The central theme of this book is that the U.S. political system is failing to ensure that medical treatments are based on evidence about what works best for particular conditions. Too many patients receive treatments that are less beneficial than other treatment options or even harmful. In many instances, we simply don’t know what treatments are most effective because rigorous studies have never been done. There are no mechanisms to ensure that gaps in the medical evidence are quickly addressed and that clinically significant questions are answered. Even when scientists determine that an existing treatment has little value, patients may continue to receive it. The integration of research into clinical practice occurs slowly. It can be hard to get doctors to abandon a treatment contradicted by evidence, especially if their peers use the treatment and it constitutes a large share of their practices.

The use of treatments not supported by strong evidence not only wastes resources. It also has real consequences for patients. Consider the treatment of angina, or stable chest pain. Each year, more than a half million heart patients worldwide undergo a procedure (percutaneous coronary intervention, or PCI) in which a doctor inserts a stent to open up clogged arteries. The procedure costs $11,000 to $41,000 in U.S. hospitals. One might think that the benefits of PCI would have been carefully studied before it entered clinical practice forty years ago. In fact, the first blinded, placebo-controlled trial of PCI was published only in 2018. The study (called ORBITA) found that PCI did not increase exercise tolerance (the primary end point) by more than the effect of a placebo procedure. (All patients in the trial received drugs like beta blockers before they were randomly assigned to receive a drug-eluting stent or a sham procedure.)1

The landmark ORBITA study offers a window into the dynamics we explore in this book. At a conference at Yale University in 2018, we brought together leading cardiologists to discuss ORBITA and how the medical community is responding to its findings.2 Dr. Darrel P. Francis, the study’s lead investigator, explained that peer reviewers blocked the study from receiving external grant funding, even though ethics boards were supportive. Interventional cardiologists regarded ORBITA as “an attack on our specialty,” he said. “Professional societies lashed out at the trial and the investigators with angry and inaccurate retorts.”3 According to Francis, professional societies raised a number of criticisms of ORBITA’s methodology that he maintains are invalid or exaggerated—that the study was too short, or focused on the wrong patients or end point, for example. Dr. William E. Boden, a Veterans Administration cardiologist, predicted that professional resistance to ORBITA’s findings will be strong. “Likely medical practice will not change in response to ORBITA,” he said.4 The negative reaction of interventional cardiologists to ORBITA is strikingly similar to the behavior of other specialty groups (see the case study of arthroscopic knee surgery in chapter 2) when studies have challenged the efficacy of treatments in their respective practice areas.

Unhealthy Politics shows that the coalition for evidence-based medicine is weak. It includes too few doctors and attracts too little energy and political entrepreneurship from policy makers. We explore evidence-based medicine not only to illuminate the performance of the U.S. health care sector, but also as a vehicle to investigate a basic question about democratic politics: Can government serve as a problem-solving institution in sectors, such as medicine, in which prestigious professional groups are a dominant actor? In our modern society, we depend on professionals—doctors, lawyers, engineers, accountants, architects, economists, and others—to manage a wide range of complex tasks. But what happens if professionals do not consistently and energetically use their training, expertise, and social influence to address demonstrated failures in the performance of the sectors in which they play a leading role? Can government catalyze change—or do political forces serve to perpetuate the status quo?

We delegate authority over medicine to physicians for a good reason. Physicians undergo specialized training and are much more knowledgeable than patients about how to treat illnesses. Even in an era when many of us search the Internet for our symptoms and therapeutic options, most people believe doctors know best. But the delegation of medical authority rests on a social contract. According to the American Board of Internal Medicine Foundation, this social contract encompasses many commitments: “placing the interests of patients above those of the physician,” “uphold[ing] scientific standards,” and “working collaboratively with other professionals to reduce medical error, increase patient safety, minimize overuse of health care resources, and optimize the outcomes of care.”5 Unhealthy Politics shows that when the medical profession fails to fulfil these commitments, the U.S. political system struggles to respond.

To be sure, efforts are being made to strengthen the evidence base of medicine.6 The most noteworthy public policy development of the past decade has been the creation of the Patient-Centered Outcomes Research Institute (PCORI) as part of the Affordable Care Act (ACA). PCORI’s mission is to help patients, physicians, payers, and policy makers make informed decisions by producing and disseminating comparative effectiveness research (CER) on medical services. While CER is a public good, the creation of PCORI sparked controversy during the ACA debate. Drug and medical device companies, along with many Republicans in Congress, feared the agency would ration care and interfere with the doctor-patient relationship. To allay these concerns, the ACA set up PCORI as an independent, nonprofit agency (not part of the Department of Health and Human Services) with a narrow mission. For example, it was prohibited from using study results to mandate coverage and reimbursement decisions or practice guidelines. In sum, Congress recognized that evidence about what treatments work best is often missing and created PCORI to encourage its production. At the same time, however, Congress was wary of giving PCORI regulatory powers or other tools to make it more likely that this information would influence delivery patterns and resource allocation decisions.

From the standpoint of bringing about the transformations required for an evidence-based health care system, PCORI faced three key challenges during its initial, ten-year authorization period: to overcome the controversy of its birth and escape early termination, to begin to have a meaningful impact on clinical practice, and to build a public reputation among key stakeholders for relevance. Unlike the Independent Payment Advisory Board (which Congress repealed in 2018) PCORI has passed the political survival test. Congress renewed PCORI for another ten years in December 2019 on a bipartisan basis as part of a year-end spending package. The renewal included only minor tweaks to PCORI’s mission, financing, and governance.7 Republicans’ support for PCORI has increased as they have gained confidence that public payers will not use its research to make coverage or reimbursement decisions.8 PCORI’s reauthorization also reflected support among medical schools and researchers who have received PCORI grants and wish to see this funding stream continue.

While PCORI has endured, it is still struggling to have a significant impact on clinical decisions and build a reputation among the public. Thus far, PCORI-funded studies have generated evidence on topics such as the value of daily self-monitoring for people with type 2 diabetes and the effects of more frequent versus less frequent screening for colorectal cancer, but its work has not yet had a major influence on the use of unproven or low-value treatments in most areas of clinical practice or on the trajectory of national health spending.9 PCORI has almost no public visibility, and many observers believe that CER has not yet had a significant impact on stakeholders’ decision making.10

Studies of the development of bureaucratic capacities at agencies like the Food and Drug Administration suggest that building an organizational reputation for relevance and vigilance takes time. It requires strategic leaders who can navigate a challenging political environment, finding ways to avoid creating a backlash among interest groups and esteemed professionals while simultaneously fashioning new rules, frameworks, and incentives that promote the public’s diffuse interest in effective governance.11 As bipartisan support for PCORI’s mission is consolidated and it develops a reputation for being an honest information broker, it may gain the support of key stakeholders, including not only researchers and patient advocacy groups but also physicians, payers, and policy makers. Without the incorporation of such critical stakeholders, PCORI will lack the political backing required to sponsor studies like ORBITA that investigate the value of popular treatments that have significant implications for patient welfare and national health care spending.

