Reforming Medicaid

Medicaid is enormously expensive. For the second year in a row, spending on Medicaid (for the poor) will exceed spending on Medicare (for the elderly). At $280 billion this year, Medicaid costs almost $1,000 for every man, woman and child in the country – or $4,000 for a family of four. Indeed, it is likely that many taxpayers are paying more in taxes to fund health insurance for the poor than they pay for private health insurance for themselves and their own families.

In the future, the taxpayer burden will get worse. Left unreformed, Medicaid will bankrupt every state in as little as 20 years, possibly absorbing 80 to 100 percent of all state revenues. Delay is not an option. States and the federal government must act now to avoid a real human and fiscal disaster.

Reason for Rising Costs. Why are Medicaid costs rising so rapidly? Part of the reason is that many states are paying for health care in ways that needlessly contribute to rising health care costs. Another problem is that most states have not taken advantage of cost-control techniques widely used in the private sector. For example:

• Because Texas' method of paying for hospital care is largely cost-based, Medicaid pays some Dallas hospitals three times as much as other hospitals for the same services.
• Because Ohio's method of paying for nursing home care is essentially cost-based, the state is paying for 13,000 empty beds.
• In virtually every state, Medicaid pays for inputs rather than outputs. This means that the more physicians and facilities do, the more they earn – even if patients would have been better off if less were done.
• Medicaid patients also have incentives to waste resources; the only way they can realize more benefits is by consuming more health care.

Medicaid from top to bottom is organized in ways that create perverse incentives for those who are supposed to benefit from the system. For example:

• Because Medicaid benefits are conditioned on having a low income, the program penalizes those who succeed; individuals can lose eligibility and therefore health coverage for themselves and their families simply by getting a workplace promotion or a raise.
• Because Medicaid benefits are conditioned on having few assets, the program encourages people to spend rather than save their income.
• Because Medicaid is an alternative to private insurance, the program encourages people to drop coverage for which they or their employers pay and turn to "free" insurance for which all taxpayers pay.

Studies estimate that as much as 75 percent of the expansion of Medicaid nationwide has been offset by a reduction in private insurance. This is one reason why the number of uninsured in America keeps rising, despite the expansion of costly public programs.

Traditional Solutions. In order to participate in Medicaid and receive matching funds from the federal government, states must provide minimum benefits to certain categories of people. Nationwide, however, about 70 percent of all Medicaid spending covers either beneficiaries who do not have to be covered or services that do not have to be covered or both. Thus one alternative is to eliminate "optional" people and "optional" services. There is also an ever-present pressure to lower payments to providers. Cutting costs in these ways, however, is easier said than done:

• Most of the optional people are likely to be uninsured after they are dropped from the rolls. And evidence from Texas suggests that regions that spend less on Medicaid spend more on free care for the uninsured and vice versa.
• A major optional benefit is coverage for prescription drugs. Yet because drug therapy often substitutes for more expensive hospital and doctor therapies, eliminating coverage for drugs may not save the state any money.
• Another large optional benefit is mental health care. But mental health services are also often a substitute for other health services, and without Medicaid the state may have to spend money on many of the beneficiaries anyway – but without federal matching funds.
• Reducing payments to providers will disproportionately hurt rural and inner-city areas at a time when the nation's safety net of local clinics, hospitals and charities is overburdened and threadbare.

A New Approach. The time has come for a "pro-patient" approach to health care. The following are some elements of that approach.

Choice. Medicaid is a defined benefit program, under which the state determines the covered health services, and patients and doctors decide how much to spend on those services. It should be converted into a defined-contribution system, under which the state determines how much it is willing to spend and patients (along with their doctors) choose how to spend it.

Competition. Most of the problems with Medicaid and Medicare stem from the fact that these two programs have changed very little since they were created in 1965. By contrast, private insurance is continually evolving in response to changes in medical science and changes in consumer wants and needs. Medicaid enrollees would benefit enormously if they had access to private sector plans, including the plans of their employers.

Portability. At least a quarter of all Medicaid eligibles move in and out of the program each year, as patients are tossed from physician to physician multiple times. As a result, many find the hospital emergency room more convenient than having a primary care physician. A better approach is private, portable insurance – where enrollment continues even after eligibility for Medicaid has lapsed.

Empowerment. Arkansas, New Jersey and Florida have obtained federal waivers under which disabled Medicaid beneficiaries manage some of their own health care dollars and make decisions on how to spend those dollars. These experiments have been highly successful and much more needs to be done. Enrollees need access to private sector Medical Savings Account (MSA) plans. Those who remain in traditional Medicaid need access to something similar: a Medicaid Benefit Account (MBA).

Paying for Results. Whenever possible, the state should pay for outputs, not inputs; for results, not for efforts made to achieve those results. At a minimum this means abandoning variations on cost-plus reimbursement and adopting some of the payment methods used by private insurers instead.

Devolution. For those who remain in traditional Medicaid, another useful idea is to give local communities more control, say through a block grant of funds. Under the current system, Texas reimburses El Paso hospitals at rates that are above the state average, while a typical primary care clinic lacks an X-ray machine or even a lab and must send patients to a hospital for simple diagnostic tests. As a result, El Paso's low-income population overutilizes hospital care and underutilizes outpatient care. The city and its people would be better off it were free to reallocate funds from inpatient to outpatient care.

More Radical Devolution. Despite a raft of studies claiming that being insured affects access to health care, there is no difference in services rendered to those who seek care – at least at Parkland Hospital in Dallas. The insured, the uninsured and the Medicaid patients who are treated there enter the same emergency room door, see the same doctors and receive the same care. As a result, patients have no reason to fill out the lengthy forms and answer the intrusive questions that Medicaid enrollment requires. Similarly, the doctors and nurses who treat these patients get paid the same, regardless of the patients' enrollment in an insurance plan. This is part of the reason why at least 10 million people who are theoretically eligible for Medicaid have not bothered to enroll.

If the patients and the doctors who treat them do not care who is enrolled in which program, why should policymakers care? Why not just give Parkland Hospital a sum of money each year and let it deliver indigent health care? Our primary goal should be to enroll people in private sector plans. But to the degree that we fall short, we should make access to care as easy as possible.

The Need for Federal Waivers. In order to take full advantage of private sector techniques and private sector opportunities, states should apply for a federal waiver, called a HIFA waiver. For example, a waiver might work like this:

• All "mandated" Medicaid enrollees would have the opportunity to enroll in employer plans or other private sector plans with premium subsidies from the state.
• To qualify, the private insurance would have to be similar to the plans currently offered to state employees.
• Those beneficiaries who do not qualify for an employer plan would have the opportunity to enroll annually in a plan of their choice through an insurance exchange (a health mart), organized and operated by the state.
• At least one of the plans offered would make use of a Medicaid Benefit Account (MBA), which would be similar to a medical savings account, except that the funds could be used only to pay health care expenses or health insurance premiums, now and in the future.
• Projected savings (in a static sense) from these changes would be used to enroll additional, "optional" people in Medicaid (this is a necessary condition to get approval for a HIFA waiver).
• The optional enrollees, consisting of additional people with disabilities and low-income families, could be offered a limited set of benefits (e.g., primary care benefits only) or they could be given the same options as other Medicaid enrollees – with premium support from the state diminishing as family income rises.

Expected Results. The HIFA waiver, combined with other recommendations in this report, would allow the state to take advantage of the full range of techniques employed by the private sector. It also would allow the state to move large numbers of people from state-funded insurance to insurance largely paid by employers and – since employer-provided benefits are earned by working – by the beneficiaries themselves.

[page]

¹Medicaid, the joint federal/state health care program for the poor, is in crisis. At a time when state revenues are frozen or declining, Medicaid costs per enrollee have almost doubled in the last five years – with projected spending increases rising even faster.² Medicaid, unreformed, threatens to take every new state tax dollar. As a result, an out-of-control Medicaid program is a threat to education, agriculture, highways, the environment, parklands and every other program funded by state government. Medicaid reform is everyone's concern because it affects everyone and every issue.

"Medicaid is in crisis"

Problem: Escalating Costs. The National Governors Association claims that 49 states, faced with stagnant revenues and exploding Medicaid costs, are in a real fiscal emergency.³ Medicaid and other health expenses already account for about 20 percent of state spending nationally, and those costs rose 13 percent last year – quot;the largest increase in a decade,quot; according to a National Governors Association report.⁴ Left unreformed, Medicaid will bankrupt every state in as little as 20 years – absorbing 80 to 100 percent of all state revenues.⁵ Delay is not an option. States and the federal government must act now to avoid a real human and fiscal disaster. What can be done?

Traditional Solutions. The Kaiser Commission on Medicaid and the Uninsured has reported that 49 states plan to cut Medicaid benefits, restrict eligibility, increase copayments or reduce provider payments.⁶ Already, 37 states have cut a combined $12.6 billion from their Medicaid budgets.⁷ More cuts will follow. According to the Kaiser Commission:⁸

• 18 states plan to tighten Medicaid eligibility rules in fiscal year 2003, compared to 8 in 2002;
• 15 states are cutting Medicaid services this fiscal year, compared to 9 last year; and 
• 40 states are reducing the amount they will pay for prescription drugs or are implementing preferred drug plans.

