6

Electoral Competition, Polarization, and the Breakdown of Elite-Led Social Learning

Contemporary American politics is characterized by ferocious partisan combat. While bipartisan cooperation has not disappeared entirely from the halls of Congress, many important roll call votes see nearly all Democrats on one side and nearly all Republicans on the other. There are at least two reasons for the high level of partisanship in Washington. The first is sincere ideological disagreement about the role of government. According to quantitative methods for measuring ideological placement, the distance between Democratic and Republican officeholders is much greater today than it was in the 1950s and 1960s, and the two parties have become more internally cohesive.¹ The political center has all but vanished in Congress.² Second, partisan conflict also reflects strategic behavior to gain electoral advantage. As political scientist Frances Lee observes, “The period since 1980 stands out as the longest sustained period of competitive balance between the parties since the Civil War. Our politics is distinctive for its narrow and switching national majorities. Nearly every recent election has held out the possibility of a shift in party control of one institution or another.”³

In this chapter, we argue that elite polarization and a near parity of partisan competition degrades government problem solving in two ways. First, it creates incentives for politicians to transform what plausibly could be consensual “valence” issues, on which nearly all candidates and parties adopt the same stance, into contentious “position” issues, on which candidates and parties take different stances in a zero-sum competition for voter support.⁴ Second, elite partisan polarization can stimulate polarization among ordinary voters. Because politically aware members of the public tend to follow the signals of their party leaders, when the parties diverge on an issue, attentive citizens often do so as well, even if the issue has little intrinsic ideological content.⁵

Taken together, these twin dynamics can undercut the processes of elite-led social learning and technocratic problem solving on which social progress to no small extent depends. We show here how these distortions played out in 2009–10, when the Obama administration moved forward with its proposal for a major investment in research on the comparative effectiveness of medical treatments, despite the lack of public buy-in for this reform project.

The Risk of Incomplete Incubation in an Era of Polarization

Writing about the politics of policy innovation, the late political scientist Nelson W. Polsby described a process of “incubation,” in which advocates take up an idea, conduct research about its justifications and effects, publicize it, and slowly build acceptance for it among relevant constituencies so that the idea has strong patrons when circumstances permit it to be considered.⁶ Incubated ideas often appear in party platforms decades before their formal enactment. The benefits of incubation are considerable. As Hugh Heclo writes, “The very ability to sustain the policy argument over time helps persuade people to the view that there is a real problem that will not go away until something is done. Almost unnoticed, a presumption for policy action can grow.”⁷

However, not all innovations are incubated. Sometimes, when the political stars align, new ideas emerge hastily on the public agenda and their supporters find themselves in the right place at the right time.⁸ It is perfectly understandable that advocates take advantage of those propitious circumstances to move their ideas forward in the legislative process, but the decision to do so is not without strategic costs.

As described in chapter 5, the incubation process for the idea of building a new CER institute was underway by the time Barack Obama won election in November 2008, but it was clearly incomplete. Policy formation and agenda setting was less the result of a mobilization of public support around the need for a more evidence-based medical system than it was the product of an insider process dominated by experts who enjoyed close access to decision makers.⁹ CER was left vulnerable to the baleful influence of partisan forces when proposals to invest tax dollars in research on what works best for patients unexpectedly gained prominence on the president’s domestic agenda. While CER funding initiatives did win enactment by being incorporated into two omnibus laws that congressional Democrats viewed as “must pass”—the 2009 economic stimulus measure and the Affordable Care Act—the measures became objects of ideological derision and partisan attack, undermining their technocratic image and political sustainability.

There is a fundamental dilemma at play: Nonpartisan, technocratic solutions to collective problems, grounded in scientific knowledge and neutral expertise, may lack for effective political advocates who can build a broad coalition when the proposals challenge the autonomy of esteemed professional groups, especially when public trust in government is low, as it has been in the United States since the late 1960s. This creates a strong incentive for advocates to employ “under the radar” legislative tactics in the hope that the measures can progress before the opposition can mobilize. Yet embedding these technocratic solutions within broader legislative proposals (like health insurance expansion) before the political ground has been prepared is risky. Opponents may in fact organize and exploit the issue, thus disrupting the elite-led process of social learning and undercutting the consolidation of reform.

