Aligning the System to Promote Value— Now and in the Future
The nation and our healthcare system are at a critical juncture. If overall national expenditures for health care continue to rise, other priorities such as education and energy will suffer (White House, 2009). Consensus among stakeholders—patients, healthcare providers, payers, manufacturers, and legislators—continues to grow that the healthcare infrastructure in the United States needs transformative change (Oberlander, 2007). Investments, both short and long term, involving the public and the private sectors, will be crucial to achieving control over escalating healthcare expenditures while preserving incentives for innovation and enhancing quality of care.
The U.S. healthcare system needs many fundamental changes in order to create a coherent, efficient care delivery system that delivers value to Americans (IOM, 2007). Many of the tools necessary to achieve this vision—such as value-based insurance design, tiering, workplace wellness programs, electronic health records, patient-centered medical homes, and disease management programs—were discussed at this workshop. However, mobilizing stakeholders into agreement about the specific policy levers necessary to obtain increased value in health care also depends on the framing of the message, John C. Rother asserted as part of a panel that discussed the challenges and opportunities confronting the value movement. He encouraged focusing on clinical excellence and improvements in health rather than just cost savings; otherwise he feared a repetition of prior mistakes that occurred with the implementation of capitation and managed care.
As part of that concluding panel, Ezekiel J. Emanuel cautioned that no single one of the tools discussed during the entire workshop is a “magic bullet.” He described how many economists believe that proper provider incentives and payment design will fix the system; how some stakeholders speak of the panacea of health information technology (HIT); how health services researchers tout comparative effectiveness as a generator of data that will inform healthcare decision making and reshape the delivery of health care in this country; and finally, how others believe that educating consumers about the total costs of healthcare interventions and increasing cost sharing with patients will stimulate changes in behavior that will yield cost savings.
Emanuel posited that all of these elements simply constitute pieces of a toolbox that need to be engaged simultaneously and synergistically to create disruptive evolution in health care. The difficulty lies in engaging all of these tools effectively to produce incremental changes that eventually coalesce into enduring system-wide, synergistic innovations.
In this chapter’s first paper, Sir Michael Rawlins provides lessons for U.S. efforts drawn from his experiences leading the United Kingdom’s National Institute for Health and Clinical Excellence (NICE). He describes four particular challenges faced by NICE as it attempted to obtain value for patients, families, and UK society: (1) the need for direct comparative effectiveness studies between interventions, (2) the limitations in translating clinical trial results to real-world settings, (3) the difficulties in valuing treatments across clinical conditions, and (4) the incorporation of cost-effectiveness into value determinations. Although these challenges are global, he cautions that ultimately value assessments must consider the societal context, culture, and preferences of the country in which the decisions are made.
Christine K. Cassel continues with a consideration of likely future advances in medicine and a framework for understanding the value proposition in the context of these innovations, suggesting that the current environment might present a moment of disequilibrium that could catalyze reengineering of the healthcare system. Ezekiel J. Emanuel, Samuel R. Nussbaum, and John C. Rother discussed short- and long-term investment opportunities that would foster synergistic innovations and disruptive evolution toward increasing value in health care.
PLENARY ADDRESS: PERSPECTIVES ON VALUE FROM THE UNITED KINGDOM
Sir Michael Rawlins, M.D., National Institute for Health and Clinical Excellence
The United Kingdom’s National Health Service (NHS) came into existence in 1948. Funded from general taxation, it provides care from “the
cradle to the grave” and is free at the point of use. The small private sector provides less than 5 percent (mainly elective surgery) of the nation’s healthcare needs.
Although the structure and political context of the NHS is unique, the problems it faces are global. The tension between the increasing demands for health care and the limits on the available resources underscores—for all nations—the importance of obtaining value for patients, their families, and society.
NICE was created, as part of the NHS, in 1999. It is required to provide NHS healthcare professionals with advice on achieving the highest quality of care for patients at an affordable cost. In doing so, NICE faces four particular challenges:
The dearth of direct comparative effectiveness studies between interventions;
The limitations in applying the results from clinical trials to the real world;
The translation of clinical effectiveness into value; and
Drawing conclusions that are based, in part, on considerations of cost-effectiveness.