These traditional responses all seek to further restrict access to quality health care to save money. At best, they are stopgap measures. At worst, they will fail to halt what may be Medicaid cost increases as high as 15 percent this year.⁹

States have considerable flexibility in determining the number of people and the types of services that are covered by Medicaid. Nationwide, about 70 percent of all Medicaid spending is "optional," covering either beneficiaries who do not have to be covered or services that do not have to be covered or both.¹⁰ Since the average state currently spends 19.6 percent of its budget on Medicaid,¹¹ almost 14 percent of Medicaid spending could be saved in an average state by eliminating optional people and optional services. There is also an ever-present pressure to lower payments to providers. But are these options wise?

Traditional Option: Reduce the Number of People Who Are Covered by Medicaid. People who must be covered by Medicaid under federal law are "mandatory" populations. Those enrolled at each state's discretion are the aforementioned optional populations. For example, as of September 2002, states are required to grant eligibility to all children living in poverty, regardless of their age.¹² Unfortunately, paring people from the Medicaid rolls may save less than the states expect. For example, among the 11 health regions of Texas spending per Medicaid recipient in 1998 varied from a high of $4,425 to a low of $2,101 – a difference of about $2,300. But the variation falls to about half that amount when per capita spending on free care for the uninsured is considered. [See Table A-1] It appears that non-Medicaid health care spending substitutes for Medicaid spending, as regions that spend less on Medicaid tend to spend more on free care and vice versa. Reducing the amount spent on one program just increases the amount spent by the other.

Traditional Option: Reduce the Services Covered by Medicaid. Each state must offer 14 mandatory benefits (e.g., hospitals, physicians, etc.) but can decide which of 34 optional benefits it wishes to cover, including prescribed drugs, diagnostic screening, preventive and rehabilitative services, clinic services, dental care, dentures, physical therapy and related care, prosthetic devices, TB-related care, and primary care case management.¹³

"Traditional responses try to reduce access to care and/or lower provider payments."

Trying to eliminate these expenditures, however, may prove penny-wise and pound-foolish. For example, drug therapy is often a less expensive and more effective alternative to doctor therapy and hospital therapy. A study by Columbia University professor Frank Lichtenberg found that in the health care system an increase of 100 prescriptions is generally associated with 1.48 fewer hospital admissions, 16.3 fewer hospital days and 3.36 fewer inpatient surgical procedures. Overall, a $1 increase in pharmaceutical expenditures is associated with a $3.65 reduction in hospital care expenditures.¹⁴ Mental health and physical health services are also often substitutes for each other. In one study, an employer who reduced spending on mental health saw more than offsetting cost increases in other health services.¹⁵

Traditional Option: Reducing Payments to Providers. As financially stressed as states and taxpayers are, the newspapers daily report on struggling hospitals, nursing homes and physicians who see Medicaid patients. Reducing their reimbursements will disproportionately hurt rural and inner city providers. It will also have an economic ripple effect, since Medicaid dependent providers are also the chief (and sometimes only) economic engine for growth in these same rural and inner city neighborhoods.

Traditionally, hospitals have covered losses that arise from people who can't pay for their care and less generous reimbursement from Medicaid and Medicare by overcharging other patients. But as the medical marketplace becomes more competitive, these overcharges are shrinking. Cost shifting is virtually impossible in so competitive a market. There is ample evidence that this problem is not trivial. For example, preliminary findings from a RAND study show that safety net spending by the nation's hospitals is not keeping pace with the overall increase in per capita spending.¹⁶ A National Academy of Sciences Institute of Medicine study found that the safety net of local clinics, hospitals and charities is "overburdened and threadbare," and "could collapse with disastrous consequences."¹⁷

A New Approach. What is needed is a new solution, one that does not seek to solve the states' Medicaid fiscal crisis on the backs of the poor, disabled and blind. What is needed is a "pro-patient" solution.

[page]

Medicare (Title 18 of Social Security) and Medicaid (Title 19) were passed in 1965 at the height of President Johnson's Great Society and War on Poverty initiatives. They signified the high-water mark of postwar liberalism. The appearance of former President Truman at the signing ceremony signaled that they were intended to be first steps, culminating in a universal national health insurance program similar to the state-run systems in Europe.

"What is needed is a pro-patient solution."

The year 1965 was also the high-water mark of faith in the federal government and strong mistrust of the states. Only two years after Martin Luther King's "I Have a Dream" speech, one year after the Civil Rights Act and in the same year as the Voting Rights Act, "state's rights" was almost synonymous with racism and segregation. This explains Medicare and Medicaid's strong federal orientation.

By 1966, we were in an unpopular war that eroded confidence and consensus. The administration's "guns and butter" approach attempted to fight a war and proceed on Johnson's ambitious domestic agenda. It was not successful and costs and inflation got out of control. By 1970, Medicare and Medicaid were spending what Congress had estimated they would cost two decades later, in 1990. This was only the first of many underestimates of future spending.

In efforts to control costs in the 1970s, states centralized health planning, issued certificates of need, and imposed wage and price controls and moratoriums or restrictions on hospital and nursing home beds. All of these initiatives sought to control the supply of health care while the demand for it increased. Many politicians were surprised when costs skyrocketed.

In the 1980s, President Reagan's "New Federalism" proposed a swap: the federal government would take all of Medicaid and return other programs to the states. The idea was discarded when the Office of Management and Budget Director Stockman looked at Medicaid's astronomical out-year costs. Other regulatory reform proposals were defeated by a combination of seniors' lobbies and liberal legislators. In the 1982 elections, Sen. Claude Pepper sent out a nationwide letter stamped "Important Social Security Information Inside" and looking exactly like an official Social Security Administration document. It claimed that Republicans were trying to dismantle Social Security (and by implication Medicare and Medicaid). Most election analysts believed this was a primary reason why the Republicans lost 26 congressional seats and why the Reagan administration abandoned further Medicaid and Medicare reform efforts.

The 1980s also was a decade in which states tried numerous bookkeeping schemes to "enhance" the federal match, using methods that are now illegal or discouraged. [See the sidebar on "The Texas Experience."] Enhancing or maximizing the federal match has become a cottage industry for some Washington, D.C., think tanks, but the real culprit is the match itself.

In the 1990s, concern about Medicaid funding was superseded by concern for the uninsured. Health care became a major national issue when, in 1991, Pennsylvania's little-known Harris Wofford defeated the sitting U.S. attorney general for a U.S. Senate seat. Wofford's campaign manager, James Carville, took the issue to the governor of Arkansas for use in his campaign for U.S. president in 1992.

The result of these three decades is an astonishingly bare record of successful reform. Although America has changed greatly in 37 years, Medicare and Medicaid remain essentially the same programs that were enacted in the 1960s, that distant and different social and economic time.

[page]

"The nation is now spending more on Medicaid (for the poor) than it spends on Medicare(for the elderly)."

The Office of Management and Budget estimates total Medicaid spending will be $280 billion in Fiscal Year 2003, serving about 40 million poor, elderly and disabled. Medicare, by comparison will serve 40 million at a cost of $230 billion.¹⁸ This means Medicaid will spend more than Medicare for the second time in history (FY2002 was the first).

Medicaid versus Medicare. Medicare is a single federal health care program with a single national eligibility, benefits and reimbursement structure. By contrast, Medicaid is actually 56 separate state and territorial welfare programs with distinct eligibility, benefit and reimbursement rules.¹⁹ A person eligible for Medicaid in one state will not necessarily be eligible in another. Since upward of 70 percent of services and people are optional, the mix of eligibles, benefits and costs are unique to each state.²⁰

Another difference between Medicare and Medicaid is that Medicare pays almost exclusively for acute care. It has no long-term care or drug benefit and is insulated from many of the problems of an aging America. Medicaid, by contrast, pays for two-thirds of all nursing home residents and 50 percent of all long-term care costs nationally. Although most people tend to think of Medicaid as the health insurance plan for welfare mothers, most of the money is spent on others. For example, long-term care accounts for more than 43 percent of all Medicaid costs but is used by only 9 percent of all enrollees. The elderly, blind and disabled combined account for about one-fourth of enrollees and almost three-fourths of all costs. Any successful reform must address this massive imbalance.²¹ [See Figure I.]

"The aged, blind and disabled account for one-fourth of the beneficiaries, but almost three-fourths of the spending."

In theory, Medicare is supposed to be a federal program, paid for by the federal government. However, there is a class of Medicare recipients called dual eligibles. They initially qualify for Medicare, but because of their low incomes and few assets they can also receive Medicaid. Although Medicare is the primary payer, states must pay for any benefits Medicare doesn't cover if Medicaid does cover them.

