Retracing the Oregon trail: the experience of rationing and the Oregon health plan

A decade ago the state of Oregon attracted worldwide interest when it began an ambitious attempt to set priorities for health care on a systematic basis. Stimulated by the death of a 7 year old boy who had been waiting for a bone marrow transplant operation, and led by John Kitzhaber, a doctor turned politician, Oregon passed legislation in 1989 designed to provide access to health insurance for all residents. A key part of the strategy was to increase eligibility for Medicaid, a publicly funded programme of health care for people with low incomes, by including in the programme all citizens with an income at or below 100% of the federally defined poverty level. To keep the costs of this policy within affordable limits, the legislature determined that the services provided should constitute a basic healthcare package, and it sought to ensure that Medicaid recipients were, whenever possible, enrolled in managed care plans. These plans, often known as health maintenance organisations, have developed rapidly as an alternative to fee for service medicine, and provide services to those enrolled on a prepaid basis. By managing the use of services by both patients and doctors, health maintenance organisations seek to reduce the overall cost of providing care while maintaining high standards of provision.

Summary points

• The basic healthcare package available to those eligible for Medicaid coverage under the Oregon health plan has been expanded. Enrolment in Medicaid has increased by over 100 000 since 1994 but eligibility criteria have been tightened in response to rising costs

• The proportion of the population that remains uninsured has fallen from 18% in 1993 to 11% in 1996 as a result of implementation of the plan and economic growth

• Providers of safety net services may not be able to continue to provide such services under the plan’s payment arrangements. The delivery of services through managed care plans may put those who receive Medicaid at a disadvantage

• The employer mandate to provide health insurance lapsed in 1996 and the problem of providing health insurance to people who are working and yet are still uninsured remains

• Explicit priority setting tends to result in inflation of the basic healthcare package. Defining a list of services to be covered must go together with the development of clinical guidelines

As well as focusing on expanding Medicaid enrolment, politicians in Oregon legislated to provide assistance to residents with pre-existing medical conditions who were unable to buy insurance coverage, and they also took action to help people working in small businesses obtain insurance. More importantly, the employer mandate was introduced, which was designed to ensure that in time the basic healthcare package offered under Medicaid would be available to workers who were not entitled to Medicaid. The purpose of the employer mandate was to move towards universal coverage by requiring businesses to offer protection to uninsured people who were working and whose incomes placed them above the federal poverty level. The mandate was also intended to avoid a reduction in the services available to those eligible for Medicaid by building a wider constituency of support for the package.

Among this diverse range of initiatives, it was the attempt to define the basic healthcare package that attracted the most interest outside Oregon. The architects of the Oregon health plan sought to raise the threshold for inclusion in Medicaid from people who earned around 50% or less of the federal poverty level to include those who earned up to 100% by restricting the services that would be funded. The task of determining what should be on the list was entrusted to a health services commission whose 11 members comprised professional and lay people. The conclusions of the commission were reported to the legislature which then had to decide whether to accept the conclusions and also agree the level of funding that would be made available. The work of the commission was conducted in public and included considering the advice of experts and consulting with the community. It was partly for this reason that the Oregon experience acted as a magnet for those policy makers and health professionals from other healthcare systems where priority setting or rationing were undertaken covertly rather than overtly.¹

Genesis

During its development, the Oregon health plan attracted a range of commentary.^2–5 One of the main concerns of critics was that by concentrating on Medicaid, Oregon was rationing services only for the poor (most of whom were women and children) and was therefore accentuating the inequities in the US healthcare system. Critics believed that the development of the Oregon health plan distracted attention from the bigger and more important question of how to ensure universal healthcare coverage, a question that called for examination of more radical options, such as the introduction of national health insurance.⁶ Supporters of the plan, while acknowledging the force of these arguments, contended that reform had to start somewhere and that politicians in Oregon deserved credit for their willingness to make progress by seeking to extend insurance coverage to the most vulnerable residents.⁷ Other commentators maintained that explicit priority setting was preferable to implicit priority setting because it resulted in greater public accountability for decisions.⁵ Yet other commentators drew on the experience of Oregon to highlight the technical challenges of priority setting⁸ including the complexity of the formula originally used by the health services commission to rank services, the anomalies that resulted from cost effectiveness analyses, and the difficulty of organising consultations with the public to ensure that those involved were representative of the population.

Undeterred by these challenges, the commission refined its methods; research evidence and professional opinions on the effectiveness of different treatments were used together with the values derived from the public consultations to draw up a list of around 700 pairs of conditions and treatments to be given priority for funding. These conditions were categorised according to the ICD-9 (international classification of diseases, ninth revision) and they were then associated with their appropriate treatments from Current Procedural Terminology.⁹ When the first list was submitted by the Oregon legislature to the federal government for approval, it was rejected as being inconsistent with the Americans with Disabilities Act 1990.

