Health Promotion International Advance Access published November 4, 2014 Health Promotion International doi:10.1093/heapro/dau094

# The Author 2014. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]

Confidence in government and vaccination willingness in the USA 1

Department of Sociology, The University of Haifa, Haifa, Israel and 2The Ohio State University, Columbus, USA *Corresponding author. E-mail: [email protected]

SUMMARY The most recent internationally widespread disease outbreak occurred during the flu season of 2009 and 2010. On April 2009, the first cases of influenza A (H1N1) (Popularly called, Swine Flu) were confirmed in the USA and UK following a novel virus that was first identified in Mexico. As the virus spread rapidly, the risk of morbidity and mortality increased in several countries. In this paper, we rely on the social cognitive theory of risk to assess the willingness of the US public to comply with vaccination and reduce the risk of sickness and death from the flu. We conduct a secondary data analysis of the Pew Research for the People and Press October 2009 and investigate the factors associated with

willingness to take the swine flu vaccine (n ¼ 1000). The findings indicate that the decision to take the swine flu vaccination was highly polarized across partisan lines. Controlling for education, income and demographic factors, the likelihood of taking the vaccine was associated with party identification. Individuals that identified themselves as Democrats were more likely to be willing to take the swine vaccine than individuals that identify themselves as Republicans and Independents. Confidence in the ability of the government to deal with the swine flu crisis seems to explain party identification differences in the willingness to take the vaccine. The implications of the findings are discussed.

Key words: H1N1; vaccination; sociology; health

INTRODUCTION The most recent pandemic occurred during the flu season of 2009 and 2010. On April 2009, the first cases of influenza A (H1N1) (Swine Flu), which was first identified in Mexico, were confirmed in both the USA and UK. On 11 June 2009, The World Health Organization (WHO) declared the H1N1 outbreak to be the first worldwide pandemic in more than 40 years (Cohen and Carter, 2010). By the year end, 203 000 people had died worldwide from H1N1 (McNeil, 2013). In the USA, the H1N1 virus had killed 12 500, hospitalized 275 000 and sickened 61 million (Mesch et al., 2013). In spite of the fact that vaccination is one of the greatest public health success stories and the

best defense medicine has against the risk of acquiring infectious diseases (Stern and Markel, 2005), there was no vaccine available for protection against the H1N1. However, by August 2009 several US companies were underway in researching, manufacturing and field testing a vaccine with a view to having it available for distribution early fall 2009 (Ravnovich, 2009). During the winter of 2010, the vaccination program had become a heated public issue, as an increasing segment of the public became hesitant to take the vaccine. Between May 2009 and January 2010, the percentage of unvaccinated adults intending to be vaccinated fell from 50 to 16% (Gidengil et al., 2012). Public health practitioners as well as government officials were surprised at both the widespread and critical discussion of the advantages and Page 1 of 9

Downloaded from http://heapro.oxfordjournals.org/ at Michigan State University on February 16, 2015

GUSTAVO S. MESCH1* and KENT P. SCHWIRIAN2

Page 2 of 9

G. S. Mesch and K. P. Schwirian

VACCINATION, OPPOSITION AND THE MEDIA The reluctance of many to take the Swine Flu vaccination in 2009– 2010 mirrored earlier responses in the history of vaccination. Indeed, some have argued that the anti-vaccionists of today in the USA and UK are the direct intellectual descendants from those in the late 19th century (Wolfe and Sharp, 2002). When seen from the standpoint of cultural cognition theory, there has been a long-run fundamental tension over vaccination between the individual actor’s agency and the collective agency of the state and the public health authorities. These collective actors have attempted to provide individual actors with what they see as the means to safeguard or improve individual actors’ health even when the individual actors refuse or resist those means. The history of the tension between vaccination and opposition to it is rooted in the long-standing war against smallpox as early as 1840 (Fitzpatrick, 2005; Gardner, 2011). Many believed that the vaccine was dangerous and more risky than awaiting infection. The press at the time played on these fears by publishing news articles, commentaries and cartoons in widely read sources such as the London Times and more scholarly publications such as the Westminster Review and Lancet (Maugh, 2009).

