Matthew 9:17

Disease management is one of several population-based approaches to medical care currently being used in managed care circles. It is offered as a new strategy for controlling medical costs that is superior to laissez faire physician-patient-based medicine or third-party micromanagement of the delivery process. Like new wine, which is best suited to new bottles, disease management and other population-based concepts are better suited to new integrated delivery models than to the traditional, fragmented health care delivery system. To understand the promises and the pitfalls of disease management, it is important to recognize the changing character of the delivery system.

The New Bottles of Health Care Delivery: Organized Systems

The cottage industry of health care delivery in the United States is disappearing in a uniquely American way. Most industrialized countries have adopted government-run or government-regulated health care financing systems. These systems serve as a brake on the growth of health care costs, but because they are rooted in the existing delivery system, they serve to perpetuate that system's structure. In contrast, the U.S. system of health care financing is not as closely tied to a particular structural approach of health care delivery. Our unregulated, pluralistic financing system has permitted and even fueled the explosive growth of the U.S. health care industry. However, the tide has now shifted, and the financing system is collectively sending the strong message that the way to economic survival in health care is down the path of decreased costs. Without the protective umbrella of a universal financing scheme for health care based on preserving the status quo, the delivery system is being reorganized and reconfigured before our eyes. This reconfiguration is being manifested in two forms: Smaller health care organizations are giving way to larger ones, and boundaries that previously separated different segments of the industry are blurring [1].

All players in the U.S. health care delivery system are clearly getting bigger. Health insurance companies (I include most health maintenance organizations in this category) are merging, as are home care companies, hospitals, pharmaceutical companies, and medical groups. In addition to increasing in size (horizontal integration), health care organizations are buying or developing tight relations with other organizations that operate in related, but previously separate, market segments (vertical integration). An example of the latter is the purchase of pharmacy benefit management companies by pharmaceutical companies.

Of more immediate importance to physicians is that physician practices are being integrated into the functions of other organizations. Old-line health maintenance organizations, such as Kaiser, have always linked an insurance function with a captive delivery system. Large commercial insurance organizations, such as CIGNA Healthcare, Aetna Health Plans, U.S. Healthcare, Pacificare, and Prudential, are aggressively seeking to develop tight long-term relations with physicians or physician groups. Other types of health care providers, such as home care companies and hospital systems, are purchasing physician groups. These rapidly developing, integrated, medical organizations are the delivery systems of the future for the United States. There may be concerns about the industrialization or the corporatization of the delivery system; nonetheless, these organizations are the new bottles into which the wine, the clinical side of medicine, will be poured. What will this wine look like, and who will do the pouring?

The New Wine of Health Care Delivery: Population-Based Medicine

The historical clinical model of medicine has been individualistic and physician-patient based. The cumulative effect of this approach, for better or worse, has been the output of innumerable individual decisions. This model has many potential benefits, but it has also been the only possible model in our fragmented delivery system. The new delivery system, made up of integrated organizations, offers the potential for a different clinical approach known as "population-based" medicine. This approach, which is sharing an increasing amount of space with physician-patient-based medicine, has several code words, all of which end in "management." The current code words are "health risk management," "demand management," "disease management," and "outcomes management." How new is this approach, really? Who has been involved in its development? What are its pitfalls and promises?

Although it is rooted in traditional epidemiologic and public health principles, the evolving population-based approach to clinical medicine is not the same as public health. The important operational differences between the two stem from the different economic incentives facing payers and public health authorities. Because health insurance is designed to deal with illness episodes and because most payers finance health care for populations of persons—whether they are employees, persons eligible for Medicare, or families with dependent children—payers' expenditures are tied to the amount of health care that these populations use. Superficially, this presents resource allocation problems similar to those faced by a community that undertakes a public health program such as smoking education. There are, however, important differences in program scope and duration between payer-driven population-based medicine and public health.

Unlike traditional public health programs, most population-based medicine efforts are restricted to persons whose health care is being financed by particular payers and is generally limited to medical (as opposed to social or hygienic) intervention. Payers are only willing to invest in programs that result in reduced health care costs within a limited period of time. Consequently, many traditional public health programs—those with important indirect benefits such as herd immunity or increased productivity; those in which the interventions are in nonmedical areas such as sanitation; or those that only reduce morbidity far in the future, such as those that decrease cholesterol levels—are likely to remain the province of public agencies. These public health programs are not attractive to private payers. The term "commercial population-based medicine" distinguishes payer-based approaches from public health efforts, although the two may share common ground, particularly in their methods.

