|
|
| |||||||||||||||||||||||||||||||||||||||||||
Underlying our analysis are three basic assumptions. First, health care spending has high intrinsic social value, and the primary driver of cost increases is technical progress1,2 — for example, new tests and therapies or new knowledge about the benefits of existing ones. This perspective is supported by the observation that health care costs are increasing throughout the world, regardless of the system for financing health care. The aging of the population and increasing numbers of patients with chronic illnesses contribute to the problem,3 but the increasing numbers of effective therapies for these populations are major factors in cost trends.
Second, the value obtained for health care expenditures must be enhanced. Unconstrained growth in medical spending is threatening the incomes of individual patients, the cost structures of employers, and the fiscal balance of government. Third, the high social value of health care limits policy options for containing health care spending.
In short, we want cost control, but we also want broad access to health care and continued innovation in medical science. Trade-offs among these goals are inevitable, and they can be minimized only through thoughtful policies.
Table 1 lists 12 major options for reducing health care spending, with comments regarding barriers to their implementation. Rigorous experimental studies of the effect of these options are scarce, and estimates of their independent effects are not available. For example, estimates of the savings that might be derived from the use of electronic medical records include savings from other options, including improved care for patients with chronic conditions.4
|
Our belief is that there is no single "magic bullet" among these choices; our goal is to promote discussion leading to effective policies that support several approaches. We do not think responsible health care leaders can be against all of these options; indeed, we think it is insufficient for leaders to support only one or two. Policymakers must identify an array of choices with sufficient cost-savings potential to moderate financial pressures on health care.
Greatest Potential for Cost Savings
Several types of payment reform have been suggested5,6,7,8 and are being tried throughout the country. All of them are potentially disruptive to providers whose businesses are based on fee-for-service payments. Nonetheless, improving quality and efficiency in a pure fee-for-service environment is so challenging9,10 that we believe the question is not whether payment reform should be pursued, but how to pursue it without precipitating major discontent or disruptions in care.
The most potent version of payment reform is budget-based capitation, in which providers receive a fixed amount of money to cover all health care needs of a population of patients. Experiments with capitation in commercially insured populations demonstrate reductions in cost, but they have often resulted in consumer and provider dissatisfaction. Patients have rebelled against limitations on their choices of providers, and providers have rebelled against capped budgets and inadequate risk adjustments to payments.11 Although capitation is successfully used in some staff-model delivery systems, efforts to extend this payment approach more broadly have had limited success.
Short of full budget-based capitation are a variety of options, including partial capitation (e.g., a fixed payment to primary care physicians for their populations); case rates, in which a lump sum is provided for specific procedures; and pay-for-performance systems, in which bonuses for improved quality and efficiency are available to augment fee-for-service payments. Despite the limited data on the effect of such approaches, we cannot conceive of a meaningful attempt to decrease the trend in costs that does not include some form of payment reform. We also believe that payment reform is likely to be most effective when providers are organized into delivery systems that can accept responsibility for cost-mitigation goals.12
Another promising approach to cost containment is strengthening effectiveness reviews for new drugs and forms of technology. Some candidates and many policy experts support a new national institute to conduct such analyses, which could be required before decisions regarding reimbursement are made.13 Concern about this approach comes from members of industry, who worry about the possible effects of such reviews on the time and costs associated with getting products to market.
Health information systems that include electronic records have significant potential for cost savings and enjoy strong political support. Policymakers often focus on the personal health record (e.g., a small data-storage device carrying key clinical information), but we believe the greatest cost-reducing effect of electronic records will result from improved coordination among health care providers and from decision support that improves clinicians' use of tests and treatments. Such decision support has the potential to decrease variation among physicians in the use of health care services, thereby reducing both baseline costs and cost trends.
