Can Public Policy Fix What Ails Managed Care?
Stephen M. Davidson
Whether problems associated with the rush to managed care are subject to a public policy remedy depends on the answers to the following questions: First, what are the problems to be solved and are they both systemic and serious enough to require such a solution? If so, could a public policy remedy be crafted that, if adopted and implemented, would in fact solve them? Then, if the answer to the last question is yes, can a sufficiently strong version emerge from the political process and actually be adopted and implemented?
Robert Blendon et al. (1998) have argued (1) that there is no evidence that medical care has deteriorated under managed care and (2) that the backlash is largely the fear people have about their own future ability to obtain care. They already feel constrained in their use of the system and they know the horror stories that receive so much media attention. As a result, many are afraid they will not be able to get the services they need when they are sick. Blendon and his colleagues may, indeed, be correct. Certainly, there is little persuasive evidence to demonstrate that, in the aggregate, care has deteriorated because of managed care. Yet, real suffering has occurred, and some of it is attributable to bad decisions made by officials of managed care organizations (MCOs). But errors and insensitivity to patients were part of the indemnity landscape as well, and in spite of the publicity, there is no evidence that either has increased substantially.
As analysts who like to be guided by data, we might therefore be tempted to say that proposals for "reform" are premature because large-scale [End Page 1051] managed care-specific problems have not yet surfaced. Yet, even without evidence of widespread problems, the public's anxiety is not misplaced given the defining characteristic of managed care: organizations responsible for care are paid in advance, and since their net income (whether surplus or profit) is what remains after paying their costs, doing less goes right to the bottom line. Focusing on how managed care works instead of on what it produces, the backlash can be seen as evidence of real tension in a dynamic system. Even if systemic "problems" are not yet manifest as a result of those stresses, knowing the forces at work may enable us to anticipate the future and use public policy to avoid predictable harm.
From this perspective, the issues are not that access and quality have been compromised, but that the promise of managed care has not yet been fulfilled and that the stresses revealed to date are cause for concern. In this context, public policy may have two roles to play: first, to reduce predictable future risks to individuals and, second, to increase the chance that the conditions needed to ensure managed care success (and to avoid the risks) are present.
What the actual role of public policy should be depends on understanding the conditions required for managed care to produce benefit for the society and using that knowledge to assess the emerging situation. In the pages that follow, I will first describe briefly the promise of managed care and argue that the extent to which that promise is realized depends on the characteristics of MCOs. Then, I will provide a framework for examining MCO characteristics that includes a way of understanding why they sometimes take forms that undermine their ability to achieve their goals. Finally, I will conclude by returning to the questions asked at the outset.
The Promise of Managed Care and the Reality to Date
Managed care means that a single organization furnishes insurance coverage for subscribers and is responsible for providing or arranging for their care. The simple idea is that linking the two concepts changes the incentives for the better. As insurer, the MCO's gross income is determined before services are delivered, thereby providing a powerful incentive to limit its expenditures in meeting subscriber needs. As caregiver, it also has the means to improve the efficiency and effectiveness of care so that even though the MCO's income is predetermined, it has the capacity to serve patients well. [End Page 1052]
Proponents of managed care believe the combination of those incentives and a competitive market will lead MCOs, first, to lower premiums in order to attract subscribers and, then, to find less expensive, more effective ways to serve their clienteles. It is this last assumption that experience leads us to question. Competing MCOs have grown, not by offering better ways of delivering care, but by keeping their premiums low and persuading individuals that the new arrangements are not much different than the old ones, so choosing managed care is not something they need to fear. Most MCOs have kept down their costs not by finding improved methods for delivering care, but by using a variety of strategies to pay less for services. MCOs have compensated for the inherent weakness of this approach with administrative measures (like requiring prior authorization of certain services), thus, creating additional strains in the relationships between physicians, subscribers, and MCOs.