This was because the ranking of condition-treatment pairs was based in part on an assessment of the potential for restoring a full quality of life, an outcome not possible for people with disabilities. Only after extensive revision to avoid discrimination of this kind was agreement given by the Health Care Financing Administration for the list to become operational from February 1994. The Oregon health plan was launched with funds available from the legislature to provide 565 out of 696 treatments on the final priority list. The treatments included the bulk of preventative and curative services, with high priority being attached to palliative care as a result of values identified during the public consultations. The principal exclusions were the treatment of self limiting conditions and conditions where no effective interventions were available.

Four years on, how has the Oregon health plan fared? Have the objectives of its architects been achieved? And what experience has been accumulated along the way?

Implementation

Priority setting in Oregon did not end in 1994. There is a legal requirement to keep the basic healthcare package under review; this continues to be the responsibility of the health services commission. Supported by a full time project director and a part time medical director, the members of the commission, who are volunteers, undertake a formal review every two years and propose changes to the package based on their review. Much of the work of the commission is done in subcommittees that evaluate outcomes, mental health and chemical dependency services, and dental services. The commission is advised by 70 provider groups that bring together doctors from throughout the state, and they in turn make use of recommendations emanating from national expert groups such as those run by the National Institutes of Health. Any changes proposed by the commission and agreed by the legislature have to be approved by the Health Care Financing Administration. This process can be both time consuming and extended, as illustrated by the fact that in January 1998 the Health Care Financing Administration had still to approve revisions to the funding list drawn up by the health services commission in May 1997.

Changes to the list

One of the most important changes made in the last four years has been the full integration of mental health and chemical dependency services into the basic healthcare package. This has increased the length of the priority list from 696 to 743. Elderly and disabled people whose incomes are below the federal poverty level have also been integrated into the plan. A number of changes have occurred as services have been moved up or down the list in the light of experience and as a result of changes in the coding of diseases. More importantly, adjustments have occurred as new evidence has become available on the effectiveness of particular treatments. Cochlear implants, for example, have moved from being unfunded to being funded services, as have bone marrow transplant operations for breast cancer. These transplantations were included in the list for the first time in 1997 and will be funded when the list is approved by the Health Care Financing Administration; for these treatments there is a stipulation that women should agree to enter into a clinical trial of the treatment’s long term effectiveness. Inclusion of bone marrow transplantation for treatment of breast cancer was controversial and was only agreed by a narrow majority of members of the commission.

The health services commission has begun developing guidelines for the provision of some services on the list. In certain cases, these guidelines are meant to ensure that the cost of providing services is kept within the available budget. For example, the guidelines for adult dental care—a treatment that does not have to be provided under Medicaid—are intended to avoid the overprovision of services. In other cases guidelines have been developed to ensure that services are provided in accordance with the recommendations of national expert groups, for example in preventive services, where the guidelines are based on a report by the US task force on prevention. Guidelines also have a role in the provision of services that fall below the cut off point for funding where it has been recognised that more severely ill patients would benefit from earlier intervention rather than waiting for treatment of complications arising from the original diagnosis. Such services include treatment of uncomplicated hernias in adults, treatment of severe rhinitis, tonsillectomies, and adenoidectomies. The adoption of guidelines for these services demonstrates the difficulty of ruling out whole categories of care or treatment from funding.

The experience in Oregon suggests that strict adherence to a defined set of core services is likely to be problematic. This observation is reinforced by the fact that doctors in managed care plans have the freedom to decide whether treatments not listed in the basic package should be provided. Only those doctors delivering care to the 15% of Oregon’s population who are covered by Medicaid and receiving services under fee for service arrangements are subject to scrutiny to ensure that they are not providing treatments that fall below the threshold for funding. For the bulk of patients covered by Medicaid, this means that the list of funded services acts mainly as a cost containment instrument. Managed care plans carry the risk and expense of providing services that fall below the cut off for funding when doctors nevertheless determine that such treatments are needed; the plans are responsible for making their own arrangements for monitoring whether this happens.¹⁰ No data exist to assess the extent to which doctors in managed care plans offer unfunded services to Medicaid patients, but the existence of this freedom is a potentially important safety valve for clinicians and patients.

The changes to the basic package and the adoption of guidelines have increased the scope of the services that are funded. In part, this has resulted from developments in healthcare technology and the availability of evidence which challenges earlier decisions to exclude treatments. It has also been driven by the experience of implementing the original list of services and the need to make adjustments based on advice received from clinicians. This advice has been instrumental in developing guidelines that give doctors greater discretion in offering services that initially fell below the funding threshold. Expansion of the basic package has been helped by the Health Care Financing Administration’s reluctance, as the ultimate arbiter of the content of the package, to approve further restrictions in funded services. Taken together, these factors mean that in Oregon explicit priority setting has expanded the range of services provided.

