This Article Abstract Submit a response to this article Services E-mail this article to a friend Alert me to new issues of the journal

M E D I C A R E  R E F O R M
M E D I C A R E  E X T R A 

4 October 2005  
  

Medicare Extra: 
A Comprehensive Benefit Option 
For Medicare Beneficiaries

A proposed Part E benefit would eliminate the need
for multiple supplemental programs in Medicare.

By Karen Davis, Marilyn Moon, Barbara Cooper, and Cathy Schoen

ABSTRACT:

The proposed Part E, Medicare Extra, outlined in this paper adds a comprehensive benefit option to Medicare, eliminating the need for beneficiaries to purchase a private drug plan and Medigap supplemental coverage. Financed by a budget-neutral beneficiary premium, it has the advantages of greater simplicity, efficiency, and value without adding to federal costs. Beneficiaries now enrolled in Medigap plans would save money, as could employers by choosing a lower-cost alternative to current retiree health plans. Eliminating some of the excess payments to Medicare Advantage plans would yield savings that could be used to help finance premium subsidies for low-income beneficiaries.

The medicare prescription drug, Improvement, and Modernization Act (MMA) of 2003 not only created the largest expansion of benefits in Medicare’s history, it also ushered in major changes to the program’s structure. One of the key structural decisions made in enacting the law was to offer prescription drug coverage only through private plans, either stand-alone private drug insurance plans or Medicare Advantage (MA) managed care plans. This decision marks the first time in the program’s history that a Medicare benefit will not be available through the basic program. With the new, separate Part D drug benefit, beneficiaries wishing to remain in the traditional fee-for-service (FFS) program and still have comprehensive coverage will now need three separate plans: basic Medicare Parts A and B, for hospital and physician services; Part D, a private prescription drug plan; and supplemental private coverage to help cover Medicare’s high cost sharing and protect against catastrophic costs.

This patching together of multiple plans creates confusion for beneficiaries; creates the potential for risk selection; and leads to higher administrative expenses, because of the multiple insurance carriers involved and the lack of integrated claims administration.¹ Medicare beneficiaries wishing to have a plan that integrates insurance benefits now must choose a private option such as MA.

Another approach would be to allow traditional Medicare to offer an option of comprehensive benefits in one plan. The Medicare Payment Advisory Commission (MedPAC) also examined such options before MMA was enacted; it concluded that more comprehensive benefits could be introduced and that the program would benefit from reduced insurance administrative expenses.² MMA and its further fragmentation of coverage intensify the need for a comprehensive benefit option under FFS Medicare, which we call Medicare Extra, or Part E. Our proposal draws on the benefits typically featured by employer plans, particularly the Federal Employees Health Benefits Program (FEHBP).³ We estimate the incremental cost per beneficiary of expanding traditional Medicare benefits and contrast those costs with Medigap premiums. We then discuss the advantages that beneficiaries would gain from having this additional option, and how such an option might improve the value and efficiency of health care.

Benefit structure. For purposes of modeling the cost of an illustrative comprehensive Part E benefit option, our analysis assumes that covered benefits would be the same as those now covered by Medicare, including hospital and physician services, home health care, and skilled nursing facility (SNF) care, as well as the new drug benefit that is scheduled to begin in January 2006.

The cost-sharing structure of Part E would be different, though. A single $250 deductible per person would replace the current deductibles for Parts A and B.⁴ Part B coinsurance would be reduced from 20 percent to 10 percent; Part A coinsurance for hospitals would be eliminated; and home health and selected preventive care would continue to be exempt from coinsurance. There would be no deductible for prescription drugs, and an average coinsurance of 25 percent would be assessed with no gaps in coverage.