The care-related goal of MCOs is to serve patients in ways that respond to
their felt needs by providing services that are appropriate, reliable,
effective, integrated, and efficient. We assume that MCOs need to
have certain specific capacities in order to be able to accomplish
that goal. Since MCOs differ from one another on important dimensions
and since it is likely that some combinations of factors have greater
chances of success than others, it should be possible to compare the
performance of MCOs with different characteristics and, thereby, to learn
which combinations of factors contribute most to achieving the desired
results. Then, public policy may be able to play a role in increasing
the probability that MCOs do, indeed, have those characteristics. The
framework in Figure 1 can serve
as a guide to the needed empirical work.
Components of Managed Care Organizations
Although managed care is a common term, the catalogue of arrangements included under its rubric is quite varied. Since each MCO function (if not all of them) can be performed using several different methods, particular MCOs can be defined and differentiated from one another by the choices they make regarding these functions.
MCO characteristics can be grouped under four main headings: (1) provider characteristics, (2) financial conditions, (3) utilization controls, and (4) service-enhancing tools and methods. The underlying assumption is that the array of arrangements for a particular MCO will determine results relating to costs, utilization, service integration, quality, and efficiency. [End Page 1053] [Begin Page 1055] Since individual MCOs often compete in their markets by offering multiple products (e.g., group model HMO, independent practice association [IPA], and point of service [POS] plans), the characteristics for each such product need to be identified separately. With that caveat, the arrangements include the following:
Provider Characteristics
An MCO needs a variety of providers to be able to satisfy its caregiving obligation. Since all MCOs have a stake in the delivery of services and may therefore attempt to influence the care, their choice of influencing methods will vary and be affected by their provider characteristics. In general, we would expect those with large numbers of primary care physicians in small community-based offices to rely more on methods that limit the numbers or cost of services provided (e.g., prior authorization requirements) than on measures that attempt to improve efficiency by increasing the integration of services among providers. Similarly, MCOs will have more leverage to bargain over rates with hospitals in communities with an excess supply in multiple facilities than in other areas.
Financial Conditions
The nature of a plan's financial arrangements will also influence its level of spending and the tools it can use to influence clinical decisions. Principal among these arrangements are the methods and rates by which its various provider types are paid, the nature and extent of cost-sharing by subscribers (if any), the use of risk-adjusted rates to pay the plan and/or providers, and its ownership (see Figure 1).
Ultimately, it is individual physicians and other clinicians who serve individual patients and those professionals must be compensated. Individual physicians may be paid using several methods (principally, salary or a variation of fee-for-service), each with its own incentives. MCOs that pay physicians on a fee-for-service basis, even at lower rates and with withholds, will be using a payment method that encourages physicians to provide services. They will, therefore, need to find other methods for inducing restraint in service delivery. One strategy may be to shift some of the risk to others by contracting with physician groups, thus, passing to the groups the challenge of compensating individual physicians. [End Page 1055]
Utilization Controls
Managed care organizations must find ways to change utilization patterns because most of the available financial tactics, even if they help to restrain utilization, can work only for a limited time, and the factors that have been responsible for runaway health care spending will reassert themselves. Strategies to limit utilization include using a primary care physician to act as gatekeeper to other services, requiring that patients seek services only from those providers on a list; requiring that physicians obtain permission before providing certain services, and limiting the use of prescription drugs to those in a formulary. The primary goal of these strategies is to reduce the utilization of certain services or drugs in order to reduce spending. They do not by themselves improve the delivery of services although some have that potential when used in combination with other strategies.
Service-Enhancing Tools and Methods
The fact that MCOs depend heavily on restrictions to accomplish the primary goal of limiting spending may account for much of the suspicion and dissatisfaction reported in the press. To achieve the full potential of managed care--and to avoid much of the risk--requires enhancements of the service-delivery process. Some tools and processes for managing care toward this end are (1) automated clinical information systems that provide real-time patient data to clinicians, (2) evidence-based clinical practice guidelines that assist clinicians in providing appropriate treatment, and (3) disease management plans for certain chronic conditions.
These four lists are intended only to illustrate some of the characteristics implied by the term managed care organization. The important points are that managed care plans will vary considerably in their capabilities and that variation in plan components results in variations in both the nature and sustainability of the results produced--related to expenditures, utilization rates, patient and provider satisfaction, service integration, service quality, and service efficiency.
