Abstract

Context: Public opinion on the performance of health system actors is polarized today, but it remains unclear which actors enjoy the most or the least trust among Democrats and Republicans, whether the COVID-19 pandemic has influenced how people view their own physicians, and whether doctors have retained the ability to influence public beliefs about policy issues.

Methods: The authors conducted two national surveys in 2022 and 2023 to examine these questions.

Findings: Democrats rated the performance of medical research scientists and public health experts during the pandemic more highly than did Republicans and independents. About three in ten Republicans said the pandemic decreased their trust in their personal doctors. Nonetheless, most Americans reported confidence in physicians. The authors replicated the findings of Gerber and colleagues (2014) to demonstrate that respondents continued to have more positive views of doctors than other professionals and that public opinion was responsive to cues from a doctors’ group.

Conclusions: What polarizes Democrats and Republicans today is not the question of whether medical scientists and public health experts are competent but whether the advice offered by these actors is in the public interest and should guide policy makers’ decisions. Democrats strongly believe the answer to these questions is yes, while Republicans exhibit skepticism.

As experts consider how to strengthen the nation's public health capacity in the aftermath of the COVID-19 pandemic, the public's declining confidence in key health system actors and institutions is receiving growing attention (SteelFisher et al. 2023). In 2021, only 34% of Americans gave positive ratings to the nation's public health system, down from 43% in 2009 (Harvard University/RWJF 2021). The erosion in public confidence is a concern for the nation's ability to manage future public health crises, not only because mistrust is widespread but also because attitudes toward many health care actors and institutions are highly polarized (Blendon and Benson 2022). Republicans were much less supportive than Democrats of many policies recommended by public health leaders during the pandemic, including mandatory vaccination and nonpharmaceutical interventions, such as social distancing and masking requirements (Barry et al. 2021; Gadarian, Goodman, and Pepinsky 2021; Sylvester et al. 2023). Virtually all of the decline in public support for the public health system between 2009 and 2021 occurred among Republicans; Democrats’ views changed little during this period (Findling, Blendon, and Benson 2022). To be sure, partisanship is not the only factor associated with differences in public opinion. Education, religion, race, and geography (urban vs. rural) are also important cleavages (Cox et al. 2023). But the partisan divide is stark. Acknowledging this divide is critical for designing effective public health communication strategies and reaching people where they are (Gollust, Nagler, and Fowler 2020; Larsen et al. 2022) and for encouraging self-reflection among scientific and medical leaders about what they can do to ensure their institutions are “worthy of the public's trust” (Cox et al. 2023: 5).

As Margaret Levi and Laura Stoker (2000: 476) argue, “Trust is relational; it involves an individual making herself vulnerable to another individual, group, or institution that has the capacity to do her harm or to betray her. Trust is seldom unconditional; it is given to specific individuals or institutions over specific domains.” The degree to which the public (the truster) has confidence in the decisions or guidance of physicians, policy makers, and other leaders (the trustees) during public health crises such as the COVID-19 pandemic turns on two main factors. The first is whether leaders are perceived to be competent in their respective areas of expertise. The second is whether leaders are perceived to be acting in the best interests of the public (Levi and Stoker 2000).

To investigate the degree to which public attitudes became polarized along these and other dimensions during and after the emergency phase of the COVID-19 pandemic, we conducted two national surveys in 2022 and 2023 on the public's confidence in health system actors, including doctors, public health leaders, and medical research scientists. While most of our analysis focuses on cross-sectional comparisons between Republicans and others (including Democrats and independents), our 2023 survey repeats two survey experiments from a study published in this journal a decade ago (Gerber et al. 2014), providing an opportunity for over-time comparison on key items. The study by Gerber and colleagues (2014), conducted well before the COVID-19 pandemic, found a high level of public confidence in doctors compared to other prestigious occupational groups. Specifically, the study demonstrated that Americans considered physicians to be “harder workers,” “more trustworthy,” and “more interested in helping people” than other professionals such as teachers, lawyers, or members of Congress. In addition, Gerber and colleagues (2014) used a survey experiment to compare the persuasive influence of doctors and political actors (such as congressional Democrats and Republicans), and they found that endorsement cues from physicians have a greater influence on aggregate public opinion on a generic health care issue than do cues from elected officials. Taken together, Gerber and colleagues’ (2014) findings provide an empirical basis for the widely held belief that “doctor knows best.” It remains unclear, however, whether the controversy over responses to the COVID-19 pandemic has diminished confidence in doctors among certain demographic groups or reduced doctors’ ability to influence aggregate public opinion.

Overall, this study addresses the key question of whether the pandemic has eroded the public's confidence and trust in a variety of health system actors. To answer this broader research question, we explore five major topics: (1) Americans’ confidence that health system actors are competent and are acting in the public's best interest; (2) partisan differences in perceptions of health system actors’ trustworthiness and performance during the COVID-19 pandemic; (3) the COVID-19 pandemic's effect on the public's self-reported trust in their personal doctors; (4) the public's evaluations of the motivations of physicians relative to other professionals, and whether such evaluations are less positive today than they were found to be a decade ago (Gerber et al. 2014); and (5) physicians’ capacity to persuade the general public to support a position on a public policy issue and whether such persuasive influence has changed since the Gerber and colleagues (2014) study.

Data and Results

We report data from two nationally representative, opt-in internet-based surveys that YouGov conducted under a contract with Yale University and that Yale's institutional review board approved. The first survey was fielded January 28–February 2, 2022, and the second survey was conducted July 6–18, 2023. YouGov ensured that respondents in the 2022 survey were representative of the total US adult population along dimensions including age, gender, race, education, region, political party affiliation, and income level by matching its panelists to a target sample from the US Census Bureau's 2019 American Community Survey (ACS). The final sample consists of 750 respondents. For the 2023 survey, YouGov developed a representative national sample by using the ACS, the 2020 Current Population Survey Voting and Registration supplements, the 2020 National Election Pool exit poll, and the 2020 Cooperative Election Surveys. The final sample consisted of 800 respondents.

The remainder of this article examines trust in health system actors; the public's views on the performance of these actors during the COVID-19 pandemic; public beliefs about whether medical research scientists and public health experts act in the public interest; self-reported assessments of the impact of COVID-19 on trust in personal doctors; beliefs about doctors’ motivations and attitudes; and the influence of cues from a doctor's association and other actors on public support for a health policy proposal.1

Trust in Health System Actors

We begin by exploring Americans’ trust in nine health system actors: medical doctors, nurses, medical research scientists, pharmaceutical companies, health insurance companies, the Food and Drug Administration (FDA), the Centers for Disease Control and Prevention (CDC), public health experts employed by the federal government, and “your” state public health agency. There is some degree of overlap among the actors in the functions they perform and the settings in which they work. For example, “medical research scientists” are employed in both pharmaceutical companies and public health agencies. In addition, medical research scientists may perform public health tasks, such as making public communications about the risks of diseases. We use the full list of actors to examine whether public opinion is sensitive to differences in terminology. The order of the nine actors was randomized. Figure 1 presents the results.

Figure 1

Mean (standard error) trust in a number of health care actors.

Source: YouGov original survey, July 2023.

