Health Affairs, 23, no. 2 (2004): 155-167 doi: 10.1377/hlthaff.23.2.155 © 2004 by Project HOPE	New Online   Pay Cuts For Medicare Docs   Access To Care Woes   Public Coverage More Efficient   Empowering Consumers

This Article Abstract Figures Only Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to Citation Manager Reprints & Permissions Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (9) Citing Articles via Google Scholar Google Scholar Articles by Draper, D. A. Articles by Short, A. C. Search for Related Content PubMed PubMed Citation Articles by Draper, D. A. Articles by Short, A. C. Related Collections Access To Care Managed Care - Medicaid Medicaid Safety-Net Systems State/Local Issues

Medicaid

Medicaid Managed Care: The Last Bastion Of The HMO?

Abstract

 
States rely on health maintenance organizations (HMOs) for their Medicaid beneficiaries because they offer guaranteed access to comprehensive benefits at a predictable cost. This is true despite movement away from HMOs, or at least the more restrictive variants, in the private sector. Plans that focus on Medicaid are becoming more central to states’ programs as commercial plans exit. Publicly traded, Medicaid-focused plans are also emerging. Medicaid participating plans are aggressively managing costs and care, contrasting sharply with commercial insurance where the trend is toward less intrusive managed care. In this context, state Medicaid managed care programs are facing important policy challenges related to plan participation, mainstreaming, and product design.

Reliance on managed care. Increasingly, states have chosen to rely on managed care arrangements to deliver coverage to both expanded and existing Medicaid populations. Since 1990 the nationwide Medicaid population has grown by nearly 60 percent, reaching more than forty million beneficiaries by 2002. This includes twenty-three million Medicaid beneficiaries now enrolled in managed care, including health maintenance organizations (HMOs) and primary care case management (PCCM) programs, a tenfold increase in the number of beneficiaries in these types of arrangements since 1990.²

Increased dependence on managed care, predominantly HMOs, comes at a time when there is movement away from this model in commercial insurance and Medicare.³ Between 1990 and 2002, Medicaid enrollment in HMOs grew from more than one million beneficiaries to more than seventeen million (Exhibit 1).⁴ Although total enrollment in HMOs nationally also grew during this time (by close to forty million), enrollment peaked in 1999 and has declined in each subsequent year.⁵

View larger version (38K): [in this window] [in a new window]   EXHIBIT 1 Enrollment In Medicaid Participating Health Maintenance Organizations (HMOs) Compared With Total HMO Enrollment Nationally, 1990–2002

Policymakers’ challenge. As HMOs serve more Medicaid beneficiaries, states’ efforts to ensure sufficient plan participation have become even more challenging. The flush state budgets funding much of the Medicaid expansion activity of recent years are gone. States are considering various cost reduction options for their Medicaid programs, including cutbacks in plan payment rates. Understanding the potential impact of such a change on plan participation is important for policy-makers so that they can develop appropriate strategies to minimize disruption.

Concerns, however, are not just limited to the number of participating plans but extend to the types and capability of surviving plans. At one time, states were intent on providing Medicaid beneficiaries access to mainstream plans and providers using commercial plans as the vehicle. However, the elimination by the Balanced Budget Act (BBA) of 1997 of Medicaid’s 75/25 rule (which limited a plan’s Medicaid membership to 75 percent of total membership), coupled with the extensive use of federal waivers by states to move large numbers of beneficiaries into managed care, helped plans that specialize in Medicaid to become more central to states’ programs.⁸ What this shift means for Medicaid beneficiaries—in terms of care delivery, access, and quality—is important for states to answer. It requires states to have the appropriate processes and systems in place to monitor and gauge the impact of these changing delivery systems. This may be especially challenging for some states where monitoring efforts historically have been problematic.⁹

It is also important for policymakers to better understand the effects of state and federal regulatory requirements that often dictate Medicaid managed care product design and features, and clearly differentiate this line of business from commercial insurance and Medicare products. For example, the components of the Medicaid benefit package that states design and purchase are generally fixed and nonnegotiable, and cost-sharing opportunities are limited. In the commercial insurance sector, purchasers (employers) actively modify benefit designs to control cost increases, and cost sharing is a widely used and accepted tool to manage use. Consequently, Medicaid participating plans must rely to a greater extent on tools other than cost sharing to manage care and control costs.