The Importance of Environmental Factors
Recognizing that MCOs differ along important dimensions naturally raises the question of why they differ. From open systems theory we learn that the answer lies mainly in the organization's environment (Davidson, McCollom, and Heineke 1996). [End Page 1056] Strategic decisions regarding provider characteristics, financial arrangements, utilization management methods, and efficiency enhancing measures are heavily influenced by the organization's location, the availability and cost of technology, and characteristics of the market (see the left hand column in Figure 1.
The Legal Environment
The nature and performance of managed care organizations are influenced by certain legal considerations. Most directly, the contract between the purchaser (usually an employer) and the MCO specifies what services are covered, terms of coverage, costs, and other things. Some large employers also require that MCOs report to them periodically on their employees' experience with the plan so that they can identify troublesome trends and plan actions to alter them (Maxwell et al. 1998). Many other employers are interested primarily in the premium cost and have neither the administrative capacity nor the inclination to monitor a plan's performance closely.
MCOs are subject to regulation, as well, and since most regulation occurs at the state level, interstate variations can be found regarding insurance reserves, consumer protections, and other matters. This effect is mitigated by the Employee Retirement Income Security Act (ERISA) preemption, which excludes self-insured plans from state regulation, reserving the oversight function to the federal government. One result is that MCO actions will be affected by the extent to which they bear the risk of coverage themselves or are hired by large employers to manage the care-delivery and bill-paying functions.
The Technical Environment
Although computerized information systems have much theoretical potential to aid clinicians and managers in improving the delivery of services, even plans that are willing to invest in clinical information systems are limited to those that are available on the market or systems they develop themselves. Ideally, a system should permit a physician treating a particular patient to have instant access on a computer screen to the patient's medical history, especially recent events that may bear on his or her current condition. To do so, however, requires first that the MCO have access to data from multiple providers and have also the physical capacity to get the data to clinicians in real time and in a form that can be [End Page 1057] used to avoid duplication of services and to improve the integration of care during an illness episode.
Another technical consideration that affects MCO performance is the development of risk adjustment methods for paying plans and providers (Kuttner 1998; Kronick and Dreyfus 1998). Payments based on patients' health status could theoretically benefit plans, providers, and patients, but for these benefits to accrue, reliable methods of predicting future use of service are needed. The best guide to the future may be the immediate past, but it has not yet become a very precise guide. Therefore, until more progress is made, many plans will compensate by limiting the number of older subscribers or those with prior medical conditions.
The Market Environment
In a market, a firm's choices are affected by economic conditions and the actions of competitors (see the lower left hand box on the figure). Key factors include the degree of MCO penetration in a market, the number and characteristics of competing MCOs, measures of health care system capacity, the availability of capital to competing plans, and the extent to which large employers dominate the health insurance market. These conditions affect the pricing, cost-containment, and utilization-reducing strategies available to MCOs.
To illustrate: In a market with relatively low MCO penetration, a plan will be competing largely with higher-priced indemnity insurers. If that market also has an abundance of providers, MCOs will be able to obtain favorable prices from the providers they need to meet their obligation for care. By underpricing indemnity insurers and paying low rates to providers afraid of losing patients, MCOs can do very well financially without even trying to reduce historically high utilization rates. But as managed care penetration grows, MCOs will be competing more with other MCOs than with indemnity carriers; and they will make money less by growth and more by changing the care delivered in order to reduce their costs.
The Central Role of Managers
Finally, although we talk about managed care organizations, we must not lose sight of the fact that organizations are defined by the decisions and actions of the people who work under their auspices. Paying capitation [End Page 1058] rates to MCOs does not automatically translate into less duplication of services, lower utilization rates, improved quality, or even lower expenditures. Physicians do not suddenly change clinical decision patterns developed over the course of their medical careers simply because they have managed care patients, especially since most are still paid fee-for-service. Usually, it is managers who cause those changes to occur.
