Reinventing Health Care -- The Role of States

Original Link: http://www.ppionline.org/ppi_ci.cfm?knlgAreaID=450020&subsecID=900204&contentID=254877

By David Osborne

Health care is bankrupting America. Since 1960, costs have accelerated 10 percent a year -- doubling every 7.5 years.1 We spend 16 percent of our gross domestic product on health care, almost double the European average,2 yet the World Health Organization in 1997 ranked the United States 37th in the world for the overall quality of its health-care system.3 Large American firms are at a competitive disadvantage because they spend so much more on health care than their foreign competitors.

But the fiscal squeeze is particularly debilitating in the public sector. By 2005, state and local governments were spending 21 percent of their money on health, almost double the figure from 1972.4 States spent one-third of their money on health by 2005,5 and within a decade, if present trends continue, the health sector alone will devour one-half of state spending. Similarly, Medicaid and Medicare under current policy would consume one-half of all federal revenues within 15 years.6

Yet all this money does not always buy us high-quality health care. According to the Rand Corporation, patients receive the right care only about half of the time. Between 48,000 and 98,000 patients die annually because of medical errors in hospitals.

Meanwhile, 45.7 million people have no health insurance, and the Institute of Medicine reports that 18,000 of them die prematurely every year because they cannot get the non-emergency care available to those with insurance. Many more suffer poor health because they lack care.

Our health-care system is in slow-motion collapse. No one is happy with it: not doctors, not nurses, not hospital administrators, and surely not patients. Tinkering around the edges will not fix it, and simply expanding access to insurance is not enough. Unless we are prepared to pay significantly higher taxes, we have to control costs while we expand access and improve quality.

The federal government must be involved in the solutions, because it finances roughly 56 percent of all health care in America.7 But health-care markets differ from state to state: Urban states have different practice patterns than rural states, for example. Some states have significant numbers of large, integrated health plans, such as the Mayo Clinic, Kaiser Permanente, and Intermountain Healthcare; in others, such integrated plans barely exist. Hence, successful reform models will differ from state to state.

While the federal government should lead, it can use the states as its laboratories. Federal action should encourage the states to experiment and identify best practices, so we all learn which models are most effective in which types of markets. Washington should provide some up-front funding, and then set goals for the states, with regard to both cost control and healthcare outcomes. The federal government should reward states that exceed the goals with extra funding to offset their costs for covering the uninsured. For states that fail to meet the goals, the federal government should implement a backup plan that would allow consumers to pick high-quality, lower-cost health plans sponsored through federal reform.

Recent examples show just how successful this approach can be. State action with a federal backup plan is the same approach Congress used to improve health-insurance regulation under the 1997 Health Insurance Portability and Accountability Act. Allowing states to act as laboratories of innovation was also the methodology behind welfare reform; by the time the federal bill passed in 1996, thirty-six states had already achieved welfare reform, using federal waivers.

With federal support and incentives, states can cut the Gordian knot of our profligate, dysfunctional health-care system. To cut costs by 25 percent, improve quality, and cover every citizen, states should pursue seven big strategies:

Invest in public health and encourage healthy behavior;

Replace fee-for-service payment with managed competition among health plans that charge annual per-patient fees;

Encourage integrated, managed service-delivery systems that use information technology and best practices;

Encourage health plans to purchase "cycles of care" for medical conditions;

Create statewide electronic health-records systems;

Institute new policies to encourage rational end-of-life care; and

Develop a new system of health courts to contain medical-malpractice costs.
This paper will explore all seven of these strategies. First, though, it will briefly survey the present state of health care in America -- and the consequences of adhering to a failed status quo.

The Costs

Health care has turned into the Pac-Man of the public sector, eating up tax dollars that once went to education, transportation, even public safety.

Figure 1 shows the 35-year trend at the state and local level, where growth in health-care expenditures has accompanied declines in spending on other vital needs.

Because states pay for almost one-half of Medicaid, the problem is most acute at the state level. Figure 2 shows that if current trends continue, health care will devour half of all state spending within 10 years.8 Every time health care gobbles up another percentage point -- every 1.2 years -- we lose the equivalent of 150,000 teachers.9

The worst cost explosion is occurring at the federal level, because Washington pays for Medicare and more than one-half of Medicaid. In 1972, these two programs consumed 5.6 percent of the federal budget; by 2006, the figure was 20.9 percent.10 Figure 3 shows the most frightening data: According to projections made by the Congressional Budget Office, under current policy Medicaid and Medicare would consume one-half of all federal revenues within 15 years, while Social Security and interest on the debt would consume the other half.11

As these graphs demonstrate, solving the insurance problem without solving the cost problem is a mirage. Pennsylvania Governor Ed Rendell had it exactly right when he told The New York Times, "If we're ever going to have accessible health insurance for all Americans, we have to begin by containing costs. If costs continue to spiral out of control, there is no way the government can afford to pay for it."12

Fortunately, high-quality health care costs less, because it keeps people healthier, eliminates waste, and ensures that patients get the right care the first time.

