Containing Health Care Costs
Given the many pressing demands on finite national resources and the rapid increase in the share of those resources devoted to health care, policymakers, business leaders, and ordinary citizens must be concerned with both the country's high absolute level of health care spending and its rapid rate of growth relative to the overall economy. As noted in the preceding section, cost concerns have been a major obstacle to efforts to expand access to health care for those who are now uninsured or underinsured. High health care costs are also a major contributor to growing anxiety among the middle class about the adequacy and continued availability of their health coverage.
COSTS IN CONTEXT
To serve health and access as well as cost containment objectives, policies to limit the rate of cost escalation need to be grounded in the concepts of:
value—how health care spending relates to the achievement of desired outcomes;
affordability—how health care spending relates to individual and societal resources; and
equity—how the financing and distribution of health services affects different groups.
Further, to be sustainable, cost management tools and monitoring structures should encourage and emphasize individual, professional, and organizational accountability. Detailed efforts to regulate prices, services, and other aspects of day-to-day health care delivery run two major risks. First, some health care practitioners and providers, health plans, and consumers may be preoccupied with manipulating the system rather than achieving more efficient and effective health services. Second, such manipulation may inspire ever more complex and voluminous rules that would ultimately defy sensible management or compliance by even the most well-intentioned participants. The current balance between delegation and regulation generates considerable tension and dissatisfaction, but some disagreement about the probable costs and benefits of any particular balance appears inevitable.
ELEMENTS OF A STRATEGY
As explained in the Preamble, the committee has not developed a comprehensive proposal for health care reform. It also has not formulated a comprehensive cost containment strategy to reduce the rate of increases in health care costs. One reason is that evidence about the effectiveness of specific tools supports a rather cautious view about their prospects for significantly limiting the rate of increase in health care costs or even achieving some ''once-and-for-all'' savings. Reforms should not overlook opportunities for such savings (and the associated shift to a lower trajectory of cost escalation), but they should avoid overselling strategies of modest or unproved effectiveness. (For reviews of the literature on the impact of cost containment strategies, see CBO, 1992a, 1992b; EBRI, 1991; IOM, 1989b, 1993b; Jencks and Schieber, 1991; Newhouse, 1992; Thorpe, 1992.)
Moreover, although some individual committee members had strong philosophical views about managed competition and other relatively sweeping prescriptions for controlling health care costs, the committee did not reach consensus on such prescriptions as part of this particular effort. For policymakers and the public at large, however, the next few months will bring an intense effort to forge, from contending views, an effective consensus about short-term and long-term strategies to reduce
the rate of increase in health care costs and help finance expanded access.
The remainder of this section briefly describes the elements of a cost containment strategy on which the committee focused and then discusses certain other options. The next section considers some of the questions that would need to be resolved in moving from general to specific policies.
The committee agreed that several specific elements should have a role in a strategy for health care reform. These elements focus, for the most part, on the link between resource use and health outcomes, and they include:
Movement toward provider payment methods that encourage efficiency and economy in the provision of health services as well as quality of care and good outcomes. Examples include payments partially based on an episode of illness (rather than individual services) and capitated payments to health plans (Newhouse, 1989), but the committee encourages experimentation (rather than commitment to a single approach). To avoid incentives for underservice to more seriously ill individuals, per case, capitated, and similar kinds of payments to providers or health plans should be adjusted to reflect the risk presented by a patient or health plan member. As noted earlier in this report, no fully satisfactory set of risk adjusters now exists.
Some cost-sharing by most patients (e.g., deductibles, coinsurance, premiums). Patients should be motivated to weigh the costs of specific services and health plan options against their perceived benefits. Again, as noted previously, cost-sharing requirements for those with low incomes or high expenses should not constitute a barrier to needed care.
Better information on the price, quality, and expected outcomes of medical services (see the section on infrastructure for further discussion). Such information should be presented in forms that are useful and timely (e.g., provided prior to decisions about care) for
ordinary patients and consumers as well as for practitioners, managers, and policymakers. Strategies to increase consumer price sensitivity need to be balanced by actions to help people make informed decisions about complex issues even under often stressful circumstances (IOM, 1992b). To the extent, however, that better information leads sicker individuals to health plans or practitioners that can care more effectively for them, it will encourage adverse selection and put more pressure on the risk-adjusted payment approaches discussed elsewhere in this report.
Methods for quality and utilization review that help practitioners, patients, and others learn how actual care conforms to criteria for appropriate care and why care varies in effectiveness and efficiency. These methods should be as unobtrusive as is feasible and consistent with the entire set of structures and incentives included in a reform proposal (IOM, 1989b, 1992b).
A tax cap that limits the amount of employer-paid health benefits excluded from individual taxable income. The intent is to increase individual and employer sensitivity to the actual cost of different health plans, improve tax equity, and provide some revenues to help finance reforms (IOM, 1993b).
Further movement to standardize many administrative practices and eliminate many costs associated with the immense diversity of current billing, payment, audit, reporting, and other practices.
