HEALTH IS A UNIQUE and personal subject. People everywhere, and certainly Americans, respond almost instinctively to health needs. A natural or man-made catastrophe somehow always seems to bring out the best in our human nature, and we reach out to those in need. We expect medical services to be broadly available and of high quality.
But a tipping point is now at hand. The costs of the U.S. health care system are wildly high by international standards and are rising rapidly. At more than 16 percent of the GDP, the cost of U.S. health care exceeds that of any other developed country by one-third and is, for instance, roughly double the percentage spent in Great Britain.1 Costs have escalated so rapidly in recent years that by 2006 they had doubled as a percentage of GDP from their 1975 level. This trend is not confined to the United States. In fact, health care costs as a percentage of GDP are rising steadily in all developed countries.
Looked at another way, health insurance premiums in most years since 1988 have increased at rates double or triple those of general inflation or earnings.2 This means that each year health insurance premiums are taking up a greater percentage of an individual’s income (whether paid directly by the employee or indirectly through the employer), leaving a smaller share for all other expenditures. Certainly, one reason that all too many people lack health insurance coverage is that they are simply priced out of the market. Health insurance costs cannot continue to grow faster than wages forever.
The increase in health care spending has two possible explanations, one benign and the other worrisome. The benign explanation is that as a society grows richer and older and develops new technology, individuals choose to spend more money on health care, including such things as new antibiotics, vaccines, and pacemakers. If this benign explanation is accurate, then rising spending on health care might not be a problem at all. But the worrisome explanation is that health care spending is rising because of an inefficient system that wastes money and forces us to spend more on health care than we really wish to spend. If the second explanation is correct and there are no brakes on rising health care spending, then eventually a financial crisis will ensue as other important economic expenditure categories are crowded out.
The source of rising health care costs undoubtedly involves both explanations. A significant body of evidence suggests that while much new health care spending is worthwhile, a substantial amount of it is wasted. A research study on regional variations in Medicare payments suggests that approximately 30 percent of Medicare expenditures lead to no improvements in health.3 Projections of Medicare costs reveal a system that is out of control and not even close to living within its means. Likewise, health insurance premiums rise dramatically year after year, and 40 percent of Americans are dissatisfied with their own health care costs.4
In light of these facts, can the quality and accessibility of the U.S. health care system be preserved and enhanced while escalating costs are brought under control? The answer must be yes because cost projections show that health care expenditures will take over this country’s large and rising gross domestic product unless changes are made.
A vast body of literature exists on the development of health care in America.5 This chapter, drawing on that literature, traces the historical pattern leading to current developments, presents a picture of where matters stand today, and identifies areas of large and unnecessary costs. Chapter 8 calls attention to the dramatic impact of domestic and international research and development on health. Chapter 9 examines areas where cost pressures are producing changes in the health care system and considers the questions raised most insistently about the effectiveness of that system.
Then chapter 10 evaluates emerging alternatives to the present health care system to determine which of them deserve support and further action. Chosen with care, one or more of these alternatives would lead to an improved system that can provide high-quality health care and bring costs under control.
Some History
The history of health insurance has its beginnings in the 1920s. It evolved during the 1930s as insurance plans were established to cover hospital costs, and it grew during the Great Depression. Eventually various plans were combined under the auspices of the American Hospital Association (AHA) and became known as Blue Cross. These Blue Cross plans, considered socially desirable because they provided benefits to people in need, benefited from special state-sponsored legislation that made them tax-exempt, nonprofit corporations. As such, they were freed from insurance regulations demanding high reserves, but they were burdened by their obligation to sell health insurance to anyone who wished to purchase a policy. High-risk individuals were more likely to purchase insurance coverage because they saw the plan as a good deal. This tendency, known as adverse selection, had the effect of forcing Blue Cross to set a high price for insurance coverage.6
Commercial insurance companies had long been reluctant to enter the health care market. They did so, however, when they realized how to avoid adverse selection: They offered insurance through employers to groups of working individuals, most of whom were relatively young and healthy.