By far the most difficult long-term challenge will be to integrate the Medicare program into the evidence-based medicine reform project. Medicare is the largest federal health program, and its policies have a significant influence on the commercial insurance market. Thus far, Congress has been unwilling to grant Medicare the authority to deny coverage to therapies that are expensive but deliver marginal health benefits. There is growing interest among health experts associated with both parties to allow Medicare to take steps to incorporate considerations of value into policy decisions while preserving patients’ access to services, such as experimenting “with reference pricing—a practice under which insurers pay a single reference price for all treatments with similar therapeutic effect, allowing patients who want less-cost-effective treatments to pay any difference out of pocket.”12 As concern over Medicare’s financial sustainability grows, we expect to see policy makers’ attention turn to a discussion of the pros and cons of different cost-control approaches.

Ultimately, transitioning to an evidence-based health care delivery system will require changes not only in public policy and financial incentives, but also in the norms of the medical profession. Physicians and medical societies must embrace a more robust understanding of their professional responsibility to identify gaps in the evidence base, to work with researchers on studies to expeditiously fill those gaps, and to embrace research findings (whether they confirm or challenge existing clinical practices) as an opportunity to promote patient welfare. There are some signs that the culture of medicine is beginning to change, but many doctors and medical societies continue to adopt a defensive posture. Our analysis shows that if physicians were to accept leadership responsibility for ensuring that patients receive care supported by evidence, the public’s trust in physicians would allow the medical profession to have a large influence on public opinion and the policy debate. The data and concepts developed in Unhealthy Politics can help policy makers and advocates better understand the barriers to reform of the U.S. health care delivery system—and to use this understanding to develop strategies to overcome them.


1. Rasha Al-Lamee, David Thompson, Hakim-Moulay Dehbi, Sayan Sen, Kare Tang, John Davies, Thomas Keeble, et al., “Percutaneous Coronary Intervention in Stable Angina (ORBITA): A Double-Blind, Randomised Controlled Trial,” Lancet 391, no. 10115 (2018): 31–40.

2. See Alan Gerber and Eric M. Patashnik, “Introduction: How Do Physicians, Policymakers Respond When a Procedure Proves No Better Than a Placebo?,” Health Affairs (blog), October 9, 2018, doi: 10.1377/hblog20181001.464018.

3. Darrel P. Francis, “ORBITA: What It Means for Coronary Stenting in Stable Coronary Artery Disease,” Health Affairs (blog), October 9, 2018, doi: 10.1377/hblog20181003.333377.

4. William E. Boden, “Will ORBITA Change Clinical Practice?,” Health Affairs (blog), October 9, 2018, doi: 10.1377/hblog20181003.522540.

5. ABIM Foundation. American Board of Internal Medicine. “Medical Professionalism in the New Millennium: A Physician Charter,” Annals of Internal Medicine 136, no. 3 (2002): 243.

6. A development that bears watching is the growing influence of studies conducted by the Institute for Clinical and Economic Review (ICER), an independent, nonpartisan research organization. The ICER’s analyses of the clinical and economic value of prescription drugs have become a factor in drug pricing and coverage negotiations in some commercial plans and state Medicaid programs. See Denise Roland, “Obscure Model Puts a Price on Good Health—and Drives Down Drug Costs,” The Wall Street Journal, November 2, 2019, A1.

7. The modifications include a change in PCORI’s revenue sources (transfers from the Medicare Trust Funds to the PCORI Trust Fund were eliminated and general appropriations were increased); the option for the General Accountability Office (GAO) to add two more seats to PCORI’s board for payers/purchasers; and authorization for PCORI to consider the “full range of outcomes data” including “potential burdens and economic impacts of the utilization of medical treatments, items, and services on different stakeholders and decision-makers respectively.” See the media release, “Appropriations Package Reauthorizes PCORI For Ten Years!” Friends of PCORI Reauthorization, December 17, 2019,

8. Joyce Frieden, “Outcomes Research Group Likely to Be Reauthorized,” Medpage Today, November 1, 2018,

9. See the relatively modest examples (given the scale of inefficiency in the U.S. health care system) highlighted in a fact sheet that PCORI circulated during the debate over its renewal: “PCORI-Funded Studies: Delivering Value,”

10. See Jocelyn Kaiser, “Institute that Aims to Reshape Health Care Seeks Renewal,” Science 366, no. 6470 (2019): 1179-1180.

11. See Daniel Carpenter, Reputation and Power: Organizational Image and Pharmaceutical Regulation at the FDA (Princeton, NJ: Princeton University Press, 2014).

12. Nicholas Bagley, Amitabh Chandra, and Austin Frakt, Correcting Signals for Innovation in Health Care (Washington, DC: Hamilton Institute, 2015), 6.


This book has had a long gestation. It is a pleasure to express our appreciation to the many individuals who have supported our work.

We are especially grateful to the scholars who read all or part of the manuscript, including Alan Cohen, Carolyn Engelhard, David Mechanic, Jonathan Oberlander, Mark Schlesinger, and Jon Skinner, as well as to two readers for Princeton University Press (one of whom, Ann Keller, revealed her identity to us after the manuscript was accepted). Their detailed comments, incisive criticisms, and thoughtful suggestions improved the book immeasurably. We also learned a tremendous amount from interviews, conversations, written feedback, and e-mail exchanges with many other scholars, experts, and policy makers, including Tom Allen, John Appleby, Carol Ashton, Eugene Bardach, Shawn Bishop, Michael Cannon, Dan Carpenter, Carolyn Clancy, Paul Cleary, David Cutler, Ara Darzi, John Ellwood, Sherry Glied, Phyllis Greenberger, Colleen Grogan, Michael Gusmano, Jacob Hacker, Ruth Hanft, Jennifer Hochschild, Karen Ignagni, Gary Kupfer, Miriam Laugesen, Julian Le Grand, Frank Levy, Peter Littlejohns, Rick Mayes, Etienne Minvielle, Bruce Moseley, Jonathan Nichols, Bernard Patashnik, Mark Peterson, Harold Pollack, Sir Michael Rawlins, Lise Rochaix, Jill Rutter, Herman Schwartz, Topher Spiro, Kathy Swartz, Peter Swenson, Gail Wilensky, and Nelda Wray. We apologize to anyone we have forgotten to thank and wish to stress that we alone are responsible for any errors of fact or interpretation.

Our work has been supported by talented research assistants. In a class by herself is Catlan Reardon, now a doctoral student at UC Berkeley. Other graduate students and postdoctoral fellows who contributed to this project include Daniel Biggers, David Hendry, Seth Hill, and Maya Mahin at Yale University; Victoria Catanese, Kenneth Lowande, and Jesse Rhodes at the University of Virginia; Rachel Lowenstein and Aaron Weinstein at Brown University; and Nichole Gligor at the University of Mississippi. The public opinion studies presented in this book draw on several journal articles that were coauthored with David Doherty. Thank you, David, for your major contribution to this project.