The Diagnosis

To develop a workable solution, it helps to understand four basic problems our system faces.

First, the core purpose of our current system is to treat symptoms, not to sustain health. We spend most of our energy and money responding to illness, rather than preventing it.

Second, most medical institutions are designed to provide episodic care for acute illnesses, but the real burden has shifted to chronic problems that need continuing and coordinated care, such as diabetes, asthma, cancer, heart disease, and AIDS. "In fact," report health-care experts Alain Enthoven and Laura Tollen, "about 45 percent of noninstitutionalized Americans have chronic illnesses, and they account for 75 percent of personal health-care spending."13 More than 40 percent of them have more than one chronic condition.14 Unfortunately, our health-care institutions were not designed to provide coordinated care for chronic conditions.

Third, our system is so fragmented among myriad medical practices, hospitals, and insurance companies that it produces enormous waste. Complex administrative processes consume 25 percent to 30 percent of all health-care dollars.15 Because multiple specialists dealing with the same patient rarely coordinate their care, patients fall through the cracks. According to a 2004 report by the President's Information Technology Advisory Committee, one out of five lab tests must be repeated because previous records are not available, and one of seven hospital admissions occurs for the same reason.16

Fourth, our fee-for-service payment system creates perverse incentives. Providers make more money by performing more services. Indeed, if a hospital makes a mistake or omits something important and the patient has to be treated again, that hospital makes more money than it would have if the procedure had gone well.

Studies prove that fee-for-service payment leads to more care but worse outcomes. It actually creates disincentives to improve quality or make care more cost-effective, because practitioners who find ways to cut back on procedures make less money.

John Wennberg and Elliott Fisher at Dartmouth University have studied Medicare data for decades. Their research shows that in regions with more doctors, rates of hospitalization and procedures are far higher -- often twice the level of regions with fewer doctors. Yet this higher spending does not yield better outcomes or more satisfied patients -- it yields just the opposite.17 Other studies show the same pattern with Blue Cross Blue Shield insurance.18 Christine Cassell, president of the American Board of Internal Medicine, sums it up succinctly: "There is a stark correlation between reduced utilization and better outcomes."19

Wennberg believes that up to a third of the $2.4 trillion we spend on health care each year is wasted on unnecessary treatments, overpriced drugs, and end-of-life care that yields nothing. "The Medicare system could reduce spending by at least 30 percent while improving the medical care of the most severely ill Americans," he argues.20

The Cure: Seven Strategies for States

Reformers have put forth two general models to fix this ailing system. Some propose a single-payer, single-administrator system to cut administrative costs. They cite the fact that Medicare spends only 4 percent of its budget on administration, and their ideal model often looks like Medicare for all. Unfortunately, this model would leave fee-for-service payment in place -- insuring continued rapid cost inflation. Nor does it address the fragmentation that creates so much waste. On a practical level, it would require us to put the health-insurance industry out of business, a daunting political task.

Conservatives typically prefer to intensify the market dynamics in the system by making consumers more sensitive to the price of medical services. They propose to do so by maintaining insurance for catastrophic costs, but with high deductibles. Many conservatives also favor health-savings accounts (HSAs), which give individuals money they can use to purchase medical services. Experience suggests that this would change consumer behavior, in both good ways (more efficient use of health care) and bad (less use of needed drugs and therapies by the chronically ill, leading to hospitalization and higher costs).

This "consumer-directed" strategy fails to address the real cost drivers in the system, because it leaves fee-for-service medicine in place. It also fails to change the incentives guiding the vast majority of health-care purchases. By most estimates, 90 percent of health-care costs are incurred by only 30 percent of the population.21 These people will quickly exceed their deductibles, at which point 100 percent of their medical bills will be covered, and any incentive to shop for cheaper care will disappear.

The truth is, market dynamics in health care are unlike those in other industries, and simply pushing harder on those dynamics will not solve the problem. Third-party payment is inevitable if we want people to get health care when they become seriously or chronically ill, but third-party payment also destroys a consumer's incentive to shop for the best deal.

