THE OREGON HEALTH PLAN --

LESSONS FOR THE NATION

Second of Two Parts

THOMAS BODENHEIMER, M.~.

EXPANSION PLANS

Under Oregon's 1989 legislative package, the Medicaid portion of the Oregon Health Plan was only one of its two critical pieces. The other was the requirement that employers provide insurance to employees, with the prioritized list as the minimal benefit package. Small businesses lobbied for the repeal of this part of the legislation, and on January 1,1996, it died. Gone was the hope of nearly universal health insurance in Oregon, with the prioritized list used for persons with incomes above the federal poverty level as well as Medicaid beneficiaries.
With the defeat of the requirement that employers provide insurance, Governor Kitzhaber sought alternative ways to insure the remaining 360,000 uninsured Oregonians. In November 1996, voters approved the governor's proposal for a 30-cent increase in the tobacco tax to generate additional funds for the expansion of the Oregon Health Plan. The funds are expected to allow 25,000 children to be added to the Oregon Health Plan and to provide an additional 21,000 people with state subsidies to purchase private health insurance. Kitzhaber's strategy is to shrink the ranks of the uninsured step by step; the goal is to reduce the proportion of uninsured persons to 5 percent, which may be as low a level as a nonmandatory system can achieve.
The use of state subsidies for private insurance has serious laws. The insurance policies may have high deductibles and limited benefits, and the prioritized list does not apply to these policies. The Oregon Health Action Campaign, a coalition of 110 senior citizen, community, church, and labor organizations that has been a supportive critic of the Oregon Health Plan since its inception, believes that the subsidies will be inadequate to allow low-income families to obtain private health insurance.2' The coalition had proposed that the new enrollees receive a
From the Department of Family and Community Medicine, University of California at San Francisco School of Medicine, San Francisco. Address reprint requests to Dr Bodenheimer at 1580 Valencia St., Suit 201, San Francisco, CA 94110.
~1997, Massachusetts Medical Society.
broad benefit package determined according to the same prioritized list that governs health services for current enrollees in the Oregon Health Plan.
PROBLEMS

Although the prioritized list has thus far created few difficulties within the Oregon Health Plan, there are numerous other problems related to the state's reliance on managed-care organizations to provide services for its poor citizens. Such problems raise questions about Medicaid programs in all 50 states.
The Safety Net
Like most states, Oregon has a safety net of health care providers for uninsured persons and Medicaid recipients who have difficulty obtaining care from private physicians or HMOs. Oregon's safety-net providers include the Oregon Health Sciences University, clinics run by the public health department, and a variety of urban and rural community health centers. Many of these clinics are federally qualified health centers receiving federal grants; others, known as "look-alikes," meet the federal requirements but have not received federal grants.22 In many states, city and county hospitals are the largest safety-net providers. In Oregon and nationwide, the explosive growth of Medicaid managed care threatens the financial viability of these safety-net institutions.
Medicaid managed care creates three problems for safety-net providers. The first problem concerns the provision in the Medicaid law that entitles community health centers (federally qualified health centers and look-alikes) to receive Medicaid payments on the basis of the actual cost of providing services. The 900 federally qualified health centers throughout the country, which serve a total of 10 million people at 3000 sites,23 often receive from $50 to $200 per visit for Medicaid recipients. These ample reimbursement rates have allowed community health centers to use Medicaid funds to subsidize care for the uninsured. However, section 1115 waivers may eliminate the requirement that community health centers be reimbursed on the basis of cost. In a number of states that have received section 1115 waivers, including Oregon, Tennessee, and Ohio, community health centers--most of which have subcontracts with private Medicaid managed-care plans--are receiving drastically reduced Medicaid reimbursement, threatening their survival.'6 Federally qualified health centers in Oregon's most populous county (Multnomah) are trying to cope with reimbursements as low as 25 percent of the level before the Oregon Health Plan was introduced. At safety-net clinics, medical visits are often complex, involving several acute or chronic physical or mental illnesses (sometimes related to infection with the human immunodeficiency virus [HIV]) exacerbated by psychosocial difficulties and the need for translation services. In Oregon, a number of safety-net providers may soon curtail services to persons who are uninsured or close down entirely. As one physician put it, "Those who remain uninsured may be worse off than before the Oregon Health Plan." In 1997, Oregon's legislature appropriated $3 million for the safety-net clinics, a small but welcome first step in addressing this problem.
