In the United States, the Medicare Prescription Drug,Improvement, and Modernization Act (MMA) of 2003defines a series of new Medicare private health planoptions to be added in January 2006. Health policy analystshave begun to weigh in on the challenges thatthese new options present to Medicare providers andbeneficiaries, raising issues regarding program complexity,risk selection, stability, geographic inequity, andplan overpayment.1-3 Despite the myriad concerns, theBush Administration appears committed to implementationof MMA in 2006.
One issue that has received little attention to date is theimpact of the health plan changes on Medicare coveragepolicy. When the act is fully implemented, beneficiariescan choose to stay in fee-for-service Medicare and opt fora stand-alone drug benefit, or select among availableMedicare Advantage (formerly called Medicare+Choice)plans that include drug coverage. Medicare Advantage willoffer a new regional Preferred Provider Organization(PPO) option. PPOs will contract with providers, butenrollees can seek services outside the provider network,paying additional out-of-pocket costs. The Centers forMedicare and Medicaid Services (CMS) recentlyannounced 26 PPO regions and 34 regions for the freestandingprescription drug plans.4 Medicare Advantageregional plans must serve the entire region. Existing county-wide local plans can continue, but the intent of MMA isto encourage the development of regional PPOs.
Whatever delivery model a beneficiary selects, allplans must offer the standard Medicare benefit package.The existing Medicare coverage process will continue todetermine whether and under what conditions newtechnologies and other services will be covered.
MMA offers challenges and opportunities for the coverageprocess. In some ways it further complicates the procedures.Fortunately, the MMA also directs CMS to makeimprovements in national and local coverage determinations.We urge CMS and the US Congress to focus on theseissues and strongly recommend some additional steps.
1. Align Existing Contractor Regions Withthe PPO Regions
Medicare has been drifting into geographic incoherence,raising significant challenges for programaccountability and equity. The vast majority of coveragedecisions are made by local contractors—carriers for PartB and Fiscal Intermediaries for Part A. In 2003, CMS contractedwith 20 carriers and 27 Fiscal Intermediaries,who develop local coverage policies. Many of these organizationshold more than 1 contract creating informal multicontractregional networks.5 The local coverage policyprocess is highly variable from contractor to contractorin terms of resources, productivity, use of evidence, andtimeliness.6 In addition, consolidating contractor networkshas eroded the values of "local"responsivenessbecause these networks are now geographically diverse.5
The addition of PPO regions to this already complexmix will create myriad difficulties for local coverage policy.The PPO regions are not consistent with the existingcontractors'regions. Presumably, the fee-for-service sideof the program will continue to be subject to local coveragepolicies. However, what about a regional PPO? Inmany instances, a PPO region may overlap different contractorareas, with different coverage policies required forthat PPO, depending on where the services are rendered.
For example, Minnesota is currently served by a carriernetwork that includes Minnesota, Illinois, Michigan,and Wisconsin. Minnesota's Fiscal Intermediary holdscontracts in Minnesota and North Dakota.7 Minnesota'snew PPO region includes Montana, Wyoming, NorthDakota, South Dakota, Nebraska, and Iowa. These statesare served by 5 different Fiscal Intermediaries and 4different carriers, all of whom issue coverage policies.Additionally, these states report to 3 different CMSregional offices (Iowa and Nebraska to Kansas Cityregional office; Montana, North Dakota, South Dakota,and Wyoming to Denver; and Minnesota to Chicago).8Because some larger states comprise a single PPO region,not all will experience the same level of complexity.
We recommend that the program simplify its regionalorientation by redrawing contractor regions to alignwith the PPO regions and reflect more rational boundaries.In addition, we recommend that CMS combinethe functions of Part A and Part B contractors to achievea single contractor per PPO region. As a result, allMedicare plans and providers within a region would besubject to the policies of one contractor.
2. Develop a Process for Allocation ofLocal and National Decisions
The MMA calls for CMS to consider how to obtaingreater consistency and rationality in its local coverageprocess. Alignment of contractor regions, if adopted,will provide more geographic balance, but will notinherently provide greater consistency in coverage policiesacross regions.
Currently, most coverage decisions are local. Aninterested party can request a national coverage decisionor CMS can initiate one.9 We recommend that CMSadopt a process to guide the locus of coverage decisionsin order to obtain greater consistency. A series of factorscould be considered.