In addition, many consumers are incapable of shopping effectively. Health care is extraordinarily complex, and patients often find it difficult to make informed choices among treatments and doctors.

Finally, while technological advances drive costs down in most markets, they drive costs up in health care. When we buy a car or a computer, we use it for a time and then dispose of it to purchase another model, one that happens to be better and cheaper, because technology has advanced. But health care is different. Each of us gets only one body, and we naturally do everything we can to keep that body going.

In health care, when new technologies succeed, they keep people alive. When those people get old, their systems begin to fail. At that point, we use more technology, at great expense. This timeless but inconvenient fact explains much about the economics -- and the politics -- of modern health care.

We need a third way, an approach to health care that goes beyond the ideologies of left and right. It is built on the seven proposed initiatives introduced earlier, each of which will now be discussed in detail.

1. Invest in health.

One way of describing America's health-care problem is that we experience too much care and not enough health. Our most important goal, after all, is not health care, but health. And the biggest obstacle to good health for many Americans is not a lack of care, it is their own behavior.

According to the U.S. Centers for Disease Control and Prevention, four big factors influence our health: personal behavior; the environment (elements in our air, water, homes, communities, workplaces, and food); access to health care; and our genetic makeup.

Of these four, personal behavior accounts for 50 percent of the variance in our health. The environment (physical and social) and genetics account for about 20 percent each, and health care for only 10 percent.22

Nevertheless, we spend 88 percent of our health resources on treatment, but only 4 percent on changing personal behavior.23 (See Figure 4.)

If we want better health, we must invest in changing behavior and cleaning up the environment. We all know that smoking causes cancer, heart disease, and other health problems, and for 40 years, our governments have been working to reduce smoking rates, with substantial success. But obesity has become an even greater problem. According to the Centers for Disease Control, by 2006, one in four Americans was obese and 36.5 percent were overweight.25 Obesity leads to diabetes, heart disease, and many other health problems. Medical research proves that consistent exercise lowers the risk of obesity, diabetes, heart disease, cancer, osteoporosis, depression, and high blood pressure,26 yet fewer than one-half of American adults get the minimum exercise necessary to enjoy these rewards -- 30 minutes of aerobic activity most days of the week.27

State leaders should pick the top five behaviors that undermine health in their states -- such as poor diet, inadequate exercise, smoking, drinking, drug use, and poor parenting -- and lead massive public campaigns to change those behaviors.

For instance, a state might create a Healthy Lives Trust with the power to define the 25 food and beverage products that create the most harm and levy taxes on them based on the amount of health-care spending they generate. The trust could then be empowered to use that money to create financial incentives for people to exercise, stop smoking, and the like.

We could require health plans to include similar incentives -- such as discounts for people who regularly use an exercise facility, do not smoke, and have healthy weights -- and to charge extra for those who smoke, are overweight, or drink to excess. (United Healthcare, the nation's largest health insurer, recently unveiled just such a plan, based on successful experience with this approach by some employers.28) We could also give employers financial incentives to create effective "wellness programs" for their employees, with the same goals.

Schools, both public and private, can play a role. Twenty states fall below the national health objective of providing seven recommended vaccinations to 95 percent of kindergartners. In those states, and in any other states that fall below the national standard, public-health departments could organize universal vaccinations for schoolchildren.29 All states could provide annual dental checkups in school, because only one-half of children ages 2 to 17 get them.30 Finally, why not ask schools to require participation in physical education or sports every day, as we did a generation ago? If the school day must be lengthened to accomplish this, so be it. Creating healthy habits at a young age will pay dividends for a lifetime.

Another necessary element of a health-promotion agenda is "secondary prevention": Once a condition occurs, we must ensure that it is managed well, to minimize its effects. For instance, we should make sure that people with diabetes change their diets and receive proper care. Many diabetics find their illness difficult to control, and some develop expensive complications: Between 1988 and 1994, diabetes accounted for nearly 9 percent of adult hospitalizations, 12 percent of nursing-home admissions, and 10 percent of adult deaths.31

Programs such as Enhance Wellness, in Seattle; Active for Life, in Austin; and a Matter of Balance, in Boston; educate patients and their families and help them manage such chronic conditions. Canada, the United Kingdom, and Australia have adopted such strategies.32 States could do likewise, contracting with proven disease-management programs to implement statewide efforts. They could also create incentives to encourage health providers to focus on secondary prevention.

2. Replace fee-for-service reimbursement with managed competition among health plans that charge annual per-patient fees.