The second problem concerns a law passed by Congress in 1987 to assist safety-net hospitals, which are often public, linked to universities, or both. The law channels extra federal and state Medicaid dollars to hospitals serving a disproportionate number of low income persons.24 By 1995, these payments to disproportionate-share hospitals accounted for $19 billion, or 12 percent of the total Medicaid budget.25 But section 1115 waivers in such states as Oregon and Tennessee allow the states to add these dollars to their general Medicaid funds without disbursing them to the disproportionate-share hospitals. As a result, Oregon's largest disproportionate-share hospital, the Oregon Health Sciences University hospital, is losing over $5 million per year in such payments. States such as California, which has a large proportion of Medicaid baneficiaries enrolled in managed-care plans but has not received a section 1115 waiver, must continue making the compensatory payments to hospitals with large numbers of Medicaid and uninsured patients. But these payments--a key subsidy to safety-net providers-- may be a major casualty of the Medicaid cuts currently being debated on Capitol Hill.
The third threat to the safety net is the effort to shift Medicaid patients from safety-net providers to private managed-care plans. As more and more states require Medicaid recipients to enroll in such plans, the plans are trying to attract those who are relatively healthy, leaving public hospitals and community health centers to provide care for Medicaid recipients with complex conditions requiring high-cost services (e.g., substance abuse and HIV infection) plus the uninsured--the ultimate in adverse risk selection.~5
To prevent the loss of Medicaid patients, the safety-net providers are attempting to establish Medicaid managed-care plans. In Oregon, a group of safety net providers--Oregon Health Sciences University, the Multnomah County Health Department (in Portland), and a number of community health centers--formed an HMO, CareOregon, to provide care for patients in the Oregon Health Plan. Similar alliances have been established in California. Community health centers in Connecticut, Rhode Island, New Jersey, Florida, Illinois, and Michigan are creating statewide HMOs.22 Twenty-six safety-net HMOs exist throughout the country, but they have not dodged the perverse logic of the marketplace. Michigan's safety-net HMO, for example, has been trying to obtain a state contract to provide Medicaid managed care in the state's most populous counties. Be
cause the federal government requires the safety-net HMO to be reimbursed at relatively high rates, however, the state prefers to contract with private HMOs, which can be reimbursed at lower rates. Difficulty obtaining contracts can also jeopardize public hospitals that form HMOs. In Houston, for example, the Harris County Hospital District came close to losing its bid for a Medicaid managed-care contract, which would have meant the loss of over $100 million in Medicaid reimbursements and disproportionate-share payments. The hospital district uses these funds to help care for 80 percent of uninsured residents in the Houston area.26
Adverse Risk Selection
CareOregon is Oregon's safety net turned HMO. Because its providers have been treating many seriously ill patients, CareOregon has serious disadvantages in the Medicaid marketplace.
Before the Oregon Health Plan was introduced, Oregon Health Sciences University and the Multnomah County Health Department each sponsored an AIDS clinic; the patients seen at these clinics became CareOregon enrollees. Small wonder that CareOregon, with only 9 percent of the enrollees in the Oregon Health Plan, provides services to 50 percent of the plan's HIV-infected patients who have CD4 counts below 500 per cubic millimeter. Yet the Oregon Health Plan has not provided CareOregon with higher capitation rates to compensate for this clear case of adverse risk selection.
Moreover, CareOregon provides interpreters for the 33 percent of its Oregon Health Plan enrollees whose primary language is not English. Interpretation services cost an extra $16 per patient visit. For most other health plans, non-English-speaking patients make up less than 10 percent of Oregon Health Plan enrollees. CareOregon's clinics and Oregon Health Sciences University were previously reimbursed for translation services but no longer are.
The example of CareOregon, along with the nationwide threats to safety-net institutions, raises questions about the privatization of health care for the poor through Medicaid managed-care plans. Private HMOs may scoop up the healthiest patients, leaving the high-risk Medicaid enrollees plus the uninsured to the public-sector providers. CareOregon continues to act according to its principles of public service, but the competitive marketplace exerts unrelenting pressure on that mission.
To address the problem, CareOregon has pushed hard for the Oregon Health Plan to adjust its capitation payments on the basis of risk. The state is responding and hopes to alter capitation payments to the health plans according to a complicated riskadjustment formula. The problem with risk adjustment is that physicians in health plans can manipulate the system by overreporting high-cost diagnoses.27 Oregon--along with Maryland, Colorado, Washington, Minnesota, and Michigan28--provides the nation with yet another health policy experiment in its plan to adjust capitation rates according to risk.
Premiums
To stabilize the Oregon Health Plan fiscally, Oregon instituted sliding-scale premiums for enrollees in late 1995. Families with no income are exempted from paying premiums; a family of four with an income of up to 49 percent of the federal poverty level pays $7.50 per month; a family of four with an income at the federal poverty level pays $28 per month. Families that have failed to pay their premiums are not re-enrolled in the Oregon Health Plan.
