Controlling Costs and Changing Patient Care?: The Role of Utilization Management (1989)

Page 143

6
Conclusions and Recommendations

Utilization management is part of a complex balancing act created by society's struggles with two important questions. First, how do we ensure that people get needed medical care without spending so much that other social objectives are compromised? Second, how do we discourage unnecessary and inappropriate medical services without jeopardizing necessary, high-quality care?

Experience indicates that these questions have no fixed answers. Rather, as suggested by the first two chapters of this report, we find a series of working hypotheses and partial solutions that are continually revised, discarded, and even reinvented as medical technology, social values, economic conditions, and other circumstances change. In this preliminary report, the committee has examined one current working hypothesis—that externally applied assessments (those occurring outside the relationship between the patient and physician) of the appropriateness of proposed medical services can improve how care is provided and, as a consequence, help constrain health benefit costs. The validity of this hypothesis is of interest to everyone involved with health care—patients and potential patients, providers, employers, unions, insurers, and public policymakers.

Although attempts to manage the use of health services on a prospective case-by-case basis are not inventions of the 1980s, the application of utilization management techniques in private health benefit plans has become widespread only during the last half-dozen years. Its growth reflects purchasers' dismay over continuing rapid rises in health care costs and their perception that much care is unnecessary.

Page 144

The Institute of Medicine is studying utilization management not only because the field is expanding but also because it has important implications for the delivery of medical care that have not been systematically investigated. On the one hand, utilization management has features that make it an attractive strategy for managing health benefits. When successfully implemented, utilization management takes the clinical circumstances of individual patients into account and gives patients and providers its assessments about whether services are appropriate before rather than after care is provided. It does not involve across-the-board limitations on health plan coverage. On the other hand, utilization management has certain unattractive aspects. It acids to the administrative demands on both patients and providers and employs criteria for decision making that are sometimes vague or even secret.

In this first phase of its work, the committee has examined two major methods of utilization management: (1) prior review of the appropriateness of proposed medical services and (2) high-cost case management. The first category includes several techniques—preadmission review, admission review, continued-stay or concurrent review, discharge planning, and second surgical opinion.

One of the committee's basic findings is that there is limited empirical evidence on which to make authoritative conclusions about the impact of utilization management at this time. Most private sector programs are relatively new, and the field has been changing rapidly. Therefore, the committee has tried initially to frame the important questions, offer reasoned judgments about the state of utilization management, and suggest directions for the future. As stated in the Preface, the committee has based its judgments and projections on reviews of published and unpublished reports on the workings and effects of utilization management, site visits by the committee to several utilization management organizations, and hearings and discussions involving consumers, providers, purchasers, and others. Judgments also reflect the committee members' extensive and diverse experience in health care delivery and financing.

This chapter summarizes the committee's findings about the current status of utilization management and its evolution. It identifies some issues raised by that assessment and recommends what prudent decision makers can do to promote the positive potential of utilization management

Current Status of Utilization Management

Several features of utilization management are important to keep in mind: (1) its great diversity, (2) its limited scope, and (3) its relative lack of regulatory oversight. First, utilization management, as it currently operates in the private sector, is highly variable, which makes generalization about it

Page 145

difficult. Some variations may reflect the industry's youth, for example, the growing but uneven extent of computerization. Other variations, such as those involving links with providers and claims payment arrangements, may depend on whether the utilization management function is freestanding or integrated into an insurance or prepaid health plan.

Second, utilization management has, until recently, emphasized the site, duration, and timing of medical care. The unnecessary and inappropriate use of the hospital, and not the actual need for a particular procedure, has been the main focus. Review organizations vary in their willingness to tackle new issues of appropriate care, but they generally appear to follow the lead of researchers and medical groups in adopting criteria or selecting targets for review. Also, as a general rule, prior review programs have not made case-by-case assessments of the comparative costs of alternative treatments or sites of care. In contrast, high-cost case management programs have built such assessments into their evaluations of proposed courses of treatment. The main strategy has been to discuss and negotiate appropriate care rather than to refuse prior authorization of benefits explicitly. Denials rates appear to run about 1 to 2 percent of cases with some later changed upon appeal.