Managers are responsible for the organization's viability, which depends largely on services provided to a plan's subscribers. The primary responsibility for actually delivering the clinical services, however, remains with the physicians and others trained for those roles. A key task of the managers, therefore, is to influence the work of the clinicians so that it achieves the organization's goals of providing good service to patients at reasonable cost. Although a variety of tools are available to help managers encourage clinicians in that regard, the ability of particular managers to use specific tools is affected by environmental factors, organizational capacities, and their own experience and skill.
Left on his own, each physician may provide his patients with the best care he knows how to deliver, but he will not be able to integrate the care he provides with that of other clinicians in other segments of an episode. Nor will efficiency be a particular goal, especially if his compensation is still fee-for-service. Integration of services and attention to resource use do not occur by themselves; they require the acts of managers to encourage them, facilitate them, or if necessary, demand them. Moreover, as noted, managers' capacity to accomplish those goals will vary with local conditions. For example, in a market characterized by a large number of IPAs, a system of prior authorization may be the only viable choice. But that is a strategy with limited potential partly because, although it may be tolerated for a time, it will irritate physicians and patients alike. To create conditions that permit their strategies to have a longer life, one possibility is to contract less with individual physicians and more with larger multispecialty group practices and to do so on a capitated basis, transferring much of the financial risk for the delivery of services to the group. Then, the managers of the multispecialty group and not the MCO will have the primary responsibility for encouraging integration and efficiency. In an IPA-dominated market, they will have access to a wider range of tools than the MCO's managers because the group practice's single, unified interest transcends the MCOs': provide responsive care that is both efficient and effective, regardless of who the payer is. [End Page 1059]
Conclusion
Many of the ideas discussed in this essay will be familiar to observers of the managed care scene. The contribution of the framework, therefore, is not to identify new influences on MCO behavior but to organize them into a systematic whole. Returning to the questions asked at the start of the paper, the framework can help analysts and policy makers to (1) describe each MCO, (2) identify ways in which different combinations of factors influence MCO performance, (3) anticipate what aggregate results are likely to occur in a given market, and (4) identify policies which can increase the chances that a particular plan or all of the plans in an area will achieve the desired goals.
It would be naive to treat managed care as if it were a single construct. And while readers of this journal are unlikely to make that mistake, those for whom health care is just one of many topics vying for their attention might inadvertently fall into that trap. Legislation adopted to accomplish particular goals will need to take account of the differences found in a particular state or throughout the country as a whole. The framework provided here is a useful guide to collecting data and conducting analyses that show what results to expect from particular combinations of factors.
Boston University School of Management
Stephen M. Davidson is a professor of management policy and health care management at Boston University's School of Management and Director of Research at John Snow, Inc. in Boston. He is an active researcher and consultant, concentrating on the organization and delivery of health care services, as well as related public policy issues. He is coeditor, with Stephen A. Somers, of Remaking Medicaid: Managed Care for the Public Good (1998) and coauthor, with Janelle Heineke and Marion McCollom, of The Physician-Manager Alliance: Building the Healthy Health Care Organization (1996), as well as the author or coauthor of several other books and many journal articles and monographs.
References
Blendon, R. J., M. Brodie, J. M. Benson, D. E. Altman, L. Levitt, T. Hoff, and L. Hugick. 1998. Understanding the Managed Care Backlash. Health Affairs 17(4):80-94.
Davidson, S. M., M. McCollom, and J. Heineke. 1996. The Physician-Manager Alliance: Building the Healthy Health Care Organization. San Francisco: Jossey-Bass.
Kronick, R., and T. Dreyfus. 1998. Health-Based Payment and Other Challenges of Medicaid Managed Care. In Remaking Medicaid: Managed Care for the Public Good, ed. S. M. Davidson and S. A. Somers. San Francisco: Jossey-Bass.
Kuttner, R. 1998. The Risk-Adjustment Debate. New England Journal of Medicine 339(26):1952-1956.
Maxwell, James, Forrest Briscoe, Stephen M. Davidson, Lisa Eisen, Mark Robbins, Peter Temin, and Cheryl Young. 1998. Managed Competition in Practice: "Value Purchasing" at Fourteen Large U.S. Employers. Health Affairs 17(3):216-226.
