American Psychosis (33 page)

It is apparent, then, that the first—and most important—thing that must be done before mental illness services can be improved is to fix responsibility for the services at a specific level of government. The optimum level of government for such responsibility is to be determined. Clearly, such responsibility should not be lodged at the federal level, given the federal failure of the past half-century. Assigning the responsibility to states would seem logical, given that they had such responsibility for over a century. For small and medium-sized states, this may be the optimal level. Large states such as California, Texas, New York, Florida, and Pennsylvania may be too large and populous to administer at the state level, and such states may wish to devolve responsibility to counties or blocks of counties. Even some counties may be too large, however. For example, Los Angeles County has 9.8 million people, more than the population of the 10 smallest states combined, with a million people left over. Administering human services for 9.8 million people in such a county has nothing in common with, for example, Garfield County in Montana, which has 1,184 people in an area approximately the same size as Los Angeles County. Thus, there should be great flexibility among regions in how such programs are administered and how responsibility is assigned. The one absolute given is that responsibility
must
be assigned.

10. The single biggest problem with the present anarchic system of mental illness services is that nobody is accountable. It will be necessary to assign responsibility to a single level of government, and to then hold such individuals accountable, before any improvement can occur.

HOW SHOULD MENTAL ILLNESS SERVICES
BE ORGANIZED AND FUNDED?

How should mental illness services be organized and funded? The short answer to this question is that we do not know. The history of the last half-century has illuminated many organizational and funding mechanisms that do
not
work. For example, such services should not be organized as mental
health
services and should not be delivered by for-profit organizations. But beyond such lessons, we have learned very little regarding the best way to organize and fund such services. To find out, we should allow states and counties to experiment with different systems and then carefully assess the outcomes.

Those who are knowledgeable about the organization and funding of mental illness services in European countries often invoke them as models. To be sure, almost every European country does a better job of providing care for mentally ill individuals than does the United States. Services in countries such as Sweden and the Netherlands are significantly superior to anything that can be found in this country, but they cannot be easily imported. The organization and funding of mental illness services in countries such as Sweden and the Netherlands are intimately tied to their broader organization and funding of medical services, and this is fundamentally different from the American system. It should also be remembered that countries like Sweden and the Netherlands are much smaller and more homogeneous than the large and diverse United States. Sweden has approximately the same population as the state of Georgia, and the Netherlands has a population less than that of Florida. We can learn from the successes of these countries, but we cannot simply import their models. We have to start with what we have and ask how it can be made into a functioning mental illness services system.

Although it was never planned as such, Medicaid is the fiscal giant that dominates the funding of mental health and mental illness services. As a federal program with state-matching funds, it mandates core services that must be provided by the states and allows states to use Medicaid funds to provide additional services if they choose to do so. The share of total mental health and mental illness spending covered by Medicaid has increased from 17% in 1986 to 28% in 2005 to more than 30% today. This includes inpatient care in general hospitals, nursing homes, outpatient psychiatric and medical care, and prescription drug costs. For individuals with a diagnosis of schizophrenia, Medicaid and Medicare paid for the care of 19% of these individuals in 1977, 63% in 1996, and perhaps 90% today. The state-matching costs for Medicaid are the second largest item in most states’ budgets, behind only education.
⁴⁵

The Medicaid program as currently constructed is the single largest fiscal impediment to improving services for mentally ill persons in the United States. Services at
the state and local level are organized exclusively to maximize Medicaid reimbursement by the federal government, with little regard for organizational efficiency or what patients actually need. Medicaid officials in Washington have tried various strategies to control federal costs, but in every instance states have found ways to defeat these efforts. A classic example is the institution for mental disease (IMD) exclusion, by which Medicaid refuses to pay for inpatient costs in state mental hospitals. The states responded by simply emptying the state hospitals and shifting inpatient admissions to the psychiatric units of general hospitals, which are covered by Medicaid. The fact that the state hospitals already had the patients’ records and were much better set up to provide care for seriously mentally ill individuals was not considered. Medicaid reimbursement, not patient needs, has been the driving force behind the organization of public psychiatric services for four decades.

States have also utilized various organizational schemes in attempts to control state Medicaid costs. At least 34 states deliver “some or all mental health services through managed care arrangements, including both carve outs and comprehensive MCOs [managed care organizations].” States such as California, Utah, Colorado, Pennsylvania, New York, and Massachusetts have used capitation funding, under which providers are paid a fixed amount to deliver all necessary services.
⁴⁶

Such funding programs have three things in common. First, the bottom line for these programs is cost savings, not patient care. Almost none of these programs make any attempt to assess quality of care or patient outcomes. Second, the sickest mentally ill patients are the ones who suffer most under such funding programs. The reason is that individuals with mental illnesses constitute only 11% of all Medicaid beneficiaries, but this 11% accounts for one-third of all high-cost beneficiaries. As described in previous chapters, seriously mentally ill individuals incur high expenses as they migrate from program to program in the present disjointed care system. When funding programs want to save Medicaid money, therefore, denying services to seriously mentally ill individuals is the easiest way to do so. Such individuals are unlikely to complain, they do not have an effective lobby of family members to advocate on their behalf, and they often end up in public shelters or jails, where they are out of sight.
⁴⁷