One option is to redefine which decisions can bemade locally, allocating some to the national level basedon category. Research has shown that there are 3 typesof coverage decisions—new technology, technologyextensions, and utilization management.10 In relation tonew technology, these authors found that local contractors:have fewer resources or less expertise to performevidence-based technology assessments; rely heavily onother carriers and Fiscal Intermediaries; and vary widelyin terms of when the contractors implement policiesrelated to new technology. While the policies emerge atdifferent times, often they are quite similar—evidencethat the contractors copy the decisions of others. Thisoutcome contrasts with utilization management policies,which show more variation because they respond to concernsabout utilization of widely diffused procedures tiedto variations in practice patterns from region to region.
Other factors CMS could consider include programcosts (ie, more costly technologies or widely used technologiesundergo national evaluation), uncertainty (ie,interventions of uncertain value go through the morerigorous national assessment process), and the likelycosts and benefits of collecting additional information.There has been concern that centralization could potentiallystifle innovation.11 Existing creative nationalcoverage policies along with the reforms discussed inthe next recommendation will go a long way towardaddressing these concerns.
It is time to abandon decentralization for its ownsake, use national policy to encourage efficiency andequity, and develop regional policymaking in morefocused and specific ways.
3. Reform the National Process
Medicare also makes national coverage decisionseach year for some 10 to 15 technologies projected tohave a major impact on the program, or for which thereis considerable clinical controversy.
In recent years CMS has implemented a series ofreforms aimed at making its national coverage processmore open, participatory, and evidence-based.9,12-14The reforms have worked to a good extent: the availabilityof information on the CMS Web site has made itsubstantially easier to follow national coverage decisionsand the rationale underlying them; the creation ofthe Medicare Coverage Advisory Committee has providedCMS with external expertise and advice, as well as apublic forum for deliberation. CMS decisions have generallybeen consistent with the strength of evidence.15The MMA seeks to open and improve this processfurther by imposing stricter standards for CMS reviews,by requiring CMS to make its rationale for decisionspublic, and by adding a public comment period beforenational coverage decisions are implemented.16 Thelegislation also creates a council for technology andinnovation within CMS to coordinate coverage, coding,and payment for new technologies.
Still, more could be done. For one thing, the MMAwill tighten timeframes for making decisions, yet CMSis not currently meeting its own self-imposed timeframes.More resources earmarked for national coveragedecisions would help, akin to giving the US Foodand Drug Administration more resources to speed drugapprovals. Medicare could also move to establish clearcriteria for coverage, including a role for formal cost-effectivenessanalysis. This could help CMS targetresources to improve health more efficiently.17
Yet history suggests that adding cost-effectivenesswill be difficult. Medicare has tried on several occasionsto add cost-effectiveness criteria but was thwarted bystrong political opposition.18-21 MMA is conspicuouslysilent about cost-effectiveness, but Congress and CMSshould continue to contemplate ways to use this technique.For example, Congress should consider cost-effectivenessanalysis when adding new services tothe basic Medicare benefit package, and CMS shouldconsider such analysis when implementing congressionallymandated benefits. CMS should also requirecost-effectiveness analysis for new demonstrationprojects (as it does with MMA's replacement drugdemonstration, under which Medicare will temporarilycover oral drugs prescribed as substitutes for injectabledrugs).
Congress should also provide additional resources tothe Agency for Health Care Research and Quality forresearch to inform local and national coverage decisions.Research shows the quality of much of the evidenceavailable to CMS for coverage decisions is fair or poor.15Section 1013 of MMA provides funds for "comparative-effectiveness"(not cost-effectiveness) research, but only$15 million of the original $50 million targeted for thispurpose was appropriated. This amount should be increasedand the scope expanded to include cost-effectivenessanalysis, as well as value-of-information analysis,which could help determine priorities for research.22
Finally, Medicare should expand its use of creativeapproaches, such as "coverage under protocol,"wherebyCMS conditionally covers a technology while theagency enrolls beneficiaries in clinical studies to determinethe effectiveness of the technology.23 Coverageunder protocol continues the agency's longstanding policyto cover new technologies with conditions. That is,CMS rarely issues blanket coverage policies, but insteadrenders more nuanced judgments that place restrictionsbased on patient characteristics or setting of care.Coverage under protocol also recognizes an importantreality about evidence: all decisions reflect a judgmentabout the costs and benefits of collectingadditional information. CMS should also consider otherinnovative arrangements, including requiring manufacturersto pay for patient registries and exploring risksharingarrangements whereby CMS covers atechnology but holds manufacturers at risk if expectationsabout its effectiveness do not develop.