Data And Methods

Top Data And Methods Plan Participation Trends In... Medicaid Product Designs,... Policy Implications Editor's Notes NOTES

 
This paper is based on data collected from the CTS, a longitudinal study conducted by the Center for Studying Health System Change.¹⁰ The CTS uses multiple data sources, including site visits to twelve metropolitan areas every two years, to examine changes in local health care systems. The states in which the CTS markets are located are representative of states’ Medicaid managed care strategies. They vary in their experience with Medicaid managed care, including enrollment size and penetration rates (Exhibit 2). Two markets have particularly well-established, high-profile programs, including the Arizona Health Care Cost Containment System (AHCCCS) in Phoenix and CalOPTIMA in Orange County.¹¹ The states also vary as to the types of Medicaid managed care programs they operate—HMO, PCCM, or both (Exhibit 3). All but one state, Arkansas (with a PCCM program only), have Medicaid enrollment in HMOs. Overall, the Medicaid managed care population in the CTS states represents nearly half of all Medicaid managed care beneficiaries.

Plan Participation Trends In Medicaid

Top Data And Methods Plan Participation Trends In... Medicaid Product Designs,... Policy Implications Editor's Notes NOTES

 
In the twelve CTS markets, there are 30 percent fewer plans that participate in Medicaid now than in 1996 (Exhibit 4). The most dramatic change has been a decrease in the number of commercial plans (including Blue Cross Blue Shield plans) that participate. Although this shift partially reflects mergers and acquisitions among these plans since 1996, it also reflects these plans’ declining interest in Medicaid. In some cases, such as in Orange County, this shrinkage has been orchestrated by the Medicaid agency itself to reduce the administrative complexity of managing a large number of plan participants. While the number of participating plans that are Medicaid-focused has remained relatively steady since 1996, the plans that have remained in the program have expanded their memberships as Medicaid enrollment has grown and have strengthened their hold on this line of business as other types of plans exit.¹³

Medicaid-focused plans gain prominence. The majority of Medicaid participating plans across the CTS markets are Medicaid-focused. Ownership of these plans varies, but providers—often safety-net hospital or clinic systems—own the majority, followed by investors and other nonprofit concerns (Exhibit 5). In three markets—Boston, Greenville, and Indianapolis—Medicaid-focused plans are the only plans participating in Medicaid managed care. In other markets, including Cleveland, Lansing, northern New Jersey, and Phoenix, Medicaid-focused plans are the majority of participating plans or enroll the largest number of beneficiaries, or both.

New breed of Medicaid-focused plans emerges. A new "breed" of Medicaid-focused plans has recently emerged in the CTS markets: for-profit, publicly traded health insurance companies that specialize in Medicaid and are fast becoming major players in states’ programs. Amerigroup, for example, now operates in Miami and northern New Jersey; Centene operates in Indianapolis and northern New Jersey; and Molina, the most recent Medicaid-focused managed care plan to go forward with a public offering, operates in Seattle. AmeriChoice—which operates in Miami, northern New Jersey, Phoenix, and Syracuse—is similar in structure in that it is publicly traded and Medicaid-focused, but it was acquired by United-Healthcare in 2002 and operates as a separate (and brand-distinct) subsidiary.

Some publicly traded plans enter a market by acquiring other Medicaid-focused plans or the Medicaid membership of exiting plans. Centene, for example, entered the northern New Jersey market when it acquired provider-sponsored University Health Plans Inc. in 2002. Amerigroup used a similar strategy to enter the Miami market. Molina greatly expanded in Seattle when Aetna withdrew. By 2003 approximately one in five Medicaid participating plans in the CTS markets were owned by either Amerigroup, Centene, Molina, or AmeriChoice.