Assessments of comparative effectiveness, between alternative therapeutic or diagnostic strategies, are critical for examining both clinical and cost-effectiveness. A placebo-controlled trial provides information about the absolute efficacy of an intervention but does not necessarily give much indication about a product’s performance in comparison to other available treatments for the same condition. Although head-to-head trials with an active comparator offer richer data for informing routine clinical care, most drug regulatory authorities do not require them for registration purposes. This occurs, in part, because comparative studies will necessarily be larger and hence more costly, but also because what constitutes the current standard of care varies widely both within, and between, healthcare organizations.
In the absence of direct evidence of comparative effectiveness, NICE is often forced to use indirect comparisons in its assessments. An indirect comparison between drug A and drug B can be undertaken by analyzing the results of each against placebo. While there is evidence to show that some types of indirect comparison provide more reliable evidence than others, the approach is often challenging. More head-to-head comparative studies would allow physicians and patients to select appropriate treatments. This particularly applies to studies involving comparisons of pharmaceuticals with devices, physical therapies, or psychological approaches.
Limitations of Randomized Controlled Trials
Randomized controlled trials usually involve relatively small numbers of people from homogeneous patient populations for brief periods of time. The generalizability of these findings to the use of an intervention in routine clinical care is often uncertain. Many drugs, for example, are given to elderly patients with multiple conditions for which they are receiving other necessary medications. Polypharmacy therefore creates additional difficulties that will rarely be resolved from the results of randomized controlled trials. Physicians are thus left to weigh these potential interactions with little empirical data to inform their decisions.
The increasing use of interim and subgroup analyses, in the analysis of the results of randomized controlled trials, poses further problems. There is no consensus among statisticians about how—or, indeed, whether—these should be undertaken. Interim analyses, in particular, more frequently lead to early termination of studies in the field of oncology. While emerging evidence of serious toxicity or significant benefits during a study is a valid reason for premature termination of the trial, there are also concerns that interim findings may represent—by chance—either false positive or false negative results (known, respectively, as “random highs” or “random lows”).
Measuring ultimate end points in clinical trials can also be difficult, especially over the long term, and surrogate markers are often an attractive proposition. In oncology, for example, investigators have frequently used tumor response rates as predictors of long-term clinical outcomes. Tumor response rates, however, tend to be weak markers of ultimate end points such as overall survival or quality of life, and greater faith has been placed in them than is warranted.
Valuing Treatments Across Clinical Conditions
NICE faces challenges when translating clinical effectiveness into “value.” There is, however, an imperative to do so in order to avoid the provision of cost-ineffective treatments for a few patients with one condition at the expense of others with another condition, for whom cost-effective care would be denied. NICE therefore expresses the health gain produced by a particular intervention as a gain in “utility.” This takes account of both the increased quality of life and the time for which it is enjoyed.
There is substantial research to support the use of the EuroQol-5D (EQ-5D)1 as an appropriate approach to assessing utility gains. More
The EuroQol-5D is a standardized instrument used for measuring healthcare outcomes that produces a single measure (utility) of health status. Increasingly used as a stand-alone measure in many countries, the United Kingdom weights the instrument’s variables to reflect the values of the population as a whole. For more information see http://www.euroqol.org/.
recently the Short-Form 36 (SF-36)2 has also been shown to be capable of assessing utility gains. The EQ-5D considers societal preferences (derived from large population surveys) for various health states that enable comparisons of health gains to be made of different interventions across different conditions.
The use of measures of utility, derived from instruments such as the EQ-5D and the SF-36, is not without problems. Randomized controlled trials rarely include the capture of overall quality-of-life measures let alone ones that can be expressed as utilities. Measures of utility may sometimes fail to capture significant elements that are critical for some conditions (e.g., fatigue, exhaustion, poor cognition). Societal preferences about the burden associated with a particular condition may also differ markedly from those of individuals with the disorder. For example, fully sighted people regard the loss of one eye as causing a significant reduction in the quality of life. For those with monocular vision, however, blindness in one eye is much less of a burden.