The issue of dual eligibles sounds obscure, but recall that Medicare does not have a long-term care benefit and Medicaid does. Also, Medicare does not have a drug benefit, but Medicaid usually does. As a result, these huge costs are born by the states. The states have no control over Medicare eligibility, but they are automatically responsible for Medicaid payment. The National Governors Association estimates that more than one-third of all Medicaid costs are for dual eligibles and has made this their principal issue with the federal government.²²

Medicaid's Federal Match. As stated above, Medicaid is designed as a joint federal/state program. Currently, the federal government pays about 57 percent of all Medicaid costs (or $159 billion in FY2003) with states paying the rest (about $121 billion). The percentage each state pays varies. A complex formula called FMAP (Federal Medicaid Assistance Percentage or the "federal match") is used to calculate each state's relative per capita income. In "rich" states like New York, state spending is evenly matched by the federal government (each pays 50 percent). In "poor" states like Mississippi, the match has been as high as 83-17 percent.²³

The average cost per Medicaid beneficiary nationwide is about $7,000.²⁴ But because New York offers almost all optional benefits to all levels of enrollees, it spends almost double the national average. Mississippi, which has a less generous benefit package and confines coverage mostly to the "mandatory" poor, spends just about half the national average.

"New York receives twice as much federal money per enrollee as Mississippi – even though Mississippi's need is much greater."

The result is that New York receives about twice as much federal money for each enrollee as Mississippi, even though the need is much greater in Mississippi. This is why "block granting" Medicaid funds to the states has proved difficult politically. Block granting federal funds now being spent would lock in place a system in which the federal government continues to send the most money to the places that least need it.²⁵ It is also the reason the Government Accounting Office (GAO) in 1995 declared the FMAP a failure in reallocating resources from rich to poor states.²⁶

Medicaid's Impact on Economic Behavior. Although Medicaid was designed to assist people who need assistance through no fault of their own, the program offers perverse incentives. For example:

• Because Medicaid benefits are conditioned on low income, the program penalizes those who succeed; individuals can lose eligibility and therefore health insurance for themselves and their families simply by getting a workplace promotion or a raise.
• Because Medicaid benefits are conditioned on having few assets, the program encourages people to spend rather than save their income.
• Because Medicaid is an alternative to private insurance, the program encourages people to drop coverage they or their employers buy and rely instead on "free" Medicaid.

Very little research documents the ways in which people have responded to these incentives. But that modest research confirms what common sense would predict. Medicaid beneficiaries have behaved in a rational manner. They have dropped private insurance coverage, saved less and consumed more.²⁷ Any attempted reform of state Medicaid plans needs to understand the perverse effects of the program already in place.

There is also empirical documentation of a "crowding out" effect of Medicaid on private health insurance. Nationwide, the percentage of children who can receive Medicaid increased by more than 50 percent between 1987 to 1992 and the number of women eligible for Medicaid if pregnant more than doubled. Thus, Medicaid coverage increased by more than 2.3 million. However, this increase was accompanied by a significant drop in private insurance, offsetting from 50 to 75 percent of the increase in Medicaid coverage. The vast majority of this reduction came from workers deciding to drop private coverage (particularly for dependents) rather than because their employers stopped insurance coverage.²⁸

Medicaid's Impact on Health Behavior. An oft-cited argument for Medicaid is that by making health care virtually free at the point of consumption the program encourages preventive care and potentially reduces overall health care costs. Unfortunately, there is little evidence that this occurs.

"From 50 to 75 percent of the increase in Medicaid coverage is offset by a reduction in private coverage."

Studies suggest explicit attempts to encourage Medicaid beneficiaries to use preventive care are generally unsuccessful. For example, outreach programs in North Carolina found a statistically significant but very small impact on utilization.²⁹ Another study found that providing Medicaid benefits for a year increased the probability of children receiving checkups by only 17 percent. The researchers concluded that "factors other than insurance and income, such as the low educational attainment of low-income mothers, explain approximately 80 percent of the gap between low-income and other children in their well-child visits."³⁰ Research from the University of Washington found no evidence that prenatal care pays for itself by reducing future health care costs.³¹

Additionally, no definitive evidence links infant mortality rates to Medicaid coverage, and no evidence shows that becoming eligible for Medicaid significantly improves immunization rates.³²

[page]

A reformed Medicaid system must be based on a new philosophy. The following are the central elements of that philosophy.

Choice. All insurance programs have three essential elements: eligibility, benefits (coverage) and cost (reimbursement). How they are organized will determine the ultimate structure of the plan.

In a defined benefits entitlement plan, eligibility and benefits are fixed and costs are the variable. In other words, if you are eligible then you are "entitled" to all the benefits regardless of cost to the government or taxpayers (hence the term "entitlement program"). Supporters of defined benefit welfare entitlements have always stressed the ethical need to cover the needy and uninsured, regardless of cost.

This is why cost containment has failed in Medicaid. The problem is structural rather than administrative. In the last 37 years, both political parties have exacerbated this defect by increasing benefits and broadening eligibility. As a result, Medicaid has grown from $1 billion and 10 million enrollees in the 1960s to $280 billion and 40 million enrollees during four decades of almost uninterrupted prosperity. And it kept right on growing through the 1990s, even though the welfare rolls were cut by more than half.

The alternative is a defined contribution program. Under this approach, eligibility and costs are fixed and it is the benefit package that is variable. Enrollees choose between a variety of competing private health benefit plans that fit their individual needs. For example, nursing home residents don't usually need maternity benefits and infants rarely need bypass surgery.

"The government should fix the cost of Medicaid, then let enrollees choose thier benefits."

Usually, the amount spent by the government or an employer is the same, regardless of the choice of plan. But plans differ in how they allocate the funds among alternative benefits. If a plan costs more than the defined contribution amount, enrollees usually must make up the difference with their own funds.

An example of a defined contribution approach is the Federal Employees Health Benefits Program (FEHBP), a plan almost a half-century old. Even though the federal workforce is not necessarily a good risk pool – it is generally older and has more health problems – the premiums have been comparable to other, private sector insurance and federal employees have an annual choice of a dozen or so health plans.

One objection to allowing Medicaid beneficiaries to exercise choice is that the poor, elderly, blind and disabled either lack the ability to choose between plans or may be hood-winked by unethical sales people. Although this may be true of certain populations, it isn't true for the vast majority of Medicaid recipients. Evidence shows that for certain services, the poor have just as much ability to choose as the middle class.³³

In a Medicaid program, a counselor/independent broker can be responsible for enrollment and education to assure proper access to the health care system. As an intermediary between the state and insurers, the counselor/broker can give independent advice and education. Also, they can continually screen the eligibility roles and make certain the potential enrollee is in the proper program.

Competition. One of the problems in both Medicaid and Medicare is the absence of a variety of private sector alternatives. This is the principal reason why both programs have failed to keep pace with developments in private health insurance.

For example, when Medicaid and Medicare were created in 1965, most private insurance plans (including the most common Blue Cross/Blue Shield plans) did not include coverage for prescription drugs. Since the government programs largely copied the design of private insurance, they also excluded prescription drugs. After a third of a century of developments in medical science, private insurance today is quite different. Virtually all major medical plans in the private sector cover drugs and, for reasons given above, most see drugs as an inexpensive alternative to hospital and physician therapies. But because the government programs were creatures of politics, insulated from competitive pressures, they have not changed. Medicare still does not cover most prescription drugs, and drugs are still an option under Medicaid.

In order for Medicaid enrollees to participate in, and benefit from, insurance that meets the market test and evolves as medical science evolves, beneficiaries need access to the same options that are routinely available to other citizens. How can they gain that access?

"Founded in 1965, Medicaid has not adapted to changes in medical science."

On entering office, President Bush and Health and Human Services (HHS) Secretary Tommy Thompson announced a streamlined section 1115 waiver process named the Health Insurance Flexibility and Accountability (HIFA) initiative. The section 1115 waiver, granted by the Centers for Medicare and Medicaid Services (CMS), has traditionally been the chief research and demonstration process to test innovative comprehensive Medicaid reform. The idea is to enable states to perform as "laboratories for democracy." Since August of 2001, states have had the opportunity through waivers:³⁴

• To reduce some benefits in return for increases in other benefits;
• To reduce benefits in return for increases in the number of people eligible for those benefits; or
• To reduce benefits for some people in order to create a new set of benefits for others.

Suppose a state wants to expand eligibility to a new population (and qualify for federal matching funds for its spending on that group) without increasing the total amount of federal spending on health care. Under a HIFA waiver the state can have access to three sources of funds to pay its share of costs for the newly eligible:

• Disproportionate Share Hospital (DSH) funds, which are federal and state funds available to hospitals treating a disproportionate share of Medicaid and charity care patients.
• Unspent State Children's Health Insurance Program (S-CHIP) funds.
• Savings from the reduction of Medicaid benefits for currently eligible populations or the reduction in eligible populations.

Furthermore, the benefits created for the newly eligible group can be more limited than the benefits that were available to the previously eligible group.

There are certain restrictions on the waivers.³⁵ Usually, they are valid for three years (although they can be renewed). They must be budget neutral (the federal government must not expect to expend any additional funds). The state must be trying to "research an idea" (not just cutting costs). Certain populations must be "held harmless" (usually pregnant mothers and children). Certain benefits must be protected.

"States can attain flexibility through waivers."