One of the benefits of a long-term research project is the opportunity to share preliminary findings and arguments with diverse audiences. Especially helpful were the constructive feedback and insightful criticisms we received from people who attended workshops and symposia held at or organized by the NYU Wagner Graduate School of Public Service, Harvard University RWJF Scholars Program, Virginia Commonwealth University, University of Minnesota, University of Washington, University of Virginia School of Medicine, the Miller Center of Public Affairs, Senior Statesmen of Virginia, Institute of Governmental Studies at UC Berkeley, University of Wisconsin, University of Chicago Center on Health Administration Studies, Dartmouth Institute for Health Policy and Clinical Practice, Yale School of Medicine, and Brown University. We also benefited from presentations on panels at the American Political Science Association and the Association for Public Policy Analysis and Management.

We also wish to thank colleagues for their personal, professional, and administrative support at various stages, including Jennifer Costanza, Howard Goodkin, Pam Greene, Jed Horwitt, Greg Huber, Pam Lamonaca, Dr. Heinz I. Lippmann, Rick Locke, Jim Morone, and Carrie Nordlund. We thank the Albemarle County Medical Society for its assistance in helping us to field a pilot survey of physician beliefs.

For financial support, we extend our profound thanks to the Robert Wood Johnson Foundation Investigator’s Award in Health Policy Research and to the Smith Richardson Foundation. We are grateful to Alan Cohen and Mark Steinmeyer for their patience, guidance, and encouragement during the completion of this project. We also are grateful for financial support from the Bankard Fund for Political Economy at the University of Virginia and the Institution for Social and Policy Studies (ISPS) at Yale University.

We also wish to express our gratitude to the following academic institutions: the Frank Batten School of Leadership and Public Policy at the University of Virginia, the Watson Institute for International and Public Affairs at Brown University, ISPS and the Center for the Study of American Politics at Yale University, and the College of Liberal Arts at the University of Mississippi. We especially thank David Breneman, John Bruce, Jacob Hacker, Harry Harding, Chris Ruhm, Wendy Schiller, Ed Steinfeld, and Craig Volden for creating stimulating and supportive environments in which to conduct our writing and research.

At Princeton University Press, we thank the two outstanding senior editors we worked with. Charles Myers helped us with the project’s initial framing. And Eric Crahan, our primary editor during the book’s writing and completion, has been a consistent source of insight, wisdom, and sound advice. We thank Ellen Foos for managing production, Kathleen Kageff for copyediting, and Steven Moore for preparing the index.

An earlier version of chapter 2 appeared as Alan S. Gerber and Eric M. Patashnik, “Sham Surgery: The Problem of Inadequate Medical Evidence,” in Promoting the General Welfare: New Perspectives on Government Performance, edited by Alan S. Gerber and Eric M. Patashnik (Washington, DC: Brookings Institution Press, 2006): 43–73. Chapter 3 draws on earlier versions of the following articles: Alan S. Gerber, Eric M. Patashnik, David Doherty, and Conor Dowling, “Doctor Knows Best: Physician Endorsements, Public Opinion, and the Politics of Comparative Effectiveness Research,” Journal of Health Politics, Policy and Law 39 (1): 171–208; Alan S. Gerber, Eric M. Patashnik, David Doherty, and Conor Dowling, “The Public Wants Information, Not Board Mandates, From Comparative Effectiveness Research,” Health Affairs 29 (10): 1872–81; Alan S. Gerber, Eric M. Patashnik, David Doherty, and Conor Dowling, “A National Survey Reveals Public Skepticism about Research-Based Treatment Guidelines,” Health Affairs 29 (10): 1882–84. Chapters 5 and 6 draw on earlier versions of Alan S. Gerber and Eric M. Patashnik, “Problem Solving in a Polarized Age: Comparative Effectiveness Research and the Politicization of Evidence-Based Medicine,” Forum 8 (1): Article 3; Alan S. Gerber and Eric M. Patashnik, “The Politicization of Evidence-Based Medicine: The Limits of Pragmatic Problem Solving in an Era of Polarization,” California Journal of Politics and Policy 3 (4): Article 4.

Finally, we wish to thank our family members for their love and support: Debbie Gordon and Michael and Josh Patashnik; Rachel, Eva, Miriam, Ben, and Helen Gerber; and Carey, Larkin, and Anorah Bernini Dowling.


In 2002, the New England Journal of Medicine reported that a common operation, performed on millions of Americans suffering from osteoarthritis of the knee, “worked no better than a sham procedure in which patients were sedated” while a surgeon merely “pretended to operate.”1 The patients who underwent the real surgery got better. They had less pain and could climb stairs more easily. But the patients who received the fake surgery experienced just as much pain relief and improvement in joint function as those who had undergone the real operation. In sum, the benefits of the procedure were a product of the placebo effect. Leading experts stated that 80–90 percent of these procedures should not be done.2

We were stunned when we learned about the sham surgery study. We assumed that hard evidence must have existed for the knee operation’s medical benefits. To our surprise, it did not. We carefully reviewed the medical literature and found that claims for the efficacy of the procedure rested on studies with weak research designs and that the theory behind the surgery was speculative at best.3 Yet the dearth of evidence and lack of an accepted causal mechanism to explain the procedure’s purported effects did not prevent the operation from diffusing widely into practice. “There’s a pretty good-sized industry out there that is performing this surgery,” Dr. David T. Felson of Boston University said. “It constitutes a good part of the livelihood of some orthopedic surgeons. That is a reality.”4

The knee surgery case is not an aberration. Some experts believe that less than half the medical care in the United States is based on or supported by evidence of its effectiveness.5 As leading health services experts Carol M. Ashton and Nelda P. Wray write,

The goal of every treatment is to make the patient’s outcome better than it would have been without any intervention. Because of advances in research in the past six or so decades, clinical scientists are able to estimate with considerable precision whether a particular intervention will lead to net benefit over harm in groups of individuals possessing certain characteristics. That said, much of clinical practice lacks a supporting evidence base, and what research evidence exists is predominately of poor quality.6

Procedures and tests are regularly prescribed on the basis of limited scientific information.7 Once doctors decide that a particular treatment “works,” it can become “locked in.” Randomized controlled trials—the “gold standard” for determining the effects of a treatment—are almost impossible to carry out on a treatment that has already diffused into practice.8 It is common for patients in one geographic area or region to receive different medical interventions than patients with the exact same condition in another part of the country, without a reliable mechanism to learn which approaches are best and translate this discovery to patients and clinicians.9 And when solid scientific evidence does emerge about the benefits and risks of a treatment, the informational uptake often proceeds slowly. Studies have found that more than a decade can pass before the evidence alters clinical practice.10

It is unsettling to find out that widely used treatments and tests rest on little or no evidence. Even more troubling is learning that this situation is an open secret among health care experts. Recognition of the “medical guesswork” problem (together with growing awareness of geographic variation in utilization and medical spending) has prompted calls for rationalization of health care delivery for decades. These issues have been the subject of government reports, articles in leading newspapers, and cover stories in popular magazines, such as Businessweek.11