Third, utilization management in the private sector operates under few explicit regulatory restrictions. There is, however, considerable awareness among review organizations and major purchasers of the legal risks inherent in efforts to influence patient care decisions and operationalize the terms of health benefit plans. And there appears to be growing recognition of the conventional—but not infallible—protections offered against liability by good management, good judgment, good faith, and good documentation.

As discussed in Chapters 4 and 5, empirical evidence on the effects of utilization management is limited and suffers from a number of methodological weaknesses. Although the number of comparative multivariate analyses is increasing, there still is little or no evidence about the consequences of specific differences in the way utilization management is conducted by different organizations.

Despite these limitations, the committee believes that the available reports and studies, taken together, suggest that purchasers who have adopted utilization management have experienced some changes in health care use and costs. Specifically:

• Utilization management has helped to reduce inpatient hospital use and limit inpatient costs for some purchasers beyond what could be expected from other factors such as growth in outpatient resources, changes in benefit plan design, and shifts in methods for paying hospitals. Groups with higher initial levels of hospital use tend to show more change than groups with lower initial hospital utilization.

Page 146

• The impact of utilization management on net benefit costs is less clear. Savings on inpatient care have been partially offset by increased spending for outpatient care and program administration. Some of this offset is an expected and acceptable result of utilization management (and other factors), and some is an unwanted consequence of moving care to outpatient settings where there are fewer controls on use and price.

• Although utilization management probably has reduced the level of expenditures for some purchasers, it—like other cost-control strategies— does not appear to have altered the long-term rate of increase in health care costs. Employers who saw some short-term moderation in benefit expenditures are seeing a return to previous trends.

Systematic evidence about the impact of utilization management methods on the quality of care and on patient and provider costs (both economic and noneconomic) is virtually nonexistent. Purchasers have not demanded such evidence, and researchers have found the measurement of these effects more costly, time-consuming, and uncertain than measuring effects on purchaser's costs. During the course of this study, the committee did not locate documented cases to suggest that prior review programs in the private sector are jeopardizing patient safety, although some physicians believe they are. They may also cause anxiety and inconvenience to some patients. Utilization management does add to the administrative burdens on practitioners and institutional providers and contributes to resentments about reduced professional autonomy and satisfaction.1 More positively, the committee has some confidence that high-cost case management, as currently practiced on a voluntary basis, is easing some financial and emotional burdens on catastrophically ill patients and those who care for them.

The lack of good research on the effectiveness and impact of utilization management is a frequent theme in this report. The gaps and deficiencies in clinical research on the effectiveness of many medical procedures also must be emphasized. As utilization management moves toward more review of the actual need for specific procedures, the clinical bases for these kinds of assessments become more important. (This is not, however, to assert that utilization management programs are making sufficient use of clinical information that is already available.) Later in this chapter, the committee recommends greater public and private involvement in both program evaluation and clinical outcomes research.

1. However, the committee found it difficult to be certain (1) when complaints about utilization management programs stemmed specifically from private programs rather than Medicare, Medicaid, or internal hospital programs or (2) when they really involved prior review or case management rather than basic exclusions in benefit plans, retrospective claims denials, or provider payment methodologies.

Page 147

How Utilization Management Is Evolving

The continuing evolution of utilization management is most evident in two areas: its scope and its operational efficiency. There are several reasons for this evolution. First, the initial savings from shifting the site and timing of care have largely been realized in may places, and the survival of utilization management programs may depend on their ability to continue to have an impact on use and costs. Second, review organizations are being influenced by researchers' belief that a significant amount of care is still inappropriate and unnecessary. Third, the administrative costs of review programs and the provider and patient relations problems they create make simplification and efficiency important objectives.

Scope of Review

The emphasis of utilization management is beginning to expand from the site and duration of care to include more evaluation of the actual need for specific types of inpatient and outpatient services. The question is not only whether the hospital is the appropriate site for a proposed procedure but whether the procedure is clinically indicated. The procedures targeted for this kind of assessment are a mix of very expensive and relatively high-risk services, such as coronary artery bypass surgery, and less expensive and less risky procedures, such as bunionectomy. The assessments themselves tend to involve more questions about patient history and previously tried treatments than do site-of-care reviews.