Implementation of the above elements—particularly the first and second—should reduce some of the strain placed on the health care system by reducing the demand for care. It should simultaneously encourage informed evaluation of health care options so that reductions in demand are focused on care of marginal value. On balance, however, empirical evidence suggests that although the above strategies may produce some savings, they are not likely to be strong instruments for reducing the underlying rate of increase in health care costs (IOM, 1989b, 1993b).
Elements Not Mentioned
Some omissions from the above list of elements warrant comment. These omissions include managed competition, global budgets, health planning, and clinical practice guidelines.
The committee notes a strategy that is frequently promoted as a centerpiece of health care reform: "managed competition." Most versions of this strategy call for a regulated system of competition among health plans that limit coverage to care provided by restricted networks of practitioners and providers, for example, staff-model health maintenance organizations that rely on salaried physicians. Members of the committee disagree about whether a patient's choice of provider must be relinquished to contain costs, and they doubt that managed competition is feasible outside major metropolitan areas. Furthermore, empirical evidence about the cost containment effects of managed competition is essentially nonexistent because the full strategy has not been implemented. These reservations notwithstanding, the committee members agree that well-managed, "seamless" systems that integrate an array of medical services can improve the quality and efficiency of care.
The committee's list of cost containment measures also omits any reference to two other oft-mentioned strategies: a global limit or cap on health care spending and price and wage controls. The committee recognizes that budgets and caps may be essential or useful for some segments of the health care system, most obviously for government spending (including revenues lost through tax breaks for private spending). The committee, however, as a matter of philosophy, believes that individuals should be free to spend their after-tax dollars on health care or health insurance. Such freedom is inconsistent with a collectively determined and imposed limit on total spending. When, however, private choices affect the cost of the basic program, policymakers may consider limiting certain choices. An example would be discouraging the purchase of supplemental insurance designed to cover basic program deductibles and copayments; such coverage might erode the effectiveness of patient cost sharing as an incentive for lower health care utilization.
With respect to price and wage controls specifically, the committee recognizes the attraction of short-term price and wage controls (1) to achieve short-term savings that can be used to extend coverage to the
uninsured, (2) to limit cost increases until longer term reforms begin to achieve results, or (3) to supplement long-term strategies, such as managed competition, that may not by themselves be sufficient to limit cost increases. Some committee members, however, doubt the practicality and desirability of even short-term price controls. They cite: the possibility that increased service volume will offset a freeze on unit prices; the complexity of determining what services, settings, and providers will be subject to the cap; the unfairness of subjecting, for example, janitors and food service workers in the health sector to wage controls when their peers in other sectors are unaffected; and the potential for black markets to arise for some services. Another worry is that short-term controls might undermine longer term reforms by discouraging organizational flexibility and innovation and failing to reward efficient practitioners and providers.
At a more specific level, the committee's list of elements of a cost containment strategy mentions neither health planning nor clinical practice guidelines. The former has risen and fallen in favor among policymakers (IOM, 1989b, 1993b) whereas the latter has only recently been promoted as a tool for cost containment. Health planning is cited in this report's discussion of infrastructure as one option for influencing the supply, distribution, and kinds of physical capital to meet a variety of policy objectives. A related discussion focuses on health workforce planning, in particular, the need to increase the proportion of primary care physicians who provide the preventive, primary care, and gatekeeping services that many reform proposals emphasize as means of increasing efficiency in health care delivery.
Clinical practice guidelines are also discussed later in this report as an element of strategies to improve quality of care. With respect to cost containment, policymakers should have reasonable expectations for practice guidelines (IOM, 1992b). Guidelines based on systematically evaluated scientific evidence and professional judgment currently apply to a relatively small proportion of all health services and specific patient problems, and such guidelines are expensive and time consuming to develop. Moreover, the net effect that guidelines may have on costs is hard to predict. Some guidelines, for example, those on proper management of pain, are likely to increase the use of some currently underused services; other guidelines are likely to discourage the use of
currently overused services; in sense cases, increased or decreased utilization will be offset by intended or unintended changes in care that has been or can be used in lieu of the target services; not uncommonly, guidelines are ignored.
MOVING FROM GENERAL TO SPECIFIC POLICIES
The design of specific policies to constrain costs and improve the value received for health care spending will require decisions about a large number of complex questions, many of which do not have obvious answers. Two particularly contentious questions involve the degree to which cost containment can work through market forces rather than regulation and the extent to which regional needs or wants can be accommodated. In general, the committee believes that reforms will require a pragmatic mix of regulatory and market strategies and that some degree of local flexibility and discretion is desirable. Further, a reform strategy oriented toward experimentation and learning in both the public and the private sectors holds more promise of long-term success than an approach that attempts to answer all key questions a priori. Proposals should specify whether (and how) they would phase in certain components on a test-and-revise or compare-and-select basis.
A few other prominent questions that proposals for health care reform should address are discussed below. The committee has a number of observations and a few specific recommendations regarding administrative costs, tax caps, cost sharing, benefit design, and biased risk selection.