Beneficiaries’ out-of-pocket outlays for all covered services, including prescription drugs, would be subject to a $3,000 ceiling. The ceiling would replace the current catastrophic drug benefit, which covers 95 percent of drug expenses after beneficiaries have paid $3,600 out of pocket for prescription drugs alone.
The benefit structure described above is comparable to the median preferred provider organization (PPO) coverage through an employer and the FEHBP Blue Cross Blue Shield standard option (Exhibit 1).⁵

Medicare Part E would have somewhat less comprehensive hospital and physician coverage than Medigap plans. Medigap Plan F, for example, has no hospital or physician deductible and no coinsurance on Part B services. If a $250 overall deductible and 10 percent coinsurance on physician services were imposed, beneficiaries electing the Part E option and dropping Medigap would experience greater cost sharing for hospital and physician services. However, the proposed Part E would have better prescription drug coverage than the new Medicare drug benefit and than current Medigap plans that include drug coverage (Exhibit 1).⁶

Estimated premiums and impact on beneficiaries. Our premium estimates and impact analyses are based on data from the 2000 Medicare Current Beneficiary Survey (MCBS).⁷ The survey includes information on beneficiaries’ use of services and out-of-pocket costs by type of service, as well as their sociodemographic information. Medicare claims data are linked to respondents. The MCBS adjusts total Medicare outlays for respondents to reflect actual Medicare program outlays for 2000. We corrected for underreporting of out-of-pocket costs by simulating required cost sharing under Medicare benefit design rather than using reported survey amounts. This approach allowed us to simulate consistent estimates of beneficiaries’ out-of-pocket spending and premiums. We began with Medicare’s reported spending for each type of service—based on administrative records—and used that to calculate beneficiaries’ “liability.” That is, our liability estimates reflect calculations of Medicare’s cost-sharing rules. We did this instead of using reported out-of-pocket spending and spending by insurers because each of the spending categories are self-reported and can capture noncovered services (particularly in the case of provider services). We also corrected for presumed underreporting in prescription drug use in a manner similar to that used by the Congressional Budget Office.⁸

We assumed that everyone would enroll in the new drug benefit under traditional Medicare. Having a baseline estimate of Medicare liability under current law, we then calculated a consistent estimate of Medicare liability that would result from the Medicare Extra option.
|
We used estimates from the Centers for Medicare and Medicaid Services (CMS) Office of the Actuary for 2004 to project current Medicare-covered services.⁹ We used differential growth rates for various Medicare components based on reported per capita averages. For drugs, we deflated 2006 estimates of spending and premiums to 2004 dollars. We modeled the new drug benefit as if it were in full effect in 2004. We used 2004 so that we could have reliable estimates of total Medicare spending.

We excluded some beneficiary groups from the analysis, including those in managed care, the institutionalized, end-stage renal disease (ESRD) patients, and Medicaid beneficiaries (dual eligibles). Doing so provided us with an estimate of beneficiaries who would be likely to participate in the Medicare Extra benefit and for whom we have reliable data.¹⁰

To keep the analysis simple, we did not attempt to simulate how use of hospital and physician services would change, but we do provide a range based on induced changes in use of prescription drugs. There are several reasons for these assumptions. Most importantly, cost sharing for hospital and physician services under Medicare Extra would both increase and decrease for different beneficiaries. While cost sharing would be reduced for those beneficiaries now covered only by Medicare, few low-income, “Medicare-only” beneficiaries are likely to enroll, given the premium. For other enrollees, cost sharing would go up for physician and other ambulatory services for anyone who switches from Medigap or employer-sponsored coverage to Medicare Extra. On the other hand, cost sharing would go down for drug coverage for all except those with generous employer coverage. Approximately 85 percent of beneficiaries affected by Medicare Extra would have higher Part B cost sharing—where spending would average about $350 per person. For drugs, coverage would improve for about 80 percent of beneficiaries. In this case, the impact could be larger, because $498 in spending on drugs would be newly covered under Medicare Extra, not Part D.