Essentially, however, states can adjust almost all the benefits, eligibility and reimbursement standards. They need only CMS approval, not any congressional or judicial approval. If the waiver proves unsuccessful at any time, the state can unilaterally cancel after not adding any new enrollees for six months.

The intention of the 1115/HIFA waivers is to find ways to expand eligibility while not spending any more money. Tried as a defined contribution approach, it can be successful and provide quality care and patient satisfaction. For example, a reformed system under a wavier might work like this:

• Compare the present Medicaid program (benefits and cost) to some other health plan, such as a private plan routinely made available to state employees.
• Calculate what Medicaid would save if it paid only for the benefits of the private program (assuming no change of behavior); say the savings is 10 percent.
• Then for every 10 Medicaid enrollees that choose to join the private plan, the state could afford to extend coverage to one more potential Medicaid enrollee and stay budget neutral.
• If savings were greater than 10 percent (say, because of actual behavioral changes), the state could return the surplus to the taxpayers, support "safety net providers," or spend it on something else.
• If the Medicaid enrollee is able to obtain the same type of plan from an employer, the state could significantly reduce expenses by paying only the employee's share of the premium.

Any governor or legislature considering these waivers should note two caveats. First, these programs must be monitored carefully to see that projected savings appear. Otherwise, all the waiver will do is increase enrollment and costs.³⁶ Second, a bureaucracy can kill any new initiative.³⁷ It is usually not in the interest of an agency to admit that a better way exists or to implement it. The waiver should be administered directly from the governor's office or some new independent agency/board. For example, the Federal Employees Health Benefits Program is not administered at CMS. It is administered in the White House Office of Personnel Management by an appointee reporting directly to the President. If a new Medicaid program is structured using private insurance, perhaps it should be administered by the State Commissioner of Insurance rather than the present Medicaid bureau.

Portability. There are serious quality and continuity of care problems in the present Medicaid program. At least a quarter of all Medicaid eligibles are in and out of the program each year.³⁸ As a result, patients can be tossed from physician to physician multiple times a year. Many find using a hospital emergency room more convenient than having a primary care physician. Lack of portability also is a disincentive to find work and leave the Medicaid rolls.

"Ideal health insurance is portable insurance."

Of course, the easiest way to obtain portability is to allow Medicaid beneficiaries to enroll in plans their employers offer. The key to good health care is continuity of care: keeping one's physician, knowing one's hospital and understanding how best to access the system for oneself and one's children. Quality also depends on providers wanting to provide that care, which is more likely if patients are exercising choice. It is even more likely if patients are managing some of their own health care dollars.

Empowerment. Arkansas, New Jersey and Florida were the first states to receive Section 1115/ HIFA waivers (described above) that provided disabled Medicaid recipients direct cash payments to purchase needed services. This program, "Cash and Counseling," encompasses the essential elements of a defined contribution approach. The patient is given a set dollar contribution and is free to choose his or her providers.

The program also involves counseling to assure that the patient is well-informed. Previously, the states selected the providers without patient input. Now the patient can choose his or her own provider. Unbelievably, patient satisfaction is almost 100 percent.³⁹

"Ideal health insurance empowers patients."

Another promising idea is the use of a Medical Benefit Account (MBA), an idea similar to Medical Savings Accounts (MSAs), except that funds in an MBA would be restricted to health care and health insurance premiums and could not be withdrawn as cash.⁴⁰ Through this account, beneficiaries would manage some of their own health care dollars and thus would have incentives to make prudent health care choices. Opponents believe that the poor will forgo needed health care to accrue more cash. However, through a debit card, the state could ensure that the recipient completed certain medical procedures such as child immunizations or prenatal care before accessing any cash. The recipient could then use his or her remaining MBA funds for other health, social, child education or job training needs.

Paying for Results. The cost control methods the private sector plans use will be those that survive the market test. However, even for those who remain in traditional Medicaid, more can be done to insure that taxpayers get their money's worth.

The first and most important goal is to pay for services rendered rather than paying for the cost of producing those services. Only a few decades ago, our entire health care system was based on cost-plus finance.⁴¹ Blue Cross and most other private insurers paid hospitals based on their costs. So did Medicare. So did Medicaid.

Today, almost no private insurers pay hospitals based on their costs. And the federal government pays hospitals under Medicare on a DRG system – with fixed fees for treating different categories of illness, regardless of actual costs.⁴²

The exception is Medicaid. All too often state Medicaid programs pay hospitals, nursing homes and other facilities based on some version of cost-plus finance. For example, Table A-I shows the basic rate Medicaid pays hospitals in Dallas County. As the table shows, the state pays different hospitals very different amounts for what are essentially the same services. Private insurers, by contrast, tend to contract with hospitals selectively, steering patients to facilities that charge the lowest prices, other things equal.

Nursing home care reimbursement also is in urgent need of reform. Under Ohio's cost-based method of payment, for example, the state is paying for 13,000 empty beds. To make matters worse, Ohio's method of reimbursement encourages churning (wherein operators sell nursing homes back and forth to each other) and even encourages bankruptcy. [See the sidebar on "Paying for Long-Term Care in Ohio."]

"Medicaid should pay for results, rather than reimbursing for costs."

The second goal is to take advantage of evidence-based disease management and care coordination⁴³ to assure that services provided are necessary and appropriate.⁴⁴ Where treatment protocols do not exist, the state should foster their development in conjunction with academic institutions and practi-tioners. Doctors still would retain the right to make treatment decisions for their patients, but when they chose to prescribe treatments other than those in the protocols, they would know they have a higher burden of justification. [See the sidebar on "Smart Buying in Texas."] Protection from liability, or at least limited liability, could be the reward for following protocols.

A third goal is to eliminate errors. Some of the more costly problems include drug misuse, antibiotic overuse, preventable hospital-acquired infections, and the under-diagnosis and mistreatment of chronic conditions.⁴⁵ Medical errors are dangerous and costly. Elimination of errors in diagnosis and care provides better treatment at lower cost. Everyone wins.

A fourth goal is to create evaluation and payment systems with incentives to lower cost and achieve desired outcomes. If the state paid only for outcomes it deemed worthwhile,⁴⁶ it could reap significant savings in such areas as mental health care, substance abuse treatment and purchases of durable medical equipment.⁴⁷

Some might argue that the achievement of these goals was supposed to be the purpose of managed care. Yet during the 1980s, all too often managed care consisted of little more than negotiating price discounts, even though the managed care organizations (MCOs) claimed they were instituting efficiencies and eliminating waste. Actually, a lot of dollars can be saved without a large managed care bureaucracy by simply using common sense. [See the sidebar on "Smart Buying in the Texas S-CHIP Program."] [Full disclosure: One of our authors, Ronald Lindsay, is involved with this project.]

"Our primary goal should be to place people in private health plans."

Devolution. For those who remain in traditional Medicaid, another useful idea is to give local communities more control, say through a block grant of funds. An example of a community that could benefit from more flexibility is El Paso, Texas.⁴⁸ As Table A-I shows, El Paso (the Upper Rio Grande area) has a low utilization of Medicaid, but a high level of spending for uncompensated care. Moreover, under Texas reimbursement formulas, El Paso hospitals are paid at rates that are above the state average, while primary care clinics are starved for resources. A typical primary care clinic, for example, lacks an X-ray machine or even a lab and must send patients to a local hospital for simple diagnostic tests.

As a result of these factors, El Paso's low-income population overutilizes hospital care and underutilizes outpatient care. The city and its people would be better off it were free to reallocate funds from inpatient to outpatient care. But restrictions on the use of funds apparently make this difficult, if not impossible.

More Radical Devolution. How far can devolution go? In searching for an answer to that question, consider first one of the most perplexing questions about Medicaid. An estimated 3 million adults in the United States are eligible for Medicaid but do not bother to enroll. An estimated 6.8 million children who are eligible for S-CHIP or Medicaid are not enrolled.⁴⁹ Yet if these health insurance programs are valuable, and given the amount of money they spend, they must be, why do people choose not to enroll?

"For those who remain in the public sector, we need block grants to local communities."

A visit to the emergency room of Parkland Hospital in Dallas suggests an answer. This is the primary portal into the health care system for Dallas area residents who are on Medicaid and also for those who are uninsured. However, Parkland often fails to achieve its enrollment goals for the uninsured. The question for public policy is this: If the patients and the doctors do not care who is enrolled in which program, why should policy-makers care? Why not just give Parkland hospital a sum of money each year and let it deliver indigent health care?

Most communities receive health care funding from both the state and the federal government. These funds flow through such programs as Medicaid, S-CHIP, DSH, etc. Each program has its own set of rules, narrowly prescribing how the funds may be spent. Local communities have no power to allocate resources in ways that would maximize their impact.

For example, in one community the greatest return on health care spending may come from fluoridating the water supply. In another, the greatest return may come from improving sanitation. Yet local communities have no authority to use their diverse health care funding to achieve these goals. Ideally, cities should be able to combine all their public health care dollars and freely allocate them so as to achieve maximum health impact.⁵⁰ Our primary goal should be to enroll people in private sector plans. But to the degree that we cannot, we should make access to care as easy as possible.