Provider sponsorship remains strong. In several of the CTS markets, provider (often safety-net provider) sponsorship of Medicaid participating plans remains strong, even though this type of sponsorship has declined in other sectors. This likely reflects a greater compatibility of the mission of these organizations with serving low-income, disadvantaged populations, as others have suggested.¹⁴ Providers’ involvement also reflects a strong desire to retain market share for the sponsoring safety-net or other system. In Miami, for example, Medicaid-focused JMH Health Plan is sponsored by the largest safety-net hospital system in the market: Jackson Memorial Health System. MDWise in Indianapolis is another Medicaid-focused plan that is sponsored by local health care systems. In Phoenix the county-owned public hospital system, Maricopa Integrated Health System, sponsors Maricopa Health Plan.

Medicaid Product Designs, Features, And Practices

Top Data And Methods Plan Participation Trends In... Medicaid Product Designs,... Policy Implications Editor's Notes NOTES

 
Plans participating in Medicaid managed care aggressively use conventional HMO designs, features, and practices, because of both state contracting requirements and the belief that these techniques are necessary and appropriate to effectively serve their members, which include an increasing number of disabled people. This differs greatly from the commercial HMO sector, where there has been extensive movement away from restrictive product designs in recent years.¹⁵ The restrictive approach in Medicaid appears relatively sustainable since beneficiaries have less financial or political clout than private consumers have to effect change. In addition, enhancements that have come about with Medicaid managed care, such as improved access to providers, help to sustain these models of care.¹⁶

The growing differentiation between Medicaid and commercially focused HMOs is made even more pronounced by regulatory limitations imposed by Medicaid. For example, Medicaid sharply limits cost sharing. In addition, federal regulations limit states’ ability to modify Medicaid’s comprehensive benefit package. As a result, the widely used practice of "buying down" benefits to reduce premium costs that commonly occurs in the commercial sector does not work in Medicaid.¹⁷

Cost sharing. Cost-sharing options are more limited in Medicaid than in the commercial insurance sector, where cost sharing is actively used to encourage consumers to use fewer services or to take a more active decision-making role in service use and product choice. Across many of the CTS markets, however, state officials reported that they are now exploring ways to increase copayments in their Medicaid managed care programs. The purpose is multifold—not only to create incentives to encourage Medicaid beneficiaries to become more involved in their health care decisions, but as an effort to better control utilization and to respond to high program cost growth and tight state budgets. In 2002, for example, Massachusetts implemented a $2 prescription copayment for Medicaid. In addition, states such as Arizona that are facing major budget pressures are proposing to expand cost sharing in their Medicaid managed care programs to help offset escalating costs.

Building cost sharing into the design of Medicaid managed care programs can, however, create tension. For providers, cost sharing may result in reduced reimbursement if they fail to collect from the patient. For Medicaid plan and state officials, the key is finding the right balance. On the one hand, these informants said that it is important to set the cost-sharing requirements at a level that is mindful of the low-income populations served and that does not erect barriers to the use of needed health care services. On the other hand, they added that cost-sharing requirements should also be set so that they control use appropriately. For example, a $10 copayment for the nonemergency use of hospitals’ emergency departments is being proposed in Washington State.

Utilization management. In contrast to the commercial sector, where there has been a strong movement away from more aggressive utilization management, Medicaid participating plans continue to rely extensively on the traditional tools of managed care, such as primary care gatekeeping, prior authorization, and concurrent review. This aggressiveness reflects in part Medicaid’s limited ability to use cost sharing as a way to manage use. However, it also reflects the unique needs of the Medicaid population, including fewer social supports and more complicated medical problems such as those often presented by the disabled. In response, participating plans often develop customized utilization management programs. In Miami, for example, one plan has nurses for Medicaid members on site in hospitals and skilled nursing facilities to coordinate care and manage service use. Nurses also conduct on-site concurrent reviews and collect discharge information to send to the enrollee’s primary care physician to ensure better coordination of information. Similarly, in Phoenix a Medicaid participating plan has on-site nurses doing concurrent review. A participating plan in northern New Jersey staffs a heavily used emergency room from 11 a.m. to 7 p.m. daily with a case manager to triage its Medicaid enrollees.