According to one consumer coalition, thousands of people have lost coverage or decided not to enroll in the Oregon Health Plan because they cannot afford the premiums.2i The Ore,gon Health Fornm reports that the premium requirements are knocking 1000 members of the enrollment list each month.29
Mental Health Services
Arranging for accessible, cost-effective services for the mentally ill has been a perpetual challenge.30 In the 1980s, for-profit corporations built psychiatric hospitals and made profits by hospitalizing patients for inappropriately long periods. Under managed care, these corporations have switched tactics and now earn profits by reducing hospitalization rates and lengths of stay and providing limited treatment for outpatients. Some state governments have contracted with for-profit mental health organizations to provide lower-cost mental health services for Medicaid patients, justifying the ensuing reduction in services by claiming prior overutilization. Reasonable voices in the mental health community oppose overly zealous cutbacks in mental health services, arguing that "two wrongs don't make a right."3'
Oregon has added generous mental health benefits to its prioritized list and in October 1997 will make these benefits available to all Oregon Health Plan enrollees through managed care. Mental health advocates in Oregon fear that the use of managed care to provide mental health services will lead to the situation that Tennessee is now facing. In 1996, TennCare contracted with two mental health organizations controlled by for-profit chains. By early 1997, Tennessee newspapers were publishing stories of turmoil in the mental health system. The community mental health centers with which the mental health organizations contracted to provide services saw their budgets cut by an average of 40 percent while the for-profit organizations kept 12 percent of the state mental health budget for administration and profits. One community mental health center closed; others were forced to reduce their staffs and services dramatically. These centers, which had been caring
for 30,000 uninsured people at little or no cost, are unable to continue serving this population. During 1996, the number of severely disturbed children cared for in therapeutic nurseries throughout the state fell from 840 to 72, and many day-treatment programs have been terminated. According to the director of one mental health center, thousands of people with severe and chronic mental illness are being denied even minimal services.32 Many of these findings have been corroborated by a report on a federal site visit.33 According to an executive of a mental health corporation that does not do business in Tennessee, the state's "mental health system is in the process of unravelling."34
Oregon is moving more cautiously. Twenty-five percent of the state's population has already participated in managed-care pilot projects for mental health services. When managed mental health care is instituted throughout the state, in October 1997, counties contracting with community mental health centers, the traditional safety-net providers for persons with severe and chronic mental illnesses, will be offered a major role. However, four of the Medicaid managed-care plans will accept mental health capitation and subcontract for mental health services. These subcontracts open Oregon's door to the national for-profit mental health organizations.
One such organization is Green Spring Health Services, which will provide mental health services under a subcontract with Blue Cross and Blue Shield's HMO Oregon, the largest HMO participating in the Oregon Health Plan. Green Spring was started by several Blue Cross and Blue Shield plans, and in 1995 Magellan Health Services, the nation's largest mental health provider, bought a majority interest in the company.35 Green Spring subsequently acquired Ceres Behavioral Health, an Oregon company. Nationwide, Green Spring has enrolled 16 million people in commercial contracts and Medicaid arrangements. Rhode Island has cited Green Spring for numerous deficiencies in services.36 The Washin~ton Post recently reported that Green Spring denied inpatient psychiatric services to a seriously disturbed Maryland adolescent, who killed himself three weeks after the denial.37 Green Spring is also one of the participants in TennCare's troubled managed-care program for mental health services.
Providing capitated mental health services to people with severe and chronic mental illnesses is risky. One study found that patients with schizophrenia had slightly worse outcomes under nonprofit capitation.38 Similar studies are needed in the for-profit sector. The introduction of managed mental health care in Oregon bears close watching.
Treatment of Elderly and Disabled Persons
Historically, Medicaid managed-care plans targeted poor families with children for enrollment. Nationally, this population accounts for 72 percent of Medicaid recipients but incurs only 29 percent of Medicaid expenditures.25 Services to elderly and disabled persons account for 59 percent of Medicaid expenditures. Enrolling this population in managed care plans is desirable for containing costs.
Oregon's section 1115 waiver allows the state to require most elderly and disabled Medicaid recipients to join managed-care plans for acute care. Most elderly and disabled persons receive comprehensive coverage through both Medicare and Medicaid. For many people with dual eligibility, capitated plans may result in reduced services--for example, requirements for prior authorization of emergency room and specialty care, early hospital discharge, restricted provision of durable medical equipment, and limited panels of providers. These requirements are particularly onerous for elderly and disabled people. In Oregon, a patient with dual eligibility enrolled in a Medicaid health plan that is also a Medicare HMO is required to join the Medicare HMO. Yet a recent national survey found that 26 percent of MedicareHMO enrollees would not recommend their health plans to family or friends with serious health problems.39 Moreover, in the large, multisite Medical Outcomes Study, low-income elderly patients with chronic illnesses who were enrolled in HMOs had markedly greater declines in physical health over a four-year period than such patients in fee-for-service systems of care.40 Disregarding these findings, state governments are increasingly requiring persons with dual eligibility--many of whom are poor, elderly, and disabled--to join Medicaid managed-care plans.