[page]

A HIFA waiver alone, without any cost-control efforts, would be unwise. But an HIFA waiver could be coupled with market incentives to improve quality and reduce costs and with other appropriate cost-control activities. Furthermore, phasing in parts of the HIFA waiver and limiting eligibility expansions until each phase has been evaluated can limit unanticipated financial risks.⁵¹ What follows is a specific proposal.

1. Medicaid Benefit Changes. Medicaid has a very rich benefit package. Consistently, Medicaid benefits exceed most private insurance benefits in most states. As a result, taxpayers generally have lower benefits in their own health insurance plans than those provided to Medicaid enrollees at taxpayer expense.

This is unfair and unwise. It is unfair because taxpayers should not be forced to provide others with health benefits more generous than they can afford to purchase for themselves and their families. It is unwise because the Medicaid population is largely insulated from many of the cost-controlling, quality-improving innovations that are available to private sector plans.

"Medicaid enrollees should be allowed to join private plans, including employer plans and individually owned insurance."

Medicaid enrollees should be allowed to enroll in private sector plans, including employer plans and individually owned insurance plans. To qualify to accept Medicaid enrollees, a health plan should have to offer benefits at least as generous as any of the plans currently offered to state employees. In addition, the state should authorize a new type of plan that incorporates a health care savings account.

Although the private sector plans may appear less generous on paper than the current Medicaid program, they usually would allow enrollees access to a greater range of providers and facilities. Put differently, this proposal would allow Medicaid enrollees to participate in the same kinds of health plans as other citizens.

2. Eligibility Changes: As noted above, the benefits provided to a new group of eligibles under a federal waiver do not have to be identical to the benefits provided to the currently eligible. Nor does the state subsidy have to be the same. One option is to follow Utah's example and provide primary care only to the newly eligible population. [See the sidebar on "The Utah Experience.] Another option is to allow the new eligibles access to an insurance exchange (see below) but subsidize the insurance on a sliding scale, with additional income reducing the premium subsidy.

To discourage crowding out of private coverage, the state could impose a waiting period between the time individuals lose private coverage and the time they are eligible for Medicaid. In a more extreme manner, the state could deny Medicaid coverage to any individual whose employer offered private insurance. Since Medicaid payments often are made on behalf of individuals who have private health insurance, workmen's compensation and/or liability insurance, the state should vigorously attempt to recover these funds from the appropriate carriers.

3. Marked-Based Cost Control Devices. In order to take full advantage of cost-controlling, quality-improving innovations in the private sector, Medicaid enrollees need access to health plans that compete for customers in the marketplace. One way to provide that access is to allow Medicaid beneficiaries to enroll in employer plans for which they qualify at Medicaid's expense. Provided the plan satisfies the benefit requirements described above, Medicaid should be willing to pay the standard employee contribution or 50 percent of the premium, whichever is less.

[Note: To enable employers to offer the same type of insurance to their employees that is available to state employees, such employer plans must be free from state mandates and other cost-increasing regulations.]

"Structured properly, competition can help meet patient needs."

Those beneficiaries who do not have access to employer plans should be able to participate in a state-operated insurance exchange, or health mart, to directly enroll in individually owned health insurance plans. [See the sidebar on "Making Health Marts Work."] The carriers participating in the insurance exchange would offer a host of competing health plans. Participants could purchase an HMO type plan, a preferred provider plan or a plan combining significant patient cost-sharing with a Medicaid Benefit Account (see below).⁵²

The insurance exchange would have the following features. The role of the state would be to organize the exchange, solicit information from a host of health plans and make information available to the beneficiaries. This would make the plan similar to the federal government employee health insurance system.⁵³ Health plans would be required to accept enrollees at risk-adjusted premiums based on age and sex. Those failing to make a plan choice would automatically be enrolled in the "no frills" plan with limited choices (probably a closed-panel HMO). The exchange could be contracted privately, with Medicaid overhead reductions financing the operation. Health plans offered through the exchange would be exempt from state small-group market reforms and laws mandating health benefits. This could significantly lower the cost of coverage.

The introduction of an insurance exchange would put the marketplace to work to reduce costs and encourage innovation, allowing beneficiaries to choose among private sector plans. This would also make it easier for those leaving public assistance to keep medical coverage, since they or their employer could simply substitute private premium payments for the state subsidy. Other advantages of this plan: reasonably priced coverage for the poor and near poor, continuity that permits the poor to keep their coverage when they leave public assistance, an entry point for Medicaid beneficiaries into the private insurance system, and the publication and distribution of easy-to-understand information about each plan.⁵⁴

The state would move from being a health insurer to a health financier. Its role would be to set the ground rules, determine what minimum benefits private insurers would have to offer, solicit insurers to join, collect information about plan benefits and costs, and provide the information to prospective enrollees through the state insurance exchange.

"Patients should be able to control some of their own health care dollars through Medicaid Benefit Accounts."

4. Patient Power Cost-Control Devices. When patients have first-dollar coverage for health care services, they have no incentive to avoid waste or insure that they get a dollar's worth of value for each dollar they spend. To the contrary, if the out-of-pocket costs are zero, patients have an incentive to utilize health care services until their value approaches zero, at the margin. Similarly, doctors treating patients with first-dollar insurance coverage have an incentive to provide services as long as those services offer any positive medical benefit (or probability of benefit), even if the value of the benefit is well below its cost.

Managed care arose to try and counteract these perverse incentives. But all too often managed care consisted of an impersonal bureaucracy that put cost control ahead of patient welfare. As an alternative, many employers across the country are empowering employees by letting them manage some of their own health care dollars and experience the costs and benefits of prudent health care consumption.⁵⁵

State Medicaid programs need to follow this example. Beneficiaries should have access to plans that combine patient cost sharing with a health care saving account called a Medicaid Benefit Account (MBA).

Since these accounts would be wholly or partly funded with taxpayer dollars, they should probably by restricted to the payment of medical bills and insurance premiums. Beneficiaries who consumed health care wisely and saw their MBA balances grow through time would not be able to withdraw these balances for non-health care spending. However, they would be able to use the funds for medical services not covered by their health plan. And in the future, they would be able to use unspent balances to pay insurance premiums and buy medical care directly after they left the Medicaid rolls. [See the sidebar on "Medicaid Benefit Accounts."]

Expected Benefits from the HIFA Waiver. The HIFA waiver recommendation is based on the twin principles of free markets and personal responsibility.⁵⁶ A number of significant benefits can result from this approach:

• First, moving to a market-based system can provide greater freedom and efficiency than the present system allows.
• Second, individual recipients and providers can be given incentives to improve quality and lower costs, to benefit them and other taxpayers.
• Third, the effects of these improvements and resulting savings can continue to accrue over time.
• Fourth, Medicaid can begin moving from Washington to the states, which better know their own needs.
• Finally, draconian cost increases can be reduced or avoided during a period when available revenue is limited.

[page]

Covered Families with Children (CFC) is a fairly homogenous group of Medicaid beneficiaries. It is also a fairly healthy group. This is one reason why states have had considerable success in moving this group into private, managed care programs. It is also the reason why we anticipate success in moving this group into other private sector insurance, including employer plans. People with disabilities pose a harder problem because their problems range from schizophrenia to mental retardation to blindness. Further, this has been an area of exploding costs. [See Figure II.]

"To care for people with disabilites, we need to contract with the private sector."

Contracting with the Private Sector. The many different health needs of these groups make it unlikely that a private insurer will insure the whole group, as a group. The state of Florida has had some success in finding private sector entities who will contract to care for people with specific types of disabilities. States need to explore such opportunities. For example, disparate providers could serve the mentally ill, the physically disabled, the drug addicted and so forth. The comparative advantage of these various providers would serve to reduce costs and possibly increase the quality of service.⁵⁷

Empowering Patients. As noted above, several states have had some success with "Cash and Counseling," which gives beneficiaries control over some of the health care dollars and allows them tdiscretion in using those dollars. More such empowering programs are needed.

Monitoring Eligibility. A large number of people with disabilities access Medicaid benefits by qualifying for federal Supplemental Security Income (SSI). This program has a history of being riddled with fraud and abuse.⁵⁸ It is not difficult to understand why. Since coverage often is related to medical conditions that are fairly easy to fabricate or exaggerate, individuals and parents have incentives to misrepresent their medical conditions. Unlike the poor and near poor, whose income and assets can be documented, these recipients can "game" the system to obtain coverage.

"Care for the people with disabilities has experienced the fastest growth."

Most states could better determine who qualifies for Medicaid disability coverage by separating it from SSI coverage. States can provide additional resources to the appropriate screening bureau or create a panel to judge whether individuals are truly disabled. Savings from removing the unqualified from the rolls would serve as the funding source.

Tailoring Benefit Packages. Given varying degrees of disability, the benefits package could be constrained so as to pay only those health costs related to the disability, depending on the severity of the health problem, and could offer premium support on a sliding scale.

[page]

As noted above, many states could substantially reduce their long-term care expenses by abandoning cost-plus reimbursements and all its variations. The state should pay for services received rather than paying for the cost of delivering those services. The following are some additional suggestions.