Figure 1

Mean (standard error) trust in a number of health care actors.

Source: YouGov original survey, July 2023.

Close modal

Consistent with previous research (Blendon and Benson 2022; Findling, Blendon, and Benson 2022), we found that the public has the most trust in nurses (M = 7.17; SE = 0.10) and doctors (M = 6.91; SE = 0.10) on a 10-point scale. Medical research scientists are the third-most-trusted actor (M = 6.60; SE = 0.20). The two least-trusted actors are pharmaceutical companies (M = 3.99; SE = 0.13) and health insurers (M = 4.31; SE = 0.16). The much lower rating of pharmaceutical companies than medical research scientists (p < .001) implies that the public may associate pharmaceutical companies more with high drug prices than with scientific research, although this requires additional study. Ranking in between on the public trust scale are several government actors, including the FDA, the CDC, federal public health experts, and state public health agencies. State public health agencies are the most trusted among this group (M = 5.32; SE = 0.11), while federal public health experts are the least trusted (M = 4.97; SE = 0.13).

Next, we explored partisan differences in public trust in health system actors (online-only fig. A1).2 Democrats generally expressed a medium or high level of trust in the nine actors (with mean values ranging from a low of 4.71 for pharmaceutical companies to a high of 7.61 for nurses), while Republicans were substantially less trusting (with mean values ranging between 3.73 for federal public health experts and 6.87 for nurses). While the greater level of trust among Democrats in the nine health system actors than Republicans is notable, it masks important points of bipartisan consensus. Both Democrats and Republicans expressed trust (at least 6.2 on the 10-point scale) in doctors, nurses, and medical research scientists. At the same time, respondents from both parties expressed a low level of trust in pharmaceutical companies and health insurers. In sum, the main partisan difference is that Democrats reported much greater trust in public health actors than did Republicans. The partisan gap in trust in doctors, nurses, and medical research scientists is significantly smaller (p < .001) than the partisan gap in trust in the CDC, the FDA, federal public health experts, or state public health agencies.

The polarization in trust in public health actors is consistent with large partisan divides in public confidence in a host of major national institutions. Democratic respondents in our 2023 survey were much more likely than their Republican counterparts to express a great deal/quite a lot of confidence in the public school system (47% Democrats to 18% Republicans), newspapers (34% D to 12% R), and organized labor (37% D to 16% R). In contrast, Republicans were more likely than Democrats to say they have a great deal/quite a lot of confidence in small business (59% R to 43% D), the military (62% R to 46% D), the police (65% R to 31% D), the church (49% R to 23% D), and the Supreme Court (53% R to 20% D). All these differences are statistically significant at the p < .001 level. We also found a 10-percentage-point partisan divide in confidence in the medical system (50% D to 40% R, p < .05). There were no partisan differences in public confidence in banks (31% R to 26% D, p = .31) and Big Tech (23% R to 23% D, p = .91). In these comparisons, the views of independents are much closer to those of Republicans than to those of Democrats. These results are consistent with accounts showing that independents are often partisans in disguise. Many people call themselves independent to avoid conflict in a political landscape marked by partisan animosity and because independents are often seen as more trustworthy and likable than partisans (Klar and Krupnikov 2016).

We also examined trust among respondents who self-reported that they did not receive the COVID-19 vaccine. The unvaccinated generally displayed much lower trust in the nine health system actors than did respondents who said they received the vaccine. For example, the mean value of trust in federal public health experts among the vaccinated was 5.8; among the unvaccinated, it was 2.7. The unvaccinated trusted medical research scientists at M = 4.6 compared to M = 7.4 for the vaccinated. The unvaccinated also gave lower trust ratings to physicians (M = 5.4) than did the vaccinated (M = 7.5). All these differences between vaccinated and unvaccinated respondents are statistically significant at the p < .001 level.

Unvaccinated respondents are distinctive in reporting a lack of confidence in virtually all major national institutions. In contrast to the overall views of Republicans in our sample, only 37% of the unvaccinated said they had a great deal/quite a lot of confidence in the military, 38% in the police, and just 26% in the Supreme Court. A near majority (49%) of the unvaccinated did express confidence in small business, the only institution that ranked relatively highly for this population. Many strategies were tried during the pandemic to encourage the unvaccinated to get the vaccine, including messaging about personal and societal health risks (Motta et al. 2021), copartisan source cues (Sylvester et al. 2023), and behavioral strategies (Omer et al. 2021) such as leveraging the psychology of scarcity (Del Ponte, De Dominicis, and Canofari 2023). Persuasive strategies that were shown to work in studies conducted during the initial phase of the pandemic were sometimes found be no longer effective in subsequent studies, suggesting the importance of retesting interventions over time (Bokemper et al. 2021). Overall, our understanding of how to increase vaccination rates among politically and socially diverse communities remains incomplete and is an important challenge for both researchers and the public health community.

Health System Actors’ Performance during the COVID-19 Pandemic

The COVID-19 pandemic was a global health emergency that caused the deaths of more than one million Americans, disrupted the economy, and created tremendous hardship for individuals, families, and communities. The pandemic was also a stress test for the performance of the health system. In both our 2022 and 2023 surveys, we asked respondents to rate the performance of key health system actors during the COVID-19 pandemic on a 5-point scale from “poor” to “excellent.” The 2023 survey asked respondents to rate the performance of the same nine actors mentioned above. (The 2022 survey included the full list with the exception of federal public health experts). The order of the actors was randomized in each survey.

Overall, we found a close correspondence between public trust in health system actors and ratings of each actor's performance during the pandemic (fig. 2). In January–February 2022, in the midst of the omicron wave of COVID-19, 73% of participants rated nurses’ performance as excellent or very good, which was significantly higher than the rating of medical doctors (60%, p < .001), and medical research scientists (52%, p < .001). The public gave the lowest ratings to pharmaceutical companies (despite the remarkably rapid development of COVID-19 vaccines) and health insurers. Just about one in four participants rated the performance of pharmaceutical and health insurance companies highly. In our July 2023 follow-up survey—which was conducted during a period when the majority (64%) of Americans said they believed the pandemic was over in the United States (Brenan 2023)—overall public ratings in the performance of doctors and nurses declined (all p < .05) and stayed the same for medical research scientists (p = .29), but still remained notably higher than for all other actors. Pharmaceutical and health insurance companies received the lowest ratings, with just one in four or one in five respondents, respectively, rating their performance as excellent or very good.

Figure 2

Percentages of participants who rated each health care actor's performance as “excellent” or “very good.”

Source: YouGov 2022 and 2023 original surveys.

Figure 2

Percentages of participants who rated each health care actor's performance as “excellent” or “very good.”

Source: YouGov 2022 and 2023 original surveys.