In a Rand Corp. study, group medical practices that charged a set prepaid fee cost 25 percent to 30 percent less than those operating on a fee-for-service basis.33 The fundamental reason was that the prepaid physicians had clear financial incentives to become more cost-effective, as Enthoven and Tollen explain:

"Prepayment rewards doctors for keeping patients healthy, for solving their problems in economical ways, and for avoiding errors. It encourages superior ambulatory care for patients with chronic conditions, thereby reducing their need for hospitalization. In contrast, the fee-for-service payment system gives doctors powerful financial incentives to do more (and more costly) procedures, which may not be in patients' best interests, financially or clinically."
Enthoven and Tollen go on to argue that prepaid delivery systems, such as health maintenance organizations (HMOs), have similar incentives and even more weapons to improve quality:


"... [A] system prepaid for total costs can examine the full spectrum of care to find opportunities for cost reduction, not just shifting costs to other parts of the system. For example, a prepaid delivery system can evaluate new technologies for their cost-effectiveness and impact on quality and can deploy them as needed. ...
"Prepaid (and partially prepaid) integrated delivery systems are far ahead of small groups and individual doctors in the use of quality-enhancing decision support tools, disease registries, guidelines, automated reminders, performance feedback, patient self-management, linkages to community resources, and electronic medical records."34
To capture some of these benefits, states should create purchasing pools involving all their programs -- Medicaid, the State Children's Health Insurance Program (SCHIP), state employees' and retirees' plans, and others -- and invite health plans to compete for this market on a prepaid basis, based on both price and quality. They should partner with any private and nonprofit employers willing to join the pool.

Wisconsin's insurance program for state employees offers a good example of how states can use prepayment to stimulate price competition between health plans. The program defines a basic benefit package, asks health plans to submit bids specifying the annual dollar amount they would charge for this package, and then ranks those bids.

The Wisconsin program uses price and quality measures to define three tiers. Plans in Tier 1, which are low in price and high in quality, cost the least for state employees. If they prefer a more expensive plan -- because their family physician is not part of a Tier 1 plan, for instance -- they are free to choose it and pay part of the difference. The vast majority of members choose Tier 1 plans, and this fact creates an incentive for health plans to lower their prices. (Members can switch plans once a year.)

Wisconsin put this three-tier approach into effect in 2003. In Dane County, which includes Madison, the state employee plan covers 25 percent to 30 percent of the private (non-Medicare and Medicaid) market. By 2006, costs in Dane County for individual and family plans had fallen 14 percent below the statewide average and 30 percent below the most expensive regions of the state.35

Cost increases would probably slow even more in Dane County if the competition were sharpened and people had to pay more for Tier 2 and Tier 3 plans. Pool participants should have to pay the full cost difference between the Tier 1 plan and any Tier 2 or Tier 3 plan they choose, in order to maximize the incentive to pick a Tier 1 plan.

The one danger of prepayment is that health plans and providers could make money by simply withholding care from members, such as by refusing to do certain procedures or making it difficult to schedule appointments. Plans that did this would quickly lose most of their subscribers, of course. To prevent such behavior in the first place, states could include points for quality in the formula used to rank health plans. They could also generate quality and customer-service rankings of the health plans that bid for state business each year, and make them available to the public. This would create a powerful market incentive for prepaid plans to deliver sound customer service.

If a state crafted such a purchasing pool for all its programs, each plan in the system would provide the same essential benefit package, although individuals and employers would be free to purchase additional coverage. The state and its private-sector partners would define the benefits package based on hard evidence about which services were most cost-effective.

To control drug costs, the package would include a preferred drug list. Generic drugs would require the lowest co-payments; non-generics on the preferred list would be slightly more expensive; and those not on the preferred list would be the most expensive. The state would use the size of its pool to negotiate steep discounts for drugs on the preferred list. By creating joint purchasing consortia with other states, it could increase its bargaining power even more.

The benefits package could have a deductible and require co-payments, to discourage wasteful use of medical care. Preventive care and chronic care should be covered without deductibles or co-payments, however, because such charges can discourage people from getting the treatment they need. (A recent study of three managed-care plans with co-payments of more than $10 for drugs showed a 20 percent reduction in the use of diabetes medications.36) This leads to more emergency-room visits, more hospitalizations, and more nursing-home care. As the New England Journal of Medicine explained in an editorial:


"Attempts to save money through the redesign of insurance plans -- involving caps on benefits and increases in out-of-pocket spending for prescription drugs -- result in the delivery of poor care to chronically ill patients. ... We should be reducing the barriers to treatment and encouraging patients to take appropriate medications for the recommended duration, rather than increasing these barriers by limiting benefits."37
Health savings accounts (HSA s) could also be useful elements of the benefit package, for those with jobs and sufficient incomes. By creating a $1,200 annual deductible and offering an $800 annual health-savings account, for example, a state could give residents an incentive to shop carefully for health services, as well as the ability to select from an expanded menu of choices, because they could use the HSA to buy services not covered by the benefits package. To make this mechanism more effective, states would need to help consumers understand the prices of different medical services, perhaps by requiring that providers post their prices on a state-sponsored web site.