Encouraging Cost-Effective Alternatives. Cost savings could be realized in some cases with home health care services instead of a medical center. However, efficient screening for eligibility would be needed. For example, an individual with disabilities might resist applying for coverage if it implies residence in a nursing home but opt to do so if eligible for home health care. When less costly services also are more attractive, they can have the effect of increasing demand.

Enforce the Asset Tests for Eligibility. The asset tests for eligibility for Medicaid long-term care should work as intended by the legislature. Although legal, establishing Miller Trusts to circumvent income tests for eligibility for Medicaid long-term care is an abuse of legislative intent and is driving up costs.⁵⁹ If Miller Trusts are not what the legislature intended, the law should be changed. If existing trusts are what were intended, additional statutes should exclude trusts that reduce current income and should set clear income eligibility standards.

Encourage Community Care Rather than Institutional Care. In general, care in a home or small facility is preferable to care in a large institution. At a minimum, no bias towards institutional care should exist. Where comparable services are more cost-effective in the community, services should be provided there.

"We can get better long-term care if we pay for results."

To appreciate the opportunities for reducing costs, consider that Oregon, Washington and Wisconsin expanded home- and community-based care to help control rapidly increasing institutional care expenditures. These states were able to provide services to more people with the available budget. Home- and community-based services have helped them control the growth in overall long-term care spending. For example, between 1982 and 1992 the combined number of beds in Oregon, Washington and Wisconsin declined by 1.3 percent, while total nursing facility beds in the U.S. increased by 20.5 percent.⁶⁰

Base Payments on Outcomes. In general, it is better to pay for outputs (i.e., quality of care and consumer satisfaction) than for inputs (i.e. beds, staff, etc.). To get results, a payment system should only pay for what is desired. Quality and satisfaction is what recipients and their families want. An index of quality-of-care indicators should be constructed and used to distribute total funds. Three factors could be averaged to create the index. One-third could be based on resolved complaints, one-third on customer satisfaction surveys of patients or their guardians and one third on such performance measures as changes in number and severity of pressure sores (adjusted for acuity level of population), use of restraints, odors, food, etc.⁶¹ A facility that scored 50 percent higher on this index than another facility would get a 50 percent larger share of payments (per patient, adjusted for acuity). Nothing improves quality quicker than paying for it.

This recommendation is consistent with a recent National Academy of Sciences report, which recommended that all federal health programs pay providers based on objective assessments of quality for the treatment of 15 health conditions.⁶²

[page]

Since its inception, fraud, waste and abuse have plagued Medicaid programs. For example, a 1993 investigative report of the Illinois Medicaid system by the Chicago Tribune found that:

• In one year, 71,064 Medicaid patients had more than 11 visits to a doctor's office (compared to a national average of six visits per year), while four patients had more than 300 visits in one year.
• In one day, one patient saw five doctors, made seven visits to a pharmacy and had 22 prescriptions filled with 663 pills.

The report also uncovered some "Medicaid mills," whose freely prescribed drugs, syringes and other medical products were bought with American tax dollars and sold on the street.

A decade ago, opportunities for reducing fraud nationwide appeared to be much larger. But in the intervening years, activity in this area has been minimal. Efforts to eliminate fraud, waste and abuse should be constantly reviewed to make sure they are up-to-date and cost-effective.

"Since its inception, Medicaid has been plagued by fraud, waste and abuse."

Fraud. Medicaid is especially vulnerable to fraud. It is a large program with a rapidly growing budget. It generates more than $1 billion in medical claims per year, nationwide. The General Accounting Office estimates that fraud and abuse may be a high as 10 percent of Medicaid spending.⁶³

State Medicaid agencies claims data and other medical information could be used to identify fraud abuse, overuse and unnecessary care, but it seldom is. Most abuse is identified through tips or other unreliable means. The numerous jurisdictions having responsibility in a fraud case confounds detecting and prosecuting fraud.

In addition to the little chance of being caught, the penalties for fraud have been light. Perpetrators often have plea bargained or accepted pretrial diversion wherein their court records were sealed if they abided by court-approved probation for a short time. Financial penalties have been very light, even for providers who have billed into the millions. More than 50 percent of cases resulted in restitution of $5,000 or less, which the providers could easily pay. Where higher restitution has been set, actual collections have usually amounted to only a small percentage of the total.

Further, those convicted of fraud are usually free to re-offend. There apparently is little follow-up to ensure that the perpetrators of fraud are barred from the health care system. A major problem is that these individuals/groups relocate and become providers in new, unsuspecting states. Establishment of a state Medicaid-provider information exchange would be useful.⁶⁴ The que tam provisions of current law that allow private citizens to obtain up to triple damages for any proven fraud also may be of use.

What, then, is legally required to prosecute and convict a perpetrator? First, the activity must be clearly illegal. Second, it must be persistent. Third, it must have significant impact. Aside from worker compensation-type fraud, very little recipient fraud exists. The difficulty of overcoming all three hurdles may explain why there are so few convictions. For example, in Texas, recipient fraud has exceeded 1 percent of claims only once. Provider fraud may be more prevalent, but it tends to be committed by small groups that move from area to area and state to state before routine audits identify them.

Abuse. With respect to fraud, clear legal standards must be met. The same is not true of abuse. All too frequently, "abuse" is defined as what someone finds offensive. For example, putting stainless steel caps on the baby teeth of a child is abuse to some. But according to the Medicaid rules, it is not.

Medicaid rules are very complex and can be less than clear. (Remember: The same kind of bureaucracy that develops income tax forms develops Medicaid rules and regulations.) One result is that many providers are unsure of what is required of them. The ensuing confusion and attendant publicity give rise to a popular conception that abuse is pervasive in Medicaid. However, Medicaid provider abuse is rarely proven and should not be expected to generate significant savings.

"The potential to reduce costs by controlling waste is significant."

Waste. The potential for reducing Medicaid costs through controlling waste is real and significant. Medical errors are more prevalent than anyone would like to admit, and the threat of tort liability makes it dangerous to be entirely candid when an error occurs. Some believe the reduction of errors may be the greatest opportunity for reducing health care costs.⁶⁵ As noted above, evidence-based evaluation systems should be developed, starting with those procedures that generate the most spending. Done properly, disease management and care coordination can reduce costs and improve quality of care in ways acceptable to providers and patients alike. Major savings could be realized.

Also, as in the case of fraud detection, the introduction of MBAs would give Medicaid patients an economic incentive to detect and reduce some of the waste that occurs.

[page]

What is the ideal amount of federal oversight and control? None. Since Medicaid is a welfare program, it would be logical for states to include Medicaid into their "welfare package." Indeed, leaving Medicaid as the exception to returning welfare to the states undercuts welfare reform.

Those who seek to continue federal control often cite the need for retaining federal quality standards. This argument should be reversed. Quality of care has not been the hallmark of 37 years of Medicaid. It certainly should not be the only criterion for continued federal meddling in the program. The states are capable of administering Medicaid and are more interested in quality than is the remote, Medicare-oriented CMS.

Even short of a no-strings-attached block grant to the states, federal oversight should be significantly reduced. States should not have to go through a lengthy, time-consuming waiver process to enact essential reforms.

[page]

"An alternative to uncontrolled Medicaid growth exists."

An alternative to uncontrolled Medicaid growth exists. It is not simple, and in many ways it is unsettling. It will require expanded eligibility for certain populations. It will require standing up to powerful interest groups. It will require dedicated staff. The alternative will not stop Medicaid budget growth, but it will lower the rate of that growth. It will provide policy-makers with significantly greater control over costs and health outcomes. It will introduce some of the efficiency of the marketplace into Medicaid programs. And it will allow patients and providers to make more of their own decisions.

Out-of-control increases in Medicaid costs are not inevitable. But if reforms are not made soon, the question in a few years will be: Why didn't policy-makers take control of our destiny when they had the chance?

NOTE: Nothing written here should be construed as necessarily reflecting the views of the National Center for Policy Analysis or as an attempt to aid or hinder the passage of any bill before Congress.