The new model of commercial population-based medicine rests on the premise that if the health care needs of a population are known, then programs to decrease the costs of medical care received by that population can be designed. Properly designed programs would therefore be more efficient than similar efforts that might depend on individual steps taken during countless physician-patient interactions. This argument has some merit. For example, population-based immunization programs have repeatedly been shown to save costs, even in the short run and even in the elderly [2-4].

The management code words of commercial population-based medicine—"health risk," "demand," "disease," and "outcomes"—refer to population-based interventions that are delivered at different points in the health care cycle. The cycle of care for the individual person begins with one or more health risks before any care is sought and ends with one or more outcomes after care is received. The sum of the individual risks, demands, diseases, and outcomes of the members of a population can be approached systematically by interventions that favorably alter the population's health and health care costs. Although medically oriented, these interventions may not require physician participation. However, the new code words are popular and often mean exactly what the speaker wants them to mean, no more and no less.

Commercial population-based medicine is a natural outgrowth of the forces that are changing our delivery system. Health care payers buy health care for populations, and they want to manage these costs. New provider organizations are accepting financial risk for care provided to these populations and have the organizational resources to develop programs for specific population needs. In many ways, the new medical care model is possible because of the new organizational model: new wine into new bottles. Despite its promise, commercial population-based medicine is beset by the economic forces that created it. Using disease management as the focal point of discussion, we can examine the pitfalls and promises of the new medical care model.

Disease Management

Commercial disease management strategies are rooted in the well-established observations that a small portion of the population generally consumes a great deal of health care resources [5] and that short-term costs in one area may be offset by longer-term savings in another [6]. Disease-based interventions that are targeted to specific populations and are able to manage the total costs of care can therefore decrease overall costs while maintaining or improving quality. Potential interventions range from "low-tech" patient-focused efforts (such as education for chronic obstructive pulmonary disease [7]) to multifactorial interventions (such as those for asthma [8]) to specialized treatment facilities (such as those for stroke [9]). These observations lead either to the conclusion that a lot of money could be saved by developing the right disease-based interventions or to the conclusion that many successful disease management models are not necessarily physician driven.

Disease Management Pitfalls

Because of its attractiveness as a cost-reducing strategy and its potential for broad application, disease management has become a popular approach in the health care system. Many hands are seeking to pour this wine. Pharmaceutical companies are particularly active, but home care companies, patient education companies, and even supermarket chains are also involved. Because of their visibility, pharmaceutical companies have been accused of using the promise of disease management as a way to sell particular drugs [6, 10]. For example, companies with large investments in asthma drugs are leading the charge to develop asthma disease management programs. The involvement of these other organizations increases not only the possibility that disease management may become more of a marketing tool than a substantive contribution to the health care delivery process but also the possibility that control over health care may move from physicians to drug companies [11].

Although physicians may worry about loss of control, it is hard to argue with a disease-based program that delivers improved outcomes and decreased costs. More important than who runs a disease management program is whether there is good reason to believe that the proposed program would work. Pharmaceutical companies, for example, make the credible argument that more compliance with a drug regimen is always better than less. However, many widely accepted drug regimens are based on the original Food and Drug Administration approval studies or on existing practice patterns and are not necessarily optimal. Recent experience has shown that longer or more conservative drug treatment regimens for certain conditions are no better than shorter, less expensive approaches [12-14]. Before embarking on a commercial disease management strategy designed to enhance drug compliance, physicians (and others) should assure themselves that the proposed drug regimen is optimal; they should not accept the sponsor's assertions on faith.

The extent to which disease management programs are accepted on faith can be considerable. An excellent example of misplaced faith in disease management is the widespread use of social-support programs for prenatal care by managed care organizations. These companies now offer educational videos, handouts, and other materials directly to their pregnant members. The logic is something like, "If only we can prevent one premature birth, the program will be worth it." In truth, there is little reason to believe that social support programs aimed at women who already have prenatal care do anything to improve outcomes or decrease costs [15-17].

A thoughtful review of prenatal support programs [18] conservatively concluded that little evidence supports such programs. Why do these programs continue? One reason is that they have become markers for organizational quality, conferring an indirect marketing benefit (a halo effect) on the organization offering the program. Another reason is that members of the organizations like these programs, which confers a direct marketing benefit on the organization. A third, more insidious, reason is that these programs may confer an actuarial advantage to a risk-bearing organization, be it an insurance company, a health maintenance organization, or an integrated delivery system. The way this actuarial advantage works is that prenatal support programs are attractive to young families—a desirable low-risk group. Organizations that offer such programs can use them to enroll these families and can thereby improve the level of risk that they must manage.