The health insurance system took off during World War II, when wages and prices were controlled but benefits were not. To compete for employees in a tight labor market, employers looked elsewhere and realized they could offer to provide health insurance, among other inducements, as a substitute for banned wage increases. By this time, commercial insurers were ready to go. These alternative forms of compensation were not regarded as taxable income, so employers bought insurance policies and their employees could access the health care system using pretax dollars without incurring significant out-of-pocket expenses.7 As the system evolved and as the tax advantage of employment-based health insurance stimulated its use, the dominant players came to be the employers (those who provided the money) and the insurers and medical professionals (those who provided the services). Beneficiaries of the health care system, the patients, had little incentive to evaluate relative costs or to take the outlook of a normal consumer, while they had every incentive to use the system without restraint because they did not directly bear a significant fraction of the costs.
With the jump start provided by the World War II wage and price controls, the support of labor unions, and the tax-advantaged status of benefits, employment-based health coverage grew by leaps and bounds in the second half of the twentieth century. By 2003, employment-based health insurance covered approximately 160 million people, or nearly three of every five nonelderly Americans.8
Beginning in 1965, the federal government emerged as a major financier of medical services. Established by legislation that President Johnson signed with a flourish in 1965, Medicare is a health program for individuals over the age of sixty-five, and Medicaid is a joint federal-state health program targeted toward the poor and disabled. Initially, both programs followed the general pattern of employment-based insurance in the sense that an eligible person was entitled to receive a set of covered medical services while paying only modest fees.
Medicaid
By 2004, Medicaid spending had risen to $287 billion, with $115 billion spent by the states and $172 billion by the federal government.9 Just two years later, in 2006, the total had grown to approximately $350 billion. More than one in every six Americans is a recipient of Medicaid.
Nonetheless, not all individuals who qualify for Medicaid are enrolled in the program. In fact, some studies estimate that approximately 20 percent of uninsured children are eligible for Medicaid but not enrolled.10 It is not surprising that uninsured adults who are eligible for, but not enrolled in, Medicaid have fewer health problems than adults who are enrolled in the program.11 Many individuals and families do not realize they are eligible for Medicaid or wait to sign up until they encounter health problems.
Medicaid is actually composed of fifty-six separate entities because each state and territory administers its own unique program.12 In order to qualify for federal support, a state’s Medicaid program must cover a number of services such as inpatient and outpatient hospital treatment, prenatal care, children’s vaccines, and nursing homes. The state then receives federal matching funds for all the Medicaid services it delivers. States also receive federal matching funds if they provide a variety of optional services such as diagnostic tests, prosthetic devices, and optometry services. States must apply for waivers if they wish to depart, even slightly, from the basic structure of Medicaid.
Medicaid does not provide money to Medicaid beneficiaries but pays doctors and health care providers directly for services rendered. Copayments or other cost-sharing arrangements are limited by law to extremely low levels.
Eligibility requirements for Medicaid differ by state, but in general they are based on income level, value of personal assets, age, and whether an individual is blind, disabled, or pregnant. Children make up the largest group of Medicaid enrollees (48 percent) whereas the disabled receive the largest component of spending (43 percent). A significant portion of Medicaid funds (28 percent) goes to the aged, including dual-eligible individuals—those low-income elderly who are eligible for both Medicare and Medicaid. Medicaid also plays a large role in such specific health care areas as prenatal and long-term care, and it provides financial support for more than one-third of all births in the United States.13
Flush with cash during boom times in the late 1990s, many states expanded their Medicaid programs by reducing eligibility requirements and increasing coverage for optional services. Since then, health costs have continued to rise while increases in tax revenue to states have declined from the giddy days of the late 1990s. One unnerving result is that 2003 was the first year in which states, taken as a whole, spent more on Medicaid than they did on K–12 education. Medicaid spending growth has slowed recently from 12 percent per annum in 2000–2002 to 8.1 percent per annum in 2003–2004, but these rates of increase are still well in excess of the rates of inflation or the growth in wages.14
In 2004, Medicaid paid for 42 percent of the $158 billion spent on long-term care while Medicare paid for another 20 percent.15 Nursing home care, the single largest component of long-term care, totaled $115 billion in 2004.16
In one of many efforts to control spending growth, states and the federal government have been tightening rules to combat various Medicaid-planning strategies by which patients in need of long-term care transfer their assets to family members in order to qualify for Medicaid. The federal government’s Deficit Reduction Act (DRA) of 2005 lengthened various income look-back periods and tightened eligibility rules to combat this type of abuse. It is not yet clear if the DRA will substantially alleviate the problem or if, over time, people will again outfox the system.
