9
How Easily Do Health Care Systems Adopt New Knowledge, and What Are the Likely Future Developments?



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 169
--> 9 How Easily Do Health Care Systems Adopt New Knowledge, and What Are the Likely Future Developments?

OCR for page 169
This page in the original is blank.

OCR for page 169
--> Introduction Michael R. McGarvey Larry Manheim was invited to address the question, How easily do health care systems adopt new knowledge? His charge was further broadened to address the questions What are the likely future developments in the care of rheumatoid arthritis and lupus, and in managed care? Manheim's remarks and those of the invited reactors (Mark Robbins and Michael McGarvey) are formulated against the overall theme of the conference: What overall impact on the care of patients with chronic diseases are the dramatic alterations occurring in the health care system, particularly the rapid growth of managed care likely to have? Like other speakers, Manheim reflects on the positive and negative incentives inherent in the basic modes of physician or provider reimbursement, fee for service, and capitation. His own review of the literature suggests that actual outcomes for patients with chronic diseases in fee-for-service (FFS) and managed care systems yielded little difference, with the possible exception of slightly decreased patient satisfaction in managed care associated with access to services. Offsetting this is the potential opportunity within the managed care setting for the more organized and creative use of nonphysician providers to assist in continuity of care for patients with chronic illness. However, the current rapid consolidation among managed care organizations and the propensity of premium payors to shop for the plan with the lowest price work against continuity of doctor-patient relationships and of services for patients. Future developments in the care of patients with lupus and rheumatoid arthritis were, from a scientific and technology point of view, predicted to be incremental. Invited reactor Mark Robbins emphasizes the impact of the rapid evolution

OCR for page 169
--> of managed care, including consolidation in the industry. Michael McGarvey points out that the organized nature of managed care provides at least a theoretical opportunity for a more orderly approach to technology assessment and adoption than currently exists in the FFS system. Both Robbins and McGarvey point out that managed care can neither substitute for nor adequately address health care issues that are essentially social in nature, including coverage of the uninsured.

OCR for page 169
--> Invited Address Larry M. Manheim How easily do health care systems adopt new knowledge? There are a number of issues here, but I will try to keep the focus on how financial incentives are likely to impact adoption of new organizational knowledge by managed care organizations. I also will review the characteristics of evolving "model systems" that appear to provide good, integrated services at reasonable cost to frail elderly patients in managed care. I would suggest that these might serve as a model for rheumatologist service delivery. Although discussions of managed care are often phrased in terms of comparisons of capitated (HMOs) versus other service models, this can blur the critical questions of who the decision-makers are and what their incentives, constraints, training, and interests are in learning about, and adopting, new knowledge. I am an economist, not a physician, so I will not talk much about questions of training and the nature of medical knowledge in rheumatology. I cannot really speculate on the specifics of likely biomedical advances in treatment, except to say that breakthroughs are likely to be on the margin rather than in the form of cures. Treatment of rheumatoid arthritis and lupus usually involves multiple drugs. Changing medical technology often affects physician decisions with respect to whether one or multiple drugs are administered in situations in which one drug works better than another in limiting progression of disease, minimizing symptoms, or minimizing toxicity. Choice among which of these qualities of a drug to stress will often change depending on the ebb and flow of the disease. For rheumatoid arthritis, and even more for lupus, providers worry about systemic effects, monitoring organ system involvement, and changing the treatment as required.