As the focus of utilization management expands, the development of sound clinical criteria for assessments of medical necessity is becoming increasingly important. When the question is not where someone will get care but whether or not a particular treatment is necessary at all, the stakes for the patient, the practitioner, and the purchaser tend to be higher and the soundness of the criteria and their application more critical. The generation of useful criteria requires at least three steps: (1) the development of clinical and other evidence of treatment effects for particular procedures and services; (2) the formulation of guidelines for treating specific conditions or using specific procedures; and (3) the delineation of review criteria based on clinical evidence and practice guidelines. The research on feedback and education strategies to influence physician decisions on patient care suggests that the criteria used to make medical necessity assessments will be more likely to win acceptance and change behavior if they are based on good clinical evidence from respected academic and professional sources and if the medical community is involved in the process of criteria development.

Page 148

Operational Efficiency

With respect to administrative costs, utilization management organizations have a strong self-interest in measures that can reduce their own costs—particularly since the short-term impact of their programs seems to be diminishing and the ratio of savings to administrative costs appears to be dropping. Frequently mentioned priorities include greater computerization, expanded use of treatment protocols in high-cost case management, and greater targeting of reviews to high-payoff categories of problems and services. TO some extent, gains in operational efficiency for review organizations could also reduce administrative burdens on patients and providers.

In addition, some moves are being made that may standardize certain aspects of utilization management and reduce the variability and ambiguity now complained of by both purchasers and providers of care. One example, described in Chapter 4, is the effort initiated by the American Medical Association, the Health Insurance Association of America, and the Blue Cross and Blue Shield Association to develop guidelines for the conduct of prior review programs. The American Hospital Association also has been working with insurers to establish guidelines for on-site review of medical records by insurers, utilization management organizations, and others.

It seems likely that much of the clinical and administrative technology of utilization management will be increasingly integrated into health benefit programs, although the details will continue to vary across HMOs, PPOs, and fee-for-service plans. However, to the extent that clinical standards for medical decision-making are expanded and then internalized by practitioners, third parties at risk for the costs of medical care may be able to focus their utilization management programs more narrowly and reduce the burdens on patients and providers.

In addition, redesign and more careful specification of health benefits may, in certain situations, eliminate the need for some case-by-case prospective review of services and for some forms of case management.

Rationing

At this time, the committee does not see utilization management moving toward intentional rationing of clinically necessary medical services. A decision not to approve payment for a medically unnecessary service is not rationing, per se. Nurse and physician reviewers do not make explicit case-by-case assessments of whether the expected clinical benefits of a hospital admission or other proposed service for a specific patient are, in some way, worth not only the clinical risks but also the economic costs. Reviewers also do not make decisions, like those sometimes involved in emergency care or transplant services, to allocate limited medical resources

Page 149

to those most in need or most likely to benefit. The committee recognizes, however, that there may be some instances when a review nurse or physician may apply implicit cost-effectiveness judgments to discourage proposed care that may have only marginal clinical value in a specific case.

The committee notes that high-cost case management may assess whether making an exception to routine coverage restrictions will be less expensive for a specific patient than enforcing the contract provision. This might be considered a form of rationing. If so, it is a peculiar form in that it adds resources where they normally would be restricted rather than restricting resources that would otherwise be available.

It is useful to draw a distinction between micro and macro-level resource allocation decisions. Micro-level decision-making operates at the individual level, for example, to determine who gets a heart transplant when organs are in short supply. Macro-level decision-making operates at an aggregate level, as when health plans decide to cover only 30 days of inpatient psychiatric care. It is possible that prior review could become a micro-level rationing tool, but the committee found no evidence of such a development at this time. In contrast, employers and others already make a wide range of macro-level allocation judgments in benefit plan design.

Issues for the Future

In Chapters 3, 4, and 5, the committee identified some shortcomings in utilization management or gaps in our knowledge that raise concerns about patient protection, particularly given the course now being charted by purchasers and utilization management organizations. If the positive potential of utilization management is to be encouraged, then the committee believes that several issues need attention.

• Methods and resources for evaluating the impact of utilization management on health care costs, use, and quality are limited and must be strengthened. All those involved in health care delivery and financing need to feel more confident about what works and what does not under what circumstances.

