Assessing the Vital Signs for U.S. Health Care Reform
The U.S. health care debate has focused on five basic issues: increasing health care coverage, expanding access to health care, decreasing health care costs, improving quality of care, and promoting prevention and wellness. The following summary provides key insights from RAND research about the tough choices that policymakers face on these issues. RAND’s insights will remain relevant to health care reform in the long term, regardless of the short-term outcome of the current legislative debate.
AP IMAGES/DAYTON DAILY NEWS, JIM NOELKER
Margaret Wilson, who is 93, can remain in her home in Bethany Village, Ohio, instead of going into a nursing home, thanks to the help of Rita Fashjian of Home Instead Senior Care. Expansion of the Ohio program hinges on funding and the supply of home health aides.
1. Increasing Health Care Coverage
About 46 million people in the United States — representing every income level, age group, employment status, gender, race, ethnicity, and geographic region — are uninsured. People with health insurance tend to receive more preventive care, are less likely to avoid or delay needed medical care because of cost, and may have better health outcomes than individuals without insurance. The most widely discussed options for expanding health care coverage are individual mandates, employer mandates, refundable tax credits, and expanded eligibility for Medicaid and the State Children’s Health Insurance Program (SCHIP).
RAND analysts have used simulation models to predict how each of these policy options would affect health care coverage and costs. Although it is likely that several of these policies would be implemented together, RAND estimated the effects of implementing each policy individually. The size of the effects depends strongly on the specific details of the policy when it is implemented.
An individual mandate for Americans to obtain coverage or else possibly risk a penalty would likely increase the number of people with health care coverage by 9 to 34 million. At the low end, RAND estimates that about 9 million people could be added if there were full subsidies for people with household incomes below the federal poverty level ($22,050 for a family of four), partial subsidies for people with household incomes up to 200 percent of that level, and no penalty for noncompliance. Approximately 34 million people could be added if there were full subsidies for people with household incomes below the federal poverty level, partial subsidies for people with household incomes up to 400 percent of that level, and a noncompliance penalty of 80 percent of the cost of the insurance premium (see Figure 1).
Figure 1 — A Stand-Alone Individual Mandate Could Add up to 34 Million Americans to Those Covered by Health Insurance
SOURCE: RAND COMPARE microsimulation modeling results, December 31, 2008.
NOTES: Subsidies are constructed as follows: A full premium subsidy is available for households with incomes up to 100 percent of the federal poverty level; a partial premium subsidy is available on a sliding scale for households with incomes up to the maximum levels specified in the legend (200 percent, 300 percent, and 400 percent, respectively). Penalties are calculated as the percentage of the premium an individual would otherwise have had to pay to obtain insurance in the purchasing pool.
An individual mandate is not likely to substantially increase total health spending; however, government spending would probably increase by about 1–6 percent. Both total spending and government spending depend on the size of the subsidies and the penalty for noncompliance.
An individual mandate is not likely to substantially increase total health spending; however, government spending would probably increase 1–6 percent.
RAND estimates that an employer mandate would likely increase the number of newly insured people by 1.8 to 3.4 million. About 1.8 million people would be added if firms with at least 25 employees were required to offer health insurance. Up to 3.4 million people would be added if firms with at least five employees were required to offer health insurance. Both of these cases presume that a penalty of 5 percent of payroll would be levied against firms that do not offer health insurance.
A stand-alone employer mandate would likely have no discernible effect on total health care spending because relatively few people would newly acquire insurance under this policy change. Consumer out-of-pocket spending would fall slightly, as would government (Medicaid/SCHIP) expenditures because some individuals would switch from Medicaid or SCHIP to employer-sponsored insurance. The largest effect of an employer mandate would be on the firms that choose to offer insurance, although their increased premium payments would be modest, ranging from 1.3 to 2.5 percent.
Refundable Tax Credits
A refundable tax credit, which would reduce the cost of buying health insurance, would likely increase the net number of newly insured people by 2.3 to 10 million. At the low end, about 2.3 million people would be added if the tax credit were $1,000 for individuals and $2,500 for families. Up to ten million people would be added if the tax credit were $5,000 for individuals and $12,500 for families. Full credits would be given to those below the income thresholds of $15,000 for individuals and $30,000 for families, with partial credits phasing out at incomes of $30,000 for individuals and $60,000 for families.