To protect against "adverse selection," in which unlucky health plans attract older or sicker people who drive up their costs, each individual covered by the state purchasing pool would be "risk-adjusted" -- meaning that the price the state paid to health plans would reflect the individual's degree of risk upon entry. In addition, a state would be wise to offer reinsurance to health plans, to cover the cost of catastrophic cases. Without risk adjustment and reinsurance, health plans would submit higher bids, to limit their risks.

One of the most important contributions a true reform package could make would be to separate health insurance from employment. Under our current approach, employers who provide health insurance are at a severe disadvantage in the marketplace, a fact that is driving more and more of them to drop this benefit. Employees are reluctant to change jobs, for fear of losing their health insurance. People with pre-existing health conditions live in fear of losing their jobs, because they could never get insurance on their own. They can even face obstacles to getting new jobs, because their insurance is so costly.

The political hurdles to shifting from employer-based insurance to a tax-financed system would be daunting, however. Insurance companies would face a huge loss of market. Hospitals and doctors would see their power to charge higher prices severely limited by the competitive dynamics of the new marketplace. State leaders would have to convince businesses and individuals to pay higher taxes in order to fund the system.

An analysis by the Lewin Group of one such proposal in Wisconsin indicated that employers' health-care costs would decline more than their taxes would go up, in part because firms that did not provide insurance would now have to pay their share.38 Still, any proposal for a significant tax increase faces tough sledding.

The alternative is to build a pool that captures at least 30 percent of the market and use it to reshape the system, as Wisconsin has in Dane County. A state would begin with its own government employees and retirees, Medicaid, the State Children's Health Insurance Program (SCHIP), and other state health plans, and then add local government and education employees to the pool. In the typical state, these groups total more than 21 percent of the market.39

In addition, states should invite private and nonprofit employers to join the pool. If a state adopted a mandate that residents have health insurance and offered subsidies for low-income people, it could also add many of the uninsured to the pool -- another 16 percent of the population in the average state.

Once the system had demonstrated the power to restrain health-care inflation, more employers would join in order to save money. Because more participants would increase the system's power to control costs, the state could offer businesses guaranteed prices for several years, to induce them to join. If the model could slow annual price increases, more and more employers would be tempted to join the pool. Once half of the state was in the pool, the politics of moving to a tax-financed system that covered all residents would get easier.

3. Encourage integrated, managed service-delivery systems that use information technology and best practices.

The approach I described would reward integrated, managed health systems and provider groups, as it has in Dane County, because they are more efficient. Examples of such systems include HMOs like Kaiser Permanente and Minnesota's HealthPartners; medical groups like the Mayo Clinic, the Cleveland Clinic, and Geisinger Health System; and emerging networks like North Carolina's Community Care, which serves 745,000 Medicaid patients.

Some of these systems employ physicians as staff members, while others pay independent practitioners. Some, like Kaiser, are fully integrated, even owning their hospitals. Others, like Community Care, are virtual networks that knit together many private practitioners and hospitals to act as one coherent system.

According to Wennberg and Fisher, regions dominated by integrated, managed systems have costs up to one-third lower than other areas.40 One reason is that integrated plans have the financial resources to create Electronic Health Record (EHR) systems, which reduce both waste and error rates.

A second reason is that they have a built-in incentive to prevent disease, because prevention brings down their overall costs. A third is that they have both the resources and incentives to use best practices ("evidence-based standards of care," in industry lingo). Fourth and finally, they can use a more cost-effective mix of personnel: fewer doctors, more nurse practitioners, midwives, medical assistants, and nutritionists.

Several of these factors help such systems deliver higher quality, as well. Their integrated nature also helps: Patients fall through the cracks less often. In the fragmented world of most private practices, when a physician refers a patient to a specialist, there is often no coordination. For patients with complex conditions, this can lead to problems.