To its credit, the Obama administration—along with prominent Republican health policy experts—recognized the benefits for patients, payers, and providers of moving toward a more evidence-based medical system. The Affordable Care Act (ACA; “Obamacare”) launched a new, independent, nongovernment entity—the Patient-Centered Outcomes Research Institute (PCORI)—to fund and disseminate research on the comparative effectiveness of different interventions to prevent, diagnose, treat, and monitor health conditions. Comparative effectiveness research (CER) compares two or more health care interventions, such as a drug, diagnostic test, or surgical procedure, to determine which interventions work best for which patients. CER is distinct from cost-effectiveness analysis, which examines medical interventions through an economic lens. Cost-effective analysis asks how much an intervention costs, and whether the outcomes it produces are worth its cost.12

The overall mission of PCORI is to help people make informed health care decisions, and improve outcomes, by “producing and promoting high-integrity, evidence-based information that comes from research guided by patients, caregivers, and the broader healthcare community.”13 Following a contentious debate in which critics charged that CER would lead to rationing and “death panels,” Congress established a narrow medical research program that left existing patterns of therapeutic authority and health care financing largely untouched. In contrast to publicly funded comparative effectiveness and health technology assessment entities found in other advanced democracies, including Australia, Germany, France, and the United Kingdom, which have linkages to policy-making bodies, PCORI’s research findings may not include “practice guidelines, coverage recommendations, payment, or policy recommendations.”14 In addition, PCORI’s research is not required to consider whether an intervention is cost effective.15

Despite its narrow research mission, PCORI has been mired in controversy. While PCORI won reauthorization in 2019, its influence may remain modest. If PCORI has little impact on clinical practice, it will follow a long line of unsuccessful reform initiatives. Past federal efforts to promote evidence-based practices through medical research and clinical guideline development have crumbled under pressure from doctors, drug companies, and the medical device industry.16 Yet even if the regulatory protections and insurance subsidies under the Affordable Care Act are eroded over time and the United States shifts to greater reliance on an individualized system of health savings accounts, there will still be a pressing need for public funding of CER. Health markets cannot function efficiently if physicians and patients lack reliable information about the comparative effectiveness of treatment options, and the market cannot be counted on to generate the optimal level of this information without public subsidy. In sum, government has a role to play in generating evidence about what works in medicine irrespective of the organization and financing of the insurance system. We discuss the future prospects of the evidence-based medicine movement in the concluding chapter. For now, it is fair to say that PCORI has not had a major impact. The agency did not transform the everyday practice of medicine, build a durable base of political support, or solve the systemic problems we describe. All this is unsurprising given PCORI’s limited power and the restrictions placed on Medicare to not consider cost and cost-effectiveness in its coverage decisions, but it is a failure of governance just the same.

Why Evidence-Based Medicine Is Important

The sluggish incorporation of medical evidence into clinical practice is a concern for three key reasons—safety, quality, and the efficiency of resource allocation. First, the delivery of unproven care can expose patients to serious risks. For example, each year 100,000 Americans undergo a procedure called vertebroplasty, in which their collapsed vertebras are filled with bone cement, even though according to two New England Journal of Medicine studies there is no clear answer to the question of whether the procedure works better than physical therapy, injections of local anesthetic, or simply just waiting to heal.17 Second, the slow integration of evidence can lead to suboptimal outcomes for patients who receive treatments that work less well for their conditions than alternatives. Third, the failure to implement evidence-based practices encourages wasteful spending, causing the health care system to underperform relative to its level of investment. While the drive to root out inefficiency is unlikely to stir the passions of ordinary citizens, it is a key to a health care system’s long-run performance.

Evidence-based medicine is also related to cost containment, but the two are not synonymous. First, better use of evidence can and should sometimes lead to more spending, not less. A 2003 Rand Institute study found that Americans receive just half of the recommended care for a range of medical conditions.18 Also, the delivery of low-value care, while certainly wasteful and expensive, is not the only reason why the United States spends far more on health care than do other advanced nations. The United States also has much higher service prices and administrative costs than do other rich nations.19

Yet the utilization of services—that is, the amount of care delivered to patients—is still important to cost containment, especially in the Medicare program. There are two basic strategies for taming health care costs—limit how much providers can charge (control prices) or reduce the amount of low-value care (control utilization). Researchers at Dartmouth College, following the pioneering work by John E. Wennberg and Alan Gittelsohn on geographical variations in health care delivery,20 have found more than a twofold variation in per capita Medicare spending in different regions of the country. The main driver of these regional differences in Medicare spending is not differences in poverty rates, the relative illness of patients, or differences in how much Medicare pays for services, but rather variation in utilization.21 Hospitals that give more tests and treatments to their Medicare patients do not achieve consistently better outcomes than those that deliver fewer services.22 However, spending more on effective treatments does produce better results, and higher quality hospitals do tend to have higher market shares and expand more over time.23 Reducing the overuse of low-value treatments by patients over 65 could help control Medicare spending growth and federal budget deficits.

The situation is different for patients outside Medicare who obtain insurance from their employers. While Medicare uses its authority to regulate prices for hospitals, employer-sponsored plans generally pay more variable (and higher) prices than the government does. For example, hospital prices for lower-limb MRIs vary by a factor of twelve across the nation.24 In under- 65 commercial markets, prices, not utilization, are thus the driver of variations in spending levels across hospital regions.25

Yet there are still huge variations in utilization in under-65 private insurance markets.26 Indeed, the correlation between what doctors do for their under-65 patients and their elderly Medicare patients is remarkably high. That is, regional utilization patterns in Medicare provide a strong predictor of utilization rates in the private insurance population.27 To be sure, reducing unwarranted variation in treatment decisions by promoting evidence-based practices would not by itself solve the overall cost problems in American health care. It would, however, significantly improve the quality, safety and efficiency of service delivery. This is too huge an opportunity to miss.

Our Approach

Our argument blends policy analysis and political science. Our comparative advantage is to use political science methods and concepts to generate fresh answers to questions that mostly have occupied economists and health services researchers. We seek to identify the forces that sustain a political equilibrium characterized by widespread utilization of tests and treatments that are unproven and possibly useless, or, at best, produce low value for the money spent on them. In sum, we identify, analyze, and document a major national problem and ask why the solutions to this problem are not being designed, advocated for, and vigorously implemented.

The U.S. health care system is a vast topic. If we attempt to explore every facet of institutional performance, and investigate every actor who participates in the production, delivery, financing, and regulation of U.S. medical care, we will lose the forest for the trees. To keep our eyes on the fundamentals, our analysis focuses on a single performance indicator: whether credible scientific information is available on the comparative clinical effectiveness of alternative interventions, meaning evidence on what tests, treatments, and procedures work best for groups of patients with different conditions, and whether such information informs the decisions of patients, providers, and payers. In sum, our concern is whether the delivery of medical care in the United States is evidence based.