Medicare
Medicare, a federal health care program for individuals aged sixty-five and over, is composed of four parts, known by the first four letters of the alphabet. Part A provides hospital insurance and is financed by a 2.9 percent payroll tax. Part B covers doctor visits and a variety of outpatient treatments and is financed by premiums paid by the individual and general government revenues. Part C, otherwise known as Medicare Advantage, allows seniors to receive government Medicare in the structure of a health maintenance organization (HMO) and is financed by a combination of premiums, 2.9 percent of payroll tax, and general government revenues. Part D provides a prescription drug benefit and is also financed by premiums and general government revenues.
Individuals aged sixty-five years or older become eligible if they or their spouses worked in Medicare-covered employment for at least ten years. Kidney dialysis patients or those who have received Social Security disability payments for at least two years are also eligible.
As of 2006, 43 million people, including seniors and the disabled, were enrolled in Part A (hospital insurance); 40 million in Part B (doctor visits and outpatient treatments); 6.6 million in Part C (the alternative HMO structure); and 28 million in Part D (the new prescription drug benefit).17
The federal government spent $408 billion, or 3.1 percent of GDP, on Medicare in 2006, and the Congressional Budget Office (CBO) projects that this figure will reach 4 percent by 2015. Medicare receives more federal funds than Medicaid does, but in terms of combined state and federal spending, Medicaid is larger than Medicare.
Another important feature of Medicare’s cost structure is that, like Medicaid, a relatively small number of cases account for a large proportion of Medicare’s costs. The costliest 5 percent of Medicare recipients account for 43 percent of Medicare’s costs and have average yearly spending of $63,000 a person. The least expensive 50 percent account for 3.8 percent of Medicare’s costs and have average yearly spending of $550 a person. Medicaid, not Medicare, is the primary source of funding for long-term care because the low-income elderly can qualify for both programs simultaneously and because Medicare contributes only to relatively short stays in nursing homes and other long-term care facilities.
Over time, Medicare’s premiums, deductibles, and copayments have become more significant, at least for high-income elderly Americans. Part A, hospital coverage, is available without a monthly premium for those who have paid the Medicare payroll tax for a minimum of ten years. As of 2008, the Part A deductible amount was $1,024 per hospital stay, meaning that the patient pays the first $1,024 of the total bill. For the first sixty days of a hospital stay, Medicare pays all covered costs above the deductible without copayments.
Part B, which is often called medical insurance in contrast to hospital insurance covered by Part A, had a basic monthly premium of $96.40 per month as of 2008. This premium is higher for single people with incomes in excess of $80,000 and for married couples with incomes in excess of $160,000. For those high-income elderly, the monthly premium is gradually escalated as income goes up until the premium reaches $238.40 per month. Part B also features an annual $135 deductible and a 20 percent copayment for all Medicare-covered services. Part D, the relatively new prescription drug program, offers a wide variety of plans, but most involve monthly premiums, deductible amounts, and copayment charges.
Because of these deductibles and copayments in Parts A, B, and D, as well as some perceived gaps in coverage, many elderly individuals buy supplemental private so-called Medigap policies. The copayment features of Parts B and D have the advantage of encouraging people to be more thoughtful about their use of medical care inasmuch as they must make some contribution to the cost of the drugs or services. However, all reformers of entitlement programs need to keep in mind the welfare of the low-income elderly and the potential burden of the copayments and deductibles. This is one of the reasons that several of the Social Security reform proposals discussed in the last chapter protected the lifetime poor from benefit cuts and thereby increased the progressivity of the system.
Indicators of Large Unnecessary Costs
Any system that allows, and even encourages, people to use its services at a small fraction of cost to them will produce a few predictable results:
Three very different efforts to identify these issues more precisely shed light on the nature and extent of the problem. The New York Times conducted a year-long investigation of Medicaid in New York that produced a stunning set of revelations of fraud and abuse of the system.19 Examples include a dentist who claimed, and was paid for, as many as 991 procedures a day; van services for patients who cannot walk used routinely by those who can at an unjustified cost estimated at $200 million per year; and a nursing home operator who received $1.5 million in salary and profit in the same year that he was fired for neglecting the residents of the home. The authors of the investigation summarized: “The program has been misspending billions of dollars annually because of fraud, waste and profiteering. A computer analysis of several million records obtained under the state Freedom of Information Law revealed numerous indications of fraud and abuse that the state had never looked into.” The retired chief state investigator of Medicaid fraud and abuse in New York City, James Mehmet, who retired in 2001, believed that fraudulent claims made up at least 10 percent of state Medicaid dollars, while 20 or 30 percent more were siphoned off by what was termed “abuse,” or unnecessary spending that might not be criminal. Mr. Mehmet concluded that “about 40 percent of all claims are questionable.”20
Another sign of unnecessary costs is the vast difference in Medicaid spending between states. In 2004, New York had 4.6 million people enrolled in the system and $41.6 billion in Medicaid spending. By contrast, California had 10 million people enrolled and $31 billion in Medicaid spending. The per-enrollee cost is $9,435 in New York compared with $3,100 in California.