Refundable tax credits would likely have no discernible effect on total health care spending because relatively few people would become newly insured under this policy. However, net government spending would probably increase by 0.5 to 6.3 percent (see the table) primarily because of forgone tax revenue. Consumer out-of-pocket spending would also increase slightly as previously uninsured consumers and those previously on Medicaid or SCHIP would begin to pay for deductibles and copayments.
Refundable Tax Credits for Buying Health Insurance Could Increase Government Spending on Health Care up to 6.3 Percent
|Category of Spending
|Tax Credit Amount (individual/family)|
|Cost of tax credit (billions of U.S. dollars)||$5.8||$29.46||$69.0|
|Change in Medicaid spending (billions, federal and state)||($1.08)||($3.01)||($5.53)|
|Net government spending (billions)||$4.72||$26.45||$63.47|
|Percentage increase in U.S. government spending on health care||0.5%||2.6%||6.3%|
SOURCE: RAND COMPARE microsimulation modeling results, December 31, 2008.
NOTES: Total U.S. government spending on health care is currently about $1 trillion a year. Numbers in parentheses are negative, showing a decrease in spending.
Both Medicaid and SCHIP are partnerships between the federal and state governments. The two health insurance programs are merged in some states while separate in others. Expanded eligibility for Medicaid/SCHIP would likely increase the total number of people newly enrolled in these programs by 6 to 35 million, depending on the income eligibility level. More than one-third of these new enrollees, however, could be individuals who would drop their prior coverage and switch to the lower-cost Medicaid/SCHIP program. That share would grow as eligibility expands to higher income levels. At a minimum, there would be 4.1 million net newly insured people if Medicaid/SCHIP eligibility were limited to people living below the federal poverty level. About 13.9 million net newly insured people would be added if eligibility were expanded to everyone living below 300 percent of that level.
Expanding Medicaid/SCHIP eligibility would probably have little effect on aggregate national health spending. However, there would be a net increase in government spending on health care of 6 to 28 percent and a small decrease in consumer out-of-pocket spending.
AP IMAGES/ERIC J. SHELTON
U.S. Health and Human Services Secretary Kathleen Sebelius, right, tours the Tufts Medical Center in Boston with Massachusetts Governor Deval Patrick, center, and Massachusetts Health and Human Services Secretary Judy Ann Bigby in June. Sebelius announced the use of $200 million in federal stimulus funds to recruit more doctors and clinicians to underserved areas.
2. Expanding Access to Health Care
Adequate insurance coverage is only one important component of access to health care. Beyond financial access (whether someone has insurance), there are issues of potential access (whether someone has a regular provider of medical care) and realized access (whether someone actually receives care when it is needed). In short, access to health care refers to the ease with which an individual can obtain needed medical services.
One way that Congress seeks to expand access to care is by addressing socioeconomic disparities in access. RAND has examined the relationship between access and health for racial and ethnic minorities and for children, highlighting some policy priorities.
Racial and Ethnic Minorities
Racial and ethnic minorities are at greater risk of ill health than their nonminority, nonpoor, better educated peers. However, erasing disparities in health cannot be accomplished simply by achieving universal coverage. Policies that affect public health and the nonmedical determinants of health — from eliminating environmental toxins to promoting physical activity to renovating city parks — are also necessary.
Figure 2 — High-Risk Children Are Less Likely to Have a Regular Health Care Provider
SOURCE: “Disparities in Primary Care for Vulnerable Children: The Influence of Multiple Risk Factors,” Health Services Research, Vol. 41, No. 2, April 2006, pp. 507–531, Gregory D. Stevens, Michael Seid, Retish Mistry, Neal Halfon.
NOTES: To analyze the effects of multiple risks on vulnerable children, the research team studied primary care experiences in relation to five common risk factors: the child’s race/ethnicity, the household poverty status, parents’ education, whether or not the child is insured, and the child’s primary language. The study used data on children and adolescents (up to age 19) from the 2001 California Health Interview Survey.