In its estimates of the cost of drug spending, the CBO assumed about a 9 percent increase in use resulting from the cost-sharing change itself, although that was fully offset by other savings from price reductions that would be achieved by private plans.¹¹ The improvements in drug coverage that would be included here could lead to higher use of drugs for those beneficiaries in the “doughnut hole.”

It could be argued that it is better to make no assumptions about net induction for physician or drug services since the effects could be largely offsetting. But for purposes of illustration, we used upper-bound estimates of an induced use effect, resulting in 10 percent higher drug spending on average. Estimates in Exhibit 2 thus show a range of potential cost-sharing changes for prescription drugs—with lower reductions in out-of-pocket spending as a result of assumed higher overall drug spending.

It is also important to compare what a typical Medicare beneficiary might experience in terms of premium changes and out-of-pocket spending by switching to Medicare Extra from Medigap. To do this, we calculated the average share of estimated Medicare liability now paid by Medigap plans. We assumed that the remaining share comes from out of pocket. To the actuarial value of Medigap coverage, we added a 20 percent premium-loading factor.

Study Findings

Cost per beneficiary. Under the current program, total spending per beneficiary for Medicare-covered services, including the Medicare drug benefit (Part D), for a typical beneficiary is an estimated $9,236 in 2004 dollars (Exhibit 2). Of this amount, the government pays 67 percent ($6,170) and beneficiaries are responsible for 33 percent.

We assumed that the new Part E benefit would be financed primarily by beneficiary premiums. Given the federal budget deficit and the large commitment of federal subsidies for the new drug benefit, it is unlikely that new budgetary resources could be committed to an improved Part E benefit.

We assumed that Part E spending, financed by a beneficiary premium, would increase 2 percent to reflect any additional administrative costs associated with running Medicare Extra. If there are no changes in the use of services, beneficiaries’ deductible and coinsurance expenses would be reduced by $1,081 and converted into a beneficiary Part E premium of $1,103 (or $92 per month, including 2 percent administrative costs) to cover the incremental cost of reducing Medicare cost sharing. Assuming an increase in the use of drugs, beneficiaries’ out-of-pocket drug costs would increase further, by as much as $50, but beneficiaries would also be receiving improved access to drugs.

A typical Medicare beneficiary would face lower costs under Part E than under Medigap (Exhibit 3). This finding is consistent with results from an earlier study before the Part D benefit was introduced.¹² Supplemental premiums would drop from an estimated $1,400 per year under Medigap to $1,103 under Part E.¹³ In addition, that beneficiary’s yearly out-of-pocket costs would drop from $933 to $873. The total savings to a beneficiary switching from Medigap to Medicare Extra would be $357 per year. Again, both out-of-pocket spending and beneficiary premiums would be slightly higher under our induced-demand estimates.

Medicare Extra could be expected to attract beneficiaries regardless of sex, age, or race/ethnicity, since the value of the improved benefit would not vary greatly across demographic groups. The greatest variation would be by age: The benefit’s value ranges from $943 for beneficiaries ages 65–74 to $1,231 for those age 85 and older (Exhibit 4).
About half of Medicare Extra’s improved benefits represent an improvement in prescription drug coverage. The net drug benefit would average $498 in 2004 dollars. The amount would vary, of course, depending on each beneficiary’s circumstances. The value of the improved drug benefit for disabled beneficiaries under age 65 would be $571, compared with $452 for beneficiaries ages 65–74.

Likely participation and potential for risk selection. The desirability of offering Part E depends on which beneficiaries would select this option. Would there be adverse selection, with sicker people opting for these more comprehensive benefits? Risk selection is a problem any time multiple options are provided.

The beneficiaries most likely to take up Medicare Extra would be those now covered by Medigap plans. The monthly average Medigap premium in 2004 was $144 for Plan F and $230 for Plan J.¹⁴ Although people now in Medigap plans are those most likely to switch to Part E, Medigap could continue to attract younger, healthier Medicare beneficiaries because of its practice of charging them lower premiums. To prevent this from happening, Medigap plans would need to be required to return to charging community- rather than age-rated premiums.