Close modal

The public's ratings of the pandemic performance of key health system actors are highly polarized (online-only fig. A2). Democrats consistently rated the performance of health system actors more favorably than Republicans did. In 2022, the largest gaps between the ratings of Democrats and Republicans were for the CDC and medical research scientists. Fifty-nine percent of Democrats rated the CDC's performance as excellent or very good, compared to just 15% of Republicans, a gap of 44 percentage points. For medical research scientists, the partisan divide was also very large: 73% of Democrats and 33% of Republicans rated their performance as excellent or very good, a 40-percentage-point gap. Both differences are significant at the p < .001 level. There was a smaller but still significant partisan gap in respondents’ evaluations of the pandemic performance of other health system actors. For example, the partisan gap was only 9 percentage points for health insurers (31% for Democrats and 22% for Republicans, p < .01), 9 percentage points for nurses (80% for Democrats and 71% for Republicans, p < .01), and 23 percentage points for doctors (72% for Democrats and 49% for Republicans, p < .001).

Our 2023 follow-up survey produced broadly similar results. Table 1 displays pandemic performance ratings for each actor in the 2023 survey. The 2023 survey results again show a very high level of partisan polarization. Nurses are the only actor that received excellent or very good ratings for their pandemic performance from more than half of Republicans (66%) in 2023. Even doctors (44%) and medical research scientists (35%) failed to receive positive ratings from most Republicans.

Table 1

Ratings of Health Care Actors’ Performance amid the Pandemic

MDsNursesMedical research scientistsPharma companiesHealth insurance companiesFDACDCState public health agencyFederal public health experts
All respondents 54.8 67.4 48.8 26.7 20.6 30.2 34.7 31.8 29.7 
Party affiliation          
 Democrat 68.6 72.2 67.1 38.1 27.0 45.5 53.9 47.6 46.8 
 Republican 44.1 66.1 35.1 23.8 22.4 20.9 21.8 26.8 20.6 
 Independent 50.1 64.0 42.1 18.7 14.0 23.1 26.8 21.6 21.0 
2020 turnout          
 Voted in 2020 55.1 70.0 48.5 25.7 20.3 29.9 35.3 32.0 31.4 
 Did not vote in 2020 54.4 63.0 49.2 28.3 21.3 30.5 33.8 31.5 27.0 
Age (years)          
 18–37 60.0 63.8 53.8 33.4 26.6 34.0 36.9 35.5 32.8 
 38–51 56.5 67.4 48.9 23.9 17.9 26.3 36.1 32.0 31.9 
 52–64 47.5 66.3 45.6 21.8 16.8 27.9 34.7 29.7 26.9 
 ≥ 65 52.7 74.2 44.1 24.2 18.2 30.5 29.7 28.1 25.4 
Education          
 Low 49.1 62.7 43.7 24.1 20.8 31.8 33.6 27.5 25.9 
 Medium 48.9 65.3 40.4 22.1 18.0 22.9 29.0 27.0 22.7 
 High 66.5 74.5 61.9 33.6 22.7 34.4 40.8 41.0 40.2 
Race          
 White 57.7 73.4 48.2 27.9 17.0 27.4 33.7 29.3 28.4 
 Black 54.9 67.3 58.7 31.8 34.0 39.1 47.5 48.7 39.7 
 Hispanic 44.2 50.3 41.9 25.2 26.6 33.4 33.4 28.0 26.5 
Income          
 Low 51.1 59.6 44.7 22.9 20.9 31.1 34.1 31.7 29.3 
 Medium 54.9 70.1 47.7 23.0 19.1 30.5 33.3 29.6 27.9 
 High 64.3 78.2 57.7 37.8 23.0 29.4 39.4 37.8 36.5 
Residential status          
 Urban 57.3 63.3 54.9 35.0 32.0 39.2 42.6 43.8 39.4 
 Suburban 54.4 71.4 49.5 22.6 17.4 27.4 33.0 29.3 27.9 
 Rural 52.4 64.2 39.6 24.6 13.0 24.3 28.1 21.7 21.2 
COVID-19 vaccination status          
 Vaccinated 64.7 75.7 60.4 32.5 24.8 37.2 43.1 38.3 37.0 
 Unvaccinated 29.0 45.8 18.3 11.6 9.9 11.7 12.8 14.8 10.7 
Gender          
 Male 58.6 68.2 48.4 30.0 20.3 31.4 35.9 31.9 30.1 
 Female 51.3 66.5 49.1 23.5 21.0 28.9 33.5 31.8 29.4 
MDsNursesMedical research scientistsPharma companiesHealth insurance companiesFDACDCState public health agencyFederal public health experts
All respondents 54.8 67.4 48.8 26.7 20.6 30.2 34.7 31.8 29.7 
Party affiliation          
 Democrat 68.6 72.2 67.1 38.1 27.0 45.5 53.9 47.6 46.8 
 Republican 44.1 66.1 35.1 23.8 22.4 20.9 21.8 26.8 20.6 
 Independent 50.1 64.0 42.1 18.7 14.0 23.1 26.8 21.6 21.0 
2020 turnout          
 Voted in 2020 55.1 70.0 48.5 25.7 20.3 29.9 35.3 32.0 31.4 
 Did not vote in 2020 54.4 63.0 49.2 28.3 21.3 30.5 33.8 31.5 27.0 
Age (years)          
 18–37 60.0 63.8 53.8 33.4 26.6 34.0 36.9 35.5 32.8 
 38–51 56.5 67.4 48.9 23.9 17.9 26.3 36.1 32.0 31.9 
 52–64 47.5 66.3 45.6 21.8 16.8 27.9 34.7 29.7 26.9 
 ≥ 65 52.7 74.2 44.1 24.2 18.2 30.5 29.7 28.1 25.4 
Education          
 Low 49.1 62.7 43.7 24.1 20.8 31.8 33.6 27.5 25.9 
 Medium 48.9 65.3 40.4 22.1 18.0 22.9 29.0 27.0 22.7 
 High 66.5 74.5 61.9 33.6 22.7 34.4 40.8 41.0 40.2 
Race          
 White 57.7 73.4 48.2 27.9 17.0 27.4 33.7 29.3 28.4 
 Black 54.9 67.3 58.7 31.8 34.0 39.1 47.5 48.7 39.7 
 Hispanic 44.2 50.3 41.9 25.2 26.6 33.4 33.4 28.0 26.5 
Income          
 Low 51.1 59.6 44.7 22.9 20.9 31.1 34.1 31.7 29.3 
 Medium 54.9 70.1 47.7 23.0 19.1 30.5 33.3 29.6 27.9 
 High 64.3 78.2 57.7 37.8 23.0 29.4 39.4 37.8 36.5 
Residential status          
 Urban 57.3 63.3 54.9 35.0 32.0 39.2 42.6 43.8 39.4 
 Suburban 54.4 71.4 49.5 22.6 17.4 27.4 33.0 29.3 27.9 
 Rural 52.4 64.2 39.6 24.6 13.0 24.3 28.1 21.7 21.2 
COVID-19 vaccination status          
 Vaccinated 64.7 75.7 60.4 32.5 24.8 37.2 43.1 38.3 37.0 
 Unvaccinated 29.0 45.8 18.3 11.6 9.9 11.7 12.8 14.8 10.7 
Gender          
 Male 58.6 68.2 48.4 30.0 20.3 31.4 35.9 31.9 30.1 
 Female 51.3 66.5 49.1 23.5 21.0 28.9 33.5 31.8 29.4 

Source: YouGov 2023 original survey.

Note: Percentage of participants who rated a health care actor's performance as “excellent” or “very good.”