Taking Cost into Account
When considering the cost of an intervention, NICE takes into account the full range of costs and savings that are relevant to the NHS. These include the acquisition costs of the product, any special monitoring requirements, and the treatment of complications. In addition, the cost considerations also include savings, for example, from reduced hospitalization or delayed nursing home admission.
NICE does not take into account the broader costs and savings to society as a whole, such as those incurred by sickness benefits and lost productivity. There are obviously potential benefits from using such a wider economic perspective. However, NICE’s Statutory Instruments, which form its legal basis, specifically deny such an approach. Moreover, providing advantages to those who are economically active would inevitably disadvantage those who are economically inactive, such as the elderly. This, in the United Kingdom, would be unacceptable.
NICE generally expresses the additional costs, for the additional health gains, as the incremental cost-effectiveness ratio (usually as the cost per quality-adjusted life-year gained or cost per QALY). The Institute, in distinguishing cost-effective from cost-ineffective interventions, does not how-
The Short-Form 36 (SF-36) is a “multipurpose, 36-item survey that measures eightdomains of health: physical functioning, role limitations due to physical health, bodily pain, general health perceptions, vitality, social functioning, role limitations due to emotional problems, and mental health.” These can be combined into a single measure of utility. See http://www.iqola.org/instruments.aspx.
ever use a strict cost per QALY “cutoff.” Instead, it provides its advisory committees with a range of preferred values and asks these committees to exercise their judgment about whether the specific circumstances should allow the preferred value range to be exceeded. In particular, committees are asked to take account of whether the intervention adds demonstrable and substantial benefits that may not have been captured adequately in the measurement of health gain. In this approach, the cost per QALY is used as “a tool not a rule.”
Of the 350 decisions that NICE has made about the clinical and cost-effectiveness of new and established interventions (including pharmaceuticals, devices, and procedures), it has rejected only 9 percent. A further 6 percent have been recommended for use “only in research” (i.e., coverage with evidence development).
The global interest in NICE is suggested by the more than 2,000 articles about its work in peer-reviewed journals and the 450,000 monthly visits to its website. This suggests that there is an appetite for the role the Institute undertakes in distinguishing between cost-effective and cost-ineffective interventions. It is important to appreciate that such distinctions, though informed by evidence, require a degree of judgment on the part of decision makers, and the social and cultural environment in which decision makers work will, rightly, condition the conclusions they reach.
VALUE IN HEALTH CARE: THE PATH TO VALUE
Christine K. Cassel, M.D., American Board of Internal Medicine
As we look at how to achieve value in health care, I have been asked to consider the future in two dimensions: (1) to anticipate likely advances in medicine and (2) to create a framework to understand the additive value of these advances in the important context of resource constraints and value trade-offs.
Categories of Technology Advancement
As we look toward the world of health care in the future, for example in the year 2025, the advances we envision are in areas where rapid progress is already occurring, such as imaging technologies, specialty pharmaceuticals, genetic targeting and personalized medicine, nanotechnology, robotics, and telemedicine. Just as advances in these fields will continue to emerge, there are very likely to be some advances in fields that we cannot even imagine.
As we consider these future advances, I believe it will be helpful to consider the different categorical types of technological advances in medicine. The first category of advances includes interventions that prevent disease. Immunizations—which mobilize the body’s own defenses through the immune system to prevent illness—represent the ideal cost-effective technology. Immunizations have been used most effectively against infectious diseases, such as polio and measles, but recently also have been successfully employed as a prevention tool for cervical cancer, possible because it is caused by a virus.
Some advances can lead to early detection and, with appropriate treatment, a cure of illness. We see evidence of this in the early detection of many malignancies as well as early treatment of infectious illnesses, such as community-acquired pneumonias. Early detection can also identify a chronic condition for which there is not a cure, but for which early detection can lead to prolonged survival if treatment is available. Examples of these diseases are diabetes, hypertension, and coronary artery disease. The concept of using modern interventions—not for cure, but to stave off decline—is possible not just in diabetes and heart disease, but increasingly in other chronic conditions such as cancer, Alzheimer’s disease, heart disease, renal failure, diabetes, vascular disease, and treatment of mood disorders such as depression or bipolar conditions. In these conditions, advances occur not only in the early detection methods, but just as importantly in the interventions.