OCR for page 169
--> As with other chronic diseases, physicians often work to achieve marginal improvements. At least for the foreseeable future, medical improvements are likely to increase the need to manage the patient over time. Therefore, the more important questions in terms of organizations adopting new knowledge, are those concerning changes an organization can adopt to manage chronic diseases more efficiently. In this regard, the knowledge base is growing as ''model systems'' are identified and studied. Before looking at the characteristics of some model systems that I believe hold promise for the management of chronic diseases such as rheumatoid arthritis and lupus, we need to ask whether the decision makers will be interested in this new knowledge? Who are the decision makers? At a minimum, they are the patient, the physician or other care provider, and the organization responsible for managing the patient. The patient will often be dissatisfied with treatment, simply because of the nature of these chronic diseases. Patients may go through a number of drugs, have physical therapy at different times, and go through some pain management. Some of these will work now, but not later. Some may not work now, or at least may not relieve symptoms to the satisfaction of the patient, leading to the question of trying an array of alternative treatments. Patient participation in these decisions may depend on socioeconomic or educational status, and patients (rightly or wrongly) may think they know more about their needs than their primary care physician and want to be managed by a physician who specializes in the patient's particular disease, which can cause tension if the primary care physician acts as gatekeeper. The physician is another decisionmaker. In fact, multiple physicians and allied health professionals are often involved, given the systemic nature of these diseases. What is the optimal relationship among the rheumatologist, other specialists and allied health professionals, and the primary care physician? To a large extent this is determined by the structure, incentives, and hiring practices of the managed care organization. The financial value of the patient to the HMO depends on the expected financial remuneration from treating that patient, compared with the expected costs of treatment. The HMO will also worry about the financial risks of certain types of patients turning out to be very high-cost patients. Many have pointed out that chronically ill individuals are not people that HMOs will necessarily want to attract, at least if they are not paid a capitation rate that is viewed as reflective of the cost of that type of patient. A lupus patient, for example, might apply for disability benefits under Social Security Disability Insurance (SSDI). Patients receiving coverage under SSDI, and therefore eligible for Medicare capitation rates based on the average cost of a disabled person in that county, may have a capitation rate far higher than a similar individual who does not receive SSDI eligibility—and physicians say eligibility determination can be quite arbitrary for a disease

OCR for page 169
--> such as lupus. In the first case the patient may be viewed by the HMO as a financially profitable patient, on average, while at a lower capitation rate the patient may be viewed as a financial liability. There is also the incentive question: who bears the cost of bad patient outcomes? Clearly, the HMO bears these costs to the extent that it is responsible for medical costs or has its reputation tarnished among people it wants to attract as members. In many cases the managed care organization may not bear the cost of bad outcomes from care (e.g., needed domiciliary long-term care costs if access to joint replacement is limited). For lupus patients, renal failure is a major complication that may occur because of poor chronic care management. However, transplantation and renal dialysis are covered under separate Medicare programs so that the cost of providing lessthan-optimal care, which results in renal failure, may not be borne by the patient's care plan. Thus, the incentives to provide good care are reduced. There are numerous cases in which such fragmentation distorts incentives when an HMO has a chronic care patient at risk for long-term care services. More generally, Medicare patients are not locked into their current plan, and even in private plans, employees are generally allowed to change plans annually. Given the high disenrollment rates for Medicare and Medicaid managed care patients, good, comprehensive medical care that prevents expensive utilization in the long term is not likely to be viewed as of much financial value to the HMO unless it generally enhances the HMO's reputation. Indeed, by making it difficult for costly, chronic care patients to access certain types of services, the HMO may be able to increase disenrollment and decrease future enrollment of these less profitable patients. Of course, this strategy is only financially successful if healthier patients do not also leave the HMO. With reference to training, I would stress the need for physicians to understand that the appropriate use of allied health personnel is likely to be a valued attribute of managed care subspecialists. Providing the care that practitioners in a chronic care field feel is first rate, either under the current system or under managed care, will have to involve, I think, less costly providers. Regardless of provider, appropriate use of social supports is very important both in physician training and in setting incentives for the system. Finally, because the nature of new knowledge in rheumatoid arthritis and lupus is often equivocal and requires complex long-term management (e.g., use of multiple drugs with toxic effects that must be monitored), an ongoing provider patient relationship that enhances patient monitoring will continue to be essential to good chronic care management. To help manage disease, patients need access to someone who they believe specializes in their disease, but it may be too expensive if that is always a rheumatologist. A useful approach is to use the rheumatologist as an ongoing consultant and part of a multidisciplinary team that includes less expensive allied health professionals who become the primary contact for individuals