There is some evidence that polarization in the public's beliefs about the pandemic performance of key actors eased between 2022 and 2023. For example, the partisan gap in respondents’ evaluations of the CDC declined by 12 percentage points (but is still significant at the p < .001 level). The key exception to this trend of moderating polarization lies in ratings of state public health agencies. In 2022, the partisan gap was 10 percentage points (42% Democrats versus 32% Republicans, p < .001). It climbed to 21 percentage points in 2023, when 48% of Democrats rated state public health agencies’ performance as excellent or very good compared to 27% of Republicans (p < .001 for the partisan gap).

Table 1 also shows significant divides among different segments of the population in their respective pandemic performance ratings. In general, highly educated respondents provided more favorable ratings of health system actors than did those with less education. Urban respondents were generally more satisfied with the performance of health actors during the pandemic than were the residents of rural communities. And there were stark differences between the views of people who self-reported that they were vaccinated for COVID-19 and the views of those who said they were unvaccinated. All these differences were significant at the 5% level or better. The latter group gave extremely low ratings to the pandemic performance of all nine actors, with the partial exception of nurses.

Views on Medical Research Scientists and Public Health Experts

We now take a closer look at the public's views on two health system actors who played important roles during the COVID-19 pandemic: medical research scientists and public health experts. In our 2023 survey, we asked respondents three questions about each of these two actors: First, how much confidence do you have in the actor's competence? Second, how much confidence do you have that the actor's advice to policy makers is in the best interest of the public? Answers to these two questions were given as ratings on a 4-point scale (no confidence at all, not too much, a fair amount, a great deal). Finally, we asked respondents whether they agreed or disagreed that policy makers should follow the actor's advice about public policy.

Figure 3 displays the results. Overall, we found that the majority of the public has at least a fair amount of confidence that both medical research scientists (78%) and public health experts (69%) are competent. (The difference between ratings of the two groups is statistically significant at p < .001). Most participants expressed at least a fair amount of confidence that the advice of both medical research scientists (61%) and public health experts (60%) is in the best interest of the public (p = .72 for the comparison). Similar proportions of the sample said they believed that policy makers should follow the advice of these actors about public policy (61% for medical research scientists and 60% for public health experts, p = .65).

Figure 3

Percentages of participants who indicated they are very or somewhat confident in each actor's competence, each actor's advice, and in the appropriateness of having policy makers follow their guidance.

Notes: For the policy maker item, the 5-point scale ranged from “strongly agree” to “strongly disagree.” Hence the figure portrays the percentage of participants who at least agreed with the statement that policy makers should follow medical research scientists’ and public health experts’ advice, respectively. The only significant pairwise comparison between medical research scientists and public health experts is for competence (p < .001).

Source: YouGov 2023 original survey.

Figure 3

Percentages of participants who indicated they are very or somewhat confident in each actor's competence, each actor's advice, and in the appropriateness of having policy makers follow their guidance.

Notes: For the policy maker item, the 5-point scale ranged from “strongly agree” to “strongly disagree.” Hence the figure portrays the percentage of participants who at least agreed with the statement that policy makers should follow medical research scientists’ and public health experts’ advice, respectively. The only significant pairwise comparison between medical research scientists and public health experts is for competence (p < .001).

Source: YouGov 2023 original survey.

Close modal

These results mask large partisan differences between the views of Democrats and Republicans about these actors, however (table A1). To be sure, most Republicans expressed at least a fair degree of confidence in the competence of both medical research scientists (68%) and public health experts (55%). However, fewer than half of Republicans (48% for medical research scientists and 45% for public health experts) expressed confidence that the advice given by these actors is in the best interest of the public. And fewer than half agreed that policy makers should follow their advice (48% for medical research scientists and 45% for public health experts). The story is quite different for Democrats. Not only were the vast majority of Democrats convinced that these actors are competent (92% for medical research scientists and 90% for public health experts), but they also believed their advice was in the public's best interests (83%–85%, respectively) and that policy makers should follow it (79% and 81%, respectively). Independents fell in between on all measures but were much closer to Republicans, with differences ranging between 3 and 7 percentage points. Together, these findings speak to research suggesting that the two political parties are polarizing over what government should do as well as who should be in charge of developing and implementing public policy (Grossmann and Hopkins 2023).

Impact of COVID-19 on Trust in Personal Doctors

As shown in online-only figure A1, both Democrats (M = 7.6) and Republicans (M = 6.8) reported that they had a high degree of trust in physicians. Americans have greater confidence in their personal doctors than in the health care system as a whole (Goidel et al. 2023). As a result, doctors are often seen as uniquely influential actors in the communication and implementation of health policies (Goidel et al. 2023; Patashnik, Gerber, and Dowling 2018). Despite their expertise, however, physicians may not always provide accurate information to their patients. Physicians receive information about medical issues in part from the “broader communication environment that is increasingly defined by partisan polarization and ideologically driven news” (Goidel et al. 2023: 319). For example, research shows that the news coverage of the COVID-19 pandemic provided by conservative media outlets was significantly different from the coverage offered by the mainstream media (Motta, Stecula, and Farhart 2020). Politics can seep into the interactions between individual doctors and their patients. Physicians’ political identities have been found to shape their treatment decisions on politicized health issues. For example, Hersh and Goldenberg (2016) show that Republican physicians are more likely than Democratic physicians to discuss the health risks of marijuana, while Democratic doctors are more likely to advise patients not to keep guns at home than their Republican counterparts.

While the medical profession was once a predominately conservative profession, doctors have polarized in recent decades. The medical profession has trended in “a liberal direction as females have become more present in recent cohorts and as both males and females in the cohorts have become more liberal” (Bonica et al. 2020: 1054). There is also evidence of increasing polarization in the public's trust in their personal doctors (O'Brian and Kent 2023). In a 2021 Gallup survey, 60% of Republicans and Republican-leaning independents said they were confident in the accuracy of the advice they received from their doctor, down from 70% in 2002. Confidence among Democrats and Democratic leaners increased during this period, from 62% to 71% (Jones 2021).

One question is whether doctors’ positions in contentious debates around salient issues such as vaccination, masking, and social distancing during the COVID-19 pandemic have specifically contributed to polarization in the public's trust in doctors. O'Brian and Kent (2023) provide experimental evidence showing that priming respondents to think about COVID-19 through a partisan lens by reading a news article accusing Dr. Anthony Fauci of being tied to the Democratic party lowered trust in personal doctors among Trump voters, but it increased trust among Biden voters.

In our 2023 survey, we explored the degree to which partisan conflicts over issues like vaccination have spilled over into Americans’ trust in their own doctors by asking respondents how the COVID-19 pandemic has affected their level of trust in the doctor they have seen the most (table 2). For the vast majority of respondents (61.5%), we found that the pandemic had no impact on trust in personal doctors, 18% reported that it increased their trust, and 20.5% reported that it decreased their trust.3 However, there were important partisan differences. Republicans (28.8%) were much more likely than Democrats (9.3%) to say that the pandemic decreased their trust in their personal doctor (p < .001). Democratic respondents were more likely (28%) than Republicans (14%) to say the pandemic increased their trust. Democrats were also more likely than Republicans to say that their personal doctor encouraged vaccination and COVID-19 precautions such as masking and social distancing (84% D, 70% R; p < .001) and that doctors expressed an opinion on a national or local COVID policy measure such as lockdowns and school closings (54% D, 30% R; p < .001). Finally, Democrats were more likely (3.3 on a 5-point scale) to say they agreed with their doctor's views on COVID-19 than Republicans (2.7, p < .001).