Another category of technological advance is rescue technology. These are primarily therapeutic interventions where the natural history of life-threatening conditions can be interrupted and altered in order to avert death. The most dramatic of these have been utilized among the elderly. For example, previously an individual who developed a severe pneumonia or a dissecting aneurysm at the age of 80 and subsequently died of that condition would not have been considered to have died prematurely. Yet, aggressive interventions can now often prevent that death and lead to prolonged survival for the patient.
However, while rescue technologies can avert death, they do not necessarily return patients back to their premorbid baseline physiological and functional status. Patients sometimes suffer from additional post-condition disability. One stark example of this is hip fracture, which occurs with escalating frequency in the eighth and ninth decades of life. Even after repair of the hip fracture, there is a significant likelihood that the individual will not return to independent function and a more than 50 percent chance of needing prolonged nursing home care, with subsequent likely complications leading to death.
Medical advances also include interventions that are not treating a disease necessarily but do enhance normal function and, therefore, quality
of life—a current, exciting example of which is cognitive enhancers. Neuro-cognitive research has led to an increased understanding of medications that can enhance normal memory as well as normal alertness and reduce the need for sleep. Research is also improving mobility, in terms of both muscle strength without the adverse consequences of anabolic steroids and improving cartilage and synovial function in aging joints to avoid the need for joint replacement.
Another example in this category involves sensory conditions, which are rarely considered serious medical problems. Yet vision and hearing impairments can lead to lack of independence, lack of mobility, and greater risk of depression, all of which can reduce quality of life and be potentially life-threatening. Treatments for hearing disorders can dramatically improve an otherwise functional older person’s ability to continue working and engage in community and family activities. All of these interventions that enhance normal functioning lead to improved quality of life. Additionally, by keeping people active and engaged as they age, they also reduce the accretion of other kinds of chronic illnesses.
Finally, we must also acknowledge the tremendous boom in cosmetic surgery and cosmetic dermatology that has been driven by the combination of scientific advances and the market of aging baby boomers.
More Than Biomedical Progress
All that I have just discussed has considered categories of advances in biomedical science. Yet in light of the inevitable growth of additional advances, we need to consider two other developments as equally important. The first one is information technology advances and the second is healthcare delivery advances.
We already have seen major advances in the ability to manage healthcare information electronically, and yet we have not invested as a nation in making sure that this technology is used more widely and more effectively. In fact, we may have more mandates for the use of sildenafil than we have for health information technology. Effective information management cannot only reduce overuse, particularly in the diagnostic arena, but also better coordinate care across providers to reduce errors and missed opportunities for appropriate interventions. In addition to information management, information technology allows us to perform wireless monitoring of conditions such as arrhythmias. It has also made possible robotic surgery, robotic healthcare support for the home-bound, and telemedicine. We ought to think about the value that is placed on healthcare information technology advances every bit as critically and thoughtfully as we do with biomedical advances.
Delivery system advances are generally not thought of in the same way as technology—either information technology or biomedical technology—
and yet almost every healthcare expert looking at the next chapter of healthcare reform in the United States points out that our delivery system is broken, fragmented, ineffective, redundant, inefficient, and wasteful. It is also inequitable and unable to deal effectively with healthcare disparities and cultural complexity.
Yet there are established effective delivery systems in the United States as well as in other countries from which we can learn. There are also major efforts under way to reestablish systems that coordinate care on behalf of patients with complex conditions. The most notable of these are the medical home demonstrations now under way, supported by healthcare payers and purchasers and soon to be an initiated under a major Medicare demonstration. In addition, there are small sites of innovation around transition management from hospital to home or to nursing home, as well as a drive for greater integration between specialists caring for a single individual. Finally, because many aspects of the delivery system have become global, we must also include efforts to better integrate and coordinate these global services as we think about the next steps for health care.
Categorical Questions of Value
As we think about how to determine the value of new and all-too-often very expensive technologies in the categories I have just described, we can think of evaluating them on the basis of two factors. One factor is the prolongation of life, where technology prolongs survival although sometimes with significant, ongoing, serious illness or disability. The other factor is quality of life, where interventions that do not necessarily prolong life can increase quality of life—quality in terms of function at any level, such as physical mobility, mental function, emotional health, or pain control.