OCR for page 169
--> with rheumatoid arthritis and lupus. This model for managing chronic care fits with a number of examples of best-practice models for providing care to the chronically disabled and frail elderly. We heard about the system at the University of California, Los Angeles, which is one example. Other models that seem to work in managed care systems involve gerontologists. In each case, subspecialists act as consultants to primary care physicians and as head of a team that will take over management when long-term care services are required. Physician assistants and allied health professionals working in the team are assigned to given patients and have the most direct contact with them. The key to such an approach is the notion of a team of health professionals, which includes someone knowledgeable in the disease (and directly responsible to the subspecialist) that a patient can readily access. A case manager might then be viewed by the patient as a facilitator rather than a gatekeeper. There also would be someone on the team through whom social and educational programs could be coordinated. The patient then accesses this system when he or she becomes problematic in terms of independent living. The same approach would seem to be possible in the management of rheumatologic chronic diseases. This may provide a system with good access to care and with costs that are reasonable relative to what the managed care organization sees as the program's value added. However, there is always likely to be tension between the notion that value is added by these additional services and the realization that both the value added and the funds available may be modest. Like everyone, I have been focusing on traditional managed care systems. Systemic lupus erythematosus (SLE) disproportionately affects African American women—nine out of ten cases of SLE are women, and the incidence of SLE is between two and five times greater for African American women than for white women. A lot of these women seek care in public institutions or in private institutions covered under Medicaid. Given low payments and declining budgets, more attention needs to be given to the management of chronic disease in the public sector. The shift to Medicaid case management might seem to offer hope for more comprehensive care for Medicaid eligibles, but this might prove illusory if Medicaid payments are insufficient to support true case management. Thus, it is important to continue research to monitor access to medical care received by chronic care patients under Medicaid or with no insurance coverage.

OCR for page 169
--> Invited Reaction Mark L. Robbins I would like to react to Dr. Manheim's presentation by providing an overview of the forces that are shaping the health care industry, the incentives that are influencing decisions at the patient and provider levels, and the changing nature of these incentives. I think that only by understanding these forces and their dynamics can we acquire insight into the critical issue of whether competition in health care will take place on the basis of innovations designed to enhance quality, thereby encouraging the adoption of new knowledge and improved chronic disease management, or based on potentially less desirable factors. Many of the comments that follow are based on my work on a study funded by the Robert Wood Johnson Foundation that is looking at private sector health care reform in four different U.S. markets. Economic recession, health care inflation, and an increasing uninsured population stimulated national health care reform efforts, which drove early changes in the market but ultimately did not proceed. The failure of federal health care reform left the purchasers of health care—large employers such as state governments and corporations—to fend for themselves in containing health care costs. Managed care and managed competition in a variety of forms appeared to be the best systematic approach by which these purchasers of health care could contain costs without introducing substantial cuts in health benefits. However, managed competition evolved primarily for large employers and state governments. As a result, some of our discussions and concerns today about where the unemployed, unregistered, or uninsured go for continuing health care. Employers generally wanted to maximize their purchasing power, avoid internal risk shifting, and provide wide geographic access for their

OCR for page 169
--> employees. This drove managed care organizations to develop larger and larger provider networks. The accelerated and sometimes frenetic cascade of mergers and acquisitions among health plans and hospitals has led to new and sometimes dysfunctional delivery systems and has raised the critical issue of what the essential components of a truly integrated delivery system are. It has also raised the questions of what is the most effective model or mixture of models (independent practice association (IPA), preferred provider organization (PPO), staff or group health maintenance organization) to deliver integrated care. This is particularly important for the continuing care of chronic disease patients. In some health markets, such as Minneapolis, the level of consolidation was so complete that the end result was an anticompetitive oligopoly of only three major remaining health delivery systems. The Minnesota business coalition is taking active steps to reverse this trend by shifting the focus of decision making, quality measurement and improvement, and incentives away from oversized health plans back to the clinic and provider levels. In response to this challenge, insurers and health plans are reassessing the value and services they add to the process of health care delivery. A significant part of their future contribution will be in the areas of technology assessment and innovation, development of clinical guidelines, and refinement of health information systems, because of their significant technical expertise and financial reserves. Large purchasers of health care and their employees also influenced the market by demanding the widest possible choice of providers. This contributed significantly to the dwindling enrollments of closed panel staff and group model managed care organizations and to the emergence of point-of-service (POS) options and rapidly expanding IPA networks. Overlapping IPAs, in which one provider may be listed with an infinite number of insurers, potentially represent a significant force against competition based on quality and innovation. First, quality indicators measured at the health plan level lose their significance and blunt the power of consumer choice based on quality reporting. If a difference between health plans did arise on a particular quality indicator that essentially derives from the same providers in the same offices with the same systems and support, it becomes a challenge to explain the measurement difference. Second, there is little incentive for these overlapping IPA health plans to introduce innovations. Innovations introduced by the health plan, which previously might have been associated with an exclusive provider network and provided brand identity, now diffuse rapidly through the overlapping provider network. For example, if an insurer develops a new clinical guideline or pathway, offers a new service, or develops a new chronic disease management technique that the provider finds of value, these innovations are likely to be adopted for patients in the providers' practices who are covered by other insurers or health plans.