CER IN THE 2008 PRESIDENTIAL CAMPAIGN

While bipartisan support for CER was waning on Capitol Hill, both major party nominees made CER a component of their health reform agendas in the 2008 presidential campaign—a signal that, among the small crowd of serious health policy experts who feed ideas to speech writers, medical evidence gulfs were widely regarded as a problem. The Republican nominee, John McCain, who had brought on CER expert Gail Wilensky as one of his health policy advisers, stated, “We must make public more information on treatment options and doctor records, and require transparency regarding medical outcomes, quality of care, costs, and prices. We must also facilitate the development of national standards for measuring and recording treatments and outcomes.”¹⁰ In his health care campaign platform, the Democratic nominee, Barack Obama, identified CER as “one of the keys to eliminating waste and missed opportunities” and pledged to create “an independent institute to guide reviews and research on comparative effectiveness, so that Americans and their doctors will have accurate and objective information to make the best decisions for their health and well-being.”¹¹ Several advisers close to Obama were strong CER supporters. Former senator Tom Daschle (D-SD), who would be Obama’s initial nominee to be secretary of the Department of Health and Human Services, had published a book arguing that the United States needed to create an independent Federal Health Board that would make coverage decisions on the basis of scientific evidence.¹² While Daschle was forced to withdraw his nomination owing to tax trouble, Peter Orszag picked up the mantle in his new role as Obama’s expected OMB director. By all accounts, Orszag wanted to see CER used as a tool to curb wasteful spending. As he wrote in an op-ed in the New York Times,

Right now, health care is more evidence-free than you might think. And even where evidence-based clinical guidelines exist, research suggests that doctors follow them only about half of the time. One estimate suggests that it takes 17 years on average to incorporate new research findings into widespread practice. As a result, any clinical guidelines that exist often have limited impact.¹³

Orszag’s favorite idea was to create “safe harbors,” whereby doctors who followed evidence-based best practices would be shielded from malpractice liability.¹⁴ It seemed likely that CER would emerge as a White House priority regardless of who occupied the Oval Office. Given the fierce environment of partisan conflict, the real question was whether the issue would then gain support from lawmakers of the opposition party.

RISING GOP OPPOSITION

Any hope that CER would win the backing of congressional Republicans after Obama’s election victory quickly vanished. There are at least three explanations for the vociferous GOP opposition to CER during the Obama years. The first is the general claim that the modern Republican Party simply does not believe in the use of scientific evidence to guide policy decisions, or at least believes in science much less than does the Democratic Party. Despite recent arguments about a Republican “war on science,”¹⁵ we are skeptical of such sweeping claims. At times, both parties can be politically opportunistic in the research they support, depending on their evolving electoral incentives.¹⁶ Further, as previously noted, some prominent Republicans, including Bill Frist and John McCain, endorsed CER and evidence-based medicine before 2009. Republican administrations have sought to promote evidence-based health policy making within the constraints of federal law. As Tom Scully, who served as administrator of the Centers for Medicare and Medicaid Services under George W. Bush, stated in 2009:

I have always been a big fan of comparative effectiveness research if done correctly.… There’s a lot of fear from some people that it’s going to stifle innovation of drugs and devices and other things and I just don’t [think] that’s correct…. I like Senator [Jon] Kyl [who was then voicing strong opposition to the Obama administration’s CER agenda], he’s a very big free market guy, he’s got a lot of doctors in Arizona that are very worried about this. And so, like any member of [C]ongress he tries to please constituents.

Scully went on to say that the attacks of conservative pundits on CER were “just noise” and that since the Republicans are the minority party, “their job is to hurl attacks.”¹⁸

A second argument is that GOP opposition reflected increasingly close ties to the drug and device industry. In European countries where CER and technology assessment are now part of national policy, long before that was the case, drug companies in those countries were sponsoring a lot of research in order to satisfy the growing need to demonstrate “value for money.” In their review of CER in Britain, France, Australia, and Germany, Chalkidou and colleagues found that industry had adapted to the evidentiary requirements of coverage and reimbursement systems in Europe.