[page]

1. "Medicaid's architects envisioned a program that would provide poor people with mainstream medical care in a fashion similar to that of private insurance. As the decades have passed, that vision has largely faded…poor people continue to rely on providers that make up the nation's medical safety net: public and some not-for-profit hospitals and clinics [that] by virtue of their location or social calling provide a disproportionate amount of care to the poor." John K. Iglehart, "The American Health System-Medicaid," New England Journal of Medicine, Feb. 4, 1999, pp. 403-08.
2. www.whitehouse.gov/omb/budget/fy2003/bud15.html.
3. Amy Goldstein, "State Budget Woes Fuel Medicaid Cuts," Washington Post, Oct. 11, 2002, p. A1
4. "The Fiscal Survey of States," National Governors Association and National Association of State Budget Officers, November 2002.
5. American Council of Life Insurance, April 1998.
6. www.kaisernetwork.org/daily_reports/rep_hpolicy.cfm; posted Jan.14, 2003.
7. American Health Line, November 26, 2002.
8. Vernon Smith et al., "Medicaid Spending Growth: Results from a 2002 Survey," Kaiser Commission on Medicaid and the Uninsured, Report No. 4064, September 2002.
9. For example, the federal government contemplated allowing HMOs to restrict Medicaid patients' access to hospital emergency rooms but dropped the proposal in response to a political backlash. See Amy Goldstein, "Bush Abandons Rule On Limiting ER Use," Washington Post, January 23, 2003, p. A3. Maine Rx, a state program that seeks to obtain additional pharmaceutical rebates, is being litigated before the U.S. Supreme Court. See Charles Lane, "High Court Considers Cost of Prescriptions," Washington Post, January 23,2003, p. A2.
10. "HIFA: Will it Solve the Problem of the Uninsured?" National Health Law Program, HIFA Talking Points, February 28, 2002.
11. "The Fiscal Survey of States," National Governors Association and National Association of State Budget Officers, November 2002.
12. In general, optional populations include children in families with income above federal minimums, adults with children with income above section 1931 minimums (see below), people with disabilities, elders with income above Supplemental Security Income (SSI) levels receiving home- and community-based services, certain workers with disabilities whose incomes are above SSI levels, elderly nursing home residents with income above SSI levels, pregnant women with incomes above 133 percent of poverty, and the medically needy. Two recent federal changes have expanded the options available to states for covering low-income parents under Medicaid. First, the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (PRWORA) created a new category of Medicaid eligibility in Section 1931 of the Social Security Act by requiring states to grant such eligibility to those adults and children who would have been entitled to AFDC under the income and resource standards in effect on July 16, 1996. Additionally, Section 1931 gives states the option to use less restrictive income and resource standards in determining eligibility, allowing states to make families that meet the categorical requirement under the old AFDC program eligible for Medicaid at the higher incomes. You are income eligible to receive SSI if a) you work and have an income less than $1,085 per month per person or $1,587 per month per couple or b) you don't work and have an income less than $520 per month per person or $771 per month per couple, provided you meet other eligibility criteria. Medically needy programs serve those whose incomes are higher than eligibility levels under TANF or one of the Medical Assistance Only (MAO) program for children and pregnant women but are not high enough to cover their medical expenses. The coverage limits are the same as for Medicaid.
13. Examples of long-term care optional benefits include Intermediate Care Facilities – Mental Retardation (ICF-MR), inpatient and nursing facilities for individuals over age 65 in an institution for mental disease, home health care, case management, respiratory care for ventilator-dependent individuals, personal care, private duty nursing, hospice, Programs of All-Inclusive Care for the Elderly (PACE) and home- and community-based services. Under Early Periodic Screening Diagnosis and Treatment (EPSDT) rules, all of these optional services must be provided to children when shown to be needed based on a screening. See Kaiser Commission on Medicaid and the Uninsured, June 2001. See "Medicaid 'Mandatory' and 'Optional' Eligibility and Benefits," Kaiser Commission on Medicaid and the Uninsured, Policy Brief No. 2256, July 2001. Also see John Holahan, "Restructuring Medicaid Financing: Implications for the NGA Proposal," Kaiser Commission on Medicaid and the Uninsured, Policy Brief No. 2257, June 2001.
14. Frank Lichtenberg, "Pharmaceutical Innovation, Mortality Reduction and Economic Growth," National Bureau of Economic Research, NBER Working Paper W6569, May 1998.
15. U.S. Department of Health and Human Services, Mental Health: A Report of the Surgeon General (Rockville, Md.: U.S. Department of Health and Human Services, 1999).
16. Health Care Financing and Organization: News and Progress, March 2000, pp. 5-6.
17. American Health Line, March 31, 2000.
18. www.whitehouse.gov/omb/bud/fy2003/bud15.html.
19. www.cms.hhs.gov/chart/Medicaid/2Tchartbk.pdf.
20. "HIFA: Will it Solve the Problem of the Uninsured?"
21. Ibid.
22. http://www.nga.org/nga/legislativeUpdate/1,1169,C_TESTIMONY%5ED_1949,00.html.
23. www.hhs.cms.gov/chart/medicaid2Tchartbk.pdf.
24. www.whitehouse.gov/omb/budget/fy2003.
25. Over the last 37 years, reform proposals have faltered on the twin problems of long-term care (LTC) and the federal match. The fact that the federal government now transfers Medicare-related costs to the states through dual eligibles exacerbates the problem. And as usual, the real problem is politics, not policy. True LTC reform would mean confronting middle class abuse of the Medicaid benefit as their substitute for private long-term care insurance. Any comprehensive reform has always affected the resultant federal match, creating winners and losers among the states. The result has been almost four decades of incremental, not comprehensive, reforms. Comprehensive reform is only possible if these two problems are solved.
26. "Medicaid: Matching Formula's Performance and Potential Modifications," T-HEHS-95-228, U.S. Government Accounting Office, July 27, 1995.
27. The expansion of benefits allows for a controlled experiment on the impact of Medicaid on beneficiary behavior. The major work in this area is from the National Bureau of Economic Research (NBER). NBER researchers related the expanded generosity of Medicaid plans to the household wealth of recipients. They found that in 1993 Medicaid lowered wealth holdings of eligible groups by $1,996 to $2,259. They also found that the existence of asset tests, not surprisingly, more than doubled the wealth reduction from expanded Medicaid eligibility. NBER also found that the expanded generosity and eligibility for Medicaid increased the consumption of this group by 5.2 percent. See Jonathan Gruber and Aaron Yelowitz, "Public Health Insurance and Private Savings," National Bureau of Economic Research, NBER Working Paper No. 6041, May 1997.
28. David M. Cutler and Jonathan Gruber, "Does Public Insurance Crowd Out Private Insurance?" Quarterly Journal of Economics, May 1996, pp. 391-430.
29. Janet Currie and Jonathan Gruber, "Health Insurance Eligibility, Utilization of Medical Care, and Child Health," Quarterly Journal of Economics, May 1996, pp. 431-66.
30. "Learning from S-CHIP and Learning from S-CHIP II," Agency for Health Care Policy Research, June 1998.
31. Laura-Mae Baldwin et al., "The Effect of Expanding Medicaid Prenatal Services on Birth Outcomes," American Journal of Public Health, Vol. 88, No. 11, November 1998, pp. 1623-1629.
32. See Janet Currie and Jonathan Gruber, "Saving Babies: The Efficacy and Cost of Recent Expansions of Medicaid Eligibility for Pregnant Women," Journal of Political Economy, December 1996, pp.1263-96.
33. For a discussion on giving Medicaid enrollees choice, see Irene Fraser, Elizabeth Chait and Cindy Brach, "Promoting Choice: Lessons from Managed Medicaid," Health Affairs, Vol. 17, No. 5, September/October 1998.
34. In August 2001, under authority granted by Congress, the Centers for Medicare and Medicaid Services (CMS), formerly known as the Health Care Financing Administration (HCFA), announced the Health Insurance Flexibility and Accountability (HIFA) Demonstration Initiative.
35. www.hcfa.gov/Medicaid/obs7.htm.
36. If implementing a statewide waiver is politically difficult, another approach is to do so geographically. Carve out similar economic and social geographic regions. Have half continue the present program, have the other half use the new defined contribution approach. The waiver would measure costs, provider and payer participation, quality of care and patient satisfaction. The waiver could also contemplate block-granting portions of the program to local governments (see the discussion below). This could target inner-city, rural or highly dependent "safety net provider" areas to assure that they are helped or held harmless.
37. We have seen this most recently when Health Care Financing Administration (HCFA) strangled Medicare+Choice in its bureaucratic cradle. The program was greatly hampered by the restrictive nature of the initial legislation. The regulations implementing the law were so restrictive that its marketplace incentives were smothered.
38. In a study that examined caseloads in four-month increments, caseload turnover was about 10 percent during each increment, or about 30 percent per year. See Pamela Farley Short, "Single Women and the Dynamics of Medicaid," Health Services Research, December 1998.
39. www.heritage.org/research/healthcare/BG1618.cfm.
40. Under the leadership of Gov. Jim Douglas, Vermont is submitting a waiver request that would make Medical Savings Accounts part of the Medicaid program.
41. John C. Goodman and Gerald L. Musgrave, Patient Power: Solving America's Health Care Crisis (Washington, D.C.: Cato Institute, 1992), Chapter 6.
42. Diagnosis Related Groups (DGRs) include more than 500 common treatments, ailments and protocols based on historic observations. Hospitals receive a DRG payment for treating a patient, regardless of the actual cost of care.
43. An example of the possible magnitude is shown by a study in Employee Benefit News which estimates that the cost of poor quality health care services is $1,350 per employee. If even a fraction of that figure can be saved per Medicaid recipient, hundreds of millions or billions of dollars in taxes can be saved. See, Craig Gunsauley, "Estimate: 30 percent of Health Spending is Wasted," Employee Benefit News, August 1, 2002.
44. "In health care markets, geography is destiny," concluded a 1996 report from Dartmouth Medical School's Center for the Evaluative Clinical Sciences. For example 1.4 percent of elderly women in Rapid City, South Dakota, received breast-conserving surgery for breast cancer. In Elyria, Ohio, 48 percent did. An elderly patient with back pain in Fort Meyers, Florida, was four times as likely to undergo surgery as a man or woman living in Manhattan. And a man with an enlarged prostate in Newark, New Jersey, was twice as likely to have part of his gland surgically removed as if he lived in New Haven, Connecticut. Tonsillectomies provide another example, particularly over time. In the 1950s, children of insured workers had a 50-50 chance of having their tonsils removed, despite an exhaustive review of the medical literature up to that time that found no evidence that a tonsillectomy accomplished its purposes. Today, tonsillectomy is still the surgical procedure most likely to be performed inappropriately. But now, only one in four is unneeded. An 1980 article calculated that the United States could have saved $4.2 billion in 1975 dollars if the rate at which seven common surgical procedures were performed more closely resembled low-usage rates in parts of Maine and Vermont. John E. Wennberg et al., "Dartmouth Atlas of Health Care," Center for Evaluative Clinical Sciences, Dartmouth Medical School, 1996.
45. One hospital study showed that treatment caused complications in one out of five patients, and about 7 percent of the complications were fatal. As many as eight out of 10 medical practices have never been scientifically validated. See Elihu Schimmel, "The Hazards of Hospitalization," Annals of Internal Medicine, January 1964, pp. 100-10.
46. The National Academy of Sciences recommends that all federal health programs begin paying for quality care rather than paying for services rendered. Initially, the effort would focus on the treatment of 15 health conditions, including diabetes, depression, osteoporosis, asthma, heart disease and stroke. See, Janet M. Corrigan, Jill Eden and Barbara M. Smith, eds., Leadership by Example: Coordinating Government Roles in Improving Health Care Quality (National Academies Press: Washington, D.C., 2002).
47. Moving in the direction of provider networks, as the department is already doing, also can facilitate care coordination in areas that may currently be underserved.
48. See John C. Goodman, "Minority Report," Report of the Blue Ribbon Task Force on the Uninsured to the 77th Legislature, State of Texas, February 2001.
49. Estimating those eligible for Medicaid but not enrolled is difficult at best. Exact estimates are not very reliable because of the problem of "induced demand." Every time Medicaid has estimated an eligible population, any new benefit would induce demand and create many more eligibles. In one study, just over half (51.4 percent) of eligible, nonelderly adults were enrolled Medicaid in 1997. Of the remaining adults who were Medicaid eligible, 21.6 percent had private coverage while 27 percent were uninsured. Another study found that about seven million uninsured children eligible for either S-CHIP or Medicaid are not enrolled. See Amy Davidoff, Bowen Garrett, and Alshadye Yemane, "Medicaid-Eligible Adults Who Are Not Enrolled: Who Are They and Do They Get the Care They Need?" Urban Institute, Series A, No. A-48, October 2001. Of those children eligible for Medicaid or CHIP, one-third are eligible for CHIP while two-thirds are eligible for Medicaid. Eight percent of uninsured, low-income children are illegal aliens and, as such, not eligible for either Medicaid or CHIP. See Lisa Dubay, Jennifer Haley and Genevieve Kenney, "Children's Eligibility for Medicaid and S-CHIP: A View from 2000," Urban Institute, Series B, No. B-41, March 2002.
50. For information on how to devise a safety net that does not unfairly compete with private insurance, see John C. Goodman, "Characteristics of an Ideal Health Care System," National Center for Policy Analysis, Policy Report No. 242, April 2001.
51. The Utah model limits enrollment until program evaluations can be completed to determine the effect of the waiver on a number of factors such as emergency room use.
52. Smaller firms could purchase coverage through the state-operated pool. This has been the case in Arizona. See "Arizona Medicaid: Competition Among Managed Care Plans Lowers Program Costs," General Accounting Office, October 1995.
53. The Federal Employees Health Benefits Program (FEHBP) has four main features: (1) federal employees in most places can choose among eight to 12 competing health insurance plans, including Blue Cross and a number of HMOs; (2) the government contributes a fixed amount that can be as much as 75 percent of each employee's premium; (3) the extra cost of more expensive plans must be paid by the employee with after-tax dollars, and (4) the plans are forced to community rate, charging the same premium for every enrollee. Public employee health benefit options in Minnesota are similarly organized, as is the California Public Employees' Retirement System (CalPERS).
54. This is crucial for favorable health outcomes, given some evidence that "poor" families had substandard results when given high deductible coverage. See B. Lyke, "Medical Savings Accounts: Background Issues," Congressional Research Service, May 6, 1996.
55. See Ronald E. Bachman, "Giving Patients More Control," National Center for Policy Analysis, Brief Analysis No. 399, June 17, 2002; and Greg Scandlen, "Defined Contribution Health Insurance," National Center for Policy Analysis, Policy Backgrounder No. 154, October 26, 2002.
56. Utilizing a HIFA waiver does not preclude other cost-cutting opportunities.
57. See "Medicaid Managed Care: Four States' Experiences with Mental Health Carve-Out Programs," U.S. General Accounting Office, September 1999.
58. Baldwin et al., "The Effect of Expanding Medicaid Prenatal Services on Birth Outcomes."
59. Miller Trusts, also known as Qualified Income Trusts, enable people whose income is too high to qualify for nursing home care partially paid for by Medicaid. This is accomplished by assigning income to the trust, the wording of which can limit how the income is distributed. The trust, for instance, can use the income to make certain payments including insurance, support for a spouse, and $60 per month in personal needs. The excess funds, after the aforementioned payments, are then used for nursing home care with the balance being picked up by Medicaid. This allows some people with incomes too high to qualify for Medicaid to obtain subsidies equal to the proportion of nursing home care they could not otherwise have afforded. See Ronald Lipman, "Trust Helps Person Qualify for Medicaid Nursing Care," Houston Chronicle, August 11, 2002.
60. See "Medicaid Long-Term Care: Successful State Efforts to Expand Home Services While Limiting Costs," U.S. General Accounting Office, August 1994.
61. All payment systems are subject to manipulation, but using these three factors would make manipulation of all the components more expensive than improving quality. At the same time, the administrative burden would not be too large.
62. Janet M. Corrigan, Jill Eden and Barbara M. Smith, eds., Leadership by Example: Coordinating Government Roles in Improving Health Care Realtiy (National Academies Press: Washington, D.C. 2002).
63. Leslie G. Aronovitz, "Medicaid Fraud and Abuse: Stronger Action Needed to Remove Excluded Providers from Federal Health Programs," U.S. General Accounting Office,GAO/HEHS-97-63, March 1997.
64. See Sarah F. Jaggar, "Medicare and Medicaid: Opportunities to Save Program Dollars by Reducing Fraud and Abuse," U.S. General Accounting Office, GAO/T-HEHS-95-110, March 22, 1995.
65. Michael L. Millenson, Demanding Medical Excellence: Doctors and Accountability in the Information Age (University of Chicago Press: Chicago, 1997).