In addition, there are categorical questions that are useful in the determination of value from technologies. These include the following:
Magnitude of the impact: We need to consider the relative impact on various segments of the population, the impact on life prolongation and on quality of life.
Magnitude of the risk: In diagnostic studies, the risk of false positives needs to be thoroughly evaluated. Adverse consequences need to be weighed against gains on both individual and societal levels.
Downstream effects: Similar to joint replacement or other interventions leading to enhanced mobility or improvement in vision and hearing, and thus reducing depression and increasing social engagement, one has to consider the potential impact on other comorbid chronic conditions.
Ethics and fairness: We need to ask about the ethics and fairness of value as part of these considerations. Is the access to life prolongation or functional improvement part of the social contract? For example, when we insist through Emergency Medical Treatment and Active Labor Act laws that critically ill patients be treated in emergency rooms regardless of their ability to pay, are we saying as a society that we value life enough to provide rescue technology? Does this suggest that if it were more cost-effective to treat those individuals before the condition got to the extreme, we should create an environment in which prevention were possible? Not only must we ask about the social contract, we have to ask about the insurance contract. As we have multiple approaches to healthcare insurance throughout the country, should there be a basic minimum benefits package to which every insured patient is entitled? Also, how should these benefits be insulated from special interests or political concerns that add costs but do not add value for a significant portion of the population?
Value and Values in Health Care
Finally, as we think about value in health care, it is important to first ask about values in health care. When we think of value, we think of efficiency, evidence-based medicine, and cost-effective treatments. When ethicists talk about values, however, they talk about fairness, compassion, and respect for individuals.
So where do value and values intersect in health care? Every national survey shows that the American public supports major healthcare reform. In a 2008 survey, 93 percent of respondents said the quality and affordability of health care were equally important (Employee Benefit Research Institute, 2008). Yet in the United States, an unresolved issue is the degree to which social justice is part of the fairness equation, as it is in other developed countries where the debate starts with the assumption that everyone should have coverage for basic health care. The United States has never agreed to that assumption, yet the argument from an efficiency standpoint seems to suggest that this direction is a good one.
It is worth raising the question again whether this unresolved issue stems from a concern for justice and fairness or for efficiency. Regardless of the root of the issue, we know that we have the following realities to deal with: 45 million uninsured and perhaps twice as many underinsured. Both of these facts lead to medical bankruptcy continuing to be the primary cause of the collapse of families, which of course is even more important in the current, difficult economic times. We now have good documentation of
the multiple causes of healthcare disparities, but adding fairness and justice components to our determination of value could significantly narrow those disparities.
So what kind of themes can help us advance value-based health care? One is comparative effectiveness, which is receiving a lot of attention now. Although this is not a new idea, it is hopefully an idea whose time has come. Lessons of the past have shown how vitally important it is for a comparative effectiveness entity to be both authoritative and free of political pressure. Secondly, financial incentives must be aligned for value not for volume. This will inevitably lead to some provider risk mechanism, whether it is global or bundled payments or a prepaid model, such as a capitated or concierge model. It is clear that a straight fee-for-service payment system misaligns value in an unacceptable fashion. Finally, the modern context has taught us the importance of utilizing multistakeholder processes that create a level playing field in which consumers, patients, and purchasers, as well as providers and payers, can develop and implement policies and payment approaches based on a common value system. Examining and establishing values needs to be an ongoing process that is inherent and consistent with the daily work of providers, hospitals, and physicians and built around both individual patients and their communities.
Perhaps this will be a moment in time when the Nash Equilibrium breaks down. John Nash, the Nobel Prize–winning mathematician, posited that there is a social equilibrium reached when multiple parties are frozen in current relationships because no party can change its own strategy while the strategies of the other parties remain unchanged. With the current economic crisis in the United States, the Nash Equilibrium could break down when the pain of the status quo for multiple parties, including consumers, patients, and purchasers, begins to exceed the fear of the unknown. If all the stakeholders in health care can seize that moment of disequilibrium, the opportunity could provide a catalyst for clinical reengineering of the healthcare system in a manner that deems it as important as biomedical advances. When system reengineering and biomedical advances finally synergize, it will much more likely that we can have a value-based approach to the adoption of new technology.