CER entities create a more secure environment in which the naturally risk-averse medical technology industry can make its investment choices. The reason is that well-defined and consistent CER is a much more rational and predictable way for payers to make purchasing decisions than for administrators to impose price cuts arbitrarily, to shift costs to individual patients, or to ration needed technologies and services according to ability to pay.¹⁹

But this adaption may have been as much out of necessity as choice. In the U.S. market, the biggest payer, Medicare, has avoided supply-side technology constraints and price controls on drugs. Many firms wished to keep the U.S. market an outlier in these areas, and hence industrial opposition to CER (seen as a precursor to more far-reaching shifts in policy) in the United States was particularly intense, which then arguably translated into unusually high partisan opposition. Ian Spatz, Merck’s former vice president for global health policy, stated that the view of drug companies is that CER is a “slippery slope that leads to NICE—and NICE is seen as pure evil.”²⁰

A final reason for the GOP’s opposition—echoed in Scully’s remarks—is that the GOP’s turn away from CER during this period was tactical, a product of the electoral incentives created by partisan competition and the GOP’s desire to block Obama’s overall health care agenda and take back Congress. The health experts associated with the Republican Party fully grasped the magnitude of the problem and the need for a government solution—but partisan competition undermined pragmatism.

The Disruption of Elite-Led Social Learning and Problem Solving

As political scientist David R. Mayhew observes, a democratic polity can make public policy in at least three ways.²¹ The first two are familiar: distributive politics, in which my district gets a road, your district gets a road, everyone’s district gets a road, and partisan politics, in which my party coalition gets what the government has to give and yours pays the taxes. By contrast, “problem solving” entails an “empirically detectable mindset. Some person, or small or large set of persons, needs to frame a state of affairs as exhibiting a ‘problem’ and to point toward a ‘solution.’”²²

Experts play a key role in the politics of problem solving, in part because the media and politicians often take their cues from policy specialists.²³ Expert communities generate the empirical claims and general perspectives on policy issues that structure the political debate. As political scientist John Zaller has observed, when policy specialists of differing ideological orientations reach a fact-based consensus on the existence of a technical problem, this elite consensus may diffuse out to politicians and then over time to the general public.²⁴ This broad support in turn makes it easier for coalition leaders to build the legislative majorities necessary to overcome the veto points of the U.S. political system.

However, this idealized social learning process neglects the role of partisanship and electoral competition. When, as is the case today, American politics features intense partisan conflict and the majority party chooses (or is compelled) to “go it alone” on major legislation, the minority party will possess a strong incentive to discredit the empirical claims offered by the other party—even if working from a common set of technical understandings of a problem would be in the national interest. As even sound evidentiary claims are contested owing to partisan conflict, an expert consensus on the seriousness of a given problem may then fail to unify political elites and the public around a course of action. The parties may take opposing stances on the role of government in a particular policy area—even when there is no underlying ideological disagreement between them.²⁵

We argue below that this is what happened during 2009–10: CER became an issue “owned”²⁶ by the Democrats and associated with negative images such as “rationing” and “death panels.” A new research agency was established, but the national conversation about the benefits of evidence-based medicine took a giant step backward, at least for a time.²⁷

THE RECOVERY ACT DEBATE

The Obama administration’s first major push on CER came in the 2009 American Recovery and Reinvestment Act, which included $1.1 billion for CER to be spent by the NIH, AHRQ, and the Office of the Secretary of Health and Human Services.²⁸ The act also created an advisory board mostly made up of clinicians, the Federal Coordinating Council for Comparative Effectiveness Research, to steer the research effort. As political scientists James G. Gimpel, Frances E. Lee, and Rebecca U. Thorpe argue, the advertised purpose of the $787 billion legislation was to promote economic recovery, but the measure “became a vehicle for a broad array of other policy goals” important to the administration and congressional leaders who “took advantage of an especially wide-open window of opportunity.”²⁹ Among these policy goals was promoting CER.

In the context of the $2 trillion the nation spends each year on health care, $1.1 billion for research on what medical interventions work best is a small investment. Since the idea had been supported by the McCain campaign, one might have expected the Obama administration’s CER initiative to have won bipartisan support.³⁰ But that is not what happened. Republicans used the battle over the Recovery Act as an opportunity to rehearse their arguments against what they knew would be a major drive by the Obama administration to restructure the nation’s overall health care system. The Obama administration’s health reform agenda—which featured redistribution, tax increases, and an expansion of the welfare state—was bound to be opposed by conservatives on both policy and political grounds. As mentioned earlier, a Republican committee staff member told us that conservatives attacked CER, even though they knew it would advance their long-standing goal to curb wasteful health care spending, because they viewed the battle over national health reform as a “knife fight, and in a knife fight, you don’t stop in the middle to tell your opponent, ‘Hey, I like your shirt.’”³¹ As political scientist Morris Fiorina writes, in an era in which party elites are highly polarized, “[p]olicies are proposed and opposed relatively more on the basis of ideology and the demands of the base and relatively less on the basis of their likelihood of solving problems.”³²