• The criteria used by those engaged in utilization management (including hospitals and HMOs) should be available for outside scrutiny. Physicians, purchasers, and others should know the basis for judgments about the site, timing, and need for care.

• Systematic investigation of the effects of utilization management not only on purchaser costs but also on patient and provider costs and attitudes should be a higher priority. A more complete picture of costs and benefits is needed.

• The opportunities for patients and providers to appeal utilization

Page 150

management judgments should be clearly described and free from unreasonable complexity or other barriers. This is an essential protection for patients.

• The explicit links between utilization management and quality assurance mechanisms need to be clarified and strengthened. Utilization management organizations should have standard operating procedures for responding to the quality problems they uncover.

Recommendations for the Near Term

The committee believes that utilization management has sufficient promise that a number of short- and long-term efforts should be made to promote its positive potential and guard against its shortcomings. A prudent course for the near future is for the parties involved in utilization management—purchasers, review organizations, health care providers, and patients—to accept greater responsibility for the reasonable and fair conduct of utilization management and the appropriate use of medical care.

The concepts of prudence, fairness, and responsibility described below are supported by a framework of general rights, duties, and obligations created by state and federal legislation, regulation, and judicial decisions. The most important examples include tort and contract law. Depending on the model of utilization management used and whether it involves a self-insured employer, an HMO, or other arrangement, different legal rules may apply. Little case law and not much legislation or regulation apply explicitly to utilization management. However, it is reasonable to expect that utilization management will be subject to standards of prudent conduct similar to those affecting other areas of health plan administration.

Responsibilities of Employers and Purchasers

As financers of both utilization management and health services, employers are in the best position to exert influence on the conduct of prior review and high-cost case management programs. They have a direct interest in programs that work fairly and effectively to ensure the value of their investment in employee health benefits. The committee, however, recognizes that there are practical limits to what individual employers can do, particularly small employers with limited administrative resources, expertise, and leverage.

In general, purchasers should investigate the operating procedures and capabilities of the organization or organizations from which they purchase utilization management services. This investigation should include organizations such as HMOs that provide prior review and high-cost case management as part of a broader package of services. Benefits managers

Page 151

can visit review organizations, request detailed information and references from current and former clients, and seek advice from business coalitions, consultants, and similar resources.

Table 6-1 lists the types of information a purchaser may want to request. This kind of information can help purchasers assess in broad terms a review organization's experience, its investment in staff and information technology, and its sensitivity to purchaser, patient, and provider concerns. The exercise can also help purchasers to think through their own preferences for different styles of utilization management. Unfortunately, this preliminary report cannot offer purchasers the equivalent for utilization management of the crashworthiness tests for automobiles or the nutritional quality evaluations for foods that commonly appear in consumer magazines. Again, the absence of research on the impact of variations in program structure and operations is a limiting factor.

As prudent managers of a total program of health benefits, employers should also examine other aspects of their benefit plans for impediments to the appropriate use of medical services or the rational payment for these services. For example, if reviewers are judging hospitalization to be unnecessary for intravenous (IV) therapy, is outpatient IV therapy a covered benefit? Are fees paid for necessary IV therapy in an outpatient setting subject to fewer controls than for those paid for inpatient therapy? What is provided to help employees understand what their benefit plan will pay for or how much they may pay out-of-pocket for selecting one provider or service rather than another? A benefit plan's scope of coverage, levels of cost-sharing, provisions for retrospective claims review and profiling, and methods of paying providers can as affect the impact of utilization management programs on health care costs and quality.

Although employers have the fight and duty to take vigorous actions to manage the costs of employee health benefits, they should respect both the confidentiality of medical information about employees and the primary obligations of physicians to serve their patients. In their own business interest and to protect employees, employers should see that workers are informed of their responsibilities and rights. Key means of reaching employees include orientation for new employees, clearly written and eye-catching brochures explaining responsibilities, and health plan identification cards with clear notice of utilization management requirements. Human resources staff should be trained to respond to employee questions, assist with problems, and handle grievances.