Dramatic regional differences in spending occur not only in the Medicaid program but also in Medicare. John Wennberg and associates at Dartmouth Medical School conducted a study focusing on the results of these differences in the late 1990s, and their findings are detailed and nuanced.21 They identified large differences in Medicare spending between comparable groups of sixty-five-year-olds living in Miami and Minneapolis ($8,414 versus $3,341 per enrollee, respectively) and concluded that higher spending did not lead to more effective care. They attributed this result to supply-induced, as distinct from need-induced, spending. The Dartmouth group found, for example, that cardiac bypass surgery rates were “strongly correlated with the numbers of per capita cardiac catheterization labs in the regions, but not with illness rates as measured by the incidence of heart attacks in the region.” Their conclusions point to potential gains from greater use of proved treatments as well as from the elimination of unnecessary measures: “The gains from improving the quality of care are too large to be ignored. They include preventing and reducing morbidity and saving lives and money. The gains from reducing disparities in Medicare spending are also too large to be ignored. The goals are not unreasonable; after all, large metropolitan areas such as Minneapolis and Portland are getting along just fine with relatively modest Medicare expenditures.”22
Still another study that compared the relative use of health care with outcomes arrived at similar conclusions. From 1971 to 1982, RAND, a California think tank, conducted a detailed health insurance experiment that examined the effect of varying coinsurance rates and deductibles on health care expenditures and outcomes. The study, the most comprehensive of its kind, tracked 1,400 families for three years and 600 families for five years. The health of each individual in the experiment was assessed through a comprehensive battery of tests to gauge physical, mental, social, and general health. Coinsurance rates ranged from 0 to 95 percent. Deductibles, the amount the insured must pay before the insurer begins to cover expenses, ranged from 5 to 15 percent of family income up to a maximum of $1,000.
Unsurprisingly, the RAND study found that higher deductibles and higher coinsurance rates led to less frequent use of the health care system and reduced expenditures on health care. For example, a 25 percent coinsurance rate, as compared with free access, led to a 19 percent reduction in expenditures. The conclusion was one of basic economics, even in the area of health care: Higher prices lead to more careful consideration of alternatives and to lower consumption of often unnecessary services.
What surprised many, however, was that individuals in the high-deductible, high copay insurance plans were just as healthy at the end of the study as individuals whose health care was fully paid for by their employers or by government programs. As the authors of the study put it, “Our results show that the 40 percent increase in services on the free-care plan had little or no measurable effect on health status for the average adult.”23 The only significant differences between the two groups came in the categories of dental care and hypertension (high blood pressure) among the poorest 6 percent in the sample. Among the low-income individuals who had plans without dental coverage, there was a tendency to avoid visiting the dentist and therefore to develop dental problems, while low-income individuals in high-deductible plans often were unaware that they had hypertension and lived with the condition for significant periods without treatment. This problem could be eliminated simply by having individuals screened for hypertension before being enrolled in high-deductible plans.
The key result of the RAND study was that health care consumers who were responsible for more of their health care costs through higher coinsurance rates and deductibles spent less money on health care without any adverse impact on their health.
Other studies and an abundance of anecdotal evidence all point to the conclusion that unnecessary spending on health care is rampant. For example, Dr. Arnold Milstein, the medical director of the Pacific Business Group on Health, summarized in Senate testimony the results of his work on this subject as follows: “Inefficiencies in health care delivery comprise up to 40 percent of current health care spending, and these inefficiencies can be eliminated by identification and reward of better performing physicians, hospitals, and treatment options.”24
These conclusions are both disconcerting and encouraging: disconcerting because there are so many unnecessary costs in the system; encouraging because with an improved system there are vast opportunities to attain health outcomes that equal or surpass those produced by the current system at a substantially lower cost.