This potential is largely unrealized to date, however. Critical barriers include the requirements for capital investment and standardization of administrative and clinical data. Even more daunting is the need for cultural change among physicians, who must be willing to use decision-support systems if electronic records are to improve their care.14
The improved care of patients with chronic conditions such as diabetes mellitus or coronary artery disease is a promising focus for cost reduction, because about 70% of health care costs are generated by 10% of patients, most of whom have one or more chronic diseases. Improved reliability and coordination of the care of these patients could reduce their need for hospitalization. This strategy has moderate bipartisan support, reflecting awareness of the frequent failure of our health care system to deliver interventions that are likely to be beneficial to patients with these conditions.15
As is true with information technology, however, the evidence that improvement in the care of patients with chronic conditions reduces costs falls short of the apparent opportunity.16,17 Numerous interventions are known to be cost-effective — that is, they improve health at a reasonable incremental cost. However, few interventions (e.g., disease-management programs for patients with heart failure) have been shown to actually save money while improving patients' health.
Nevertheless, we believe that the cost-saving potential of improvement in the care of patients with chronic conditions may yet turn out to be meaningful. Effective care-improvement programs generally require organized systems of care,18 as compared with a fragmented system of independent practitioners who often find these programs difficult to maintain. Implementation of these programs will also require some payment reform because institutions and practitioners currently lose money by reducing preventable hospitalizations, and proactive care-management services are typically not covered.
Intermediate Potential for Cost Savings
The observation that health care costs are concentrated in the period just before the patient's death raises concern that our health system uses excessive resources to extend the life of dying patients. Political candidates are understandably wary of engaging in this discussion, but health care providers are exploring the effect of greater use of hospice and palliative care services and more complete disclosure to patients of the risks and benefits of proposed interventions.
Medicare data from Oregon indicate that the use of hospitalization and intensive care units in the last months of life can be decreased without compromising the care of dying patients and their families.19 However, these data show that any serious attempt to change end-of-life care requires deep cultural change that extends well beyond the provider community.
Two broader approaches to cost control have support from opposite ends of the political spectrum. Political conservatives have championed consumerism, expressed through insurance products with high deductibles or copayments, health savings accounts, and "transparency." Transparency means making available information about the cost and quality of health care services so that patients can become informed consumers.
Although the impact of this approach is unknown, we believe that cost savings are likely to be limited by the medical needs of the 10% of people who account for 70% of costs. These patients tend to exceed their financial liabilities associated with these products quickly, and their ability and willingness to behave like shoppers who can make trade-offs in cost and quality are uncertain at best. In addition, these insurance products have thus far proved unpopular with employees despite their lower effect on their paychecks, and enrollment to date has been low.
On the political left, advocates of the single-payer approach argue that elimination of the employer-based commercial insurance system would dramatically reduce administrative costs.20 Despite the large savings that would result, political support for this approach is currently limited. The strongest resistance to the single-payer approach comes from the commercial insurance industry, but providers worry that this approach would extend the lower reimbursement structure of Medicare and Medicaid to all patients, and these payments would not increase fast enough to cover increasing provider costs. Thus, for the time being at least, the development of a broad coalition around a single-payer system is unlikely. There is, however, widespread interest in reducing administrative costs by pursuing standardization of the claims-payment systems of U.S. private insurers (e.g., through adoption of a universal billing form).
Lowest Potential for Cost Savings
Two familiar targets for cost reduction are malpractice and drug-pricing reform, but the potential savings from these approaches are probably small. Although the current malpractice system is an inefficient way to protect patients from negligent care,21 the direct costs of malpractice premiums and estimated costs of "defensive medicine" are not major factors in overall health care spending.22 In any case, political support for malpractice reform is partisan and weak because of the resistance to major changes on the part of plaintiffs' lawyers.
Costs can be reduced through more restrictive drug formularies and tougher price negotiations,23 but the savings are modest because pharmaceuticals account for just 10 to 15% of health care spending. The political appetite for tight government control of drug pricing is also limited by concerns about its effect on the development of new drugs.
Enhanced primary prevention efforts (e.g., programs to reduce smoking, alcohol abuse, or obesity) have strong bipartisan support, and they would lead to important general health benefits. This approach makes particular sense for employers, who can enhance the health of their workforce, and also delay the onset of serious illness among their employees by many years, at which point most costs would be absorbed by Medicare.