Responsibilities of Utilization Management Organizations

Any supplier of services has responsibilities to purchasers that are intrinsic to the concept of a buyer-seller relationship. It is in the business

Page 152

Page 153

(Table continued from previous page)

interest of utilization management organizations to anticipate and respond to purchaser demands for information about the organization, its services, and its results. Further, it is in the legal interest of these organizations to manage their activities rationally, to act in good faith, and to maintain careful records.

Although good business and legal judgment should dictate prudent behavior, those who provide utilization management services also have a moral obligation not to harm the patients whose medical care they review and influence. Harm includes discouraging appropriate care and mishandling confidential information. When organizations perform prior review and high-cost case management for individually purchased insurance plans (with no employer sponsorship), they have a particular responsibility to provide good educational materials and appeals processes for beneficiaries who have no employer or other sponsor to act as their agent and aid.

With respect to physicians and other providers of care, good business sense should dictate that review organizations encourage provider acceptance and cooperation, although their incentives to do so are less strong than the incentives to respond to client wishes. Chapters 3, 4, and 5

Page 154

suggest that utilization management organizations can increase physician acceptance by

• involving the medical community in the development and evaluation of review criteria;

• making the bases for review decisions open for examination by purchasers, physicians, and others;

• clarifying and improving the process for discussing cases and appealing negative decisions; and

• minimizing the administrative burdens placed on hospitals and physicians.

Given the diversity and competition now evident in utilization management programs, significant progress in these two directions will probably require ongoing cooperation between elements of the utilization management industry and provider groups to identify problem areas and propose operating procedures that avoid unreasonable demands on providers. Currently, there is no natural focal point to coordinate and promote these efforts or monitor progress. However, the cooperative efforts of organizations like the American Medical Association, American Hospital Association, Blue Cross and Blue Shield Association, and Health Insurance Association of America to develop guidelines for review are encouraging.

The committee is aware that further steps, particularly making clinical criteria available, raise difficulties given the competitive environment of benefit plan administration. Organizations that have invested their own resources in developing criteria will be reluctant, on the one hand, to make them available to competitors with less initiative and, on the other hand, to reveal some details to practitioners and institutions for fear of their ''gaming the criteria.'' Although these are reasonable concerns, on balance, they are outweighed by the need to move toward open criteria and standards.

It is inevitable that utilization management organizations will occasionally identify cases of apparent malpractice or other quality-of-care problems. These cases may involve a proposed treatment for a single patient or a pattern of behavior. The experience that the Health Care Financing Administration has had in developing regulations and procedures for PROs that have uncovered quality problems indicates that balancing patient protection against due process for providers is a difficult political and legal issue. Nonetheless, review organizations should prepare guidelines for dealing with clear quality-of-care problems, at a minimum, having their medical director discuss the situation with the provider in question and, perhaps, raise the matter with the local PRO.

Page 155

Responsibilities of Health Care Practitioners and Institutions

The committee found the responsibilities of health care practitioners and institutions most troublesome to analyze and define. In this respect, utilization management mirrors the situation with many other questions of provider responsibility and ethics, whether the questions involve the limits of heroic efforts to maintain life, the provision of care to those who are unable to pay, or the recognition of patient preferences about treatment alternatives.

The narrowest area of controversy involves physician self-interest versus the patient's well-being, where law and ethics have long agreed that the latter should prevail. Broader controversy and lack of agreement surround the questions of whether and how physicians should consider societal interests in making patient care decisions. Intermediate in scope are issues involving the physician's responsibilities when third parties are paying for patient care. For example, is it acceptable to shade the truth in order to obtain prior certification or retrospective payment for services? Is it legitimate to refuse to disclose information needed by review organizations or claims administrators? When should a physician challenge a negative payment decision? Is the physician obligated to provide a service he or she believes is necessary when a review organization, HMO, or hospital disagrees? What role should patient preferences play?

With respect to third-party payment, the committee agreed that health care practitioners and institutions are responsible for

• cooperating with payers' reasonable efforts, including utilization management, to assure that they are paying for appropriate care within the terms of their benefit plans, but

• constructively challenging unreasonable review programs and specific decisions that threaten patient safety or damage patient privacy.