Pharmaceutical and medical device companies viewed the Obama administration’s support for a major investment in CER with suspicion, seeing it as the first step toward tighter governmental control over medical service payment decisions. “There was a real fear about the government saying: We know what’s best, that drug’s too expensive, you can’t have it,” said former Democratic congressional leader Tony Coehlo.³³ Coehlo, who has epilepsy, was chosen to head the Partnership to Improve Patient Care (PIPC). Funded by pharmaceutical, device, and biotechnology companies, PIPC emerged in 2008 as the medical product industry’s watchdog on federal CER efforts.³⁴ Former Republican congressman Billy Tauzin, president of PhRMA, explained that his industry’s mobilization against the CER funding in the Recovery Act was a signal to politicians to be careful about embracing more ambitious policy reforms. “I hope it is a clear warning,” Tauzin said. “There are a lot of beehives out there. You don’t just go around punching them.”³⁵

Allegations that Democrats wanted to use studies of treatment effects to reduce wasteful spending were supported by language in a House Appropriations Committee draft report stating:

By knowing what works best and presenting the information more broadly to patients and health care professionals, those items, procedures, and interventions that are most effective to prevent, control, and treat health conditions will be utilized, while those that are found to be less effective and in some cases, more expensive, will no longer be prescribed.³⁶

One could interpret this statement as having no clear implication for overall health care investment, but it was widely read by those hostile to CER as a signal that the Obama administration wanted to use CER as a tool to ration care.³⁷ Representative Tom Price (R-GA), a physician—who would later be secretary of the Department of Health and Human Services in the Donald Trump administration—sent out an alert through the Republican Study Committee stating that that purpose of the CER measure was to grant Tom Daschle, Obama’s nominee for secretary of HHS, “his wish of a permanent government rationing board prescribing care in place of doctors and patients.”³⁸ Former New York lieutenant governor Betsy Mc- Caughey warned on Bloomberg.com that the elderly would be hardest hit, and that the government would use electronic medical records to monitor the behavior of physicians, punishing those that did not comply with the government’s clinical practice guidelines.³⁹ Rush Limbaugh then disseminated the rationing charge on his national radio show.⁴⁰ The final version of the stimulus legislation took out the House report language about less effective, more expensive treatments not receiving coverage, and instead referenced the Medicare Modernization Act, which focused on clinical outcomes, not cost.⁴¹ The conference report also stated that “the conferees do not intend for the comparative effectiveness research funding … to be used to mandate coverage, reimbursement, or other policies for any public or private payer.”42

The partisan controversy over the CER language in the stimulus bill greatly raised the issue’s public visibility. In 2008, U.S. newspapers and wires ran only 249 articles that mentioned CER and health care or medicine. The following year, the number of such articles increased sixfold to 1,497. Twenty-two percent of the articles published about CER in 2009 mentioned “rationing,” the frame that Republcans strategically deployed to tap into the public’s long-standing unease with service restricitons and perceived interference with the doctor-patient relationship.⁴³

THE AFFORDABLE CARE ACT DEBATE

Despite the controversy over the stimulus act, the Obama administration endorsed an even greater federal role in CER in health reform legislation.⁴⁴ By all accounts, President Obama himself believed in CER and the need for stronger evidence about what works. In an interview with the New York Times, President Obama explained,

When Peter Orszag and I talk about the importance of using comparativeeffectiveness studies as a way of reining in costs, that’s not an attempt to micromanage the doctor-patient relationship. It is an attempt to say to patients, you know what, we’ve looked at some objective studies out here, people who know about this stuff, concluding that the blue pill, which costs half as much as the red pill, is just as effective, and you might want to go ahead and get the blue one. And if a provider is pushing the red one on you, then you should at least ask some important questions.⁴⁵

While Obama attempted to persuade people that implementing evidence-based medicine is common sense, the controversy over CER continued. Indeed, the backlash over the issue got so harsh that White House chief of staff Rahm Emanuel argued within the White House for abandoning the effort.⁴⁶ Ultimately, the authority and funding for the Patient-Centered Outcomes Research Institute made it into the Affordable Care Act, but not before key concessions were made.