For instance, a specialist might prescribe a medication that interacts badly with medications the patient is already on, because the specialist is unaware of the patient's medications and the patient has forgotten to tell him or her. Integrated, managed systems typically require that every member have a medical "home" -- a physician who acts as his or her primary caregiver and is responsible for coordinating his or her care. States could require that every person in the state purchasing pool have such a medical home.

Some systems even provide "case management" for patients who consume a great deal of health care. (One percent of the population uses 27 percent of all health-care dollars, and 5 percent consumes more than one-half of all medical expenditures.41) These are typically people with chronic conditions that require a great deal of care, often by multiple doctors.

Sometimes the doctors do not coordinate, as noted. Sometimes patients do not comply with their physicians' instructions -- because they are elderly, or do not speak English, or are alcoholics, or have mental problems. They do not take the medications prescribed; they do not return for follow-up visits; they do not follow the doctor's behavioral recommendations.

Community Care in North Carolina has begun to use case managers to work with these patients and their physicians to ensure better compliance, better coordination among multiple physicians, and thus better health.42 Similarly, the Monroe Plan, a managed-care organization in Rochester, N.Y. , has reduced admissions of babies to neonatal intensive-care units by helping pregnant women on Medicaid get prenatal care and avoid risky behaviors. Every dollar invested in the program has saved $2.43

States could push the adoption of all these practices faster with incentives, such as extra points in the ranking system for degrees of integration and management. They could even reward integrated systems that demonstrate efficiency and high quality with "charter health plan" status. As with charter schools, charter health plans could be given greater flexibility under state (and federal, if possible) regulations, in return for greater accountability for outcomes.

Some may say this model is simply "managed care," which has been tried and rejected. Indeed, managed care by integrated health-maintenance organizations was on the rise throughout the 1970s and 1980s, and it did help slow health care inflation in the 1990s. But it was derailed by a backlash from consumers and doctors in the late 1990s.

The use of managed care was flawed in three ways. First, most employers did not give consumers a choice of plans -- they chose one plan and offered it to all employees, who were unhappy if their doctor was not in the chosen plan.44 Second, few employers encouraged maximum price competition, because they chose one plan and paid virtually the entire premium for employees. Third, managed care did not penetrate enough of the market in most regions to drive prices down significantly.45

A few large employers got the incentives right, however. They asked various plans to bid, then paid a fixed price for plans and let their members choose any plan -- but required them to pay extra for more expensive ones. This created a powerful incentive for plans to lower costs. According to Enthoven and Tollen, "When employers pay a fixed-dollar amount and each employee can keep the full savings, experience shows that high percentages of employees choose economical care. For example, 70 percent to 80 percent of active employees and dependents covered by the University of California, CalPERS, and Wells Fargo in California choose HMOs."46

Wisconsin's experience in Dane County indicates that both patients and doctors can be satisfied in such a system, as long as they have choices. It also demonstrates that a government can restrain costs by including just 30 percent of the market in its managed competition arrangement. This is true because most health plans compete for state employees, and when they make changes to become more efficient and effective to capture that market, those changes affect the rest of their business.

4. Encourage health plans to purchase "cycles of care" for medical conditions.

While health plans would compete to reduce prices in the system we have described, many would still be stuck paying fees for services to physicians and hospitals. States should work to move them to payment for packages of care for specific medical conditions, such as nine months of obstetrical care and delivery, a knee replacement, or a year of treatment for diabetes.

In Redefining Health Care, Michael Porter and Elizabeth Olmsted Teisberg argue that we will get better, cheaper care if medical providers have to compete based on the price and quality of full "cycles of care" for medical conditions. If medical groups had to compete based on price and the results they produced over a full cycle of care (which could be multiple years), they would face very different incentives.

In short, they would be rewarded not just for performing procedures, but for driving down their costs by eliminating errors, managing chronic patients more effectively, providing care in doctor's offices rather than hospitals, helping their patients make healthy lifestyle choices, and the like. As Porter and Teisberg point out, care that produces better outcomes is often cheaper, because preserving health is cheaper than managing disease.

To make this possible, we will need to measure and report risk-adjusted medical outcomes for each condition and provider -- a big job, one that is perhaps best suited to the federal government.

As the data become available, states should not require health plans to purchase cycles of care, but they should certainly encourage it. Staff-model HMOs may not need to do this, because they own their hospitals, have physicians on salary, and do not use fee-for-service reimbursement.

As long as health plans have to compete on price and quality, those that provide the best outcomes for the money will win the competition. In the process, however, states should create incentives to encourage plans to abandon fee-for-service payment.