OPPORTUNITIES FOR CHANGE
The following is a summary comments made by the workshopís concluding panelists, which included Ezekiel J. Emanuel from the National Institutes of Health, Samuel R. Nussbaum from Wellpoint, Inc., and John C. Rother from AARP. Their comments focused on the short- and long-term investment opportunities that would foster synergistic innovations and disruptive evolution towards increasing value in health care.
The panel identified several opportunities for short-term investments that can yield improvements in the current health care system. Emanuel discussed the necessity to generate high-quality, reliable dataódata that addresses the comparative efficacy of medical interventions and minimizes the safety and risks of these interventions. Nussbaum cited a recent editorial co-authored by Billy Beane, Newt Gingrich, and John Kerry, which described the revolution in baseball enabled by the collection of data (Beane, et al., 2008), and argued that data and evidence-based medical practice could drive a similar evolution towards value in health care.
The panel further pinpointed electronic health records as an implementable tool to collect the necessary data. Nussbaum discussed the benefits of engaging all health care providers in e-prescribing, which could provide a quick gateway to broader adoption of HIT and real-time decision support. HIT could also provide a starting point for tailoring decisions to individual clinical circumstances while considering additional information on costs, outcomes and strength of the available evidence.
Nussbaum additionally emphasized the importance of directly educating consumers in the decision-making process while engaging stakeholders about the necessity to obtain greater value in health care. Rother built on this idea by encouraging trust building among stakeholders through transparency. Considering the differing perspectives on value, he also asserted the importance of providing opportunities for feedback and incorporating these viewpoints into the reform process. He suggested that engaging these stakeholdersóincluding, importantly, public consumersóin the short-term in this manner will likely yield substantive long-term gains.
Focusing on long-term opportunities, Nussbaum identified payment reform as a cornerstone to re-organizing the current health system. He discussed the necessary innovations to incentive improved outcomes rather than volume of services. Nussbaum also specifically cited the need to utilize payment restrictions to address the overuse of unnecessary services, such as the 20 to 50 percent of duplicative or unnecessary imaging studies that costs billions of dollars (Americaís Health Insurance Plans, 2008). He described payment for bundles of services as a potential method for reducing unnecessary services while refocusing clinical care on improving quality and outcomes.
The panel also discussed reorganizing care delivery systems to facilitate chronic disease management that will enable progressive quality improvement beyond current standards. Referencing Arnold Milsteinís discussion of
medical home runs during this workshop, Nussbaum spoke of the importance of emphasizing care coordination, and its potential for preventing hospitalizations and cost savings. Identifying missed opportunities for prevention and improving the delivery of these preventive services could also yield significant savings by emphasizing wellness and preventing costly disease complications.
As a final point, Rother raised the specter of funding these investments. Offering the possibility of joint public-private ventures, he encouraged an honest assessment of feasible and sustainable sources of financing. He stated that identifying these sources and investing in the tools discussed during this workshop will be necessary to realizing the long-term savings and improved clinical outcomes achievable through their implementation.
America’s Health Insurance Plans. 2008. Ensuring quality through appropriate use of diagnostic imaging. Washington, DC: America’s Health Insurance Plans.
Beane, B., N. Gingrich, and J. Kerry. 2008 (October 24). How to take American health care from worst to first. New York Times.
Employee Benefit Research Institute. 2008. The 2008 health confidence survey: Rising costs continue to change the way Americans use the health care system. Washington, DC: Employee Benefit Research Institute.
IOM (Institute of Medicine). 2007. The Learning Healthcare system: Workshop summary. Washington, DC: The National Academies Press.
Oberlander, J. 2007. Learning from failure in health care reform. N Engl J Med 357(17): 1677-1679.
The White House. 2009. Fiscal Responsibility Summit, Februrary 23, 2009. http://www.whitehouse.gov/assets/blog/Fiscal_Responsibility_Summit_Report.pdf (accessed April 6, 2009).