Some beneficiaries who are covered under retiree plans from an employer would also likely switch to Medicare Extra, particularly if employers could “buy” them into Medicare Extra by paying their premium. In 2003, large employers paid about $1,430 per beneficiary on average for retiree coverage, while the beneficiary’s share of the premium averaged $1,000. In about one-fifth of firms, the retiree paid the entire premium; in one-third of firms, the employer paid at least 80 percent.¹⁵ With high premium costs and administrative costs that typically run 10–15 percent in private retiree plans, employers might prefer to purchase coverage for retirees through Medicare Extra instead.

A major downside of expanding Medicare to include managed care plans has been these plans’ tendency to attract healthier patients and Medicare’s failure to adjust payment rates to reflect this relatively healthier enrolled population. MA premiums are typically lower than the $1,103 premium for Medicare Extra in part because they are receiving extra payments from Medicare along with favorable selection. MA plans are receiving an estimated extra payment of more than $500 per enrollee, or $2.75 billion as of 2004.¹⁶ The MA bidding system that will begin in 2006 could lower premiums even further. On the other hand, eliminating the favorable treatment of MA plans that comes from overpayment relative to FFS Medicare and failure to risk-adjust relative to FFS Medicare would likely cause plans to increase premiums or reduce benefits. As a result, beneficiaries who now enroll in MA might switch to Medicare Part E. Other beneficiaries might remain in MA because they prefer the health care delivery system through which they receive their care (for example, a group- or staff-model health maintenance organization [HMO]). Disabled beneficiaries who do not qualify for managed care plans and who rely only on Medicare might opt for Medicare Extra in anticipation of high expenses. Creation of Medicare Extra would provide an alternative to MA that permits choices to be made according to beneficiaries’ preferences and the relative efficiency of both options, rather than providing no comprehensive FFS alternative, as is the case with current law.

One means for reducing adverse selection would be to automatically enroll beneficiaries in Medicare Extra (unless they actively chose either traditional Medicare basic coverage or MA). When beneficiaries become eligible for Medicare today, they are automatically enrolled in traditional Medicare, and their Part B premiums are automatically deducted from their Social Security checks. The same could be done for Medicare Extra. An extensive literature suggests that many people prefer not to make active choices and would likely participate in Part E if initially enrolled in this option.¹⁷

Equitable access for low-income beneficiaries. Medicare’s high cost sharing is especially daunting for beneficiaries with low incomes, particularly those with no supplemental coverage. Assistance with Part B premiums and other cost sharing is available to some low-income beneficiaries through Medicaid and the Medicare Savings Programs. Medicare Savings Programs, which are run by the states and jointly financed by the states and federal government, subsidize all Medicare cost sharing for beneficiaries with incomes below 100 percent of the federal poverty level and premiums for those with incomes below 135 percent of the poverty level. Assets are limited to $4,000 per individual and $6,000 per couple.¹⁸ Fewer than 60 percent of those eligible are enrolled in these programs.¹⁹ Beginning in 2006, low-income beneficiaries meeting specified income and asset levels will be eligible for federal assistance with drug coverage premiums and cost sharing under Part D.²⁰

A basic principle of any benefit design is to help low-income beneficiaries afford to choose any of the Medicare options—the proposed Medicare Extra, MA, or private drug plans. This could be accomplished by subsidizing Part E premiums under the Medicare Savings Programs or by federalizing supplemental coverage for low-income beneficiaries. A federal Medicare Supplemental Insurance program could provide premium assistance for those with incomes below, say, 135 percent of poverty, with partial subsidies up to 150 percent of poverty—effectively replacing Medicaid supplemental coverage for acute care services. Federal administration would likely increase the participation of those eligible for low-income subsidies. Not requiring an asset test would simplify enrollment and ensure coverage of all low-income beneficiaries, most of whom have quite modest assets.

