The Cost of Admission — Tiered Copayments for Hospital Use
http://www.100md.com
《新英格兰医药杂志》
After a period of more modest increases, spending on health care is again growing much faster than the rest of the U.S. economy (see Figure 1). Between 2001 and 2002, it increased by 9.3 percent, to $1.6 trillion, or $5,440 per person. Hospital spending, which represents about a third of all health care spending, increased by 9.5 percent, as compared with an average annual growth rate of 3.7 percent between 1993 and 2000.1 Soaring hospital expenses reflect many factors, including demand for new medications and technology, the aging of the population, and increased compensation for medical personnel.
Figure 1. Average Annual Percent Increase in U.S. National Health Care Spending, 1993 through 2002.
Data are from the Centers for Medicare and Medicaid Services.
When health care spending increases rapidly, so do insurance premiums. Between 2002 and 2003, monthly premiums for employer-sponsored health insurance increased by 13.9 percent, as compared with overall inflation of 2.2 percent and a 3.1 percent increase in workers' earnings.2 In 2003, workers paid, on average, $508 (15 percent) of the $3,383 cost of premiums for individual coverage and $2,412 (27 percent) of the $9,068 cost of family coverage. Although the overall rate of increase in premiums may slow in 2004, some increases of 10 percent or more are likely.
Insurers and employers have struggled to find the best response to vastly increased spending. If premiums are simply increased, employers and employees will switch to other insurers or drop coverage. Negotiating lower rates from hospitals is increasingly difficult, because they have regained market power since the 1990s and because people want broad choices. Some believe that patients would be more prudent if they had a better understanding of the costs of services. Insurers and employers have responded to these conflicting pressures by shifting more costs to be paid by people when they actually use health care services (see Figure 2).
Figure 2. Percentages of U.S. Workers with Health Insurance with Different Types of Cost Sharing for Health Care Benefits, 2003.
Data are from the Employer Health Benefits 2003 Annual Survey conducted by the Kaiser Family Foundation and the Health Research and Educational Trust.
In the 1970s and 1980s, the RAND Health Insurance Experiment assigned about 2000 families to insurance plans that varied the price of services and found that "cost sharing markedly decreases the use of all types of services among all types of people."3 Although reduced use had "little or no net adverse effect on health for the average person," it adversely affected the health of the sick poor, and increased cost sharing placed a financial burden on people with chronic health problems.
In conventional insurance plans and preferred provider organizations, patients share costs through deductibles and copayments for office visits, hospital stays, and medications, usually with an annual maximum. Traditionally, cost sharing has been limited in health maintenance organizations (HMOs) and managed-care networks, which emphasize comprehensive and preventive care; patients have had minimal or no out-of-pocket expenses. But cost sharing in managed care is growing. Payments for office visits have increased (typically to $15 or more), and tiered copayments have been implemented for prescription drugs. Some plans have higher copayments for visits to specialists (perhaps $35), and some are introducing annual deductibles, making HMOs look increasingly like conventional insurance plans.
To control costs, managed-care organizations exclude some providers and negotiate lower prices from others, but they have historically treated all hospitals or physicians within their networks equally, and the charges to patients have been the same. The new approach of tiered copayments within hospital networks has attracted considerable interest. In "value" networks, hospitals are assigned to tiers on the basis of cost, measures of quality, or both.
In 2003, about a third of all consumers with employer-sponsored insurance, including half of those enrolled in HMOs, had a separate deductible or copayment for hospital admission, averaging about $200. Six percent had tiered provider benefits of some sort, including tiered payments for visits to physicians, hospital stays, or both.2 A patient might have no copayment or a smaller copayment if admitted to a hospital in the preferred tier and a larger copayment (perhaps $150 to $400) for a hospital in the nonpreferred tier. Programs vary widely with regard to specifics, such as copayments for emergency admissions, admissions to teaching hospitals, and multiple hospitalizations within a year. Insurers have established Web sites providing information about the performance of hospitals for specific surgical procedures and illnesses, and tiered copayments are an incentive for consumers to consult these sites.
In January 2002, PacifiCare of California started its first tiered hospital network. In congressional testimony in March 2004, Dr. Sam Ho, chief medical officer of PacifiCare Health Systems, said, "By driving market share to cost-effective providers and hospitals, based on differential premium contributions and/or copays tied to differential performance of providers, health care costs are approximately 20 percent lower and quality is approximately 20 percent higher than our standard plan."
Blue Shield of California has a tiered hospital network that covers 1.2 million of its 2.7 million members. About 85 percent of the more than 300 hospitals in the network are in the "choice" tier, for which no copayment is required. In Massachusetts, Tufts Health Plan is marketing a tiered network to state employees that includes only one major Boston teaching hospital in the tier with "better quality and efficiency" (and a $200 copayment); other prominent teaching hospitals are listed as having "good quality and efficiency" (and a $400 copayment).