How service use is managed in Medicaid has contributed to the shift to Medicaid-focused plans because many commercial plans are no longer selling what Medicaid is buying: aggressive management of and intervention in beneficiaries’ care. UnitedHealthcare’s positioning of its AmeriChoice acquisition provides one example of how this has played out. Rather than relaxing utilization tools such as prior authorization as it did quite notably for its commercial population several years ago, United has actually embraced a more aggressive approach to medical management for its Medicaid members through AmeriChoice, a separate business unit of the firm with its own brand identification and well-established, active cost and care management programs in place.

Intensive case management initiatives. Medicaid participating plans across the CTS markets reported the extensive use of disease management programs. Some of these programs are contractually required by the state; others are operated at the discretion of the individual participating plans. The programs most commonly focus on asthma, diabetes, and high-risk pregnancies. Specialty programs that target the unique needs of the Medicaid population also exist in some markets. In Boston, for example, a Medicaid participating plan reported specialized disease management programs for AIDS, multiple sclerosis, and severe mental retardation. In northern New Jersey a Medicaid plan has special programs that focus on lead poisoning, sickle cell anemia, AIDS, cancer, transplants, and end-stage renal disease. In addition, a participating plan in Phoenix reported special programs for pain and congestive heart failure.

In addition, Medicaid participating plans reported the widespread use of intensive case management for their high-risk, high-cost members. While commercial plans also increasingly use intensive case management, it appears to be more aggressively used in Medicaid. For example, a Medicaid participating plan in northern New Jersey reported conducting one-on-one case management in members’ homes.

Similar to a growing trend in the commercial sector, some Medicaid plans are beginning to use predictive modeling as a care management tool.¹⁸ A participating plan in Boston reported the use of predictive modeling to identify and track the small percentage of its membership—1 percent or so—that is responsible for the highest spending. Once identified, these members receive more aggressive care management, often including disease management programs or intensive case management, or both.

Provider networks and payment. Compared with commercial networks that have become broad and inclusive, Medicaid participating plans in the CTS markets reported narrower networks for their Medicaid products. Networks may be limited to provider sponsors or traditional Medicaid providers such as community health centers. Commercial plans often develop a specific network for this line of business to meet access requirements and other Medicaid mandates (for example, required federally qualified health center contracts), although not always. Narrow networks also exist when providers are unwilling to participate in Medicaid.¹⁹ Providers cited low payment rates and the administrative burden associated with Medicaid as the main reasons for nonparticipation.

Despite movement away from risk-based contracting in commercial insurance, it remains common in Medicaid managed care.²⁰ Medicaid participating plans in Cleveland, Miami, northern New Jersey, Seattle, and Syracuse use primary care physician (PCP) capitation. Global capitation arrangements are used in Boston and Indianapolis, and shared risk arrangements exist in Orange County. Plans and providers believe that these arrangements reflect a reasonable balance of sharing financial risk and rewarding high-quality performance. For many traditional providers, risk-based payment also guarantees a flow of "insured" patients—patients with at least some funding who would otherwise use uncompensated care.

Policy Implications

Top Data And Methods Plan Participation Trends In... Medicaid Product Designs,... Policy Implications Editor's Notes NOTES

 
In today’s context of a changing managed care marketplace and tight budgets, states face major challenges in operating Medicaid managed care programs. States must grapple with several important policy issues.

Can states maintain sufficient plan participation? Although the evidence from the CTS markets indicates that plans’ exits have been considerably more widespread among commercial plans, executives of Medicaid participating plans generally expressed their concerns about payment rates’ not keeping pace with costs and an increasing number of what they perceived to be inadequately funded program requirements. Plans exiting Medicaid managed care in the CTS markets cited these factors as responsible for or contributing to their exit decisions. For now, Medicaid-focused plans have remained relatively steadfast in terms of participation, but by definition and design, they are considerably more dependent on this line of business than commercial plans are.