Although some difficult situations with insurers and review organizations may be more conveniently and quickly dealt with in the short term by misrepresenting patient symptoms, diagnoses, or treatments, the committee believes it is in the patient's, physician's, and society's interest over the long term for physicians to deal honestly with reviewers and claims administrators and to challenge questionable criteria, procedures, and decisions directly. Manipulation and evasion can have serious risks. Specifically, incorrect information may enter the patient's medical records or insurance history with later negative consequences for the patient. Moreover, perceptions by purchasers that physicians are gaming the system undermine professional credibility and stimulate the sorts of auditing, second-guessing, and external oversight to which practitioners object.

With respect to patients, the committee sees additional responsibilities for practitioners. These include informing patients about treatment options,

Page 156

risks, and benefits and then considering their preferences. Insofar as is feasible and desired by the patient, the physician should also try to ensure that the patient gets care when a review organization denies precertification for services that the physician believes are medically necessary. This may mean locating an alternative source of care if the patient cannot pay and the physician cannot provide free care.

In the context of this committee's consideration of appropriate medical practice, an obvious responsibility for health care practitioners is to stay current with scientific literature on the necessity and effectiveness of medical services in their area of practice. (Many researchers would add that practicing physicians should, if at all possible, participate in the clinical trials and other research on which this literature rests.) As research results accumulate on the effectiveness and appropriateness of specific diagnostic and treatment methods, the committee sees an enormous implementation challenge to make knowledge more readily usable by busy practitioners. Meeting this challenge will require continued initiatives in such areas as the development and dissemination of practice guidelines, continuing medical education, and office-accessible computer resources for physicians. Utilization management can play a role in supporting these initiatives.

Responsibilities of Patients

In many respects, patients and potential patients are the weakest strand in the web of responsibilities for the prudent use of medical services. One person's view of appropriate patient responsibility can seem like "victim-blaming" to others. Although it is now a cliché among sophisticated consumers that more medical care is not necessarily better medical care, people may not be able to act in an informed and prudent way when they are ill. Furthermore, individuals may find both their benefit plans and their medical care difficult to understand and evaluate even when they are not stressed by illness. Education levels, family support systems, employer communication efforts, and other factors may all affect how much information people can understand and use. Nonetheless, health plan members should try to understand their responsibilities under their plan and their responsibilities for protecting their own health.

The challenge for those involved in health care delivery and financing is to help all kinds of patients make informed decisions about getting or not getting care. Concepts of probability, differential diagnosis, and risk-benefit trade-offs are not part of most people's everyday vocabulary. Fortunately, new tools, such as the video disk recently developed to explain prostatectomy risks, benefits, and alternatives, have promise in making complex issues more understandable for patients. It is not inconceivable that protocols used by review organizations to assess a proposed treatment

Page 157

might one day ask not only whether certain clinical steps (for example, a trial of antibiotic therapy) have been taken but whether and how the patient has been informed of risks, benefits, and alternatives for the proposed treatment.

Recommendations and Questions for the Longer Term

Early in this chapter, the committee observed that utilization management, per se, is essentially a working hypothesis, one of several partial and overlapping strategies for balancing health care expenditures, access, and quality. As knowledge advances, economic conditions change, and social values shift, these partial strategies are revised, integrated, and sometimes discarded. However, even if some of the techniques now employed by utilization management organizations are abandoned or the organizations themselves change, improvements in the criteria for judging appropriate care and for monitoring the provision of care will continue to be relevant. The longer-term recommendations of the committee focus on the foundations for such judgments: knowledge development, knowledge application, and value clarification.

The following discussion examines three issues:

• research on the effectiveness of medical services and the effectiveness of utilization management and related techniques,

• formulation and dissemination of guidelines for medical practice and criteria for utilization review, and

• oversight of utilization management.

Research on Effectiveness

Utilization management can be no better than the clinical evidence and expertise on which it is based. Although review organizations may not be effectively using all available research now, they are still constrained by the large areas of undocumented impact and clinical uncertainty involving many major medical procedures and services.

Policymakers are increasingly recognizing that the free-market system is unlikely to invest sufficiently in effectiveness research and data collection because those making the investment cannot capture all the benefit but must share much of it with those who have not invested. Since the public gains from investments in such research, a major public role in financing and priority setting is appropriate. It should add to, not replace, initiatives being undertaken by private researchers, health care organizations, and others.