Table 2

COVID-19 Impact on Trust in Personal Doctor and Agreement with Doctor's Views

Trust in MDMD encourage vaccineMD express opinionAgree with MD
IncreasedDecreasedYesYesMean
All respondents 18.0 20.5 75.1 35.6 3.0 
Party affiliation      
 Democrat 28.0 9.3 83.7 54.1 3.3 
 Republican 14.0 28.8 70.0 29.9 2.7 
 Independent 12.0 24.8 70.5 21.6 2.6 
COVID-19 vaccination status      
 Vaccinated 20.9 13.5 83.2 38.0 3.3 
 Unvaccinated 10.5 39.0 54.0 28.7 2.2 
Trust in MDMD encourage vaccineMD express opinionAgree with MD
IncreasedDecreasedYesYesMean
All respondents 18.0 20.5 75.1 35.6 3.0 
Party affiliation      
 Democrat 28.0 9.3 83.7 54.1 3.3 
 Republican 14.0 28.8 70.0 29.9 2.7 
 Independent 12.0 24.8 70.5 21.6 2.6 
COVID-19 vaccination status      
 Vaccinated 20.9 13.5 83.2 38.0 3.3 
 Unvaccinated 10.5 39.0 54.0 28.7 2.2 

Source: YouGov 2023 original survey.

Notes: N = 800. The trust in MD measure has three outcomes: increased, remained the same, and decreased. Agreement with the doctor is measured on a 0–4 scale ranging from strongly disagree to strongly agree.

Beliefs about Doctors’ Motivations and Abilities

To further explore whether the polarized response to the pandemic has reconfigured public beliefs about doctors, we repeated two survey experiments from a previous article published in this journal (Gerber et al. 2014). The first experiment examines public beliefs about doctors’ motivations and dispositions compared to other professional actors, such as lawyers and teachers. Gerber and colleagues (2014: 179) conducted a survey experiment in which respondents “were asked to rate their agreement with a series of six statements about the motivations of people from their randomly assigned profession. For example, those in the ‘doctors’ condition were asked how much they agreed with the statement ‘Becoming wealthier is important for doctors.’ Responses were measured on a five-point scale ranging from ‘strongly disagree’ (1) to ‘strongly agree’(5).” The authors found that the public generally viewed doctors as “harder workers,” “more interested in helping people,” “more trustworthy,” and “more caring” than those in the other professions.

We largely replicate the original Gerber and colleagues (2014) results, suggesting that there has been remarkably little change over the past decade in how the public sees the motivations of doctors compared to those of other professions. Specifically, we found that among our three professional comparisons (lawyers, teachers, and Congress members), Americans consistently perceived doctors to work harder, care more about them, be more deserving of trust, be more interested in helping others, and be less interested in becoming wealthy or gaining prestige (fig. 4; all differences statistically significant at the 5% level or better, two-tailed test, except for interest in gaining prestige compared to teachers, p = .111). Overall, teachers were evaluated closest to doctors, although doctors still were judged to be significantly harder working and more caring (p < .001). At the other end of the comparison, Congress members were judged to be the least hardworking, the least interested in helping people, the least trustworthy, and the least caring (p < .001). At the same time, members of Congress were judged to be the most interested in wealth and prestige, with lawyers a close second. Respondents judged the motivations of lawyers not quite as harshly as the motivations of members of Congress but still significantly worse than those of teachers and doctors.

Figure 4

Beliefs about other professions’ motivations and abilities compared to doctors.

Notes: The plot shows linear regression coefficients with standard errors. The coefficients indicate the difference from the baseline, which is doctors. Each dependent variable was recoded to range from 0 (strongly disagree) to 4 (strongly agree). Means for doctors: (1) work harder = 2.83; (2) interested in helping = 3.12; (3) interested in wealth = 2.46; (4) interested in gaining prestige = 2.03; (5) care about me = 2.81; (6) can be trusted = 2.75. All differences are statistically significant at the 5% level or better (two-tailed test) except for the prestige item for teachers (p = .111). The underlying regressions are available in online-only table A2.

Figure 4

Beliefs about other professions’ motivations and abilities compared to doctors.

Notes: The plot shows linear regression coefficients with standard errors. The coefficients indicate the difference from the baseline, which is doctors. Each dependent variable was recoded to range from 0 (strongly disagree) to 4 (strongly agree). Means for doctors: (1) work harder = 2.83; (2) interested in helping = 3.12; (3) interested in wealth = 2.46; (4) interested in gaining prestige = 2.03; (5) care about me = 2.81; (6) can be trusted = 2.75. All differences are statistically significant at the 5% level or better (two-tailed test) except for the prestige item for teachers (p = .111). The underlying regressions are available in online-only table A2.

Close modal

Notably, there were no significant partisan differences (online-only figs. A3–A8), indicating a broad consensus in the perception of these professional categories compared to medical doctors. Nor were views distinctive among the unvaccinated (online-only figs. A9–A14). The positive perception of the motivations of doctors compared to those of other professions would appear to rest on solid ground.

Doctors’ Cues Survey Experiment

Finally, we repeated a second survey experiment from Gerber and colleagues (2014) that provides a way to assess how the support or opposition of doctors affects public opinion regarding a public policy issue. Following Gerber and colleagues (2014: 183), respondents were asked to indicate their level of support for a “generic proposal to ‘help reduce the amount we spend on health care.’” We experimentally manipulated two dimensions: The first dimension varied the position of a doctors’ group, the American Medical Association (AMA). “One-third of the sample was told that the AMA endorsed the proposal; another third was told that the AMA opposed the proposal; the AMA's position on the proposal was not mentioned for the remaining one-third of respondents” (Gerber et al. 2014: 183). The experiment's second dimension “examined the effects of political cues” (Gerber et al. 2014: 183). Each respondent saw a statement that described the proposal as one of the following: “(a) ‘supported by congressional Democrats but opposed by congressional Republicans,’ (b) ‘supported by congressional Republicans but opposed by congressional Democrats,’ (c) ‘supported by congressional Democrats and Republicans,’ or (d) ‘supported by a bipartisan commission on deficit reduction.’ An additional group was randomly assigned to receive no political cue. These five conditions were randomly assigned, with equal probability independently of the AMA cue treatment” (Gerber et al. 2014: 183). “The outcome measure was respondents’ assessments of how the particular cue (or cues) would affect their own support for the proposal and was measured using a five-point scale ranging from ‘much less likely to support’ (−2) to ‘much more likely to support’” (2) (Gerber et al. 2014: 183–84).