Children who face the greatest socioeconomic risks (based on race/ethnicity, household poverty status, parental education, insurance coverage, and primary language) also have the greatest difficulty obtaining access to a regular source of primary care (see Figure 2). However, low-income children whose access to care improved through SCHIP also experienced improved health-related quality of life over a two-year period. A survey of parents found that access to a regular source of care had an even greater effect on their children’s primary care than did having insurance.
3. Decreasing Health Spending
U.S. health care spending continues to rise at an alarming rate and accounts for a growing proportion of the national budget. By 2017, about 20 cents of every dollar spent in the United States will go to health care. Trends that have contributed to higher spending include higher prices for services; the aging of the population; the obesity epidemic; and the demand for more and better medical technology, services, and pharmaceuticals. In addition, when health care bills are paid for by insurance companies, by employers, and by government programs, patients may have little incentive to demand high-quality services at lower costs or to decline unnecessary care.
Congress has considered ways to make patients more prudent consumers, for example, by shifting more of the cost burden to them. The idea is to make patients more cost conscious about health care because they are paying more out of their own pockets. RAND analysts have assessed how two policies intended to promote cost consciousness — high-deductible health plans and increased cost sharing — affect costs and health.
High-Deductible Health Plans
AP IMAGES/BIZUAYEHU TESFAYE
Emmett Curran poses with his prescription medication at his apartment in Lynn, Massachusetts. For ten years, he has been taking the cholesterol medicine Lipitor, which has no low-cost generic equivalent. His new insurer under Medicare now refuses to cover it.
A high-deductible health plan is a fee-for-service insurance plan with lower premiums and higher deductibles than a traditional health plan. The deductibles are usually $1,100 per year or more for single coverage, $2,200 or more for families. High-deductible health plans reduce both spending and the use of health care services; however, the spending could be lower because healthier individuals, who do not expect to need much medical care, enroll in these plans. The early evidence also suggests that these plans have mixed effects on quality of care: The use of preventive services appears to increase, but enrollees are less likely to receive other needed services.
Increased Cost Sharing
A classic RAND study, the Health Insurance Experiment conducted between 1971 and 1982, found that consumers respond to increased cost sharing by reducing the use of both highly effective and less-effective care in roughly equal proportions. At the time, cost sharing did not significantly affect the quality of care received by participants or the health of most participants.
More recently, RAND examined the effect of different benefit designs for prescription medications. Each 10-percent increase in copayments for these drugs decreases their use by 2 to 6 percent. This form of increased cost sharing is associated with worse adherence to drug regimens. For some conditions — such as congestive heart failure, lipid disorders, diabetes, and schizophrenia — increased cost sharing for prescription drugs is also associated with increased use of other health care services, offsetting the potential cost savings. As a result, increased cost sharing for some prescription drugs may actually increase health care use. Conversely, reducing drug copayments for sicker patients can sometimes save money by increasing their compliance with drug regimens, thus avoiding costly emergency and hospital care.
AP IMAGES/WILL KINCAID
The Standing Rock Reservation tribal clinic in Fort Yates, North Dakota, failed to diagnose Victor Brave Thunder, above, with congestive heart failure, giving him Tylenol and cough syrup when he complained of discomfort and sleeplessness. He died in April 2009 at age 54 while awaiting a heart transplant at a Bismarck hospital.
4. Improving Quality
Despite spending $2.5 trillion annually, the U.S. health care system is plagued with inefficiency and poor quality. Adults in the United States receive only about half the recommended care, and poor care probably contributes to thousands of preventable deaths each year. The need to improve the quality of care has not figured prominently in the current U.S. health care reform debate. RAND research has shed light on three proposed approaches to improving quality, which should also in theory cut the costs of care: widespread adoption of improved health information technology, pay for performance, and public reporting of performance information.
Health Information Technology
RAND’s mathematical modeling suggests that widespread adoption of health information technology could yield annual efficiency savings of $77 billion after 15 years, but only a small number of empirical studies have examined the likely savings and costs. Several small studies suggest that the use of health information technology should also improve the quality of care by reducing medical errors and adverse drug events, increasing the rates of recommended care, and decreasing the duplication of tests. However, it is not clear whether these studies can be generalized to the entire health care system. Substantial new investments in the technology infrastructure will be required to reap the projected benefits.