FISCAL UNCERTAINTY
Oregon is a fiscally conservative state, and the Oregon Health Plan is on a budgetary tightrope. The 1989 legislation creating the plan prohibits substantial reductions in the number of enrollees and does not allow cuts in reimbursements to providers; the HCFA will not allow the state to reduce substantially the number of covered treatments on the prioritized list. Ballot Measure 5, approved by Oregon voters in 1991, reduced property taxes; Ballot Measure 47, passed in 1996, further reduced property taxes; and the state has no sales tax. Moreover, the federal budget-balancing act may reduce federal Medicaid funds, which currently finance 62 percent of the Oregon Health Plan.
Thus far, the Oregon plan has enjoyed fair weather: a strong economy with higher-than-expected state tax revenues combined with a low rate of inflation in the costs of medical care. These ideal conditions will not last. When medical costs return to a higher rate of inflation, reimbursements to capitated health plans will have to match these cost increases. When the business cycle takes a downward turn, funds from the state legislature will be scarce. Without tax
reform, the Oregon Health Plan could face stormy times. The administrators of the plan are realistic people; they once placed a sign on the wall: "Cost, access, quality--pick any two."
In the first three years of the Oregon Health Plan, the state has allayed the concern that it is rationing care for the poor. The prioritized list is not being used to deny many needed services. Reasonable reimbursements to providers have kept most physicians in the Medicaid system. But if medical-cost inflation coincides with an economic downturn, the state and HCFA may be forced to move the line on the prioritized list upward, thereby reducing capitation payments to providers and driving them out of the plan. Governor Kitzhaber is intent on preventing such a situation and insuring more Oregonians. Perhaps three more years are needed before history makes its final judgment about Oregon's controversial experiment. In the meantime, the nation should look to Oregon as an important laboratory of health care reform.
REFERENCES
21. A summary of testimony provided to the House Human Resources Committee concerning HB 2207 Salem: Oregon Health Action Cam
paign, March 21,1997
22. SchaufRer HH, Wolin J. Community health clinics under managed competition: navigating uncharted waters. J Health Polit Policy Law 1996; 21:461-88.
23. Hawkins D, Rosenbaum S. The challenges facing health centers in a changing health care system. In: Altman S, Reinhardt U, Shields A, eds. The future U.S. health care system: who will pay? Chicago: Health Administration Press (in press).
24. Sparer MS. Medicaid managed care and the health reform debate: lessons from New York and California. J Health Polit Policy Law 1996;21:
433-60.
25. Where is Medicaid spending headed? Washington, D.C.: Kaiser Commission on the Future of Medicaid, December 1996.
26. Page L. Safety-net providers get a place in Medicaid managed care. American Medical News. June 16, 1997:3, 24-5.
27. Grumbach K. Physician payment and the game of risk. J Am Board Fam Pract 1996,9:360-5.
28. To create the right incentives, states move ahead with diagnosis-based risk adjustment in Medicaid. State Health Watch 1997;4(3):3, 6.
29. 1,000 people drop offevery month. Oregon Health Forum 1997;7(6): 4.
30. lglehart JK. Managed care and mental health. N Engl J Med 1996; 334:131-5.
31. Jellinek MS, Nurcombe B. Two wrongs don't make a right: managed care, mental health, and the marketplace. JAMA 1993;270:1737-9.
32. Richardson JD. Remarks to the TennCare Legislative Oversight Committee, January 13,1997
33. Tennessee mental health site visit report. Washington, D.C.: Health Care Financing Administration, March 17, 1997
34. Mental health center to close as sale fizzles. Commercial Appeal. January 1,1997:A1, A12.
35. Sharpe A. More states turn over mental-health care to the private sector. Wall Street Journal. January 24, 1997:A1, Al l.
36. Green Spring Health Services: Statement of deficiencies. Providence: Rhode Island Department of Health, October 26, 1995.
37. Boodman SG. Managed care comes to mental health. Washington Post Health (Washington Post). May 6, 1997:12-5.
38. Lurie N, Moscovice IS, Finch M, Christianson JB, Popkin MK. Does capitation affect the health of the chronically mentally ill? Results from a randomized trial. JAMA 1992;267:3300-4.
39. Nelson L, Brown R, Gold M, Ciemnecki A, Docteur E. Access to care in Medicare HMOs, 1996. Heaith Aff (Miilwood) 1997;16(2):148
40. Ware JE Jr, Bayliss MS, Rogers WH Kosinski M, Tarlov AR. Differences in 4-year health outcomes for elderly and poor, chronically ill patients treated in HMO and fee-for-service systems: results from the Medical Outcomes Study. JAMA 1996;276:1039-47