By the summer of 2009, committees in both the House and the Senate were developing CER proposals as part of health reform legislation. The key issues were whether a new agency should be created within or outside government, how it should be funded, and what role it should play in coverage decisions. The House (as well as the Senate Health, Education, Labor, and Pensions Committee, HELP) sought to create a government-based entity within the AHRQ. In contrast, Senator Mike Enzi (R-WY) was an outspoken critic of the HELP bill. He compared the entity to NICE and raised fears that lodging it inside government would lead to rationing.⁴⁷

Whether CER would even remain a feature of health reform legislation was tested during the summer of 2009, when members of Congress returned to their districts for contentious town hall meetings. To an angry, wary, and confused public, CER was of a piece with two other controversial health reform proposals then under consideration: voluntary counseling for Medicare patients about their options for end-of-life care, and the creation of a commission (the Independent Payment Advisory Board) with the mission to achieve savings in Medicare if the program’s spending growth was forecast to exceed target levels, but without affecting coverage or cutting benefits. With prompting from opponents, including 2008 GOP vice presidential nominee Sarah Palin and others, these three elements combined in the public mind to give birth to the charge that Obama was seeking to slash Medicare benefits by creating “death panels” for seniors. At one town hall meeting, Senator Charles Grassley (R-IA) stated, “We should not have a government program that determines if you’re going to pull the plug on grandma.”⁴⁸

After the summer recess, Senate Finance Committee chair Max Baucus (D-MT) proposed the concept of an independent entity, the Patient- Centered Outcomes Research Institute (PCORI). While the institute’s name (“patient-centered”) and location outside government were intended to allay medical products industry concerns, the proposal remained controversial. Republicans introduced amendments “that would strip out funding for the research institute, eliminate any consideration of cost in the research, and prohibit research results from being used to make coverage decisions or to deny coverage.”⁴⁹ The Senate Finance Committee sought to strike a balance between addressing the concerns of critics who claimed that the Senate finance bill gave the industry too much influence over the research, and others who remained concerned that the institute would ration care.⁵⁰

In the midst of this controversy over CER, the Obama administration felt compelled to distance itself from new guidelines for breast cancer screening, contained in an awkwardly timed report of the U.S. Preventive Services Task Force. A Pew survey found that 68 percent of people who followed the news about the change in these guidelines very or fairly closely disagreed with the task force’s recommendation that most women should not start routine screening until age 50.⁵¹ The Obama administration responded to the public outcry against the change by promising that government insurance programs would continue to cover mammograms for women starting at age 40.⁵²

Congressional leaders anticipated a conference committee to resolve differences with the Senate, and some staffers expected that the final version would move closer to the House vision of a less industry-influenced CER-agency located inside government.⁵³ However, Republicans took control of the upper chamber following Scott Brown’s surprising victory to fill the Senate seat vacated by Ted Kennedy’s death. To avoid a conference committee and the risk of a filibuster, House Democrats passed the Senate bill, and Baucus’s PCORI design prevailed.

As signed into law by President Obama, the ACA established a nongovernmental, “Patient-Centered Outcomes Research Institute.”

The purpose of the Institute is to assist patients, clinicians, purchasers, and policy-makers in making informed health decisions by advancing the quality and relevance of evidence concerning the manner in which diseases, disorders, and other health conditions can effectively and appropriately be prevented, diagnosed, treated, monitored, and managed through research and evidence synthesis …⁵⁴

The creation of PCORI was a significant victory in the multidecade effort to improve the evidence base of American medicine, yet the institute’s financing, design, and mission reflect legislative compromises designed to protect the entity’s independence but also circumscribe its role. To fund PCORI, the ACA created a federal trust fund, fed by annual appropriations, funds transferred from the Medicare trust fund, plus annual $1 fees (rising to $2 in fiscal year 2014) per individual assessed on private health insurance and selfinsured plans. Total funding was estimated to reach $650 million annually by 2015. However, PCORI’s authorization expires in 2019, leaving the institute only a few years “in which to produce timely, practice-changing results that will build public support for comparative effectiveness research.”⁵⁵ The legislation also gave the institute less authority than CER entities in other advanced nations. While nothing in the law prevents Medicare officials (or private insurers) from using CER research to inform coverage decisions, nothing requires it: PCORI’s findings may “not be construed as mandates for practice guidelines, coverage recommendations, payment or policy recommendations.”⁵⁶ To avoid any resemblance to NICE, PCORI was barred from using a dollars per quality-adjusted life year metric as a threshold for establishing cost-effectiveness.⁵⁷

Ultimately, PhRMA managed to win enough concessions, including representation of three medical products industry seats on PCORI’s board of governors, that the trade group came out in support of PCORI. Richard Smith, PhRMA’s senior vice president, stated:

By including the full range of stakeholders in its governance, by defining the scope of research to include the full range of treatment options and the organization, delivery, and management of care, the institute is charting a different, more positive course than agencies in other countries which focus on cost-effectiveness and impose centralized restrictions on access to care.⁵⁸

The CEO of the Advanced Medical Technology Association (AdvaMed) also supported the legislation. He applauded the commitment that PCORI “does not make coverage decisions” and that the studies would not include coverage recommendations or include practice guidelines.⁵⁹ The medical profession was not highly visible during this debate. Most medical societies were focused on Medicare reimbursement rates (“the doc fix”) and did not provide much input into the shaping of the legislation.⁶⁰

As important as what the ACA accomplishes are the more ambitious changes that it leaves out. The legislation failed to modify Medicare’s obligation to cover all “reasonable and necessary” medical services.⁶¹ Peter Orszag’s proposal to shield doctors from malpractice suits if they follow evidence-based guidelines made it into the law only as a small pilot program. In sum, while Republican and industry opposition failed to kill CER, the reform project narrowed and lost its pragmatic, technocratic character. No longer a “valence” issue on which Democrats and Republicans could publicly agree (if not necessarily act), CER became a divisive “position” issue on which there were different ideological preferences, based on people’s views of the Obama administration’s overall domestic agenda.⁶² Advocates managed to win support for CER by incorporating language into omnibus bills that congressional Democrats regarded as “must pass” legislation, but the decision-making process reflected (and amplified) partisan conflict and arguably did not lay the foundation for a durable reconfiguration of the use of evidence in the medical system.

Elite Polarization and the Leadership of Public Opinion

As we argued above, elite partisan polarization not only weakens the incentives for pragmatic problem solving in a legislature; it distorts the process by which ordinary citizens learn what and how to think about novel public policy issues.⁶³ In the U.S. political system, a citizen’s political identity (how the citizen thinks of him- or herself) frequently (not always) means a partisan identity.⁶⁴ Attentive citizens who follow public affairs typically ask themselves, “What do Democrats (Republicans) like myself think about this issue?” Frequently, the answer is readily apparent: Republicans stand for lower taxes, and Democrats support a more generous safety net, and so on. But when new issues emerge on the agenda, especially those characterized by scientific or technological complexity, the “appropriate” partisan response may not be self-evident. Under these circumstances, partisan identifiers may not initially all share the same views until opinion leaders clarify the “party line.” As Levendusky argues, “When elites are polarized, they send voters clearer signals about where they stand on the issues of the day…. As voters follow these party cues on multiple issues, they begin to hold more consistent attitudes.”⁶⁵ This is largely a top-down process, one in which citizens tend to “follow the leader.”⁶⁶ As Fiorina and Abrams write, “The more visible and active members of a party, especially its elected officials and party activists, sort first and provide cues to voters that party positions are evolving.”⁶⁷

Issue publics are not always partisan, but this party-sorting dynamic has clearly shaped the evolution of public attitudes on key topics. Consider opinion on the environment and climate change.⁶⁸ Historically, there were only modest partisan differences among the general public in support for environmental protection. Between the 1970s and 1990s, for example, “support for increased spending on environmental protection by self-identified Democrats was typically only around 10 points higher than for self-identified Republicans.”⁶⁹ But, “[t]he gap began to widen in the late 1990s, likely reflecting voters’ tendency to follow cues from party leaders and political pundits.”⁷⁰ While Richard Nixon had created the Environmental Protection Agency and signed landmark clean air legislation, Ronald Reagan’s administration sought to scale back environmental regulations. The polarization of public opinion has been even more pronounced with respect to climate change. According to Gallup polling data, in 1997 nearly identical percentages of Republicans and Democrats said that the effects of global warming were already happening. However, as Republican politicians and pundits began questioning the science behind climate change, while Democratic politicians and environmental activists made belief in climate change a litmus test, a partisan divide among the public began to emerge. As figure 6.1 shows, by 2008, Democrats were 35 percentage points more likely than Republicans (76 percent versus 41 percent) to say the effects of global warming had already begun to occur, with Independents falling in between the two. This finding of growing partisan (and ideological) polarization on climate change among the public persists in a multivariate analysis that controls for other relevant variables.⁷¹