Obviously, this policy would not be budget-neutral. States could be required to continue to contribute to subsidizing the Part E premium for low-income beneficiaries (MMA already requires this), even if supplemental coverage were federalized. Savings achieved from paying MA plans on a par with FFS Medicare and risk-adjusting payments could also help finance low-income subsidies, but additional federal funds would likely be needed. One estimate determined that federalizing Part B premiums and Medicare cost sharing just for the 15 percent of Medicare beneficiaries now enrolled in Medicaid would cost about $6.5 billion in 2002.²¹

Dual eligibles have much poorer health status than beneficiaries with retiree supplemental coverage or those who purchase Medigap coverage. If dual eligibles were included in Medicare Part E, it would be important to federally subsidize their extra costs to avoid driving up the cost of the option for everyone else.

Other issues. Beneficiaries choosing Part E would not have to participate in a separate private drug plan; rather, the drug benefit would be run in the same way that other parts of Medicare operate under the traditional program. This would greatly simplify the operation of the program. Anyone enrolling in Part E would pay a premium that would be the equivalent of the B, D, and E premiums described above. Medicare would contribute the actuarial value of Part D to this benefit in the same manner in which it is calculated for private plans. In this way, the A/B deductible would not raise any coordination issues, for example, since all E enrollees would be receiving A, B, D, and E services. To beneficiaries, the program would be one unified whole.

Enrollment in Part E would assume that the E premium would fully pay for the additional services, even though this would mean some risk selection as compared with traditional Medicare. In assessing who should receive the largest subsidies, we believe that aside from people with employer-subsidized insurance, the main enrollees in traditional Medicare would be those with lower incomes, above the level protected by low-income subsidies, who believe that they cannot afford the extra benefits. In the spirit of this approach, we would also propose that private MA plans not receive additional subsidies. Resources saved from that higher cost could go to additional low-income protections.

The new drug benefit under Part E could be administered for Medicare by one or more pharmacy benefit management (PBM) firms in each region. It is assumed that such entities might choose to use tiered formularies or other adjustments as long as they average a 25 percent copayment.

Potential Advantages Of Medicare Extra

Medicare Parts A and B require extensive beneficiary financial participation. They do not have a catastrophic limit on out-of-pocket spending, and in 2004 a $67 monthly premium for Part B physician and other services, a deductible of $876 for hospital care, a $100 deductible for physician and other services, and 20 percent coinsurance for physician and other ambulatory services were imposed. Even when MMA drug coverage begins, beneficiaries will still pay, on average, about two-thirds of their drug expenses.²² For the sickest beneficiaries, out-of-pocket expenses can be financially devastating.²³

High deductibles and coinsurance have been shown to impede access to needed care and impose the greatest hardships on people with serious illnesses. For example, beneficiaries without any supplemental coverage are 2.5 times more likely than those with supplemental coverage to report delaying care because of cost.²⁴ Low-income beneficiaries, in particular, report skipping doses or even not filling prescriptions, to make their medications last longer.²⁵ Making cost sharing and coinsurance more affordable under the proposed Part E would reduce barriers to essential care and reduce the need to purchase other supplemental coverage.

Adding a Part E option to Medicare should also reduce beneficiaries’ dissatisfaction with the new Medicare law. Part E would eliminate the doughnut hole for drugs and would put a more affordable ceiling on out-of-pocket expenses for all services. Rather than requiring beneficiaries to pay $3,600 in drug costs out of pocket to reach a catastrophic benefit that is estimated to benefit only 5–6 percent of beneficiaries, Part E would cap all out-of-pocket expenses at $3,000.²⁶