[page]

[page]

Michael Bond, Ph.D., is Director of the Center for Health Care Policy at The Buckeye Institute and a professor in the Department of Finance at Cleveland State University. His work on Medical Savings Accounts (MSAs) and health-care policy reform has received national attention and appeared in a wide range of professional and popular publications, including Health Care Financial Management, Public Personnel Management, Compensation and Benefits Review, Benefits Quarterly and Business Horizons. He is the author of the nation's first practical guide to establishing MSAs (published by The Buckeye Institute in 1997). He earned his Ph.D. and M.A. in economics from Case Western Reserve University.

John C. Goodman, Ph.D., is President of the National Center for Policy Analysis, a national think tank based in Dallas, Texas. Called the Father of Medical Savings Accounts, Dr. Goodman is author of numerous books, including the textbook Economics of Public Policy and the award winning Patient Power: Solving America's Health Care Crisis (with Gerald Musgrave). He is the author of over 50 published articles on health policy, tax reform and school choice. Prior to founding NCPA in 1983, Dr. Goodman was a professor of economics at the University of Dallas and has taught or conducted research at Columbia University, Stanford University, Dartmouth University and Southern Methodist University. He earned his Ph.D. in economics from Columbia University.

Ronald L. Lindsey is a Senior Fellow at the National Center for Policy Analysis and principal of Lindsey and Associates, a private consulting firm specializing in public-management issues. He is a former Commissioner of the Texas Department of Human Services, director of the state's budget office and health care finance analyst for the Texas legislature. Mr. Lindsey specializes in analyzing, evaluating, designing, implementing and managing a wide range of public and private programs. His expertise encompasses topics including appropriation processes, program and performance evaluation, spending and revenue requirements and options, budgetary effectiveness, change management, policy development and implementation, state finance and accounting systems, health and human services, water rights, insurance, records archiving and technology.

Richard Teske is an independent writer and consultant on political and health care policy. For a quarter century he has advised and worked with international, federal and state government leaders and many of the largest managed care, pharmaceutical, biotech, medical technology, long term care and hospital companies. He is recognized as an expert on market-oriented Medicare, Medicaid, Long Term Care and Uninsured Reform, having served for almost eight years in the Reagan Administration in a variety of capacities including Official HHS Liaison to the White House, Principal Deputy Assistant Secretary at HHS, and Associate Administrator for the Health Care Financing Administration (HCFA).