Implementation of the MMA presages major changesfor Medicare. Millions of beneficiaries and providers willface many new and confusing choices. It is essential thatMedicare mesh the new options with the existing program.This is an opportunity to update and improve theentire coverage process as well. If the recommendationspresented here are considered, we believe thatMedicare's coverage process will be more accountable,efficient, and equitable for all beneficiaries.
From Division of Health Services Research and Policy, School of Public Health,University of Minnesota, Minneapolis, Minn (SBF); and Harvard School of Public Health,Boston, Mass (PJN).
Address correspondence to: Susan Bartlett Foote, JD, Division of Health ServicesResearch and Policy, School of Public Health, University of Minnesota, 420 DelawareStreet SE, MMC 729, Minneapolis, MN 55455. E-mail: email@example.com.
Health Affairs Web Exclusive
1. Berenson RA. Medicare disadvantaged and the search for the elusive "level playingfield." [serial online]. December 15, 2004;W4-572 -W4-585. Available at: http://www.healthaffairs.org. Accessed January 20, 2005.
Health Affairs Web Exclusive
2. Biles B, Dallek G, Nicolas LH. Medicare Advantage: déjà vu all over again?[serial online]. December 15, 2004;W4-586 - W4-597. Available at http://www.healthaffairs.org. Accessed January 20, 2005.
3. Achman L, Gold M. Are the 2004 payment increases helping to stem MedicareAdvantage's benefit erosion? December 2004. Pub 795.Available at: http://www.cmwf.org. Accessed January 20, 2005.
4. HHS announces regions to administer new Medicare prescription drug benefitand Medicare Advantage program [press release]. US Dept of Health & HumanServices, December 6, 2004. Available at: http://www.hhs.gov/news/press/2004pres/20041206.html. Accessed January 20, 2005.
5. Foote SB. Focus on locus: evolution of Medicare's local coverage policy. 2003;22(4):137-146.
Health Aff (Millwood).
6. Foote SB, Wholey D, Rockwood T, Halpern R. Resolving the tug-of-war betweenMedicare's national and local coverage. 2004;23(4):108-123.
7. Centers for Medicare & Medicaid Services. Medicare coverage database.Available at: http://www.cms.hhs.gov/med/index_article_bycontractor.asp.Accessed January 20, 2005.
8. Centers for Medicare & Medicaid Services. CMS regional offices—informationfor professionals. Available at: http://www.cms.hhs.gov/about/regions/professionals.asp. Accessed January 20, 2005.
9. Medicare program: revised process for making Medicare national coveragedeterminations, 68 55634-55641 (2003).
Am J Manag Care.
10. Foote SB, Halpern R, Wholey D. Variation in Medicare's local coverage policies:content analysis of LMRPs. 2005;11:181-187.
11. AdvaMed rejects GAO findings on Medicare local coverage [press release].May 13, 2003. Available at: http://www.advamed.org/publicdocs/PR-177.htm.Accessed January 20, 2005.
12. Medicare program: establishment of the Medicare coverage advisory committeeand request for nominations for members, 63 68780 (1988).
13. Medicare program: procedures for making national coverage decisions, 6422619-22625 (1999).
14. Medicare coverage policy—coverage process. Criteria for making coveragedecisions, 65 31124-31229 (2000).
Health Aff (Millwood).
15. Neumann PJ, Divi N, Beinfeld MT, et al. Medicare's national coverage decisions,1999-2003: quality of evidence and review times. 2005;24(1):243-254.
16. Medicare Prescription Drug, Improvement, Modernization Act of 2003: Lawand Explanation. 2nd ed. Chicago, Ill: CCH Inc.; 2004: sec 1013.
N Engl J Med.
17. Gillick MR. Medicare coverage for technology innovations — time for new criteria?2004;350:2199-2203.
18. Medicare program: criteria and procedures for making medical services coveragedecisions that relate to health care technology, 54 4302-4318 (1989).
The New York
19. Pear R. Medicare to weigh cost as a factor in reimbursement. April 21, 1991:A1.
20. Neumann PJ. Why don't Americans use cost-effectiveness analysis? 2004;10:308-312.
J Health Politics Policy Law.
21. Foote SB. Why Medicare cannot promulgate a national coverage rule: a case ofregula mortis. 2002;27:707-730.
22. Claxton K, Cohen JT, Neumann PJ. When is evidence sufficient? 2005;24(1):93-101.
The New York
23. Kolata G. Medicare covering new treatments but with a catch. November 5, 2004:A1.