Medicare Extra should be able to lower administrative costs. More than one-fourth of Medicare beneficiaries (28 percent) now purchase Medigap coverage.²⁷ By combining basic and supplemental benefits in a single, comprehensive package, Medicare Extra would eliminate the need to coordinate two sources of coverage—Medicare and Medigap—for the same services. The Medicare program, furthermore, has lower administrative costs than Medigap plans. Most Medigap plans are sold on an individual basis and thus have high marketing and enrollment costs; administrative costs for Medigap policies average at least 20 percent (compared with Medicare’s 2 percent).²⁸ An earlier MedPAC study also concluded that administrative costs would be lower and that health spending would not rise if beneficiaries gave up supplemental plans.²⁹

Combining all benefits into a single benefit package might increase Medicare’s ability to design new benefits, such as disease management programs, which might cause some expenses (prescription drugs) to rise while lowering others (hospitalization). Medicare Extra would also replace the first-dollar (no deductible) supplemental coverage that many beneficiaries buy under Medigap with modest cost sharing. The net effect could be more cost-conscious use of health care services, without running the risk of underuse of effective services. However, underuse might remain a concern for lower-income beneficiaries, who would require reduced cost sharing to assure their access to effective care.

One of the major advantages of Part E would be that beneficiaries could obtain benefits in one plan, rather than needing a private drug plan and a Medigap plan to go with basic Medicare. About 17 percent of beneficiaries are enrolled in private MA plans. Their decision to enroll in these HMOs is driven primarily by the desire for lower premiums or more comprehensive benefits, including prescription drugs—not because they are preferred as a system of care.³⁰ Offering Part E would allow them to choose between a comprehensive private or public plan.
Finally, MA plans have been unreliable for participants in many markets and have cost the government extra money.³¹ The prior experience with managed care plans under Medicare+Choice, MA’s predecessor, was quite problematic.³² As a result, beneficiaries enrolled in managed care plans could not count on affordable and stable coverage.

Medicare Extra would have the advantages of lower administrative costs and lower provider payment rates. MA plans, however, might prove to be more flexible and innovative than the CMS can be—for example, quicker to implement effective, high-cost care management techniques or establish elite networks of high-quality, high-efficiency providers.

Perhaps more importantly, providing comparable benefits in the Part E and managed care parts of the program would provide a genuine market test and choice. Forcing beneficiaries to enroll in managed care plans or in multiple sources of coverage to obtain comprehensive benefits runs counter to the principle of “choice.” Doing so also artificially increases enrollment in private plans.

Employer-sponsored retiree coverage has been eroding during the past twenty-five years, and the decline is predicted to continue.³³ Fewer employers are expected to offer supplemental coverage to new retirees, and those that do are expected to reduce their contribution. As this source of supplemental coverage declines or becomes less affordable, beneficiary dissatisfaction is likely to increase. Providing an affordable option within Medicare could both offset this dissatisfaction and help stabilize retiree coverage.

Beneficiaries with retiree coverage whose employers pay substantial portions of the premium or have more generous benefits could be expected to prefer to remain with their employer coverage. Nevertheless, employers might well prefer to purchase coverage for their retirees through Medicare Extra rather than through their current plan. There are considerable savings to be realized for employers. Firms would avoid the 10–15 percent administrative loading cost, which even large employers pay for health benefits. In addition, there would no longer be duplicative administrative costs of coordinating benefits with basic Medicare. Having access to a more predictable and affordable source of retiree coverage could well induce more employers to continue retiree health benefits.

Giving employers the option to purchase Medicare Extra would also broaden the beneficiary pool in terms of age and health status. Employers that now have very generous retiree plans might continue to wrap around even Medicare Part E benefits, picking up the deductible and other services, such as dental care, that are covered by retiree plans but would not be included in Part E. Perhaps a more realistic expectation, however, is that providing employers with access to a more affordable option for retiree health benefits would at least help stabilize and protect retiree coverage.

Concluding Comments

Medicare beneficiaries have consistently expressed higher satisfaction with their coverage than nonelderly enrollees in employer health plans have.³⁴ They are much more confident that they will be able to obtain care when needed, and they report fewer problems obtaining access to care. They are less likely than people with employer coverage to report negative insurance experiences and problems paying medical bills. Improving Medicare’s benefits should build on this record and permit those beneficiaries who are genuinely pleased with the traditional FFS program to consolidate their benefits there.