Gerber and colleagues (2014) found, as expected, that partisan cues mattered. Republicans “who were presented with an endorsement cue from congressional Republicans and an opposition cue from congressional Democrats [were] more likely to say that the information would increase their support for the proposal” (Gerber et al. 2014: 186–88). The converse was true for Democratic respondents. But in a closely divided electorate, these partisan cues do not change aggregate public opinion. The support of a bipartisan commission on deficit reduction (or an endorsement cue from both congressional Democrats and Republicans) had no impact on the public's support for the proposal. In terms of moving public opinion, however, the cue that produced the highest overall public support was an endorsement cue from the AMA (alone or in combination with parties’ endorsements), suggesting that doctors have the ability to persuade citizens across the political spectrum on a policy issue. Respondents who received the AMA support cue were significantly more likely to say this cue would increase their support for the proposal, while respondents who received the AMA opposition cue were “more likely to say it would decrease their support for the proposal” (Gerber et al. 2014: 184).

The results of this survey experiment replicate the original Gerber and colleagues (2014) findings that the endorsement of the AMA has the unique potential to generate support for a health policy proposal among the public. Despite the political and economic scars from the pandemic and intense partisan polarization, the medical profession still has the capacity to engage in persuasive communication and move overall public opinion. Table 3 presents the mean values and standard deviation of the results for each arm of the experiment.

Table 3

Support for a Hypothetical Health Care Cost Control Proposal in a Survey Experiment

Hypothetical proposal support conditionMeanSD
Treatment AMA support proposal 2.63 0.96 
 AMA oppose proposal 2.25 1.24 
 D support, R oppose 2.31 1.28 
 R support, D oppose 2.31 1.19 
 Bipartisan support 2.46 0.90 
 Bipartisan commission support 2.28 0.98 
 AMA and D support 2.28 1.21 
 AMA oppose and D support 1.81 1.35 
 AMA and R support 2.49 1.02 
 AMA oppose and R support 1.91 1.36 
 AMA, D, and R support 2.61 0.93 
 AMA oppose, D and R support 2.14 0.91 
 AMA and bipartisan commission support 2.25 0.97 
 AMA oppose, bipartisan commission support 2.18 1.18 
Hypothetical proposal support conditionMeanSD
Treatment AMA support proposal 2.63 0.96 
 AMA oppose proposal 2.25 1.24 
 D support, R oppose 2.31 1.28 
 R support, D oppose 2.31 1.19 
 Bipartisan support 2.46 0.90 
 Bipartisan commission support 2.28 0.98 
 AMA and D support 2.28 1.21 
 AMA oppose and D support 1.81 1.35 
 AMA and R support 2.49 1.02 
 AMA oppose and R support 1.91 1.36 
 AMA, D, and R support 2.61 0.93 
 AMA oppose, D and R support 2.14 0.91 
 AMA and bipartisan commission support 2.25 0.97 
 AMA oppose, bipartisan commission support 2.18 1.18 

Source: YouGov 2023 original survey.

Note: SD = standard deviation.

First, we examined the results for the full sample. As a baseline, we considered the scenario in which the AMA supports the health care cost control proposal. Respondents’ average support for the proposal was quite high (M = 2.63, p < .001 on a 0–4 scale, where 4 indicates maximum support). Looking at the 13 experimental alternatives, we found that no other endorsement cue generated meaningful net support. All coefficients are negative and most of them are statistically significant at the p < .05 level or better, indicating that when other actors endorsed the proposal, public support declined compared to the baseline of AMA support (fig. 5). Remarkably, this is the case even when the support cue comes from a bipartisan commission on deficit reduction. The only endorsement cue that boosted the public's overall support for the proposal is the AMA's standalone endorsement.

Figure 5

Support for a hypothetical health care cost control proposal in a survey experiment from the YouGov 2023 sample.

Notes: The plot shows linear regression coefficients with standard errors. The plotted coefficients indicate the difference from the baseline, which is “supported by the AMA.” The underlying regression is available in online-only table A3.

Figure 5

Support for a hypothetical health care cost control proposal in a survey experiment from the YouGov 2023 sample.

Notes: The plot shows linear regression coefficients with standard errors. The plotted coefficients indicate the difference from the baseline, which is “supported by the AMA.” The underlying regression is available in online-only table A3.

Close modal

Next, we examined the results by partisanship (online-only fig. A15), comparing them to the same baseline condition as before (AMA support). We began by subsetting the sample to look at Democratic respondents in our survey. Compared to the baseline of an AMA support cue (M = 2.93, p < .001), an endorsement by congressional Democrats was no more effective at raising support than the AMA cue (M = 3.29, p = .239), nor was a combined endorsement of the AMA and congressional Democrats (M = 3.08, p = .642). Predictably, and in line with the results of the 2014 experiment, endorsement by congressional Republicans significantly reduced support among Democratic participants (M = 2.04, p < .05), also when combined with an AMA endorsement (M = 2.13, p < .05). When congressional Republicans supported the proposal and the AMA was opposed to it, Democratic respondents were even more opposed (M = 1.08, p < .001).

Looking at Republican respondents, we found that an endorsement by congressional Republicans (M = 3.11) was no more effective at raising support than an AMA endorsement (M = 2.45, p = .136). However, when both the AMA and congressional Republicans endorsed the proposal, the boost in support was substantial and statistically significant (M = 3.39, p < .01). This finding is important because it shows an asymmetry in parties’ influence over voters. The cue from congressional Democrats did not move its voters, whereas the cue from congressional Republicans did have an influence, albeit only when the proposal also garnered AMA support. Finally, we note that an endorsement by congressional Democrats reduced support among Republicans (M = 1.23, p < .01), even when it occurred in conjunction with an AMA endorsement (M = 1.57, p < .05).

Lastly, we found that for both Democrats and Republicans, endorsement by a bipartisan commission did not generate more support than an AMA endorsement alone (all p values not significant).

In sum, our results suggest that for a health policy cost control proposal to gain net support among the American public as a whole, doctors remain the most persuasive source of information. We found a similar pattern when conducting the analysis by vaccination status (online-only fig. A16). As noted earlier, the unvaccinated report less trust in doctors than do the vaccinated. The experimental results suggest, however, that despite their low level of trust in doctors, the unvaccinated gave moderate support to the proposal when the AMA provided an endorsement cue (M = 2.16). This moderate level of support was surpassed among unvaccinated respondents only when they received an endorsement cue from congressional Republicans (M = 3.05; p < .001 for the comparison). Moreover, the unvaccinated were responsive to opposition cues from the AMA when the proposal was also supported by the Democratic party (M = 1.00; p < .05). In sum, our results suggest that the medical profession has some persuasive communication capacity with the unvaccinated, at least in the context of a health care cost control proposal, although the unvaccinated remain more responsive to congressional Republicans.

Discussion

We conducted two national surveys in 2022 and 2023 to gauge how the public thinks about the trustworthiness and pandemic performance of key health system actors, including physicians, medical research scientists, and public health agencies. We also repeated two previous survey experiments (Gerber et al. 2014) to examine whether the public's views of the motivations of doctors has changed over the past decade and whether doctors are able to persuade the public to support heath policy reform. Our results suggest that the public's beliefs are polarized on many dimensions but not all of them. Both Democrats and Republicans have a high degree of confidence in nurses and doctors, and they view pharmaceutical firms and health insurers with great suspicion. For about 30% of Republican respondents the COVID-19 pandemic did decrease their self-reported trust in their personal doctor, although for the vast majority of Republicans the pandemic had no effect. The key partisan difference is that Democrats have much more favorable views on how well public health actors and medical research scientists performed during the pandemic. This partisan gap does not primarily reflect differences in assessments of these actors’ competence. Rather, Republicans are less likely than Democrats to believe that the advice of public health experts and medical scientists is in the public interest, and are less convinced that policy makers should follow this advice.