Cost is the primary reason for tiered payments. When Blue Shield of California compared the costs of hospitals in the same area, it found "huge differences" even after adjustment for the case mix and the severity of illness of patients, according to David Joyner, senior vice president of network management. Academic medical centers were not disproportionately in the high-cost category.
The use of measures of quality in assigning hospitals to tiers has been particularly contentious. A key issue is whether public "report cards" on hospitals are more likely to be helpful than misleading. There is no standard method for measuring the quality of hospital care; all existing methods, including the use of claims data, are imperfect. Many plans consider their methods to be proprietary, yet without the details, independent objective analysis is not possible. Physicians and hospital officials, while acknowledging the virtues of rewarding high-quality care, have raised questions regarding the quality of data that are used and the frequent lack of adjustment for socioeconomic status, referral patterns, and coexisting conditions.4 The relative weighting of cost and quality varies among plans. PacifiCare and Blue Shield of California have weighted cost more heavily than quality; Tufts Health Plan has weighted them equally. "There is no correlation between higher cost and higher quality using the metrics that we have," said Joyner. "There are high-cost hospitals that are high-quality and low-cost hospitals that are high-quality. The reverse is also true."
Some important questions about tiered networks remain unanswered. First, what are the effects on health outcomes, particularly for the sick poor and the chronically ill? According to a 2003 commentary in Health Affairs by Thomas M. Priselac, president of Cedars-Sinai Medical Center in Los Angeles, "People with chronic diseases, those who choose to receive their care in an academic medical center, and those who require complex and costly care will be required to pay much more out of pocket for all the wrong reasons." Claims about overall improvement in the quality of care are difficult to verify and may not reflect variations in health outcomes among individual patients. Studies to evaluate health outcomes require access to full information and years to complete.
Second, what are the effects on cost? Tiered networks affect only people who have insurance. Increased cost sharing should decrease their use of health care services and slow increases in premiums, but it is unlikely to change the underlying causes of skyrocketing health care spending. According to a November 2003 report from the Center for Studying Health System Change, most health plans that have experimented with tiered-network designs "have experienced significant operational difficulties and resistance from hospitals and physicians. While employers are interested in the concept, uptake has been modest so far. Whatever the long-term prognosis for the strategy, it appears unlikely that tiered networks will have much near-term influence on market dynamics or costs." In some areas, large hospital systems have refused to participate, threatened to drop out of networks, or used their clout to lobby for inclusion in the preferred tier. A higher-cost hospital may be the only institution that provides critical specialty services; in rural areas, there may be too few hospitals for them to be placed in tiers.
Third, do people want to assume new responsibilities for costs and for the management of their care? Employers choose the health plans they offer their employees. The health plans that have introduced tiered payments view them as an alternative to tightly managed networks, which also continue to be offered. "We are engaging consumers in something, frankly, that they would rather not be engaged in, making choices about costs and quality and having incentives to choose higher value," said Jon Kingsdale, senior vice president for planning and development at Tufts Health Plan. "The easiest, most comfortable thing is the old comprehensive-coverage days when everything was paid, with nominal cost sharing. If we could afford it, that is where all of us would be." Moreover, advocates of an increased role for consumerism in medicine often overlook the data showing that many patients cannot comprehend the information they need to make informed choices. Nearly half of all American adults — 90 million people — have difficulty understanding and using health care information to make appropriate decisions about their health.5
More cost sharing seems inevitable. It is too soon to say, however, whether tiered hospital networks will turn out to be common, a passing fad, or just another focus for competition among health plans or negotiations between purchasers, health plans, and hospitals. For example, in May, the California Public Employees' Retirement System, the third largest purchaser of employee health benefits, voted to exclude 38 high-cost hospitals from its Blue Shield HMO network; the excluded hospitals will have 30 days to rejoin if they agree to prices at the statewide average.
Although tiered payments do not reflect the full cost differences among facilities, they provide consumers with a financial incentive to choose some over others while maintaining a broader choice. Such plans may also provide hospitals with new incentives to reduce their costs or to offer health plans better rates, while avoiding the brinksmanship inherent in threats to drop powerful institutions from networks altogether.
References
Levit K, Smith C, Cowan C, Sensenig A, Catlin A. Health spending rebound continues in 2002. Health Aff (Millwood) 2004;23:147-159.
Gabel J, Claxton G, Holve E, et al. Health benefits in 2003: premiums reach thirteen-year high as employers adopt new forms of cost sharing. Health Aff (Millwood) 2003;22:117-126.
Newhouse JP. Free for all? Lessons from the RAND health insurance experiment. Cambridge, Mass.: Harvard University Press, 1993.