A comprehensive Part E benefit offered by traditional Medicare would have strong advantages for beneficiaries. Beneficiaries would get the benefits they want at lower cost and with much less confusion and complexity. It would create choices for beneficiaries and a genuine choice between Medicare Extra and MA plans. A strong role would remain for private insurers, both offering MA plans and serving as fiscal intermediaries in paying claims for basic Medicare and Medicare Extra. But savings also would be achieved, by having the government assume fiscal risk and negotiating rates with providers. In addition, employers could save money on retiree health benefits. Such an alternative should be a part of the debate over the future of Medicare and any midcourse corrections to the new drug benefit.

The authors thank January Angeles and Alice Ho for research assistance, and Stephen Schoenbaum and Jeanne Lambrew for helpful comments.

NOTES

1. Medicare Payment Advisory Commission, Report to the Congress: Assessing Medicare Benefits (Washington: MedPAC, June 2002).
2. Prior to passage of a prescription drug benefit, MedPAC described several scenarios for providing a comprehensive benefit package, although they did not specify where it would be offered. Ibid.
3. K. Davis, B.S. Cooper, and R. Capasso, The Federal Employee Health Benefits Program: A Model for Workers, Not Medicare (New York: Commonwealth Fund, November 2003).
4. A $100 per admission charge is assessed for beneficiaries not exceeding the deductible.
5. Henry J. Kaiser Family Foundation/Health Research and Educational Trust, Employer Health Benefits: 2004 Annual Survey (Menlo Park, Calif.: Kaiser Family Foundation, 2004).
6. Both Medigap H and J have 50 percent coinsurance and a cap on benefits of $1,250 under Medigap H and $3,000 under Medigap J. This coverage would be eliminated under the new Medicare prescription drug law that includes a $250 deductible, 25 percent coinsurance on expenses between $250 and $2,250, and no coverage between $2,250 and $5,100.
7. For a description of the Medicare Current Beneficiary Survey (MCBS), see G.S. Adler, “A Profile of the Medicare Current Beneficiary Survey,” Health Care Financing Review 15, no. 4 (1994): 153–163.
8. Congressional Budget Office, A Detailed Description of CBO’s Cost Estimate for the Medicare Prescription Drug Benefit (Washington: CBO, July 2004).
9. Social Security Administration, The 2004 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Disability Insurance Trust Funds (Washington: SSA, March 2004).
10. More detailed information on the estimation procedure can be obtained from the authors.
11. In fact, the CBO estimated that without a Medicare drug benefit, spending would average $2,894 in 2006. With a drug benefit, per capita spending would fall slightly to $2,878. See CBO, A Detailed Description.
12. MedPAC, Report to the Congress.
13. Average Medigap premiums are estimated to be $1,727 in 2004 (estimates based on MedPAC, Report to the Congress, and adjusted for inflation from 2000 to 2004). The higher actual Medigap premium could reflect the cost of including services not covered by Medicare, or the 20 percent loading factor assumption could be too conservative.
14. Trended to 2004, it would equal about $144 per month or $1,724 annually. The average premium for Plan J, which includes a drug benefit capped at $3,000, is conservatively about $2,764 annually in 2004 dollars.
15. Kaiser Family Foundation and Hewitt, Retiree Health Benefits Now and in the Future: Findings from the Kaiser/Hewitt 2003 Survey on Retiree Health Benefits (Menlo Park, Calif.: Kaiser Family Foundation, 2004).