Skepticism of technocratic experts is not a new phenomenon in the United States. In his book Anti-Intellectualism in American Life, historian Richard Hofstadter (1963: 37) argued that this tendency has deep roots in a “culture which has always set a premium on government by the common man.” As Brewer (2016: 253) notes, “American populism tends to be highly resentful of being told by experts ‘we know best.’” During the pandemic, public health agencies and other health system actors made far-reaching recommendations about the operation of businesses, schools, churches, and other social institutions. Many Republicans appear to believe that during public health crises, policy makers should listen to a broad set of voices, not just health experts. One of the most important legacies of the political conflicts over the responses to the pandemic is partisan disagreement about the width of public health's lane and the degree to which public health decisions should reflect expertise or the views of ordinary citizens.4

While skepticism of scientific beliefs can occur on both sides of the political spectrum depending on the issue, the Republican party's social and economic conservatism has long given it something of an anti-expert bent. What is new is the growing “diploma divide” in the American electorate. During the past few decades, Democrats have improved their electoral standing with white voters who graduated from college, while Republicans have gained strength among white voters who do not have college degrees (Harris 2018). Many Democrats today believe that that the views of credentialed experts should be respected on issues from COVID-19 to climate change, while many Republicans express skepticism toward degree-holding experts, who are often portrayed in conservative media outlets as being out of touch with the values and beliefs of ordinary citizens (see Grossmann and Hopkins 2023). Our survey results are consistent with the idea that the parties are polarizing over the role of experts. However, our findings also suggest that the key source of the partisan divide (at least in the health care arena) is not that Republicans believe that experts are incompetent, but rather that Democrats strongly believe that the advice of experts is in the public interest and should be followed, while a narrow majority of Republicans disagrees with this view.

Taken together, our findings suggest that the public health community must work harder to build “the credibility and cultural competence” required to perform its work “across vast swathes of America” (Pollack 2021). Whether the solution lies in affirmative efforts to recruit conservatives into the ranks of public health (as Pollack 2021 proposes) or in other measures, the public health community must recognize that large segments of the public are skeptical about its recommendations for public policy. At a time when life expectancy in the United States is falling behind that of peer countries (Achenbach et al. 2023), it is vital to build the broadest possible support for public health initiatives. It will be important for doctors and nurses to play a leadership role in this effort, given that the public continues to trust them more than other actors.

Our study has several limitations that should be addressed in future research. The surveys were conducted during and in the immediate aftermath of the emergency phase of the COVID-19 pandemic. It will be important to investigate whether the public's trust in health system actors changes over time. In addition, our survey experiment testing the persuasive influence of physicians on a public policy issue presented respondents with a highly simplified representation of the world and may therefore overstate the effects that would occur outside the experimental context (Barabas and Jerit 2010). In this experiment, the health care cost control proposal was generic and lacked rich substantive content. In addition, the experiment did not expose respondents to specific counterarguments from opponents, such as arguments that efforts to reduce costs will limit patients’ access to effective treatments or worsen health disparities. In an era of polarization, it is rare for major health reform issues not to spark heated partisan debate. As Gerber and colleagues (2014: 193) noted, we therefore cannot assume that a medical association's endorsement of a proposal “would result in similar effects when other information is available to citizens.”

Notes

1.

Replication files are available on the Harvard Dataverse (see Del Ponte 2023).

2.

To allow for a precise comparison with Gerber and colleagues (2014), we employ a 3-point partisanship scale in this analysis. Partisan leaners and pure independents are excluded.

3.

The share of Americans who say the pandemic affected their trust in their personal doctor increased somewhat over time. In a national survey carried out between December 29 and January 26, 2021 (ABIM Foundation/ NORC 2021), 77% of respondents said that their trust in the doctor they have seen the most had remained the same, 11% said their trust had increased, and 12% said their trust had decreased.

4.

Research suggests that Democrats have a much more expansive view of public health than do Republicans. For example, about six in 10 Democrats believe that reducing racism and preventing gun violence are within the responsibilities of public health agencies; fewer than one in four Republicans accept this view (Findling, Blendon, and Benson 2022).