For instance, states could take off points in the rating system for plans that stick with fee-for-service reimbursement of physicians and hospitals. To enable conversion to payment for cycles of care, states could create electronic exchanges -- websites on which physicians and hospitals would list their prices and outcomes for each cycle of care, enabling health plans (and individuals) to purchase the most cost-effective options.

States might even encourage plans to offer a voucher option: If a woman became pregnant, for example, she might be given a voucher for $5,800, which she could use to purchase nine months of care and delivery. If she could find the same care for $5,300, she could pocket the savings; if her preferred doctor charged $6,500, she would have to pay the extra amount. In this way, health plans could give providers an incentive to offer lower prices.

5. Create statewide electronic health record systems.

As noted, one of every five medical tests must be repeated, and one in seven hospitalizations is ordered because records are missing or information is unavailable. The price tag for all of this needless repetition is enormous: The redundant hospital admissions alone cost $30 billion a year.47

Many integrated-delivery systems now have EHR systems, but the majority of American physicians and hospitals do not yet use them, and most private systems cannot interact with one another. Perhaps the best system has come from a surprising quarter: the Veterans Health Administration (VHA). Between 1995 and 2005, while the rest of the country suffered through a doubling of health-care costs, the VHA's cost per patient remained level. With 10,000 fewer staff members, the VHA more than doubled the number of patients it served.48

The VHA's EHR system is the most advanced in the world. Leaders from Britain, Germany, and other countries seeking to create EHR systems visit to learn about it, and the government of Mexico is installing it in its largest health-care provider system. In 2006, the VHA system won a prestigious Innovation in American Government Award.

How does it work? Every VHA physician is linked by computer to the system. He or she has access to every patient's entire medical record, including all X-rays, CT-scans, and test results. The doctor enters all notes and prescribes all medications electronically.

The VHA has mined the data the system has accumulated -- and other sources of medical knowledge -- to develop best-practice guidelines, which the system provides to physicians through timely reminders. For instance, it might remind a doctor that a patient with diabetes is due for a retinal exam, or that an older patient should be immunized against pneumonia. If a doctor is prescribing a medication that will interact with another drug the patient takes -- or to which the patient is allergic -- the computer will flag it.

The system costs the VHA $78 per patient per year: less than the cost of repeating a single lab test. It has saved billions of dollars while dramatically improving quality. Errors on prescriptions -- which are bar-coded, not handwritten -- fell from the national average of 5 percent to a fraction of 1 percent. Study after study has concluded that the VHA's quality of care is superior to private-sector care. Not surprisingly, customer satisfaction has soared: On surveys using the American Customer Satisfaction Index, the VHA outscores private health-care providers.

EHR is not the only reform responsible for these improvements, but it has had a huge impact. And believe it or not, the VHA's software is available to anyone who wants it, free of charge.

The federal government is already offering states outcome data on providers from its Medicare database, if they will create electronic information exchanges. Governors should use their bully pulpits to convene the leaders of every major health care provider, insurance company, and employer into a consortium devoted to creating a statewide system as good as the VHA's within five years. (Some states might simply choose to adopt the VHA software.)

Financing could be shared among the state and those who would benefit the most: insurance companies and integrated-delivery systems. To encourage widespread adoption, a governor could announce that at the end of five years, his or her state would cease reimbursing any provider not using the system.

The potential for improvement is enormous. In addition to the kind of savings and quality improvements experienced by the VHA, EHR systems could give individuals access to their own records, test results, and the like on a web portal. (To protect their privacy, individuals could determine who was allowed access to their records.) To extend the system's reach, a patient's entire medical record could be put on a smart card that could be read by any computer.

Some day we may even be able to pay for our health services instantaneously, with a swipe of our smart cards. States could also use the EHR system to gather data on quality and outcomes and make this information available to patients on the web portal.

6. Institute new policies to encourage rational end-of-life care.

No one knows exactly how many of our health-care dollars go to elderly people in their last year of life, but 25 percent is a good guess. This is one reason American health care is so expensive: We succeed in keeping many people alive into their eighties, and then spend an enormous amount in their last few months, as their systems collapse. In many cases, this serves no rational purpose and pleases no one. But doctors and nurses are taught to do everything they can to help patients, and in the absence of specific policies that tell them otherwise, that is what they do.

Should a state pay for a knee replacement for an 82-year-old who is expected to die of lung cancer within a year? Should it pay for heart bypasses for 89-year-olds? Governors should engage the public in a wide and deep consultation process to begin to answer such questions. In addition, they should require that all members of the state purchasing pool have living wills, and define a default living will for anyone in the state pool who has not created one.