16. B. Biles, L.H. Nicholas, and B.S. Cooper, The Cost of Privatization: Extra Payments to Medicare Advantage Plans (New York: Commonwealth Fund, December 2004).
17. R.W. Kopcke, J.S. Little, and G.M.B. Tootell, “How Humans Behave: Implications for Economics and Economic Policy,” New England Economic Review (First Quarter 2004): 21–26.
18. L. Summer and L. Thompson, How Asset Tests Block Low-Income Medicare Beneficiaries from Needed Benefits (New York: Commonwealth Fund, May 2004).
19. M. Moon, N. Brennan, and M. Segal, Improving Coverage for Low-Income Medicare Beneficiaries (New York: Commonwealth Fund, August 1998).
20. Beneficiaries with incomes below 135 percent of poverty will pay no premiums and only modest copayments; beneficiaries with incomes 135–150 percent of poverty will face a $50 deductible and a sliding-scale premium. The income and asset levels (of $6,000 for an individual and $9,000 for a couple with incomes below 135 percent of poverty; $10,000 for an individual and $20,000 for a couple at 135–150 percent of poverty) are more generous than those under Medicare Savings Programs. See Summer and Thompson, How Asset Tests Block Low-Income Medicare Beneficiaries.
21. B. Bruen and J. Holahan, Shifting the Cost of Dual Eligibles: Implications for States and the Federal Government (Washington: Kaiser Family Foundation, November 2003); and MedPAC, A Data Book: Healthcare Spending and the Medicare Program (Washington: MedPAC, June 2004).
22. M. Moon., How Beneficiaries Fare under the New Medicare Drug Bill (New York: Commonwealth Fund, June 2004). Includes an annual premium of $418.
23. S. Maxwell, M. Storeygard, and M. Moon, Modernizing Medicare Cost-Sharing: Policy Options and Impacts on Beneficiary and Program Expenditures (New York: Commonwealth Fund, November 2002).
24. MCBS Cost and Use File, 2000.
25. D.G. Safran et al., “Prescription Drug Coverage and Seniors: How Well Are States Closing the Gap?” Health Affairs, 31 July 2002, content.healthaffairs.org/cgi/content/abstract/hlthaff.w2.253 (19 October 2004).
26. R. Adams, “Few Will Benefit from Medicare Coverage for Catastrophic Drug Costs,” CQ HealthBeat, 8 October 2004.
27. MedPAC, A Data Book.
28. MedPAC, Report to the Congress.
29. Ibid.
30. C. Schoen et al., Medicare Beneficiaries: A Population at Risk, Kaiser/Commonwealth Fund 1997 Survey of Medicare Beneficiaries (New York: Commonwealth Fund, December 1998).
31. Biles et al., The Cost of Privatization.
32. G. Dallek, B. Biles, and L.H. Nicholas, Lessons from Medicare+Choice for Medicare Reform (New York: Commonwealth Fund, June 2003); and M. Gold and L. Achman, Average Out-of-Pocket Health Care Costs for Medicare+Choice Enrollees Increase 10 Percent in 2003 (New York: Commonwealth Fund, August 2003).
33. L.A. McCormack et al., “Retiree Health Insurance: Recent Trends and Tomorrow’s Prospects,” Health Care Financing Review 23, no. 3 (2002): 17–34; and L. Murray and F. Eppig, “Insurance Trends for the Medicare Population, 1991–1999,” Health Care Financing Review 23, no. 3 (2002): 9–15.
34. K. Davis et al., “Medicare versus Private Insurance: Rhetoric and Reality,” Health Affairs, 9 October 2002, content.healthaffairs.org/cgi/content/abstract/hlthaff.w2.311 (1 November 2004).

Karen Davis (kd{at}cmwf.org) is president of the Commonwealth Fund in New York City. Marilyn Moon is vice president and program director at the American Institutes for Research in Silver Spring, Maryland. Barbara Cooper is a consultant in McLean, Virginia. Cathy Schoen is vice president, Health Policy, Research, and Evaluation, at the Commonwealth Fund.

DOI: 10.1377/hlthaff.W5.442
©2005 Project HOPE–The People-to-People Health Foundation, Inc.