References

ABIM Foundation/NORC
.
2021
. “
Surveys of Trust in the US Health Care System
.”
May
21
. https://www.norc.org/PDFs/ABIM%20Foundation/20210520_NORC_ABIM_Foundation_Trust%20in%20Healthcare_Part%201.pdf.
Achenbach, Joel, Keating, Dan, McGinley, Laurie, Johnson, Akilah, and Chikwendiu, Jahi.
2023
. “
America's Epidemic of Chronic Illness Is Killing Us Too Soon
.”
Washington Post
,
October
4
. https://www.washingtonpost.com/health/interactive/2023/american-life-expectancy-dropping/.
Barabas, Jason, and Jerit, Jennifer.
2010
. “
Are Survey Experiments Externally Valid?
American Political Science Review
104
, no.
2
:
226
42
.
Barry, Colleen L., Anderson, Kelly E., Han, Hahrie, Presskreischer, Rachel, and McGinty, Emma E.
2021
. “
Change over Time in Public Support for Social Distancing, Mask Wearing, and Contact Tracing to Combat the COVID-19 Pandemic among US Adults, April to November 2020
.”
American Journal of Public Health
111
, no.
5
:
937
48
.
Blendon, Robert J., and Benson, John M.
2022
. “
Trust in Medicine, the Health System, and Public Health
.”
Daedalus
151
, no.
4
:
67
82
.
Bokemper, Scott E., Gerber, Alan S., Omer, Saad B., and Huber, Gregory A.
2021
. “
Persuading US White Evangelicals to Vaccinate for COVID-19: Testing Message Effectiveness in Fall 2020 and Spring 2021
.”
Proceedings of the National Academy of Sciences
118
, no.
49
: article ID e2114762118. https://doi.org/10.1073/pnas.2114762118.
Bonica, Adam, Rosenthal, Howard, Blackwood, Kristy, and Rothman, David J.
2020
. “
Ideological Sorting of Physicians in Both Geography and the Workplace
.”
Journal of Health Politics, Policy and Law
45
, no.
6
:
1023
57
. https://doi.org/10.1215/03616878-8641555.
Brenan, Megan.
2023
. “
Most Now Say Pandemic Is Over, but Normalcy Remains Elusive
.” Gallup,
June
28
. https://news.gallup.com/poll/507800/say-pandemic-normalcy-remains-elusive.aspx.
Brewer, Mark D.
2016
. “
Populism in American Politics
.”
Forum
14
, no.
3
:
249
64
.
Cox, Daniel A., Mills, Anthony M., Banks, Ian R., Hammond, Kelsey Eyre, and Gray, Kyle.
2023
. “
America's Crisis of Confidence: Rising Mistrust, Conspiracies, and Vaccine Hesitancy after COVID-19
.” Survey Center on American Life,
September
28
. https://www.americansurveycenter.org/research/americas-crisis-of-confidence-rising-mistrust-conspiracies-and-vaccine-hesitancy-after-covid-19/.
Del Ponte, Alessandro.
2023
. “
Replication Data for: Polarization, the Pandemic, and Public Trust in Health System Actors
.” Harvard Dataverse,
October
10
. https://doi.org/10.7910/DVN/TR0Q2J.
Del Ponte, Alessandro, De Dominicis, Audrey, and Canofari, Paolo.
2023
. “
Initiatives Based on the Psychology of Scarcity Can Increase COVID-19 Vaccinations
.” arXiv,
October
5
. https://arxiv.org/pdf/2310.03689.pdf.
Findling, Mary G., Blendon, Robert J., and Benson, John M.
2022
. “
Polarized Public Opinion about Public Health during the COVID-19 Pandemic: Political Divides and Future Implications
.”
JAMA Health Forum
3
, no.
3
: article ID e220016. https://doi.org/10.1001/jamahealthforum.2022.0016.
Gadarian, Shana Kushner, Goodman, Sara Wallace, and Pepinsky, Thomas B.
2021
. “
Partisanship, Health Behavior, and Policy Attitudes in the Early Stages of the COVID-19 Pandemic
.”
PLoS One
16
, no.
4
: article ID e0249596.
Gerber, Alan S., Patashnik, Eric M., Doherty, David, and Dowling, Conor M.
2014
. “
Doctor Knows Best: Physician Endorsements, Public Opinion, and the Politics of Comparative Effectiveness Research
.”
Journal of Health Politics, Policy and Law
39
, no.
1
:
171
208
.
Goidel, Kirby, Callaghan, Timothy, Washburn, David J., Nuzhath, Tasmiah, Scobee, Julia, Spiegelman, Abigail, and Motta, Matt.
2023
. “
Physician Trust in the News Media and Attitudes toward COVID-19
.”
Journal of Health Politics, Policy and Law
48
, no.
3
:
317
50
.
Gollust, Sarah E., Nagler, Rebekah H., and Fowler, Erika Franklin.
2020
. “
The Emergence of COVID-19 in the US: A Public Health and Political Communication Crisis
.”
Journal of Health Politics, Policy and Law
45
, no.
6
:
967
81
.
Grossmann, Matt, and Hopkins, David.
2023
. “
Solving Problems or Imposing Values? Party Polarization over Policy Experts
.” Paper presented at the American Political Science Association Annual Meeting and Exhibition, Los Angeles Convention Center,
August
31–September 3.
Harris, Adam.
2018
. “
America Is Divided by Education
.”
Atlantic
,
November
7
. https://www.theatlantic.com/education/archive/2018/11/education-gap-explains-american-politics/575113/
Harvard University/RWJF (Robert Wood Johnson Foundation)
.
2021
. “
The Public's Perspective on the United States Public Health System
.”
May
13
. https://www.rwjf.org/en/insights/our-research/2021/05/the-publics-perspective-on-the-united-states-public-health-system.html.
Hersh, Eitan D., and Goldenberg, Matthew N.
2016
. “
Democratic and Republican Physicians Provide Different Care on Politicized Health Issues
.”
Proceedings of the National Academy of Sciences
113
, no.
42
:
11811
16
. https://doi.org/10.1073/pnas.1606609113.
Hofstadter, Richard.
1963
.
Anti-Intellectualism in American Life
.
New York
:
Vintage
.
Jones, Jeffrey M.
2021
. “
Republicans Less Trusting of Doctor's Advice than in the Past
.” Gallup,
December
7
. https://news.gallup.com/poll/357821/republicans-less-trusting-doctor-advice-past.aspx.
Klar, Samara, and Krupnikov, Yanna.
2016
.
Independent Politics
.
New York
:
Cambridge University Press
.
Larsen, Bradley J., Ryan, Timothy J., Greene, Steven, Hetherington, Marc J., Maxwell, Rahsaan, and Tadelis, Steven.
2022
. “
Using Donald Trump's COVID-19 Vaccine Endorsement to Give Public Health a Shot in the Arm: A Large-Scale Ad Experiment
.” Stanford Institute for Economic Policy Research,
March
28
. https://siepr.stanford.edu/publications/politics-and-media/using-donald-trumps-covid-19-vaccine-endorsement-give-public-health.
Levi, Margaret, and Stoker, Laura.
2000
. “
Political Trust and Trustworthiness
.”
Annual Review of Political Science
3
, no.
1
:
475
507
.
Motta, Matt, Stecula, Dominik, and Farhart, Christina.
2020
. “
How Right-Leaning Media Coverage of COVID-19 Facilitated the Spread of Misinformation in the Early Stages of the Pandemic in the US
.”
Canadian Journal of Political Science
53
, no.
2
:
335
42
.
Motta, Matt, Sylvester, Steven, Callaghan, Timothy, and Lunz-Trujillo, Kristin.
2021
. “
Encouraging COVID-19 Vaccine Uptake through Effective Health Communication
.”
Frontiers in Political Science
3
: article ID 630133.
O'Brian, Neil, and Kent, Thomas Bradley.
2023
. “
Partisanship and Trust in Personal Doctors: Causes and Consequences
.” Unpublished manuscript.
Omer, Saad B., Benjamin, Regina M., Brewer, Noel T., Buttenheim, Alison M., Callaghan, Timothy, Caplan, Arthur, Carpiano, Richard M., Clinton, Chelsea, DiResta, Renee, and Elharake, Jad A.
2021
. “
Promoting COVID-19 Vaccine Acceptance: Recommendations from the Lancet Commission on Vaccine Refusal, Acceptance, and Demand in the USA
.”
Lancet
398
, no.
10317
:
2186
92
.
Patashnik, Eric, Gerber, Alan, and Dowling, Conor.
2018
.
Unhealthy Politics: The Battle over Evidence-Based Medicine
.
Princeton, NJ
:
Princeton University Press
.
Pollack, Harold A.
2021
. “
Why Public Health Experts Aren't Reaching Conservatives on COVID-19
.”
Politico
,
August
12
. https://www.politico.com/news/magazine/2021/08/12/conservative-public-health-covid-conservative-affirmative-action-503448.
SteelFisher, Gillian K., Findling, Mary G., Caporello, Hannah L., Lubell, Keri M., Melville, Kathleen G. Vidoloff, Lane, Lindsay, Boyea, Alyssa A., Schafer, Thomas J., and Ben-Porath, Eran N.
2023
. “
Trust In US Federal, State, and Local Public Health Agencies during COVID-19: Responses and Policy Implications
.”
Health Affairs
42
, no.
3
:
328
37
.
Sylvester, Steven, Motta, Matthew, Trujillo, Kristin Lunz, and Callaghan, Timothy.
2023
. “
Vaccinating across the Aisle: Using Co-Partisan Source Cues to Encourage COVID-19 Vaccine Uptake in the Ideological Right
.”
Journal of Behavioral Medicine
46
, nos.
1–2
:
311
23
.

Supplementary data