Conversely, an example of a disease management program that confers an actuarial disadvantage on an organization would be a well-advertised program for the acquired immunodeficiency syndrome (AIDS), cancer, diabetes, asthma, or another serious chronic disease. Although such a program might help an organization better manage the risks of these diseases, it might also serve as magnets for persons who have them. Specifically, an organization promoting a high-quality cost-effective diabetes program might attract more than enough patients with diabetes to its plan or facilities to eliminate the cost savings from the program. Because most actuaries have no trouble envisioning this possibility, there is a constant tension in risk-bearing organizations between those who seek to develop and promote disease management programs and those who would like to develop the programs but keep them secret. The second pitfall of disease management programs, therefore, is that their indirect effects can outweigh their direct effects. This may confuse physicians who tend to look only at direct health care costs and benefits when making decisions.

A third pitfall of disease management programs is an overestimation of their likely benefit. One example, given above, is that of prenatal programs. Another is diabetes management programs. Those who treat patients with diabetes say that 14% of the U.S. health care bill is spent on diabetes [19]. Others who sell medications and supplies to patients with diabetes say that the cost of treatment is 5.8% of personal health care expenditures [20]. My internal review of data from a large insurance carrier found that a smaller percentage of claims, closer to 1%, could actually be attributed to services associated with the care of diabetes. Although it cannot be disputed that patients with diabetes have higher than average health care costs, large numbers should not seduce one into starting a disease management program without clear evidence as to which costs will be reduced and which outcomes will be improved. Would the benefits of the program actually outweigh its setup costs?

A fourth pitfall of disease management programs is fragmentation of care. Superficially, this might appear to be the natural consequence of involvement in the medical care delivery process by disease management companies, but involvement by nonphysicians is generally not an issue in practice. Although drug companies and others want to sell disease management programs, they do not want to be sued for malpractice, and they do not want to offend physicians who still drive the sales of their core products. At a practical level, the biggest threat of fragmentation comes from physician-driven disease-management programs.

Disease management, with its emphasis on specialized care for certain conditions, is a natural fit for specialists who see their incomes and control waning [21]. Beyond just using the argument that specialty-based programs improve care, physicians have begun to develop disease-based networks and to offer disease-based insurance "carve-outs" to purchasers. These programs surpass the existing managed care specialty insurance products—vision, dental, mental health, and workers' compensation care (each with its own provider networks)—by offering carve-outs for AIDS and oncology, which also have their own networks. Provider-based disease management programs, therefore, have considerable potential to further fragment the delivery process.

The current trend in managed care financing, in which integrated medical systems can accept the responsibilities of delivery, is to rebundle insurance products such as vision care, mental health care, and workers' compensation into comprehensive 24-hour insured care using one provider network. This is occurring because purchasers are concerned that one type of coverage, such as workers' compensation, may be shifting costs to or receiving costs from another type, such as standard health insurance. In addition, most payers recognize that uncoordinated and fragmented care is not clinically or economically desirable. Such is the driving force behind commercial population-based medicine. Given this high-level trend toward consolidation of care and financing, it is unlikely that disease-based insurance carve-outs will assume a large role. Where they do exist, however, they invite even greater fragmentation of the delivery system.

Disease Management Promises

Population-based medicine, including disease management, is a new approach to medical care. It will not replace physician-patient-based medicine but will coexist alongside it in the new network of delivery systems. It offers several promises.

The promise of improved efficiency has been discussed above. As noted, much literature describes successful population-based approaches to different diseases. These programs often rely on a systematic team approach to patients with the disease and have been shown to decrease overall costs while maintaining or improving outcomes. If anything, the results of reported disease-management successes need wider imitation and refinement by integrated delivery systems.

There are also two long-term promises for disease management and population-based medicine in general. The first is that this approach affords the opportunity to find systems solutions to health care problems. Systems solutions, as opposed to individual ones, can reduce the pressure for perfection currently felt by physicians. Physicians and all participants in the health care delivery process are imperfect. Careful examination of the care provided to a population of patients can show ways in which care can be improved by altering the delivery system rather than the persons who provide the delivery [22-24]. It is this kind of systems thinking that provides the organizational framework for a continuous quality-improvement process.

The second promise is that disease management programs can improve medical decision making by exposing physicians to the full ramifications of their decisions. These programs allow physicians to better understand the role others play in health care delivery and the need to balance competing priorities [25, 26]. Both of these long-term benefits will most likely be achieved in the context of an integrated medical delivery system.

Conclusion

Top Conclusion Author & Article Info References

In summary, the new wine of commercial population-based medicine has arrived just in time for the new bottles of integrated delivery systems. Much of the data with which to develop successful population-based programs currently resides in insurance company claims databases and in the records of pharmaceutical companies. Most of the practical knowledge about how to use these data and how to develop programs that work resides in the heads of clinicians. To develop successful programs, all participants must cooperate. Population-based medicine offers pitfalls and promises; it would have never fit in the old bottles.