OCR for page 169
--> In some markets such as California, the dominance of nonexclusive overlapping provider networks has led to extreme price sensitivity. There is evidence that people in these markets will switch health plans to save as little as four dollars a month on their premium payments because, despite the health plan change, they can still maintain their providers. One can certainly argue that the growth of overlapping IPAs is an effective way to force health plans to compete on price. Cost containment is something we have been talking about quite negatively today—the shift from inpatient to outpatient treatment; primary care gatekeepers limiting specialty access; the heavily discounted fee-for-service policies that were initially tried to control costs; and capitation, which followed and is gradually marching from west to east. In summary, I want again to emphasize that managed competition was not intended for and is not likely to address many key public policy and public health issues. It does not deal adequately with the public health system, social safety net issues, teaching and research, and the uninsured. It does not systematically address issues of technology assessment and innovation, and it does not provide an ethical framework for managing care while containing costs. For quality and innovation in chronic disease management to take place, a number of important developments are necessary. Purchasers of health care will have to recognize that payment for health services must include risk adjustment. Current payment systems create economic disincentives for health plans or provider networks to develop centers of excellence for the management of rheumatologic or other chronic diseases. Quality measurement related to chronic disease management will have to be improved and its focal point shifted from health plan to exclusive multispecialty provider groups. Capitation or risk payment for providers and health plans will have to be based on better chronic disease management rather than on measures of decreased utilization and costs.

OCR for page 169
This page in the original is blank.

OCR for page 169
--> Invited Reaction Michael R. McGarvey I speak as an individual with responsibility, among other things, for one of the most rapidly growing HMOs in the State of New Jersey (Blue Cross and Blue Shield of New Jersey). We currently have about 500,000 people covered either in our HMO or in our point-of-service product. You won't be surprised to hear that this is our most rapidly growing line of business. To get to the question that was just posed, whether purchasers are going to be buying largely on the basis of quality or of price, I am afraid that at the moment the answer tends to be price. A few very sophisticated, large organizations have worked a bit to see if they can't identify one organization from another on the basis of quality. However, for the most part, we are certainly finding in our marketplace that individuals and organizations, large and small, are buying on price, not on quality, even to the extent that we can manifest some degree of quality and are trying all the time to do a better job at that. I think Larry Manheim asked a very important question concerning the basic modes of adopting new technology in different managed care systems. That is, who is the decision maker? I think the fact of the matter is that in standard, traditional fee-for-service medicine, the decision maker has essentially always been the individual practicing physician. I think that depending on his or her reading habits and conscientiousness about continuing education, the individual physician has been in a good position to make a decision on whether to adopt the latest treatment modalities. Such decisions could involve whether or not, in the case of rheumatoid arthritis, to move to a fairly early trial of methotrexate, given current knowledge that its toxicity is much less than we originally feared; or whether antibiotics should be used for