Financing the plan

The addition of treatments to the list has created problems for the funding of Medicaid, especially since the number of people enrolled has increased, in line with the intentions of the politicians in Oregon. In 1994, during the early phase of implementation, the helpline set up to provide residents with information on the plan received 4000 calls each day; only 5000 calls each week had been expected. An additional 130 000 people were subsequently enrolled in Medicaid, taking the total number covered to over 400 000. The backlog of need that was revealed in areas such as dental care, reminiscent of what happened when the NHS was set up in 1948, made accurate budgeting difficult. A deficit of $18m (£11.25m) resulted and led to the consideration of radical options for change, including limiting funding to include only the first 505 treatments on the priority list; this would have eliminated a number of potentially effective treatments, such as those for oesophageal cancer.

This option was not implemented, but other changes were. Most importantly, eligibility criteria were tightened to exclude those with liquid assets of $5000 (£3100) or more, to introduce contributions of up to $28 (£17.50) each month for families at the high end of the (admittedly low) income range (a yearly income of $16 500 (£10 300) for a family of four), to exclude full time college students, and to require people’s incomes to be assessed over three months rather than just one month. These changes helped put Medicaid on sounder financial footing, but at the expense of reducing enrolment by around 15 000. The eligibility restrictions agreed by the legislature have been criticised by citizens’ action groups as being inconsistent with the fundamental purpose of the health plan.

“In Oregon explicit priority setting has expanded the range of services provided.”

Changes in the economy and the labour market, which have taken people out of poverty, have reduced the number of people dependent on Medicaid. On one level, this reflects the strong performance of the Oregon economy and the additional jobs that have been created. At another, it has resulted from action by the state government to move people out of welfare and into work. Paradoxically, these developments have in turn created problems; those above the poverty level and in work do not always have private health insurance. The position of those who are working and yet are still uninsured is a particular concern of citizens’ action groups. These groups have challenged claims made by the state government about the extent to which the proportion of the population without insurance has fallen, arguing that official figures do not reflect the true numbers.

An unintended and unanticipated effect of the implementation of the plan has been to bring into question the funding of safety net providers who deliver services to patients with special needs such as people with HIV or AIDS and migrant workers who do not speak English. These services were fully funded in the past and enabled providers to deliver appropriate care. The capitation payments available under the plan are less generous, and this has made it difficult to sustain these services at their previous level. Providers of safety net services in Oregon have responded by forming their own managed care plan, CareOregon, which is lobbying for capitation payments to be adjusted for risk to allow for the higher cost of treating those clients who need these services. Unless changes are made to the plan’s financing, there is fear that safety net services will have to be terminated, with adverse consequences for both the providers and clients of these services.

Beyond its effects on those covered by Medicaid, the plan has had a number of other effects. Progress has been made in enabling patients in high risk groups and those who are employed in small businesses to obtain insurance coverage, thereby reducing the number of people who are uninsured. However, the employer mandate lapsed in 1996 when a waiver of federal laws could not be obtained. The failure to proceed with the mandate means that people who are working and yet are still uninsured continue to pose a challenge to those seeking to ensure universal coverage.

“The main weakness of the plan is not healthcare rationing. . . but managed care”

More positively, the revenue from a new tobacco tax (30 cents (18 pence) on a packet of cigarettes) has been earmarked for the maintenance and expansion of the plan. Among other schemes, this includes taking action to bring into the plan pregnant women and children up to age 19 who earn incomes of up to 170% of the federal poverty level. Other options for expanding coverage will be considered during 1998 by the Oregon health council, a group of citizens appointed by Governor John Kitzhaber to advise on reforms. The deliberations of the council will be informed by the results of a new programme of community consultations on fairness and financing in health care that are being undertaken by Oregon Health Decisions, a non-profit making organisation which organised the community meetings in 1990 and which helped to clarify the values the public believed should guide priority setting for the basic package. As in federal policy, the aim is to gradually extend health coverage in recognition of the difficulty of achieving more fundamental change.