Lee TH, Meyer GS, Brennan TA. A middle ground on public accountability. N Engl J Med 2004;350:2409-2412.
Nielsen-Bohlman L, Panzer AM, Kindig DA, eds. Health literacy: a prescription to end confusion. Washington, D.C.: National Academies Press, 2004.(Robert Steinbrook, M.D.)
Figure 1. Average Annual Percent Increase in U.S. National Health Care Spending, 1993 through 2002.
Data are from the Centers for Medicare and Medicaid Services.
When health care spending increases rapidly, so do insurance premiums. Between 2002 and 2003, monthly premiums for employer-sponsored health insurance increased by 13.9 percent, as compared with overall inflation of 2.2 percent and a 3.1 percent increase in workers' earnings.2 In 2003, workers paid, on average, $508 (15 percent) of the $3,383 cost of premiums for individual coverage and $2,412 (27 percent) of the $9,068 cost of family coverage. Although the overall rate of increase in premiums may slow in 2004, some increases of 10 percent or more are likely.
Insurers and employers have struggled to find the best response to vastly increased spending. If premiums are simply increased, employers and employees will switch to other insurers or drop coverage. Negotiating lower rates from hospitals is increasingly difficult, because they have regained market power since the 1990s and because people want broad choices. Some believe that patients would be more prudent if they had a better understanding of the costs of services. Insurers and employers have responded to these conflicting pressures by shifting more costs to be paid by people when they actually use health care services (see Figure 2).
Figure 2. Percentages of U.S. Workers with Health Insurance with Different Types of Cost Sharing for Health Care Benefits, 2003.
Data are from the Employer Health Benefits 2003 Annual Survey conducted by the Kaiser Family Foundation and the Health Research and Educational Trust.
In the 1970s and 1980s, the RAND Health Insurance Experiment assigned about 2000 families to insurance plans that varied the price of services and found that "cost sharing markedly decreases the use of all types of services among all types of people."3 Although reduced use had "little or no net adverse effect on health for the average person," it adversely affected the health of the sick poor, and increased cost sharing placed a financial burden on people with chronic health problems.
In conventional insurance plans and preferred provider organizations, patients share costs through deductibles and copayments for office visits, hospital stays, and medications, usually with an annual maximum. Traditionally, cost sharing has been limited in health maintenance organizations (HMOs) and managed-care networks, which emphasize comprehensive and preventive care; patients have had minimal or no out-of-pocket expenses. But cost sharing in managed care is growing. Payments for office visits have increased (typically to $15 or more), and tiered copayments have been implemented for prescription drugs. Some plans have higher copayments for visits to specialists (perhaps $35), and some are introducing annual deductibles, making HMOs look increasingly like conventional insurance plans.
To control costs, managed-care organizations exclude some providers and negotiate lower prices from others, but they have historically treated all hospitals or physicians within their networks equally, and the charges to patients have been the same. The new approach of tiered copayments within hospital networks has attracted considerable interest. In "value" networks, hospitals are assigned to tiers on the basis of cost, measures of quality, or both.
In 2003, about a third of all consumers with employer-sponsored insurance, including half of those enrolled in HMOs, had a separate deductible or copayment for hospital admission, averaging about $200. Six percent had tiered provider benefits of some sort, including tiered payments for visits to physicians, hospital stays, or both.2 A patient might have no copayment or a smaller copayment if admitted to a hospital in the preferred tier and a larger copayment (perhaps $150 to $400) for a hospital in the nonpreferred tier. Programs vary widely with regard to specifics, such as copayments for emergency admissions, admissions to teaching hospitals, and multiple hospitalizations within a year. Insurers have established Web sites providing information about the performance of hospitals for specific surgical procedures and illnesses, and tiered copayments are an incentive for consumers to consult these sites.
In January 2002, PacifiCare of California started its first tiered hospital network. In congressional testimony in March 2004, Dr. Sam Ho, chief medical officer of PacifiCare Health Systems, said, "By driving market share to cost-effective providers and hospitals, based on differential premium contributions and/or copays tied to differential performance of providers, health care costs are approximately 20 percent lower and quality is approximately 20 percent higher than our standard plan."
Blue Shield of California has a tiered hospital network that covers 1.2 million of its 2.7 million members. About 85 percent of the more than 300 hospitals in the network are in the "choice" tier, for which no copayment is required. In Massachusetts, Tufts Health Plan is marketing a tiered network to state employees that includes only one major Boston teaching hospital in the tier with "better quality and efficiency" (and a $200 copayment); other prominent teaching hospitals are listed as having "good quality and efficiency" (and a $400 copayment).
Cost is the primary reason for tiered payments. When Blue Shield of California compared the costs of hospitals in the same area, it found "huge differences" even after adjustment for the case mix and the severity of illness of patients, according to David Joyner, senior vice president of network management. Academic medical centers were not disproportionately in the high-cost category.