The third thing these funding organizations for Medicaid patients have in common is that they are very profitable. Most managed care companies are part of the highly profitable health insurance industry. For example, United Behavioral Health, part of the United Health Group company, had revenues of approximately $92 billion in 2010. United Behavioral Health claims to “oversee behavioral health services for more than 23 million beneficiaries,” including Medicaid patients in California and other states. Its CEO for 17 years was Saul Feldman, a psychologist who began his career working in the NIMH Community Mental Health Centers program. In 2001 the
San Francisco Chronicle
described Feldman as being among the “super-rich” and
“ultra-wealthy . . . living in luxury above it all” in a penthouse atop the Four Season Hotel; such penthouses rent for approximately $8,000 per day and sell for up to $14.5 million. Many other managed care executives have done equally well from the profits accrued from administering programs for mentally ill individuals. The people who have not done well, as described in this book, are the mentally ill themselves, and there is a direct cause-and-effect relationship between these two disparate outcomes.
⁴⁸

The Patient Protection and Affordable Care Act (ACA), widely referred to as Obamacare, is scheduled to take effect in 2014 but is unlikely to improve care for most mentally ill individuals. Although it expands Medicaid eligibility by an estimated 16 million people and increases Medicaid benefits, it will put more pressure on states to control costs by denying services. Thus, it is likely to lead to managed care companies finding new and creative ways to not provide services to the mentally ill individuals who need the services most. The most promising parts of the ACA are the demonstration projects being set up in some states to, for example, abolish the Institutions for Mental Disease (IMD) Medicaid exclusion on an experimental basis and to make more coherent the funding for individuals who are eligible for both Medicaid and Medicare. If these demonstration projects are carefully monitored and assessed, then they could provide very useful data.

ALTERNATIVE FUNDING STRATEGIES

Fifty years have passed since President Kennedy planted the seeds of the community mental health centers movement, a massive federal experiment in the organization and delivery of services for mentally ill persons. The experiment destroyed the existing state system and failed tragically, as has been detailed in the preceding chapters. At this time, we appear to be stuck with the resulting dysfunctional system, making modifications to Medicaid reimbursement that improve the fiscal bottom line for managed care organizations but do not improve services for mentally ill individuals. All attempts to improve services over the past three decades have essentially been tinkerings with the status quo.

If we have any hope of improving such services, then we need to think much more broadly and creatively. We know that our present system is failing, but we do not know what system would work better. Therefore, we should be willing to take risks and think outside the traditional Medicaid box. In doing so, it is essential to measure the results of our trials and objectively assess the results. A research component must therefore be included in each mental illness services experiment, a component that has been largely missing from past attempts to change the system.

What should be measured? Many methods for assessing treatment outcomes for mentally ill persons have been proposed but infrequently used. They include subjective
and objective measures of the effect of the services on patients, such as quality-of-life scales and severity of symptoms. They also include measures of the effects on the community, such as the rehospitalization, homelessness, and incarceration rates. A substantial literature describing such measures and rating systems exists but is rarely used. Such measures should be included in all experiments in mental illness services, with data collected before the experiment begins and again after it is underway.
⁴⁹

What kind of trials might take place? One such experiment would involve block granting all federal Medicaid funds to two or three small states or large counties without any federal strings attached. Baseline measures would be collected before the program began, then periodically for perhaps 5 years. By the end of that time, it should become clear whether states and counties can deliver more effective and economic mental illness services without federal guidelines.

Another experiment might involve completely abolishing the state or county department of mental health and giving all Medicaid and other mental illness-related funds to the state or county department of corrections. Because the police and sheriffs have de facto become the frontline mental health workers and the jails have become the primary psychiatric inpatient units, why not let corrections take complete responsibility, along with the funds, and measure the outcome? What would most likely happen is that corrections personnel would focus resources on the most severely mentally ill patients, which would almost certainly be an improvement over the present system.

Still another experiment might involve altering the rules on the use of SSI and SSDI payments. Currently, these payments are given to disabled mentally ill individuals automatically, regardless of whether these individuals are participating in treatment programs. For selected mentally ill individuals who are aware of their illness and need for treatment, it would be a useful experiment to tie such payments, or part of such payments, to the patients’ active participation in treatment programs. This would guarantee that more patients would continue to take the medication needed to keep them well, thereby decreasing rehospitalization, homelessness, incarceration, victimization, and violence. Such outcomes would be comparatively easy to measure. Making selected mentally ill individuals more personally responsible for the outcome of their illness is similar to proposals put forth in some states to make nonmentally ill individuals more personally responsible for the outcome of their illness, such as charging higher copayments to people who smoke or are grossly overweight.

These are merely examples of many possible experiments that could be carried out with careful outcome measures. Innovation would be encouraged. All such experiments would be done on relatively small populations—small states or large counties—on a time-limited basis, so that outcomes could be accurately assessed. Such experiments should never be instituted on a national basis, as the community mental health centers program was. It may well be that, in the long run, some states may
opt for one kind of mental illness program whereas other states opt for another kind, which would be consistent with the diversity of our country and population.