Evidence-based medicine (EBM) has been defined as “the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients.”28 The aim of the EBM movement is to “evaluate the safety, effectiveness, and cost of medical practices using tools from science and social science and to base clinical practice on such knowledge.”29 There is sufficient consensus among medical experts to make the implementation of EBM a reasonable baseline expectation. As the National Academy of Medicine (formerly the Institute of Medicine) summarizes its work on developing the infrastructure required for CER, “Evidence is the cornerstone of a high-performing healthcare system.”30

To be sure, there are limits to the extent to which evidence can guide the activities of actors within the health care system. There are large uncertainties in medicine, and, as we discuss, the quest for applying standardized forms of evidence to individual treatment circumstances is viewed with great suspicion by both clinicians and the public. They are not always wrong to be skeptical. Sometimes industry funding compromises the objectivity of medical studies, and research questions can be distorted by unexamined cultural biases.31 Even if the scientific quality of research studies were assured and the production of medical evidence were much enhanced, there would still be important limits to the application of generalized rules to particular circumstances. A great deal of medical care consists of patients insisting on help when the correct pathways are not obvious.32 In many instances, patients have different reactions to drugs or other treatments in the same class (for example, psychiatric or hypertensive drugs), and the treatment that fares best on average is not necessarily the best for the individual patient.33 In short, there is tremendous value to having physicians who know their patients well, even if (as we argue) physician discretion is commonly exercised in ways that lack scientific grounding.34

In sum, we are not simplistically advocating for rule-following behavior. As Mark Schlesinger and Bradford H. Gray argue, clinical practice needs to strike a careful balance between generalized expertise (understanding the evidence generated though scientific research) and particularized expertise (applying that evidence to individual patients).35 Our starting point, however, is that by systematically ignoring scientific evidence (or the lack thereof), the United States is substantially out of balance.36 As we discuss in chapter 1, a significant fraction of the medicine that Americans consume is based on minimal scientific evidence about its comparative effectiveness. Many treatments offer only minor benefit over alternatives—and some are useless. As physician Atul Gawande wrote in the New Yorker: “Millions of people are receiving drugs that aren’t helping them, operations that aren’t going to make them better, and scans and tests that do nothing beneficial for them, and often cause harm.”37 In a national survey of primary care physicians, nearly one-half said their patients received too much medical care.38 Overutilization does not apply to all Americans, and some patients are clearly harmed by receiving too few effective treatments.39 By all accounts, the failure of physicians to practice evidence-based medicine is a serious problem in the United States, as it is in many other nations.

By having prevalent treatments that are inefficacious or even cause harm, patients are diverted away from other treatments that are actually effective. The United States can certainly “afford” to spend a fifth or more of national income on health care if its citizens wish to do so—but there are trade-offs. Spending money on useless (or low-value) treatments reduces the money available to spend on effective cures,40 or on valuable social investments in education, worker training, and infrastructure. If the nation does not evaluate the comparative effectiveness of treatment alternatives to help reduce waste and improve quality, and if such evidence does not affect clinical or policy decisions, it suggests that the defects of our institutions are first-order concerns, not minor flaws in an otherwise high-performing system.

The failure of the American political system to represent the diffuse public interest in the generation and use of medical evidence is a political economy puzzle. Given the high level of U.S. health care spending, and the desire of patients to receive the best therapies, we must understand how a democratic political system can produce inefficient outcomes year in and year out without triggering an effective response. While existing research provides many insights, new perspectives are needed.

The political scientist James Q. Wilson noted that policies with diffuse benefits and concentrated costs are unlikely to be passed without the presence of a “political entrepreneur” who can mobilize latent public support.41 A policy to promote the generation and use of medical evidence to improve clinical practice and resource allocation decisions is clearly an instance of diffuse benefits and concentrated costs: the public (and taxpayers) would gain from the policy, and drug companies and medical device makers who profit from expensive but ineffective products would lose. Our work builds on Wilson’s framework in two ways. First, using a survey experiment, we demonstrate the risks to members of Congress who would consider becoming a political entrepreneur in this area, showing that there are few electoral benefits to be gained. Second, we focus attention on the important mediating role of the medical profession. The professions do not play a prominent role in Wilson’s theory of general-interest policy making, but we believe they should be on center stage. The Progressive tradition in America has long believed in the positive role of the professions to bring scientific knowledge and expertise to bear on societal problems and improve all collective endeavors.42 One would hope that the concentrated interest of the medical profession in preserving its prestige and autonomy would make it a natural supporter of evidence-based medicine both to check the influence of concentrated interests such as insurers, hospitals, and drug companies and to preserve doctors’ hard-earned reputation as uniquely trusted guardians of patient welfare. But as we show in chapter 2, many medical societies support evidence-based medicine in the abstract but oppose research that challenges treatments in their practice areas.

Every policy system consists of pressures pushing to maintain and challenge the status quo,43 but there is no guarantee that the two sets of forces will be equally matched. In U.S. health policy, the forces supporting overutilization, unwarranted variation in utilization, and the suboptimal use of medical evidence to improve quality and efficiency appear as strong or stronger than the forces pushing for reform. One important force maintaining the status quo is the political influence of drug and medical device companies.

The pharmaceutical and health products industry is consistently near the top when it comes to federal campaign contributions. During the 2014 election cycle, for example, the Pfizer company contributed over $1.5 million to federal candidates, Amgen gave more than $1.3 million, and McKesson more than $1.1 million. The industry trade group Pharmaceutical Research and Manufacturers of America (PhRMA) spent over $16 million on lobbying in 2014.44 The industry’s goals include ensuring quicker approval of drugs and products entering the market. As we detail in chapter 6, the medical products industry’s influence affected the organizational design of the PCORI. Recently, a top industry priority has been passage of the 21st Century Cures Act, which (besides increasing funding for research on cancer and other diseases) permits manufacturers to submit real-world evidence to the FDA for approval of new drugs and devices and for new indications for existing products, including observational data arising from routine clinical use, rather than data from randomized controlled trials.45 The intention is to expedite the product approval process, but some observers point out that changing the FDA’s traditional standards of evidence may have “an unpredictable long-term effect on drug safety and efficacy.”46 Finally, pharmaceutical and health technology companies not only seek to influence the regulatory environment, but they also spend billions of dollars annually on marketing to both doctors and consumers. The pharmaceutical industry is a major presence at the meetings of many medical associations, and drug company representatives often seek to build personal relationships with physicians to influence clinical decisions.47

In short, there is tremendous pressure from powerful economic actors to maintain the health care status quo. Eliminating a dollar of waste in the health care system usually means reducing someone’s income.48 The distinctive values of the United States constitute a second force for conservatism in the health care system. As medical ethicist Daniel Callahan writes, “American health care is radically American: individualistic, scientifically ambitious, market intoxicated, suspicious of government, and profit-driven.”49 These values contribute to a political unease with explicit limits on the consumption or supply of medical technology.

Our focus in this book is not on the deep-seated economic and cultural forces sustaining the status quo. Rather, we explore why there is relatively little opposing pressure for meaningful reform to promote quality and evidence-based medicine, for it is the relative weakness of countervailing reform pressures, together with the strength of existing norms and practices, that generates the inefficient outcomes that patients, payers, and providers actually experience. To be sure, any effort to explain a “dog that doesn’t bark” must be somewhat speculative. By drawing on political science models, surveys, survey experiments, case studies, and analysis of policy history, however, we hope to generate insights into why this “unhealthy politics” persists.