Including nursing homes and other forms of long-term care in the state purchasing pool is also critical. Medicaid spends 68 percent of its money on long-term care for the aged, blind, and disabled, who make up only 27 percent of its enrollees.49 It pays for nearly half of all long-term care.50

Most long-term care is provided on a fee-for-service basis, with little coordination. Managed care for fixed prices -- in which members can choose their plans and switch annually if they are not happy -- provides incentives for health plans to find the most cost-effective setting for each person, whether in his or her residence, a nursing home, a rehabilitation hospital, a chronic long-term care hospital, or a day program.

Oregon, Arizona, Florida, Texas, and Wisconsin have already proven that such programs improve quality and cost-effectiveness.51 Oregon's case-management approach has reduced claims by roughly 50 percent.52 "Cash and Counseling" programs in several states, which give patients some Medicaid money each month to purchase their own goods and services, also show promise by helping people make choices that keep them out of nursing homes and hospitals.53

The next step would be to integrate managed care for those who are eligible for both Medicaid and Medicare -- among the most challenging populations, because they are both poor and elderly. Medicare and Medicaid policies do not always mesh well, and when both programs are paying for services, overall costs tend to be higher.54

Minnesota's Senior Health Options Program, which integrates Medicare and Medicaid in one managed-care program, has significantly reduced the number of preventable hospital and emergency-room admissions.55 Massachusetts also has a small integrated program, called Senior Care Options. The Commonwealth Care Alliance, which manages care for 2,000 people under this program, has reduced costs significantly in two years while registering impressive customer satisfaction and very low levels of disenrollment.56

After integrating care for the "dual eligibles," a state could ask Medicare for a waiver to integrate all Medicare recipients into its managed-care purchasing pool.

7. Develop a system of health courts to contain medical malpractice costs.

Finally, we need to rethink the way we handle medical malpractice. Our current approach drives physicians to practice "defensive medicine," performing more tests and procedures than necessary so they can prove they covered all the bases if they are sued. Meanwhile, more than 99 percent of those who suffer medical negligence are never compensated,57 and physicians get no clear signals about standards of care.58

This is a dysfunctional system. David Kendall at the Progressive Policy Institute has come up with a bold solution: Create a system of health courts, similar to the system that handles workers' compensation claims, to make it easier and cheaper for patients to seek reimbursement. Kendall explains:

"A health court system would be similar to the workers' compensation system in two ways. First, there would be a schedule of benefits to compensate patients for medical injuries. Second, a health court system would be designed to provide quick, consistently fair damage awards. ... An injured patient would submit a simple claim form, available through her health care provider, to a local health court review board. These boards would investigate claims and determine if they are clear, uncontestable cases of malpractice. In such cases, they would simply order the injured patient's health care provider to pay damages according to a schedule of benefits.

"After a health court review board has ruled whether or not a case is cut-and-dry, appeals of that decision, along with cases that are not clear-cut, would go to trial before a health court judge. ... Lawyers would represent both parties. But unlike malpractice cases in civil trials, health courts would render decisions that would help shape clear legal standards for medical practice. In addition, the health court judges, not the plaintiffs or defendants, would hire expert witnesses to settle questions about medical standards. When health court judges find incidents of malpractice, they would determine awards using the same schedule of benefits applied by the review board."59
Kendall believes that health courts would be less expensive than today's system because more than 50 percent of court awards go to legal costs and lawyers' fees. Medical malpractice premiums should fall, he reasons, as compensation becomes more predictable and the new system clarifies standards of practice.

Conclusion

Almost every idea in this paper is already working somewhere in the country, with the exception of end-of-life policies and health courts. All are achievable, with the proper application of political will. If, by employing the measures spelled out in this paper, a state could hold annual health-care cost inflation to 3.5 percent, while the rest of the country continued to experience the customary 10 percent inflation, it would cut health costs by 26 percent in just five years. At that point, universal coverage would be more affordable.

But the most powerful argument for this approach is the cost of not taking it. Within 10 years, if we do nothing, health care will consume one-half of state revenues and almost one-half of federal revenues.

Where would we find the money to meet such runaway costs? If the future is like the past, we will throw people off Medicaid, narrow the list of treatments covered, and increase deductibles and co-payments for state employees. We will lay off teachers and eliminate extracurricular programs in our schools. We will cut aid to local governments, who will in turn lay off police. We will defer maintenance on our infrastructure, even as our bridges collapse and more people clog the roads and rails. And we will raise taxes.

Meanwhile, 50 million Americans will be without health insurance.

This is a future none of us wants. The time to act is now.