The creation of an agenda to strengthen knowledge of what is effective in medical care is well under way, and the committee joins the many

Page 158

other public and private groups that support these efforts. Nonetheless, compared with the resources invested in biological research and health care delivery, the promised investment in effectiveness research is still relatively small.

However, while more investment in research is desirable, it is important to have realistic expectations. First, there are practical and ethical limits on clinical research—too few researchers, long time horizons, and numerous instances in which clinicians would balk at research protocols that require withholding treatments generally thought to be useful or providing treatments generally thought to be inappropriate.

Second, much research does not really focus on the effectiveness of care in an average setting or population, nor does it evaluate the impact of care on quality of life and many other outcomes that are now considered important. What is needed are longitudinal observations of natural variations in the use and outcomes of economically and clinically important medical technologies in different practice situations. Clinical, claims, and other data bases need to be designed, and parallel attention needs to be given to the methodological issues in interpreting and disseminating research findings.

Third, effectiveness research that relies on existing claims and other records, while less expensive and time-consuming than most clinical trials, is not necessarily quick nor suitable for providing answers to many questions. Effectiveness research needs to be complemented by judgmental strategies, such as consensus panels, that use but go beyond scientific literature and empirical data. Many times the practical results of these processes will be imperfect. However, in the past 10 years, we have learned that the best randomized clinical trials have weaknesses and that these weaknesses are accepted, sometimes too readily. Intelligent compromise with perfection is necessary.

Fourth, research cannot resolve some questions. For example, researchers may be able to assess the clinical benefits and risks in using cataract surgery to correct a 20/40 vision defect. They cannot tell us whether such use of resources is wise given other uses to which those resources might (but not necessarily would) be put.

Fifth, sound clinical research does not automatically affect behavior. The research on what works in medical care should be complemented by research on how best to ensure that such knowledge is used effectively and efficiently. Such programmatic research is, for the most part, a low priority today. It is expensive, methodologically troublesome, and slow to pay off. As part of the overall strategy for containing total health care costs and improving the appropriateness of health care for all citizens, the committee urges federal and private consideration of carefully targeted research

Page 159

projects to test utilization management strategies and build methodologies for documenting the effects of existing programs.

Practice Guidelines and Review Criteria

To a very considerable extent in the past, medical practice guidelines have been implicit and have been linked to the amorphous concept of community standards of practice. The medical profession has historically resisted setting standards that are both explicit and national in scope. This resistance has been challenged by the research on practice variations and inappropriate use discussed in Chapter 2. The increasing emphasis on effectiveness research is also adding to the momentum for developing better and more extensive guidelines for patient care.

Translating clinical and other research into valid, reliable, and usable guidelines for medical practice and utilization management is a complicated enterprise that can take many forms. One approach is the development of clinical protocols that specify the tests, procedures, and other medical services that should be provided to patients with a specific clinical problem. Another approach is more proscriptive than prescriptive. It essentially sets boundaries outside which specific services for specific medical conditions are defined as inappropriate. Within the boundaries, guidelines may allow considerable room for an individual practitioner's judgment about the value of alternative courses of treatment. The two latter approaches have provided the bases for explicit and implicit review criteria used in utilization management organizations, but review organizations appear to have made little use of explicit clinical protocols.

The committee has cited analyses that show that some practice guidelines do not match relevant empirical research very well, and the fit between practice guidelines and criteria for utilization management may likewise be poor. The committee has identified a number of questions about the process of developing, disseminating, and updating practice guidelines and review criteria that need attention. They include the following:

• Should there be some kind of oversight of guidelines developed by different sources—a sort of quality review mechanism that assesses both the method and the substance of specific guidelines? What should happen when different sources develop conflicting guidelines?

• To what extent should patient preferences or cost-effectiveness analyses be considered in the development of practice guidelines? How should these issues of value be dealt with in the application of guidelines or in other strategies?

• Should adherence to guidelines provide physicians with protection against malpractice charges?

Page 160

• Over the long term, should a role for community or local standards continue?