The growing role of publicly traded Medicaid-focused plans adds a new twist, one that is yet to be fully understood. It is not clear what it will take for these plans to remain financially viable and attractive to stockholders and what they expect from states in terms of support, financial or otherwise. As states continue to face serious budget problems, finding the funds to pay rates that plans perceive to be adequate to secure their continued participation will remain a challenge, at least in the foreseeable future.

Is access to mainstream plans and providers important? States are increasingly reliant on Medicaid-focused plans, as the evidence from the CTS markets suggests. This shift, however, is not necessarily a change in policy by states but rather a reflection of commercial plans’ lack of interest in participating in Medicaid managed care. However, it also reflects to some degree movement by commercial plans toward looser, less restrictive products such as preferred provider organizations (PPOs) and away from the more restrictive HMOs because of changing consumer demand in the commercial sector.²¹ As a result, states’ efforts to provide greater access to mainstream plans and providers by using commercially focused HMOs is becoming less realistic.

Whether states’ inability to provide access to mainstream plans and providers has any adverse effects on access, quality, or other patient care factors is largely dependent on the ability of Medicaid-focused plans to meet the many cost and care management challenges that Medicaid beneficiaries present while also complying with federal and state requirements. For now, most states appear to be satisfied with the current performance of their plan partners. Among the CTS markets, advocates for low-income consumers also appear generally satisfied, particularly since many states have recently expanded their Medicaid managed care programs to cover more people. However, it is difficult to determine how this view of advocates in the CTS markets compares with the views of advocates generally, since there is little, if any, information on the issue.

Do certain product designs and features matter? As findings from the CTS markets suggest, participating plans, with the support of states, aggressively manage the costs and care of their Medicaid members. They do this using many of the traditional product designs and features of earlier-vintage HMOs, including primary care gatekeeping, prior authorization, concurrent review, narrow provider networks, and risk-based provider contracting. Although such an aggressive approach to cost and care management was at the center of the private-sector consumer backlash of the late 1990s, states and participating plans believe that such an approach is both necessary and valuable in Medicaid, where resources are especially limited. This approach, state Medicaid officials said, has also allowed states to expand Medicaid coverage to more low-income people.

Continuing budget pressures will likely sustain the need of states and participating plans to aggressively manage Medicaid costs. Failure to do so may result in states’ rolling back eligibility or reducing benefits for the Medicaid-expansion populations, or both. Changes such as these would likely require states to step up their monitoring role to ensure that access and quality of care are not compromised in any way.

Is Medicaid managed care the last bastion of the HMO product? To answer this, it is important to first acknowledge that today’s Medicaid managed care is different from most commercial managed care in a number of ways. The differences exist because Medicaid’s resources are limited and becoming more so in light of states’ budget problems, so the product cannot be priced to reflect cost trends, as happens on the commercial side through underwriting. In states’ Medicaid programs, cost control is key, a factor that states find more compatible with the underlying theory of managed care—some restrictions on care in return for reasonable access and comprehensive benefits at a fairly predictable cost.

Editor's Notes

Top Data And Methods Plan Participation Trends In... Medicaid Product Designs,... Policy Implications Editor's Notes NOTES

 
An earlier version of this paper was presented at the AcademyHealth Annual Meeting in Nashville, Tennessee, June 2003. The authors gratefully acknowledge the work of the Community Tracking Study’s research teams, as well as the interview respondents who graciously contributed their time and insights. Paul Ginsburg, Cara Lesser, Laurie Felland, and Kyle Kinner provided valuable comments, as did three anonymous reviewers. The Robert Wood Johnson Foundation provided support for this research.