Rather than provide a full account of the gap between what needs to be done and what is actually happening, we investigate the incentives and behavior of just three sets of actors: physicians, politicians, and the public. Our objective is to explain how these actors interact to produce a stable system that generates large amounts of waste as a by-product, and more generally, to advance an understanding of the conditions under which a democratic polity can sustain grossly inefficient sectors seemingly indefinitely.

Our approach is somewhat selective and incomplete. As important as it is, we do not provide a detailed analysis of the antireform (medical products industry) side of the equation. Instead, we focus on how political incentives conspire to keep proreform forces relatively weak. While the importance of physicians, politicians, and the public has been recognized in existing scholarship,50 the desire of many scholars to offer comprehensive accounts of health politics and policy has sometimes caused them to get lost in the shuffle. Our distinctive contribution is to place the actions and interaction of these actors on center stage, and to explain why reform pressures can be anemic even when the potential for social progress is great.51

A Preview of Our Argument

This book explores the roles of physicians, the public, and politicians in the nation’s failure to take the steps necessary to promote a high-performing, evidence-based health delivery system. We focus on doctors because their beliefs always mediate (and often determine) what treatments patients receive, because the system cannot be reformed without their leadership, and because the scope and implications of the medical profession’s power, despite important research on the topic,52 has not been fully explored in the literature. We focus on the public because improving patient welfare is the objective of the medical system and because ordinary citizens influence Medicare policy making through their roles as taxpayers and voters. Finally, we examine the role of elected officials because Medicare and other government health programs shape the overall context in which medical services are funded and delivered, and government can support EBM through rules or incentives.

The delegation of authority to the medical profession rests on an implicit social contract: Doctors as a profession receive the “privilege of selfregulation” and financial rewards on the expectation that they will serve the health needs of individual patients and society.53 If the social contract between the medical profession and society is malfunctioning, politicians need to repair it. But can they do so? Unhealthy Politics investigates how political forces undermine the reform of medical governance.54


There is no doubt that American physicians today face much stronger external pressures than they did in the past. Their clinical decisions are routinely reviewed and sometimes questioned by third parties, and they have to justify their actions with a lot more paperwork. It is undoubtedly more stressful to be the typical doctor today than it was to be a typical physician in the 1950s.

Nonetheless, the argument that the political and economic power of physicians has eroded is arguably overstated.55 Most U.S. doctors enjoy a relatively high degree of professional autonomy in practice as a result of the dilution of strong utilization management controls following the backlash to managed care.56 Although insurers today may require prior authorization for certain therapies, they typically defer to physicians’ judgments about what treatments are medically necessary, even if the treatments are not supported by rigorous evidence.57 Off-label usage of treatments is commonplace. Moreover, efforts by insurers to deny reimbursement of questionable or unproven interventions have sometimes been overturned by courts.58 As we discuss later in the book, the views of doctors and medical societies are prominent when public controversies arise over the interpretation of research studies indicating that tests or treatments are not as effective as believed. And, as health policy scholar Miriam Laugesen observes, doctors retain enormous influence in technical, low-visibility venues, such as in the panels that make reimbursement and coverage decisions under the Medicare program.59

Why does society delegate so much authority to the medical profession? One answer is that it represents a sensible response to “market failure.” Knowledge about treatment effects is esoteric and asymmetrically distributed between doctors and patients.60 It is often impossible for the patient (even in the Internet age) to know what treatments he or she needs. Society is better off by hiring well-trained experts to prescribe therapies and shape payment rates. The sociologist Talcott Parsons held the professions in high regard, characterizing the autonomy and privileges of the professions as a “functional exchange in which society receives, in exchange, the technical competence it needs to achieve critical ends.”61 The “social trustee” model of the professions, however, has been strongly challenged.62 In his important book Profession of Medicine, Eliot Freidson argues that the social authority of organized medicine is in fact a “political commodity.”63 It was not earned as a reward for social responsibility, but rather won through political contestation and negotiation.64 How physicians use their autonomy and privileges is thus a key issue for democratic accountability.

Some physicians and health care experts argue that the implied social contract is that physicians not only will be competent technical experts, but will also ensure an appropriate distribution of finite medical resources.65 As the American Board of Internal Medicine (ABIM) states in its physician charter,

Professionalism is the basis of medicine’s contract with society…. While meeting the needs of individual patients, physicians are required to provide health care that is based on the wise and cost effective management of limited clinical resources. They should be committed to working with other physicians, hospitals, and payers to develop guidelines for cost effective care. The physician’s professional responsibility for appropriate allocation of resources requires scrupulous avoidance of superfluous tests and procedures. The provision of unnecessary services not only exposes one’s patients to avoidable harm and expense but also diminishes the resources available for others.66

However, there is always a risk that professional associations, from lawyers to professors, will begin to defend the interests of its members in addition to serving the public. The concern is that governing institutions may lack the monitoring and enforcement capacity to prevent the medical profession from abusing its delegated authority to serve its own interests. If doctors do not follow evidence-based guidelines or if medical societies use their power to discredit credible studies demonstrating that a particular treatment is not as effective as advertised, medical professionalism and self-regulation becomes a myth.

The design of the Medicare Act of 1965—the foundational statute of the U.S. health care system—crystallizes the dilemma. Medicare was created in the teeth of fierce opposition from the AMA and other medical societies.67 As political scientist Mark A. Peterson observes, “Even in ‘defeat,’ however, the physicians were influential enough to dictate … the method of reimbursement and administration under the program, demand enhancing policies and subsidies and constraints on potentially competing providers.”68

The nation essentially committed to funding seniors’ care without government scrutiny of the details and appropriateness of medical practices, which were left to physicians to determine. Medicare is permitted to exclude coverage for care that is “not reasonable and necessary for the diagnosis or treatment of illness or injury.” Attempts by Medicare administrators to read this as a requirement for cost-effectiveness have generated intense opposition from provider groups and ended in stalemate.69 In practice, billions of dollars of tax revenue flow into Medicare without the technology supply constraints or drug price controls common in European nations. The program generated vested interests and dense networks of organizations that developed material stakes in the maintenance of existing arrangements.70 As University of Michigan law professor Nicholas Bagley writes,

The basic contours of the [Medicare] program—public financing, private care—were fixed in 1965. Beneficiaries grew accustomed to subsidized coverage without meaningful restrictions, and physicians, hospitals, and other providers committed themselves to the new world order in which the government would pay the bills but assert no control.71

To be sure, Congress has taken some steps to control Medicare spending by reducing payments to providers.72 For example, Congress has adopted a prospective payment schedule for hospitals and a fee schedule for physicians.73 Since 1965, however, Congress has avoided efforts to control the volume or intensity of services, creating incentives for overutilization.74 The ACA continues this long-standing pattern. While the ACA contained a number of new financial reforms, such as accountable care organizations and bundled payments, it steered clear of reconfiguring medical governance. Medicare’s failure to incorporate a cost-effectiveness—or even comparative clinical effectiveness—requirement has far-reaching implications, as many commercial insurers follow Medicare’s coverage policies.75 In sum, policy makers have layered moderate cost-control measures atop, but have not eliminated, the preexisting Medicare governance framework.76

Doctors remain highly influential actors in the American health system despite far-reaching changes in the economic and organizational context in which medicine is practiced; their individual and collective authority over clinical decisions, as well as resource allocation, by no means goes unchallenged, but it remains quite significant. This is not to deny that there are influential reform forces within the medical profession. Like progressive reformers at the turn of the twentieth century who sought to expose the chaos, inefficiency, and corruption of medical practices and promote improvements in medical education and scientific therapeutics,77 some medical societies, leading doctors and public intellectuals such as Atul Gawande and Jerry Avorn of Harvard Medical School, among many others, are seeking to raise awareness of industry influence and the failure of many doctors to follow best medical practices. A genuine EBM movement exists today. Nonetheless, progress has been slow, uneven, and reversible. In sum, physicians still hold a position of trust and authority, even as concerns about the scientific foundation, quality, and waste of medical services have mounted.