• What considerations should apply in translating guidelines into criteria for prospective or retrospective review programs?

• Should private review organizations be encouraged to invest in criteria development if the results are open to outside examination?

These questions are relevant to much of the Institute of Medicine's agenda. Further investigation of these issues is now underway. It will draw not only on the Committee on Utilization Management by Third Parties, but also on the Committee to Design a Quality Assurance Strategy for Medicare and the Council on Health Care Technology, as well as outside groups such as the Health Care Financing Administration and the Physician Payment Review Commission.

Oversight of Utilization Management

Whether additional mechanisms for oversight of utilization management are needed to weed out incompetent organizations and programs was a question the committee discussed at considerable length. The protections offered by caveat emptor, self-regulation, and tort liability, although important, do not respond to all concerns about the impact of utilization management on patients, providers, and overall health care costs. Moreover, the current lack of oversight poses some risks to utilization management as a promising strategy for managing benefit costs. Misbehavior by a few organizations could harm patients, infuriate practitioners, and seriously damage acceptance and cooperation from patients and providers. On the other hand, premature or misguided regulation could stifle worthwhile innovations in technique and lock in inefficient or ineffective methods. Purchasers might then abandon utilization management for more onerous techniques of cost containment. As this report has emphasized, utilization management is still relatively new, is evolving quite rapidly, and works quite differently depending on the delivery and financing mechanism it serves.

At this time, the committee is not prepared to make recommendations about where oversight may be appropriate or what oversight strategies may offer a workable balancing of the conflicting values, interests, and needs of different parties. Rather, the committee has posed several questions about the oversight of utilization management that it will be considering further.

• How do we decide whether oversight is necessary and feasible?

• If an oversight mechanism is necessary and feasible, should it be public or private? If public, should all 50 states make their own policies or should the federal government impose uniform standards?2

2 As noted in the discussion of PROs in Chapter 2, federal regulation does not guarantee uniform

(Footnote continued on next page)

Page 161

• What should be the focus of oversight? Should the clinical criteria for assessing care and the mechanisms for appealing decisions be the main concerns? What about staffing, training, outcome reporting, and so on?

• Should utilization management conducted by different kinds of organizations, for example, PPOs versus staff model HMOs, be subject to different kinds of oversight? Should utilization management of some services—for example, mental health care—be subject to different kinds or levels of oversight?

• Can anything be learned from government oversight of PROs or the accreditation process managed by the Joint Commission on the Accreditation of Health Care Organizations?

These are not simple questions. Answering them demands more information and more thoughtful debate over how to judge the strengths and weaknesses of utilization management versus other strategies to control costs and influence patient care decisions and how to weigh the legitimate but sometimes conflicting needs, interests, and values of the parties involved in utilization management. At this time, the committee advises caution to public policymakers currently considering proposals to regulate and limit utilization management.

Conclusion

Utilization management needs to better demonstrate that it reduces the wasteful use of resources, improves the appropriateness of patient care, and imposes only reasonable burdens on patients and providers. It is important that it be given a chance to prove itself because if it works, it offers a means of accommodating individual patient circumstances that is more sensitive than many other cost-containment strategies, for example, those that rely on across-the-board restrictions in benefits.

However, the limits of utilization management and any other single strategy, even any combination of strategies, also need to be recognized. The issue is not whether utilization management does everything that needs to be done but whether it produces desirable results in reasonable ways at an acceptable cost. Is it, on balance, better to use it than discard it? And if utilization management is helpful, how can provider payment methods, retrospective utilization review and feedback techniques, resource planning mechanisms, and other tools be used to reinforce utilization management and deal with problems—such as distorted payment rates or excessive use of low-cost, high-volume services—not effectively addressed by utilization

(Footnote continued from previous page)

practice. And as noted in Chapter 3, proposed state regulations vary enormously and, in some cases, could make prior review and high-cost case management infeasible.

Page 162

management? This report has provided initial views on some of these questions.

The Institute of Medicine will continue its efforts to better define what role utilization management might play in helping society find an acceptable balance of efficiency, access, and appropriateness in health care. This clearly must be a shared venture. Fortunately, the quest to know what is useful and how to apply it universally are now central issues in medical research and public policy.