OCR for page 169
--> the treatment of peptic ulcer rather than continuing to dole out Maalox in copious quantities. In the managed care system, however, the decision maker, more often than not, is the medical director of the managed care organization. More often than not, a real part of his or her responsibility is trying to keep up with literature, trying to tap into technology assessment capabilities, and trying to bring that information into the organization. Again, more often than not, this individual is guided in decision making by a series of physician panels that represent enrolled physicians—subspecialty groups, specialty groups, and primary care physicians of various types, usually collected in a variety of advisory panels. Thus, new technology can be addressed, assessed, adopted or rejected in a reasonably organized manner, and transmitted and communicated in a reasonably organized manner. Then, based on how the organization does or does not monitor the performance of its physicians, their adoption and application of new technology can be assessed. On balance then, I would say, just as Karen Davis outlined a number of theoretical pros and cons of managed care, that one of the pros is that in a managed care environment there is, in fact, a structured approach to technology assessment, adoption of new technology, and investing in new technology for actual application by practicing physicians helping to shape that system. One of the myths I would like to address here is that managed care organizations behave without any clinical input. Certainly I think some have and do; those tend to be the plans that have bitten the dust. The ones that are surviving and thriving tend to have a fairly well structured approach for developing physician input into the decision making and policies adopted. A very interesting new development is the concept of disease management, fundamentally the identification of patients who have a specific disease and the treatment of these patients according to guidelines and protocols that, again, have been adopted by the clinical decision making capability of the managed care organization. Many organizations are now describing themselves as disease managers. Right from the beginning, my own belief has been that, frankly, the only organizational structure in which disease management made very much sense at all was, in fact, managed care, and the better managed care organizations have, in fact, been doing their own brand of disease management for quite some time. Another major development—several speakers have made reference to it—is specialty capitation. Our own organization is, in fact, moving in that direction. Last year we did a statewide capitated network for radiology. This year we are looking to do very much the same kind of thing in cardiology, gastrointestinal services, and allergy. We may even address orthopedics. I am very sorry to say that rheumatology has not been on our screen, simply because it has not been one of those areas that has involved many of our

OCR for page 169
--> patients or costs us large amounts. My final point is that I was really struck by the as yet unpublished data reported today by Karen Davis and A1 Tarlov, highlighting the potential vulnerability in a managed care structure of populations in this country that have always been vulnerable—the old, the sick, and the poor. I think it is important to remember that the prepaid group practices that were the progenitors of our managed care entities today developed essentially to take care of working populations—younger people who were involved in a job and needed health care, and whose organizations paid for it. It is very much in the tradition of American insurance that it has been employment based. To the extent that these organizations are now being drawn, in a variety of ways, into the care of vulnerable populations—the Medicaid population and the aged population—I think this is still unfamiliar ground and a learning experience for them. The kind of data that we are beginning to educe are extremely important if there is to be an adaptive mechanism by which these organizations can address such issues in either a private sector or a public sector regulatory environment.

OCR for page 169
This page in the original is blank.

OCR for page 169
--> Discussion ROBERT NEWCOMER: In terms of innovations, we have not talked about community care systems and social services more generally. They typically are not funded under health insurance or Medicare but sometimes are funded under Medicaid or other kinds of state-controlled services. Particularly since we are talking about vulnerable populations, do you see any incentives or disincentives, or any problems with that community care safety net and the integration of health plans into it, utilizing its resources or augmenting them for their own benefit? LARRY MANHEIM: To the extent that community care will save money on some cost that these plans would otherwise bear, capitated care systems might put money there. I think it would be limited, though, because the disenrollment rate can be so high. Most of the things you are talking about, such as long-term care, are not included in current coverage. Although we might place that requirement on the system, for the funder it is really a question of breaking even. MICHAEL MCGARVEY: From a philosophical point of view, I find utilization of community care very attractive, but it is not the sort of thing that many of the people who pay our premiums are asking for. MARK ROBBINS: The Harvard Community Health Plan has always dedicated a percentage of its premium to research and public health efforts. As a result, violence education, violence prevention, and other kinds of community efforts are taking place within the HMO. However, this percentage