Administrative Costs—and Benefits
Efforts to reduce administrative costs may, in some cases, conflict with efforts to collect more data for monitoring, educational, and other purposes. For example, as noted earlier, the committee believes that the continued growth of cost sharing, capitation, gatekeeper, utilization management, and health plan competition as cost containment strategies
requires improvements in the tools used by quality review programs for detecting underprovision of appropriate health care services and poor outcomes of care. Otherwise, efforts to delegate responsibility, reduce micromanagement of practitioner and provider decisions, and encourage restraint in the use of resources may be accompanied by unwanted increases in immediate pain and suffering and long-term dysfunction. Oversight activities, however, tend to multiply rules, information requirements, and administrative costs.
The key criterion for judging the appropriateness of administrative tasks and costs is whether the costs they impose are justified by the degree to which they serve desired objectives related to access, quality, equity, efficiency, and information. The bases for evaluating claims in these areas are neither obvious nor settled among experts. Still, the committee believes that reform proposals ought to include clear and defensible statements about expectations concerning administrative costs and savings and the prospects for increased or decreased administrative burdens on patients and health care providers.
Tax Caps and Geographic Variations in Health Care Costs
A uniform national cap on the tax exclusion for employer-paid health benefits would likely be too generous for low-cost areas and a serious burden for many in high-cost areas. The committee believes policymakers should provide for geographic indexing of any tax cap, although it recognizes that this approach challenges the conventions of tax and benefit policy and could present significant technical and practical difficulties. If, however, the tax cap excluded some percentage of the cost of locally available health plans or if the tax cap were linked to the cost of the cheapest locally available health plan, then a form of geographic indexing would occur automatically.2 Even if the details of
an indexing strategy are left to administrative agencies, a reform proposal should provide basic guidance for their work. (In its discussion of access issues, the committee recommended that the supplemental coverage not be tax-favored unless the basic coverage is very narrowly defined.)
Patient Cost Sharing and Special Problems
As recommended earlier, requirements that individuals share in the cost of health coverage and health services should exempt some services (most notably, certain immunizations) and provide for some adjustment for low-income individuals within the bounds of administrative feasibility. Moreover, in addition to limits on annual out-of-pocket cost sharing, limits on multiyear out-of-pocket cost sharing should be defined to protect those with persistent health problems and high expenses year after year. Again, the details may be left to administrative agencies or perhaps a special commission, but a reform proposal should set forth basic criteria and procedures to be followed.
Core and Supplemental Benefits
In addition to provisions for patient cost-sharing, other decisions about the scope and depth of benefits will be strongly influenced by cost containment objectives. The committee has already recommended in its discussion of access that reform proposals should define a package of core benefits that health plans should cover and that proposals should allow consumers to purchase supplemental benefits under certain conditions. It has also noted the many technical and political questions and difficulties associated with turning such provisions into implementable policies and programs. To recapitulate, these questions and difficulties involve:
the limits of available evidence about the health outcomes and costs of alternative forms of care for specific clinical problems and the consequent need for an incremental process of defining and updating coverage;
the tensions between the desirability of a reasonably comprehensive benefit package and one that limits the cost of extending such coverage to those now uninsured;
the problems of defining whether and what supplemental benefits should be permitted and the circumstances under which they might be, in part, tax-favored; and
the multiple types of health plans (e.g., network and fee-for-service) that need to be considered in designing, implementing, and monitoring a benefits package and in establishing policies for supplemental coverage.
Risk Selection versus Cost Containment
As noted in the section on access, a health care reform proposal that relies on competition among health plans should discourage health plans from competing on the basis of risk selection rather than effective management of care and costs. To promote the latter and limit the former, health care reform proposals should include provisions for standard benefit packages, risk-adjusted payments to health plans (but not risk-adjusted individual premiums), special provisions for very-high-risk individuals (e.g., reinsurance or separate risk pools), monitoring of marketing and other health plan practices, and similar measures (IOM, 1993b).
CONTAINING HEALTH CARE COSTS Key Statements
To serve health and access as well as cost containment objectives, policies to limit the rate of cost escalation need to be grounded in the concepts of:
The committee did not formulate a comprehensive strategy to contain escalating health care costs in part, because evidence about the impact of such strategies is so limited.
The committee, however, agreed that several specific elements should have a role in a strategy for health care reform:
The committee believes that reforms will require a pragmatic mix of regulatory and market strategies and that some degree of local flexibility and discretion is desirable.
Efforts to reduce administrative costs may, in some cases, conflict with efforts to collect more data for access and quality monitoring, educational, and other purposes. The key criterion for judging the appropriateness of administrative tasks and costs is whether the costs they impose are justified by the degree to which they serve desired objectives related to access, quality, equity, efficiency, and information.
To discourage health plans from competing on the basis of risk selection rather than effective management of care and costs, reform proposals should include provisions for standard benefit packages, risk-adjusted payments to health plans (but not risk-adjusted individual premiums), special provisions for very-high-risk individuals (e.g., reinsurance or separate risk pools), monitoring of marketing and other health plan practices, and similar measures.