Debra Draper (ddraper{at}mathematica-mpr.com) is a senior health policy researcher at Mathematica Policy Research Inc. in Washington, D.C. Robert Hurley is an associate professor in the Department of Health Administration, Medical College, Virginia Commonwealth University, in Richmond. Ashley Short, formerly a research analyst at the Center for Studying Health System Change, is now a student at the University of Virginia Law School in Charlottesville.

NOTES

Top Data And Methods Plan Participation Trends In... Medicaid Product Designs,... Policy Implications Editor's Notes NOTES

 

1. This expansion followed a period of declining enrollment between 1994 and 1998, which largely reflected the effects of welfare reform.
2. Data for 2002 are from the Centers for Medicare and Medicaid Services, "Medicaid Managed Care Enrollment Report, Summary Statistics," 30 June 2003, www.cms.gov/medicaid/mcaidsad.asp (3 November 2003); total Medicaid enrollment data for 1990 are from CMS, "Medicaid Program Statistics (MSIS, formerly HCFA-2082 Report)," www.cms.gov/medicaid/msis/mstats.asp (program statistics—table Mcd90t01.wk1) (3 November 2003); and Medicaid managed care data for 1990 are from N. Kaye, Medicaid Managed Care: A Guide for States, 5th ed. (Portland, Maine: National Academy for State Health Policy, 2001).
3. The declining HMO enrollment trend noted in commercial insurance (predominantly employer-based) and Medicare is attributed in part to the managed care backlash of the late 1990s. It also reflects the tight labor markets during this same period, which saw employer recruitment and retention strategies include movement to less restrictive insurance options such as PPOs to appease employees. See D.A. Draper et al., "The Changing Face of Managed Care," Health Affairs (Jan/Feb 2002): 11–23.
4. Data for 2002 are from the Centers for Medicare and Medicaid Services, "Medicaid Managed Care Enrollment Report, Summary Statistics"; and data for 1990 are from Kaye, Medicaid Managed Care.
5. "HMOs nationally" refers to all HMOs across all insurance sectors—commercial, Medicare, and Medicaid. See InterStudy, Competitive Edge 13.1, Part II: HMO Industry Report (St. Paul: InterStudy, 2002); and Competitive Edge 8.1, Part II: HMO Industry Report (1998).
6. See R. Hurley and S. Somers, "Medicaid and Managed Care: A Lasting Relationship?" Health Affairs (Jan/Feb 2003): 77–88; Draper et al., "The Changing Face of Managed Care"; R.E. Hurley and D.A. Draper, "Medicaid Confronts a Changing Managed Care Marketplace," Health Care Financing Review 24, no. 1 (2002): 11–25[Medline]; and G.P. Mays, R.E. Hurley, and J.M. Grossman, "An Empty Toolbox? Changes in Health Plans’ Approaches for Managing Costs and Care," Health Services Research 38, no. 1, Part 2 (2003): 375–393.[CrossRef][ISI][Medline]
7. A Medicaid-focused plan is one that specializes in Medicaid, with at least 75 percent of its membership in Medicaid.
8. For further details, see Kaiser Commission on Medicaid and the Uninsured, "Medicaid and Managed Care," December 2001, www.kff.org/medicaid/206803-index.cfm (3 February 2004).
9. For additional information about Medicaid managed care monitoring, see, for example, J. Wooldridge and S.D. Hoag, "Perils of Pioneering: Monitoring Medicaid Managed Care," Health Care Financing Review 22, no. 2 (2000): 61–83[ISI][Medline]; and S. Felt-Lisk, "Monitoring Quality in Medicaid Managed Care: Accomplishments and Challenges at the Year 2000," Journal of Urban Health 77, no. 4 (2000): 536–559.[CrossRef][ISI][Medline]
10. For additional details on the CTS design, see P. Kemper et al., "The Design of the Community Tracking Study: A Longitudinal Study of Health System Change and Its Effects on People," Inquiry 33, no. 2 (1996): 195–206.[ISI][Medline]
11. AHCCCS is the umbrella system for many of Arizona’s public health insurance programs including Medicaid and SCHIP. CalOPTIMA is the managed care entity in Orange County that contracts with California’s Medicaid program, MediCal. It is a county-organized health system (a quasi-government organization), which is one of several Medicaid managed care models used in California.
12. Among the seventy-one plans, fifty-seven were commercial plans and fourteen were Medicaid-focused. The majority of commercial plans interviewed, however, did not participate in Medicaid.
13. See R. Hurley, "Medicaid-Focused Health Plans: A Community Health Conspiracy," Paper prepared for the Association for Health Center Affiliated Health Plans (AHCAHP) conference in Washington, D.C., May 2003, www.ahcahp.org/publications/Working%20Papers/hurley03.pdf (21 October 2003).
14. See B.H. Gray and C. Rowe, "Safety-Net Health Plans: A Status Report," Health Affairs (Jan/Feb 2000): 185–193.
15. Draper et al., "The Changing Face of Managed Care."
16. For additional information on care delivery enhancements, see, for example, M. Gold et al., "Participation of Plans and Providers in Medicaid and SCHIP Managed Care," Health Affairs (Jan/Feb 2003): 230–240.
17. More recently, however, states are increasingly using Health Insurance Flexibility and Accountability (HIFA) waivers to pare down benefit levels. States are doing this to expand health insurance coverage to more people within the constraints of existing Medicaid/SCHIP resources.
18. Predictive modeling programs are specialized software programs that assist plans with identifying and intervening with high-risk members. These modeling programs typically use a scoring system to predict a member’s expected health care costs over a designated period of time. The specific intervention is then designed based on this score, with members identified as high risk receiving more intensive services such as case management.
19. See P. Cunningham, Mounting Pressures: Physicians Serving Medicaid Patients and the Uninsured, 1997–2001, Tracking Report no. 6 (Washington: Center for Studying Health System Change, December 2002).
20. See R. Hurley et al., "A Longitudinal Perspective on Health Plan-Provider Risk Contracting," Health Affairs (July/Aug 2002): 144–153; and D.A. Draper and M.R. Gold, "Provider Risk Sharing in Medicaid Managed Care Plans," Health Affairs (May/June 2003): 159–167.
21. Since 1999, the year that HMO enrollment peaked nationally, enrollment has declined by more than six million people, or 8 percent, in these managed care arrangements. Over this same period the number of HMOs operating nationally decreased from 643 to 500—a decline of 22 percent. See InterStudy, Competitive Edge 13.1, Part II: HMO Industry Report (2002), and Competitive Edge 9.2, Part II: HMO Industry Report (1999).