OCR for page 169
--> of premiums is under constant attack from within the plan. As the market becomes more and more competitive in terms of price, it is becoming more difficult to defend this practice to employers and others, but so far it has been protected. WILLIAM HAZZARD: It was reassuring to hear that physicians are becoming more and more involved in direction of some HMOs' policies and guidelines. What about rheumatic disease, in which we hear that patients are so critical? To what extent are patient—consumers being involved in the design of these programs? MICHAEL MCGARVEY: We do have a panel of enrolled patients who we consult on a regular basis, although I am not aware that we have as yet had anybody with a rheumatic problem. JEREMIAH BARONDESS: How do they relate to the putative topic of this section of the program: which system adapts more effectively and more efficiently to new information, new knowledge, and new techniques? MICHAEL MCGARVEY: That is a good question, but I am not sure there is much in the way of a structured method for involving patients in the nonmanaged system. They can certainly get to their individual physicians and should do so on a regular basis. JEREMIAH BARONDESS: Should we be pushing for patient engagement in selection or prioritization or adoption of new techniques? Does anybody have a view on that? LARRY MANHEIM: The only problem is that if you consider the average plan member, who is healthy, it is not clear that such a person would want to pay higher premiums to improve the system. I think it would be good to involve chronic disease patients, but you must have a payment system that supports that kind of interaction. NORMAN LEVINSKY: With regard to the issue of the disadvantaged, I would mention that there is an experiment going on, largely funded by Medicare, dealing with the frail elderly. It goes by the acronym PACE, Program for the All-Inclusive Care of the Elderly. One of the programs is related to Boston University, to one of its neighborhood health centers. This program involves socially responsible and very motivated physicians, along with a team of professionals and nonprofessional community workers. Together they provide all-inclusive care to the frail elderly. In terms of the cost potential for standard HMOs and managed care to provide that level of

OCR for page 169
--> service and dedication, I wonder whether one could expect comparable outcomes. There is a price, however, and the price is not in dollars, which is approximately the same as the standard Medicare costs. The price is very tight control of the patient's access to the health care system. Patients must use a limited group of practitioners and a limited group of hospitals; basically are putting their care totally in the hands of these—we hope—well-motivated and ethical physicians and associates. LARRY MANHEIM: I would just note that under PACE, providers are often paid as if the patients were in a nursing home, because they qualify. So, the capitation rate is quite high. LEIGH CALLAHAN: Alan Fogelman noted several things that I think are very important to vulnerable populations: patient facilitators, transportation, and education about their disease and its management. When someone asked him how he does this, the answer was capital, but a unique form of capital, in that the physicians are foregoing a percentage of their income to fund this comprehensive system. Do you think managed care organizations would put a portion of their income into government systems to enhance the care for the vulnerable populations? MICHAEL MCGARVEY: Although our HMO is technically a for-profit subsidiary, any sensible financial analyst looking at our medical loss ratio would question whether this is an accurate characterization. I think that changes are likely to come as managed care organizations acquire more and more experience with fragile populations. We have now been in Medicaid managed care activity for about two years. In fact, the leadership of that particular subsidiary of ours is now beginning to recognize that we need to have some people on staff who provide the type of facilitating role that Alan Fogelman described, and we are prepared to do it. We are even newer in the Medicare business, although we have used our traditional Medigap population as the target population for marketing. We are finding that our enrollees are considerably older than the group we had originally expected to enroll. We feel that we will probably have to make some of these investments ourselves. I think this kind of investment—this kind of programmatic development—sort of follows from what one perceives as the needs and requirements of covered populations. SUSANA SERRATE-SZTEIN: We heard today that there are no cures for either lupus or rheumatoid arthritis. I would like to get your opinions on the impact of the changing health care system on the availability of patients for clinical studies, especially large, multicenter clinical trials.

OCR for page 169
--> MICHAEL MCGARVEY: It certainly has occurred to me that managed care organizations would be the ideal places to tap in order to identify patients for enrollment in such controlled studies. I did some work last year reviewing operations at the National Institutes of Health Clinical Center. One of the recommendations, in fact, was that the clinical center be a little more aggressive about contacting managed care organizations for case finding. If things were structured properly and if people understood the ground rules, this it would be a very fertile area for patient identification.