FIGURE 6.1. Percentage of respondents saying the effects of global warming have already begun, by respondent party identification. Note: Figure displays the percentage of respondents who said “they have already begun” in response to question “Which of the following statements reflects your view of when the effects of global warming will begin to happen: [rotated: they have already begun to happen; they will start happening within a few years; they will start happening within your lifetime; they will not happen within your lifetime but they will affect future generations; they will never happen]?” From 2001–8, surveys took place in March of each year; 1997 survey was conducted in November. Question was not asked on Gallup surveys 1998–2000. Source: Gallup polling data. See Riley E. Dunlap, 2008, “Climate-Change Views: Republican-Democratic Gaps Expand,” May 29, http://www.gallup.com/poll/107569/climatechange-views-republicandemocratic-gaps-expand.aspx.

What about public opinion with respect to evidence-based medicine and CER? Ideally, we would like to have over-time data to examine whether any partisan differences have changed or remain the same. Unfortunately, the issue was not salient enough to receive attention from pollsters before the Obama administration’s major initiative. A national survey we conducted in 2010, however, reveals a split in the attitudes of Republican and Democratic voters consistent with the polarization hypothesis. As a baseline measure of how people think about CER, we asked respondents about their support for government funding of research on the effectiveness of different treatments. Specifically, we asked:

For many medical conditions, doctors use different kinds of treatments, and there is no scientific agreement on which is best. For example, a patient may be experiencing a particular type of pain and it is unclear whether the best treatment is a drug, physical therapy, or surgery.

Recently there has been discussion about the need for more research to determine which treatments are most effective for which patients. This is sometimes called comparative effectiveness research.

Would you support or oppose government funding of research on the effectiveness of different medical treatments?

Respondents were asked to choose a point on a sliding scale where the far left read “strongly oppose” (scored as 0) and the far right read “strongly support” (scored as 100). The mean rating for the sample was 61.9, suggesting that, on average, people were slightly more supportive of CER than they were in opposition. There were, however, significant differences in support levels across important demographic and political groups. These differences are displayed in table 6.1.⁷²

There were important differences among demographic groups. Blacks (68.1) and Hispanics (67.7) were more supportive of government funding of CER than whites (60.0), respondents with a college degree or higher level of education were more supportive (65.3) than those with a high school education or less (59.9), and respondents aged 18–37 were more supportive (66.3) than those aged 65 and over (55.8).⁷³ There was not a statistically significant difference between men and women or across income groups.

The most important factor in explaining support for CER, however, was partisanship. Although evidence-based medicine might seem to be a technocratic issue that should generate bipartisan support, consistent with our discussion in the previous sections concerning elite polarization, there were substantial differences between Democrats and Republicans in the mass public. Democratic Party identifiers were more supportive of government funding of CER (70.9) than Independents (59.8), who, in turn, were more supportive than Republicans (50.7). Significantly, these partisan differences hold only for respondents who reported voting in the 2008 election. Among those who voted in 2008, support for government funding of CER was 72.0, 59.1, and 49.8 among Democrats, Independents, and Republicans, respectively. Among those who did not vote, however, there was not a large or statistically significant difference in support across Democrats (63.9), Independents (61.4), and Republicans (59.8). During the debate over the stimulus bill and President Obama’s health reform proposal, evidence-based medicine evolved from a technocratic issue into a partisan issue. The two parties took opposing stances on the federal government’s role in CER, which may have led to the significant partisan split among the most attentive (voting) citizens.

We conducted a follow-up survey in 2014 to see if these partisan differences persisted.⁷⁴ We asked the same question we did in 2010, but with one wrinkle. Although the first two paragraphs of the question were identical to the question we asked in 2010, half of the respondents in 2014 were randomly assigned to receive some text that made an explicit link between CER and Obamacare. Specifically, respondents assigned to this Obamacare “treatment” condition received the following text:

Obamacare (the Affordable Care Act) includes increased funding for comparative effectiveness research. Do you support or oppose government funding of research on the effectiveness of different medical treatments?

The other half of the respondents simply received the following, “control” condition text:

Do you support or oppose government funding of research on the effectiveness of different medical treatments?

This design allows us to address two important questions. First, did overall public opinion on CER shift between 2010 and 2014? Second, does making an explicit partisan link to Obamacare exacerbate partisan differences on this issue?