Managed care

One other aspect of the implementation process merits comment. The strategy of containing the cost of Medicaid expansion by encouraging the provision of services through managed care plans has resulted in 85% of those on Medicaid receiving their care through health maintenance organisations. An explicit objective of the plan was to reimburse health insurers at a level sufficient to make it attractive for them to treat clients with Medicaid. This objective appears to have been met with the exception of safety net services. However, citizens’ action groups have expressed concern that Medicaid clients may have greater difficulties than more affluent residents in effectively gaining access to services offered in managed care plans, for example because they may lack telephones or transportation. They also argue that shortages of doctors, dentists, and other staff in some parts of the state have frustrated the attempt to translate health insurance coverage into the effective delivery of service. From this perspective, the main weakness of the plan is not healthcare rationing, since most services are now funded in the basic package, but managed care. These developments in Oregon mirror those occurring across the United States as the drive towards managed care gathers pace.

“Establishing strong purchasing bodies. . . has not been easy. . . the needs of clients are still not well articulated”

The emphasis on delivering Medicaid through a system of managed care has led to changes in the organisation of state government. The Office of Medical Assistance Programmes, the agency charged with putting the basic package into operation, has had to shift its role from paying for Medicaid services to actively purchasing them. This new role includes monitoring the performance of the 15 health plans with which it has contracts and ensuring that they are meeting the needs of Medicaid patients. The office does this by carrying out client satisfaction surveys. It is also arranging to assess standards of clinical care by commissioning an external agency to sample and review the medical records of patients with conditions such as diabetes to ensure that they are receiving appropriate care. The office also aims to develop a scorecard for monitoring performance in order to address some of the criticisms that have been levelled at managed care. Just as in the United Kingdom, establishing strong purchasing bodies that are able to negotiate on equal terms with providers has not been easy. As a consequence, the needs of clients are still not well articulated. Although this problem has been recognised, it will take time until it has been rectified to the satisfaction of those lobbying on behalf of citizens.

Outcome

Judged on its own terms, the Oregon health plan has achieved success in some areas but has failed in others. The most important achievement has been to increase enrolment among residents eligible for Medicaid by over 100 000 and to contribute to the reduction in the proportion of those who are uninsured from 18% in 1993 to 11% in 1996 (Office for Oregon Health Plan Policy and Research). In the same period the proportion of children who were uninsured fell from 14% to 8%. While some of the improvement is a result of a reduction in unemployment, the contribution of the plan, including the protection offered to people in high risk groups and people employed in small businesses, is estimated to have been responsible for around two thirds of the increase in insurance coverage (D Coffman, personal communication). Not only that, but the basic package has been expanded and encompasses a great deal more than a narrow set of services. The work that has been done on clinical guidelines has begun to define more precisely the way treatments on the list should be provided and has blurred the distinction between services that are or are not funded.

Against these achievements, the tightening of the criteria for eligibility for Medicaid has contributed to a reduction in Medicaid coverage from the peak of over 400 000 in 1995. The difficulties encountered by safety net services also show that the part of the population covered by Medicaid may have been put at a disadvantage by the implementation of the plan even though others may have benefited. A particular concern of citizens’ action groups is the position of those who are working but who are still uninsured. The failure to implement the employer mandate means that many people who work and whose incomes are above the federal poverty level still lack coverage in the event of illness.

The most important lesson to be learnt from Oregon’s experience is that explicit priority setting tends to result in inflation of a basic healthcare package. Furthermore, defining a list of services to be funded has to go together with work on clinical guidelines to ensure that treatments both above and below the threshold for funding are provided in a way that is consistent with evidence on the effective and appropriate provision of care. Because of the rapid changes in the availability of healthcare technologies and in the evidence of their effectiveness, priority lists must be continuously reviewed. In the process, those who are charged with making decisions are dependent on the advice of experts but they must also take account of the views and values of the community. Additionally, clinical discretion remains important in the implementation of the basic package, at least in managed care plans where doctors are able to decide whether treatments beyond those included in the package should be provided in practice.

Whether the outcome looks like a glass half full or a glass half empty depends on your perspective. The 1996 award to the state of the prestigious Innovations in American Government prize by the Ford Foundation brought national recognition to Oregon and testifies to the progress made by the state. It also indicates that in a country where government sponsored change in health care is notoriously difficult to achieve, Oregon has done better than most in terms of increasing access among the most vulnerable residents. However, the extent of unfinished business is daunting and illuminates the obstacles that have to be overcome even when the political commitment to change is strong.

For the future, much hinges on the strength of the economy. The Oregon health plan has been implemented in favourable economic circumstances; a downturn in the economic cycle would probably increase the number of families in poverty and at the same time put pressure on tax revenues. At that point, the balance that has been struck between the comprehensiveness of the basic package and eligibility for Medicaid would have to be reviewed. In a context in which the Health Care Financing Administration has been resistant to reductions in the list of funded services, there may be little choice but to further tighten eligibility criteria, especially if Oregon remains a fiscally conservative state.¹¹ Were criteria to be tightened the longer term sustainability of the health plan would be put to the test.