However, candidates would be ill-advised to believe they can fund broader access to health care through savings derived from primary prevention. Prevention is more likely to delay than to eliminate long-term societal costs, because longer life spans mean more years of health care adding to overall costs. Controversy persists regarding whether improved care can lead to significant savings through a "compression of morbidity" — that is, longer and healthier lives with a relatively quick, low-cost period of illness just before death.24 Regardless of what the right answer is, savings from increased primary prevention will not be substantial in the near term.
Rationing Options
Should other options fail to provide sufficient cost reductions, policymakers may be forced to consider various forms of rationing, including two types that have been proposed from different ends of the political spectrum. From the left comes the proposal for fixed, all-payer budget ceilings for health expenditures, such as those that are used in Canada and some European countries with multiple payers.25 The U.S. experiment with this approach is the Medicare funding policy that requires decreases in payments to physicians when overall spending increases.
Although there would certainly be considerable savings from this approach, inflation in health care spending in countries that use it does not lag far behind ours25 because of the constant political pressure to increase spending for essential services. Administration of these budgets would require a large government role, and such a strong government regulatory role is not likely to gain consensus in the U.S. culture.
From the right come proposals for indirect rationing by limiting Medicare and Medicaid payment for new or "discretionary" services. This approach would have Medicare evolve to provide a defined contribution toward the health care costs of the U.S. elderly instead of defined benefits. Under this framework, patients who are able to pay for the services that are not covered would do so with their own money, and patients who are unable to pay would go without. We think such a dramatic and visible increase in the two-class nature of our health system is too obviously inconsistent with our core values to be politically viable.
Discussion
We see three paths toward controlling health care costs. First, we could allow the current situation to persist. Consequences would almost certainly include increased taxation and financial burdens on individual patients and businesses, greater competition for scarce governmental resources, and a continued increase in the number of uninsured Americans. The alternative extreme would move our country toward one of the indirect rationing methods described above. This path would be practical only as a last resort. The third path would be to assemble the most reasonable package, short of rationing, using a combination of the other ideas mentioned above, and to try to bend the trend line in increasing health care costs.
While recognizing that the many stakeholders in health care will have different preferences, we suggest the following. First, modify reimbursement with the explicit goal of rewarding the practice of evidence-based medicine, reductions in variance among physicians in the use of services, and improvement in the care of patients with chronic conditions. We recommend consideration of blended arrangements including pay-for-performance programs, case rates, and even adequately funded and appropriately risk-adjusted capitation.
Second, invest in new effectiveness-review bodies. These groups would inform decisions regarding the coverage for and use of health care tests and treatments in the future.
Third, maximize support for electronic medical records with computerized decision support, recognizing that this will involve considerable national investment and cultural change. Such support can come in the form of higher reimbursement for physicians who have adopted electronic records or grants from hospitals, payers, or government to provide support for their implementation.
Fourth, enhance the standardization of health care transactions in order to drive down administrative costs. Fifth, provide support for regional efforts to improve the quality of care at the end of life. Finally, provide support for prevention programs, not because they save money, but because they lead to a better quality of life and a more productive workforce.
We recognize that many ideas for cost containment are not addressed here and that there are many potential cost-containment packages besides our approach. Our intent has been to set out a framework for considering various proposals. To deal successfully with this important issue, we must move away from cliches that fit our own political beliefs and grapple seriously with the true effectiveness and the political reality of each of these ideas. We need a real and honest dialogue on this issue — particularly in a presidential election year.
No potential conflict of interest relevant to this article was reported.
Source Information
From Partners HealthCare System, Boston.
References
| |||||||||||||||||||||||||||||||||||||||||||
Related Letters:
Slowing the Growth of Health Care Costs
Shenkin B. N., Rees M. K., Stith D. M.
Extract |
Full Text |
PDF
N Engl J Med 2008;
359:434-435, Jul 24, 2008.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