As in 2010, respondents were asked to choose a point on a (0–100) sliding scale where the far left (0) indicated strong opposition and the far right (100) indicated strong support. The mean rating for the sample was 60.2 (very close to the mean rating in 2010 of 61.9), suggesting that the American public overall remained moderately supportive of CER in fall 2014. Moreover, among the public as a whole, there was no significant difference between the two question wordings. The average rating for respondents who received the Obamacare language was 60.0, and it was 60.4 for those that received the standard (control) language in which Obamacare was not mentioned.

Note: N=2,017. ^aResponses range from strongly oppose (0) to strongly support (100) CER and are weighted sample means. ^b_p-values are from tests of whether support varied across groups. Source: May 21–24, 2010, YouGov/Polimetrix survey. A version of this table was originally published by Project HOPE/Health Affairs as exhibit 1 in Alan S. Gerber, Eric M. Patashnik, David Doherty, and Conor M. Dowling. 2010. “The Public Wants Information, Not Board Mandates, from Comparative Effectiveness Research.” Health Affairs (Millwood) 29 (10): 1872–81. The published article is archived and available online at www.healthaffairs.org.

Note: N=802. Cell entries are weighted means with standard errors in parentheses. Higher numbers represent more support for CER on 0–100 scale. For the entire sample, overall mean rating is 60.2, and for the Obamacare Cue and No Obamacare Cue conditions it is 60.0 and 60.4, respectively. Obamacare Cue: “Obamacare (the Affordable Care Act) includes increased funding for comparative effectiveness research. Do you support or oppose government funding of research on the effectiveness of different medical treatments?” No Obamacare Cue: “Do you support or oppose government funding of research on the effectiveness of different medical treatments?” “Independents” include any respondent that did not identify with one of the two major parties, including those who responded “not sure.” Source: Alan S. Gerber, Cooperative Congressional Election Study, 2014. Yale University Content. [Computer File] Release: February 2015. New Haven, CT. [producer] http://cces.gov.harvard.edu.

There were, however, large and statistically significant differences in support levels across partisan groups, differences that were exacerbated by the Obamacare language. We display these differences in table 6.2. Focusing initially on respondents who did not receive the Obamacare language, and are therefore most comparable to the sample of our 2010 survey, support for government funding of CER was 66.7, 54.7, and 53.7 among Democrats, Independents, and Republicans, respectively. This 13-point gap between Democrats and Republicans is smaller than the 20-point gap we observed in 2010, but still quite substantial. As was the case in 2010, these differences are larger among voters. However, the difference is not as stark as it was in 2010. Specifically, there is a 16-point gap in 2014 between Democrats and Republicans among voters (69.2 v. 53.6) and a 10-point gap among nonvoters (63.5 v. 53.7). Thus, it appears that polarization of opinion on CER may have followed a diffusion process—from party elites and opinion leaders in 2009 to voters by 2010 to party identifiers among the general public by 2014.⁷⁵

When explicit reference is made to Obamacare, partisan differences become even sharper. As table 6.2 shows, Democrats and Republicans move in opposite directions when CER is explicitly linked to Obama’s health reform law. Democrats become more supportive of CER, whereas Republicans become less supportive. The average rating among Democrats is 72.9 (up from 66.7 in the “control” condition, a 9 percent increase).⁷⁶ Among Republicans, the average rating is 45.0 (down from 53.7 in the “control” condition, a 16 percent decrease).⁷⁷ In short, a 13-point gap between Democrats and Republicans in terms of their support for CER when no reference to Obamacare is made more than doubles to a gap of nearly 28 points when Obamacare is referenced. This simple partisan cue experiment thus illustrates the extent to which public opinion on even technocratic issues like CER can become highly polarized when the issues are deliberated in partisan debates by policy elites.

Summary

The Obama administration’s political strategy to hitch a technocratic reform to two partisan, omnibus bills (the Economic Recovery Act and the Affordable Care Act) was both reasonable and understandable. In an area characterized by legislative stalemate and gridlock, any savvy president will look for “moving trains” on which to place agenda items. Yet there was a cost to this strategy too. While it is not easy to isolate technocratic, “good government” ideas from partisan contests over welfare state expansion and the politics of redistribution, the failure to do so can cause these solutions to become the “property” of one party and the electorally charged object of political derision for the other. The result can be to undercut efforts to cultivate an enlightened public opinion and potentially weaken the measure’s long-term impact and political sustainability. The story of how CER morphed into a symbol of rationing and government interference with the doctor-patient relationship offers a cautionary lesson about the limits of elite-led problem solving in an era of intense partisan competition and heightened polarization.