The public is the second key actor in our analysis. Although Americans’ confidence in the medical profession has fallen over time—in tandem with the decline in public confidence in most public institutions since the 1960s— doctors remain a highly trusted group within the context of American politics. People regard their doctors as knowledgeable and trustworthy agents of their own medical care and welfare, and this trust extends to matters of health policy as well.78

The catch, as the distinguished medical sociologist David Mechanic writes, is that “patients may trust blindly when some skepticism is warranted.”79 There are two potential sources of public misperceptions that can lead to excessive faith in physicians. First, the public may fail to recognize that medical societies are, at bottom, trade associations. If viewed as trade associations, the primary function of medical societies would be to protect the autonomy and advance the interests of their members. One can certainly find heartening examples of efforts by medical societies to improve the quality and cost-effectiveness of care. For example, several national medical societies have recently launched a campaign called “Choosing Wisely,” which highlights diagnostic tests, procedures, and treatments in their respective specialty areas that do patients little good.80 However, medical societies participating in this effort are mostly choosing low-hanging fruit. For example, the American Academy of Orthopaedic Surgeons includes an over-the-counter supplement but no major procedures on its list.81 The initiative also lacks an enforcement mechanism; doctors remain free to ignore the recommendations if they wish. A JAMA Internal Medicine study suggests that the initiative has had disappointing results thus far.82 In short, the record of medical societies is so replete with failures of self-governance that it is reasonable to raise doubts about the centrality of their commitment to eliminating ineffective or low-value care.

Back in the 1990s, for example, the Agency for Health Care Policy and Research (AHCPR) was launched to provide evidence-based clinical guidelines that would help physicians determine which treatments are most effective. When the agency found that there was little objective evidence to support surgery as a first-line treatment for lower back pain, however, back surgeon societies complained vehemently to members of Congress. As Bradford H. Gray, Michael K. Gusmano, and Sarah R. Collins observe in an insightful study of the politics of health services research, the agency’s budget was slashed, and its authority to make policy recommendations was curtailed.83 The drawing of negative messages from the AHCPR experience helps explain PCORI’s limited grant of authority. Many similar examples of medical societies reacting strongly to negative study results and being unwilling to change in the face of evidence could be cited.84 These medical society activities fall below the radar screen for most citizens. The psychological bonds that ordinary citizens develop with their personal doctors inhibit the public from recognizing the governance role that the medical profession plays in the health system. There is a tendency for people to assume that an organization’s collective behavior can be predicted on the basis of the character or behavior of the people they know who belong to the organization, and in our personal experience many doctors are truly extraordinary individuals.

The second misperception of the medical profession is subtler. The public may not perceive the existence of gaps between, on the one hand, the medical profession’s evolving responsibilities to patients and society, and, on the other, its willingness and capacity to fulfill those responsibilities. Such gaps may open whenever changes in the profession’s behavior and orientation fail to keep pace with changes in the wider context in which medical care is delivered. Over the past half century, there has been an explosion of scientific knowledge about health and disease, and health spending has grown dramatically as a share of the economy. These twin developments create new professional obligations for doctors under the terms of the evolving social contract. As David Blumenthal, M.D., stated in his address to the 2014 Columbia University College of Physicians and Surgeons graduation ceremony, these developments challenge doctors to “stay informed and current” as well as to “husband health care resources—to become a steward of the health care dollar—in ways never before required.”85 But it is not clear that the medical profession is developing the capacity to respond effectively to these scientific and economic challenges. Most doctors have not received extensive training in the modern research methods or the data analysis skills needed to keep up with scientific literatures on the relative benefits, costs, and risks of treatments. Yet without such training, it is difficult for doctors to make recommendations about what is best for patients. And despite concerns about the rising cost of both private insurance and public programs like Medicare, a recent physician survey found that only about a third of physicians believe they bear a major responsibility to control health care costs.86 (Doctors blamed rising health care costs on lawyers, insurance companies, drug and device manufacturers, hospitals, and patients.) This is a denial of the profession’s collective responsibility to society, but one the public—reasoning from their positive encounters with their personal doctors—may not perceive.

In sum, previous research has argued that the public shifts its individualized trust in doctors to trust in medical societies, which has allowed the medical profession to block the drive for a universal government insurance program during the postwar era and obtain favorable financial terms when Medicare finally passed in 1965.87 Unhealthy Politics considers the possibility that the consequences of Americans’ trust in the medical profession are broader and even more concerning. Despite their position as a repository of public trust in a complicated policy area, doctors and their professional societies have not consistently used their authority, standing, and prestige to promote the steps necessary to root out waste, bad science, and inefficiencies in the health care system—and too often have used their political capital to fight these steps.


Finally, we explore how politicians themselves have contributed to the persistence of the medical evidence problem. Our analysis has implications not only for an understanding of the roots of the resistance to the implementation of evidence-based medicine, but also for a more general understanding of how inefficient practices can persist in the American democratic polity. Our argument focuses on political failures at two stages: agenda setting and decision making.

The American political system is often claimed to have properties of selfcorrection. When governance veers dramatically off course, and the preferences of citizens are not being met, opportunities may arise for creative political entrepreneurs to frame problems, develop solutions, and “sell” their ideas to the public—to capture a political reward.88 Yet there is no guarantee that the need for political entrepreneurship will generate its own supply. The medical evidence problem has prompted only limited investments of political entrepreneurship from members of Congress, despite the substantial costs the problem imposes on society. If the supply of entrepreneurial problem solving is too low even in a sector as important and salient as health care, it is likely to be inadequate in other sectors as well.

We develop a theory of “zero-credit politics” to explain the undersupply of political entrepreneurship targeted at promoting diffuse interests.89 The incentive for lawmakers to engage in entrepreneurial activities to address a national problem may be especially weak when challenges to the policy status quo would threaten not just the incomes of business groups (such as the medical products industry) but also the autonomy and authority of trusted professionals, such as doctors. It may be an uphill battle for the political entrepreneur to convince the public that its understanding of reality is mistaken and that reform is warranted; the entrepreneur could decide to take his or her creative energy elsewhere, where there is more likelihood that such effort will be rewarded. This is not to say that there will be literally zero entrepreneurial investment in solving the problem, but that reelection incentives will induce much too little effort relative to the magnitude of the problem, because the risk to a reelection-minded politician of being seen as challenging doctors is too great.

