increased the federal government's share of the cost of low-income subsidies would reduce states' costs, and vice versa. The low-income subsidy would also augment or (at the state's option) replace state-run drug programs for low-income seniors who are not eligible for Medicaid. As noted earlier, 34 states either have introduced or are planning to introduce such programs. A federal subsidy of Medicare beneficiaries in those programs would directly reduce state spending.
Who Would Participate? The cost of a low-income subsidy program would ultimately depend on how many people participated. Not all eligible Medicare beneficiaries would choose to receive a low-income subsidy even if they enrolled in the drug benefit. Some might want to avoid being associated with a government "welfare" program; others might not believe that they were eligible for or needed the subsidy.
What entity was chosen to administer the low-income subsidy program could affect the level of participation. CBO estimates that only 50 percent of people eligible for QMB subsidies and 30 percent of those eligible for SLMB subsidies take part in those programs today. Most recent proposals for a Medicare drug benefit would rely on state Medicaid agencies to determine eligibility and enroll low-income beneficiaries--as the QMB and SLMB programs do now. However, another option would be to have the Social Security Administration (SSA) provide those enrollment services. Participation would be higher under that arrangement because less "welfare" stigma is associated with SSA than with Medicaid.
As was the case with subsidies for all enrollees, the size of the low-income subsidies would also influence participation. A larger subsidy would almost certainly induce more people to take part in the low-income program.
That effect would also depend on the design of the Medicare drug coverage. High deductibles or premiums might persuade eligible low-income beneficiaries to sign up for the low-income subsidy to cover those up-front costs. Similarly, the more generous the coverage of drug expenses beyond the deductible, the stronger would be the incentive to enroll.
Perhaps the major issue affecting participation by low-income beneficiaries is whether the asset standards now in place for Medicaid would be relaxed for the Medicare drug benefit. Most proposals would retain the asset standards currently used to determine QMBs and SLMBs. However, less-stringent standards would expand the number of people eligible for low-income subsidies.
The Effect on Spending for Medicaid. Adding prescription drug coverage to Medicare would alter not only federal spending for that program but also federal and state spending for Medicaid. Such coverage would reduce Medicaid's costs for dual eligibles because Medicare would pick up part of their prescription drug costs. However, some of that reduction would be offset by higher enrollment in the Medicaid program. Some people who are now eligible for Medicaid do not enroll; a Medicare drug benefit would give them a new incentive to do so, because under most proposals, state Medicaid programs would administer the low-income subsidies. Thus, people applying for Medicare drug coverage under the low-income subsidy would learn about their eligibility for Medicaid and enroll in that program at the same time.
Other Factors Affecting the Cost of Low-Income Subsidies. The effect on federal costs for low-income subsidies would depend not just on the factors discussed above but also on the interplay between the coverage provided by a Medicare drug benefit and the provisions for low-income subsidies. In general, increasing cost sharing for enrollees in the drug benefit would lead to higher federal subsidies for low-income enrollees. Conversely, reducing enrollees' cost sharing would result in lower federal costs for low-income subsidies.
Administrative Approach
The way in which a Medicare drug program was administered could also have a significant impact on its cost. Most recent proposals envision the approach--now common in the private sector--of using organizations such as pharmacy benefit managers (PBMs) to administer the benefit. The proposals differ, however, in the number of such organizations that would serve a region, the restrictions they would be subject to, the basis on which they could compete for enrollees, and whether they would assume any insurance risk. (Insurance risk occurs when the entity providing coverage is liable for payments that may not be fully covered by premiums or reimbursements or, conversely, is allowed to keep part of surpluses when costs fall short of premiums or reimbursements.)
In the past decade, PBMs have come to play a central role in administering prescription drug benefits in the private sector (see Figure 4). Their main function is to act as a health plan's agent in administering a drug benefit. PBMs do not distribute prescription drugs to patients, except when they own mail-order or community pharmacies. Instead, prescription drugs flow from manufacturers to dispensing pharmacies (often with stops along the way at wholesalers' warehouses) and then to consumers, either at pharmacies or by mail. A pharmacy pays a manufacturer for the drugs it purchases and in turn charges a price it has negotiated with the PBM. The consumer and health plan share the responsibility for paying that price.
PBMs perform various functions and use various management tools. They process and pay claims. On behalf of a health plan and its enrollees, they also negotiate price discounts with dispensing pharmacies and rebates with drug manufacturers. In return for receiving rebates, PBMs try to steer enrollees toward preferred or formulary drugs. (A formulary is a list of drugs that the health plan will cover. Nonpreferred drugs are covered and therefore included in the formulary, but they typically entail higher cost sharing for the enrollee than preferred drugs do.) PBMs' other strategies for lowering costs include encouraging the use of generic drugs and dispensing drugs through mail-order pharmacies, which many large PBMs own. PBMs may establish and enforce differential cost-sharing requirements to encourage enrollees to select lower-cost drugs. In addition, because they keep centralized records of each enrollee's prescriptions, they can help prevent inappropriate dosages and harmful drug interactions.(2)
Although PBMs in the private sector often have considerable leeway in the tools they can use, they do not assume any insurance risk for the drug benefits they administer. However, they may have a bonus added to, or a penalty subtracted from, their administrative fee on the basis of how well they meet preset goals for their performance.
Some of the Medicare drug proposals developed during the 106th Congress, such as the Clinton Administration's, called for a single PBM selected periodically to serve each region, with all insurance risk borne by Medicare, not the PBM. Other proposals, such as H.R. 4680, adopted a different approach: they would use multiple risk-bearing entities (such as PBM/insurer partnerships) that would compete to serve enrollees in each region. Enrollees would have some choice among entities, so people who were willing to accept more restrictive rules (such as a limited formulary) in return for lower cost sharing or premiums could do so, while others could select a more expensive plan with fewer restrictions. Hybrid models have been proposed in which multiple entities would compete for enrollees without bearing insurance risk. They could compete on the basis of such features as lower drug prices, fewer limits on covered drugs, and wider networks of pharmacies.
Each administrative model has its pros and cons. The next chapter describes how each model addresses key problems inherent in designing a Medicare prescription drug program--such as the risk to plans from enrolling a disproportionately large share of people with high expected drug costs, the need to control drug spending, and the feasibility of administering the benefit efficiently.
|