                      What's this?

This article has been cited by other articles:

				Practice Notes: Strategies in Health Education Health Educ Behav, June 1, 2007; 34(3): 417 - 421. [PDF]

				R. Hurley Real Health Plans Manage Care Health Aff., November 1, 2006; 25(6): 1494 - 1496. [Abstract] [Full Text] [PDF]

				R. Labonte and M. Sanger Glossary on the World Trade Organisation and public health: part 2. J. Epidemiol. Community Health, September 1, 2006; 60(9): 738 - 744. [Abstract] [Full Text] [PDF]

				J. Kincheloe, E. R. Brown, J. Frates, K. T. Call, W. Yen, and J. Watkins Can we trust population surveys to count Medicaid enrollees and the uninsured? Health Aff., July 1, 2006; 25(4): 1163 - 1167. [Abstract] [Full Text] [PDF]

				C. R. Gresenz, J. Rogowski, and J. J. Escarce Dimensions of the Local Health Care Environment and Use of Care by Uninsured Children in Rural and Urban Areas Pediatrics, March 1, 2006; 117(3): e509 - e517. [Abstract] [Full Text] [PDF]

				S. K. Long and A. Yemane Commercial Plans In Medicaid Managed Care: Understanding Who Stays And Who Leaves Health Aff., July 1, 2005; 24(4): 1084 - 1094. [Abstract] [Full Text] [PDF]