A second source of failure occurs at the decision-making stage. As the political scientist Donald E. Stokes argues, policy issues typically come in two generic forms: position issues—on which voters hold conflicting preferences, such as whether taxes on the wealthy are too low—and valence issues—on which voters hold common preferences, such as whether the government should promote prosperity or fight corruption.90 When valence issues are on the decision agenda, the electoral stakes are extremely high. No political party can afford to be seen as the “antigrowth” or “procorruption” party. In an idealized democracy, the positions of the two parties will tend to converge on such issues, and the government will make decisions that voters support. As Congress expert Frances E. Lee argues, however, in an era of partisan polarization and persistently close electoral competition, there can be strong incentives for the out party to behave strategically and block progress on what could be consensual, good government issues.91 One tactic the minority party can use is to transform a valence issue into a position issue, creating a partisan or ideological squabble where none had previously existed by reframing efficiency into a debate about the size and role of government. While this is not an entirely new dynamic in American politics, it has become a more common one in the modern era.

And this is in fact precisely what occurred during the debate over the Obama administration’s effort to identify low-value medical treatments through an investment in CER. The Obama administration was not the first to highlight the medical evidence problem. Medicare administrators in the George H. W. Bush and George W. Bush administrations, for example, had also pointed out that a large share of medical spending is wasteful and not based on sound science. One might have expected fiscal conservatives to complement the Obama administration’s support for CER, since Republican health care leaders had endorsed the CER concept in the past as a way to curb wasteful spending, not to expand government’s role.

But the fierce ideological competition between the two parties over the expansion of health insurance under Obamacare undermined the incentives for bipartisan technocratic consensus on “good government” reforms like CER.92 (One Republican congressional staff member told us in an interview that Republican support for CER was a casualty of the partisan “knife fight” over the ACA; in the middle of the knife fight, he said, you don’t pause to tell your opponent that you like his shirt.) The Republican Study Committee sent out an alert stating that the purpose of the initiative was to let the government ration care. To an anxious public, CER was linked with two other contentious proposals then under consideration: voluntary counseling for Medicare patients about living wills, advance directives, and other end-of-life care; and the creation of an independent commission (Independent Payment Advisory Board, IPAB) with the power to recommend ways to achieve Medicare savings without cutting benefits. These three elements swirled around and combined in the public mind to give birth to the charge that Obama was seeking to create “death panels” for seniors. CER thus morphed into a symbol of rationing of medical services and bureaucratic interference with the doctor-patient relationship. In sum, conservatives turned a valence issue (using science to learn what works in medicine) into a position issue (the role of the federal government in health care delivery). As we show, the “devalencing” of an issue undermines pragmatic decision making and elite-led social learning in a sustained era of competitive balance between the parties and polarization.

Stimulating Conversations

In the space of a medium-size book, we cannot hope to answer all the questions we raise. We will be content if our analysis stimulates three conversations.

First, we hope to give health care experts a better understanding of the political forces that permit waste and bad science to persist in the medical system. Generating fresh insights by applying some of the ideas, habits of thought, and tools of political science to this familiar problem will help experts—including economists, health services researchers, and policy makers—craft recommendations that stand a better chance of both winning adoption and sticking after enactment.

Second, we wish to encourage other political scientists to devote more attention to the fundamental questions of who governs, and to what ends. As Terry Moe argues, political scientists should explore the power of vested interests of all kinds, including those that strategically hide their special interests inside a public interest package.93 With respect to the second question, political scientists typically use two criteria to analyze policy outcomes: the responsiveness of government to public opinion, and distributional fairness. These are very important evaluative dimensions, but they have limitations as benchmarks of government performance on complex scientific issues like the efficiency of health care. The public’s opinions on such issues may be either uninformed,94 or thoughtlessly derivative of the beliefs of policy elites.95 Similarly, a focus on distribution may be appropriate when the government targets benefits or costs at particular groups, but it is less illuminating when government seeks to provide public goods to the citizenry as a whole.

Finally, we hope our analysis encourages reflection on the capacities and limitations of modern American government as a problem-solving institution. U.S. politicians continue to promote technocratic problem solving in many areas. For example, the biomedical research conducted by the National Institutes of Health helps to advance understanding of diseases and epidemics and has long drawn bipartisan support. Yet what is often overlooked is that the roles of science, expertise, and enlightened public opinion in governance are mediated by the power of interest groups and professional societies, the struggle for partisan control, and the dynamics of electoral competition. Our investigation of the uses and misuse of medical evidence shows how these factors can produce an “unhealthy politics” in which pragmatic problem solving in government is degraded.

In particular, our research encourages reexamination of the assumptions of the Progressive reform tradition that has provided an intellectual foundation for problem solving in the U.S.96 At the core of this tradition is a belief in the importance of expertise and delegation of many complex tasks to the professions, including lawyers and engineers, but most prominently doctors. Relying on the professional ethics of doctors to self-regulate was an imperfect if understandable choice that has broader implications for how we understand delegation and the role of the professions in a democracy. Our analysis suggests that just as Congress checks the president, so elite professions—including scientists, physicians, university researchers, and policy analysts—need to monitor one another’s performance to ensure the primacy of citizen welfare and a commitment to a rational allocation of societal resources. Our central claim is not that science and expertise alone can (or should) entirely drive policy making in a pluralistic society.97 There will always be distributional battles over “who gets what” from government, and ideological debates over the appropriate role of the state. Yet a wellfunctioning policy state creates adequate incentives for problem-solving activities and ensures that elite groups who enjoy public trust, such as the medical profession, are democratically accountable and responsive to public needs. The amount of space for technocratic expertise and pragmatism in policy making is not simply a by-product of the technical complexity of a policy area;98 rather it is shaped by the incentives created by the broader political environment. We explain why the politics in the U.S. health care sector is so unhealthy, and how shifts in the behavior of the medical profession and the design of political institutions can promote sustainable reform.

The Plan of the Book

Chapter 1 defines the “medical guesswork” problem and explains how the poor integration of evidence into clinical decision making harms the performance of the health care sector. Chapter 2 makes our argument more nuanced and concrete by presenting a detailed case study of the remarkable sham knee surgery case mentioned earlier. In chapter 3, we present the results of national public opinion surveys that illuminate how ordinary citizens think about the medical evidence problem. Chapter 4 explores the institutional roots of medical professionalism in the United States. We examine why the U.S. health care system delegates therapeutic authority to individual doctors and medical societies, with little centralized oversight in programs like Medicare. The chapter also presents findings from a national survey of physicians to gauge their views on the proper role of medical societies in medical evidence controversies.

In chapters 5 and 6, our focus turns to politicians. We explore the politics of EBM, exploring why the efforts to root out waste, bad science, and extractive behavior in health care have not engendered reliable support. Chapter 6 examines the politics of the Obama administration’s effort to promote CER as the scientific foundation of health care quality improvements and cost control. In the concluding chapter, we take stock of the EBM reform project. We draw on lessons from the literature on U.S. state building to develop strategies to increase the durability of medical governance reform in contemporary American politics.



If you find an error or have any questions, please email us at Thank you!