The use of measures of quality in assigning hospitals to tiers has been particularly contentious. A key issue is whether public "report cards" on hospitals are more likely to be helpful than misleading. There is no standard method for measuring the quality of hospital care; all existing methods, including the use of claims data, are imperfect. Many plans consider their methods to be proprietary, yet without the details, independent objective analysis is not possible. Physicians and hospital officials, while acknowledging the virtues of rewarding high-quality care, have raised questions regarding the quality of data that are used and the frequent lack of adjustment for socioeconomic status, referral patterns, and coexisting conditions.4 The relative weighting of cost and quality varies among plans. PacifiCare and Blue Shield of California have weighted cost more heavily than quality; Tufts Health Plan has weighted them equally. "There is no correlation between higher cost and higher quality using the metrics that we have," said Joyner. "There are high-cost hospitals that are high-quality and low-cost hospitals that are high-quality. The reverse is also true."
Some important questions about tiered networks remain unanswered. First, what are the effects on health outcomes, particularly for the sick poor and the chronically ill? According to a 2003 commentary in Health Affairs by Thomas M. Priselac, president of Cedars-Sinai Medical Center in Los Angeles, "People with chronic diseases, those who choose to receive their care in an academic medical center, and those who require complex and costly care will be required to pay much more out of pocket for all the wrong reasons." Claims about overall improvement in the quality of care are difficult to verify and may not reflect variations in health outcomes among individual patients. Studies to evaluate health outcomes require access to full information and years to complete.
Second, what are the effects on cost? Tiered networks affect only people who have insurance. Increased cost sharing should decrease their use of health care services and slow increases in premiums, but it is unlikely to change the underlying causes of skyrocketing health care spending. According to a November 2003 report from the Center for Studying Health System Change, most health plans that have experimented with tiered-network designs "have experienced significant operational difficulties and resistance from hospitals and physicians. While employers are interested in the concept, uptake has been modest so far. Whatever the long-term prognosis for the strategy, it appears unlikely that tiered networks will have much near-term influence on market dynamics or costs." In some areas, large hospital systems have refused to participate, threatened to drop out of networks, or used their clout to lobby for inclusion in the preferred tier. A higher-cost hospital may be the only institution that provides critical specialty services; in rural areas, there may be too few hospitals for them to be placed in tiers.
Third, do people want to assume new responsibilities for costs and for the management of their care? Employers choose the health plans they offer their employees. The health plans that have introduced tiered payments view them as an alternative to tightly managed networks, which also continue to be offered. "We are engaging consumers in something, frankly, that they would rather not be engaged in, making choices about costs and quality and having incentives to choose higher value," said Jon Kingsdale, senior vice president for planning and development at Tufts Health Plan. "The easiest, most comfortable thing is the old comprehensive-coverage days when everything was paid, with nominal cost sharing. If we could afford it, that is where all of us would be." Moreover, advocates of an increased role for consumerism in medicine often overlook the data showing that many patients cannot comprehend the information they need to make informed choices. Nearly half of all American adults — 90 million people — have difficulty understanding and using health care information to make appropriate decisions about their health.5
More cost sharing seems inevitable. It is too soon to say, however, whether tiered hospital networks will turn out to be common, a passing fad, or just another focus for competition among health plans or negotiations between purchasers, health plans, and hospitals. For example, in May, the California Public Employees' Retirement System, the third largest purchaser of employee health benefits, voted to exclude 38 high-cost hospitals from its Blue Shield HMO network; the excluded hospitals will have 30 days to rejoin if they agree to prices at the statewide average.
Although tiered payments do not reflect the full cost differences among facilities, they provide consumers with a financial incentive to choose some over others while maintaining a broader choice. Such plans may also provide hospitals with new incentives to reduce their costs or to offer health plans better rates, while avoiding the brinksmanship inherent in threats to drop powerful institutions from networks altogether.
References
Levit K, Smith C, Cowan C, Sensenig A, Catlin A. Health spending rebound continues in 2002. Health Aff (Millwood) 2004;23:147-159.
Gabel J, Claxton G, Holve E, et al. Health benefits in 2003: premiums reach thirteen-year high as employers adopt new forms of cost sharing. Health Aff (Millwood) 2003;22:117-126.
Newhouse JP. Free for all? Lessons from the RAND health insurance experiment. Cambridge, Mass.: Harvard University Press, 1993.
Lee TH, Meyer GS, Brennan TA. A middle ground on public accountability. N Engl J Med 2004;350:2409-2412.
Nielsen-Bohlman L, Panzer AM, Kindig DA, eds. Health literacy: a prescription to end confusion. Washington, D.C.: National Academies Press, 2004.(Robert Steinbrook, M.D.)