Pay for Performance
Pay for performance uses financial incentives for physicians and hospitals to stimulate improvements in quality of care and, in some cases, to reduce costs. A RAND analysis of the largest pay-for-performance experiment in the nation found no substantial gains in quality, although physician organizations involved in the program did report increased feedback to physicians about their performance and faster adoption of health information technology. A few studies have found that hospital pay-for-performance programs produce modest improvements in the reliability with which appropriate care is provided; however, studies of physician pay-for-performance programs show mixed results.
Public Reporting and Transparency
In theory, making information about the performance of hospitals, health plans, physicians, and other care providers widely available could motivate those who receive low ratings to improve their performance in order to protect their reputations and market shares. A RAND synthesis of the available evidence has concluded that publicly releasing hospital performance data stimulates quality improvement activity; however, how public reporting affects clinical outcomes (effectiveness, safety, and patient centeredness) remains uncertain. Providing performance information on physicians is not sufficient to change their behavior; rather, a combination of educational strategies might be more effective. Convincing consumers to use information about quality when making health care choices could also take time. Common barriers include limited education, limited reading skills, and age-related physical and cognitive deficits.
AP IMAGES/WADE PAYNE
Biking commuter Jim Richards rolls into work at Mast General Store in Knoxville, Tennessee. The retailer pays workers $4 a day to ride, walk, or catch a bus. A study in the Journal of Physical Activity and Health has found a correlation between “active transportation” and lower obesity rates in 17 industrialized countries.
5. Promoting Prevention and Wellness
Combating the obesity epidemic and managing chronic disease have been topics of concern in the U.S. congressional health reform debate. RAND investigators have studied the effects of obesity and disease management programs on health and on health spending.
Obesity is one of the leading causes of preventable illness in the United States. The biggest contributors to the obesity epidemic are poor diet and reduced physical activity. Having steadily increased among the U.S. population over the past two decades, obesity is linked to higher rates of chronic disease than is smoking, drinking, or poverty. Obese individuals also incur higher total health care costs: The obese spend 36 percent more on health care services and 77 percent more on medications than do those of normal weight.
Figure 3 — An Obese 70-Year-Old Would Cost Medicare $144,000; a Person of Similar Age and Normal Weight Would Cost Medicare $107,000
SOURCE: “The Health and Cost Consequences of Obesity Among the Future Elderly,” Health Affairs — Web Exclusive, September 26, 2005, pp. W5-R30–W5-R41, Darius N. Lakdawalla, Dana P. Goldman, Baoping Shang.
RAND simulation modeling projected the likely savings to Medicare from reducing obesity among 70-year-old beneficiaries. Reducing obesity was the only behavior change among this age group that resulted in Medicare savings: Although there was little difference in overall life expectancy, the elders who were not obese could anticipate nearly seven years of life free of disability (and thus were less costly to Medicare), while the obese could expect only four years of life free of disability. Starting at age 70, an obese person would cost Medicare about $144,000, while a person of similar age and normal weight would cost Medicare about $107,000 (see Figure 3). A strong link has not been established between lifetime health care costs and healthy behaviors, but RAND’s findings suggest that weight reduction among the obese should be an urgent public health priority.
Neighborhood characteristics can play an important role in stimulating exercise and combating obesity. Living within a mile of a public park significantly increases the likelihood that residents will exercise. In addition, residents of socioeconomically disadvantaged neighborhoods have less access to fresh fruits and vegetables than do their counterparts in more affluent areas.
Focused on people who already have chronic medical conditions, disease management programs are intended to reduce the long-term cost and burden of chronic disease through some combination of patient self-care, provider training, and individualized care plans. However, the evidence is mixed on disease management’s effect on spending, and some evaluations suggest that disease management actually increases spending. The effect on spending may differ depending on the disease management approach or the disease targeted. Research also suggests that disease management may increase the delivery of appropriate health care, although the health effects are uncertain. In short, disease management programs may improve health outcomes but may also increase costs.