In England, opposition to vaccination sharpened when between 1840 and 1898 the British government passed a series of laws requiring or encouraging vaccinations (Wolfe and Sharp, 2002). In America, the Anti-Vaccination Society of America was formed in New York in 1879, and this was followed by the creation of similar organizations in other cities (Wolfe and Sharp, 2002). These organizations fought the growing number of city and state laws requiring vaccination that were often brutally enforced as in 1901, when 250 New York City agents in the middle of the night went through tenement houses in the ‘Little Italy’ section of the city and forcibly vaccinated everyone they found (Willrich, 2011). The required vaccination issue came to a head in 1904 when the US Supreme Court ruled that the states had a right to require vaccination during epidemics but they also had to make provision for reasonable exclusions (Willrich, 2011). The current opposition to vaccination is informed by two major events carried widely by the media. One was a television broadcast on 19 April 1982; Washington DC television station WRC aired what became a television award-winning special documentary, ‘Diphtheria, Pertussis and Tetanus (DPT): Vaccine Roulette’ (Offit, 2011). The film caused a major uproar by claiming that the standard DPT vaccination routinely administered to children was dangerous and caused brain damage (McNeil, 2008). Following the film and the extensive media coverage of it, there was a decrease in vaccination rates in the USA and a rise in the number of lawsuits against vaccine manufactures. The second event occurred in 1998 when the Lancet published a paper, suggesting that there was a connection between the MMR (mumps, measles and rubella) vaccine and ASD (the autism spectrum disorder) (Novella, 2010). Although the study was eventually retracted, it still played a major role in the development of further resistance to vaccination (Willrich, 2011). In 2008, a US national study reported that 24% of the adult respondents said that the vaccine may cause autism and that it was safer not to have children vaccinated (Harvey, 2008). In recent years, the anti-vaccination movement relies on a greater distrust of government, science and public health than in the past, and that distrust becomes reflected in an increased opposition to vaccination (Gauchat, 2008; Tavernise, 2012). This accumulated distrust by many actors set the stage for the general response to the vaccination program developed in the fight against H1N1. This outbreak became characterized by a high level of

Downloaded from http://heapro.oxfordjournals.org/ at Michigan State University on February 16, 2015

disadvantages of taking the vaccine. Emerging from this public discourse was a significant resistance to taking the vaccine. Why did so many people resist taking the vaccine in the face of growing media reports of H1N1 high rates of mortality and morbidity? (Mesch et al., 2013). As the epidemic spread and the call to vaccinate increased, it soon became clear that the debates over the wisdom of taking the H1N1 vaccination involve tensions among political identification, trust and scientific knowledge (Gauchat, 2012). In this paper, we investigate the effect of these factors on the intent to take the Swine Flu H1N1 vaccination. In doing this, we rely on cultural cognition theory that takes an agentic perspective on behavior that views the individual actor as being embedded in a complex sociocultural context that contains numerous potential guides and motivators contributing to the decision to act in specific situations, such as the decision to take the flu vaccination (Bandura, 2001, 2002).

Confidence in government and vaccination willingness in the USA

THEORY AND RESEARCH QUESTIONS While the reasons for or against a specific vaccination in a specific circumstance may vary, we suggest that there are common underlying factors that shape attitudes and behaviors. These factors are confidence in government, political ideology and media influence. In this paper, we examine the relative effect of each of these three factors on intentions of American adults to take the Swine Flu vaccination during the 2009 –2010 pandemic. Confidence in government At the heart of the vaccination issue is trust or confidence in government and health authorities (Hobson-West, 2007). For people to follow public health admonitions, a degree of confidence in the supervising authority’s ability to deal with the risk is required. The outbreak of Swine Flu and the accompanying call for mass vaccination represented a social risk, and there was a great public concern over the production of the Swine Flu vaccine. The public distrust was further aggravated by media reports that New York healthcare workers were complaining about the regulation that they be required to be vaccinated (Steinhauer, 2009). Confidence in the government’s ability to deal with a public health crisis is not universal. In the past, opposition to vaccination mandates has been part of a more general resistance to government actions (Nelson, 2010). Indeed, as Gordon Gauchat [(Gauchat, 2012), p. 167] has argued, ‘Science has always been politicized.’ However, the politicization of opposition today is more diffuse than

earlier in that the degree of opposition to vaccination programs varies by the disease, composition of vaccine and party identification. Mooney (Mooney, 2005) has argued that over the last 40 years, the political ideology of conservatives has become increasingly skeptical and disapproving of the institution of science and of the nature of the connections among science, government and Big Pharma. Political ideology The cultural cognition theory of risk asserts that individuals selectively attend to risk in a way that reflects their membership in social groups such as political party and religious affiliation (Kahan and Braman, 2006). According to this view, individuals have a tendency to selectively credit and dismiss information in a manner that confirms previous beliefs and cultural dispositions shaped by their membership or support of social and political groups (Kahan et al., 2010). In other words, individuals have a tendency to fit their view to those of others with whom they share some important, self-identifying commitments. Studies have found that party identification was associated with the perception of risk in issues such as global warming, gun risk perception and HPV vaccination (Kahan et al., 2010). The authors conclude that party identification and worldview values affect fact perceptions of risk and their actions. In political science, the view that ideology and party identity shape citizens positions on issues has wide support. A longitudinal study that investigated the effects of ideological identity (conservative versus liberal), ideology and party identification on the readiness to adopt attitudes toward new political issues found that identification as conservative or liberal and party identity influenced responses to new issues (Malka and Lelkes, 2010). Pre-existing ideological and party identities lead people to follow cues on what stance they should adopt that is consistent with their previous ideology and party identification. In a similar vein, a study that investigated the effects of ideology and party identification on the 2008 US election candidate choice found significant effects of party identification on voting for candidates (Jacoby, 2010). A relevant issue is whether party identification shapes ideology. Studies show that party identification shapes the positions citizens take on political and risk issues (Goren et al., 2009). This position is consistent with the cultural cognition theory.

Downloaded from http://heapro.oxfordjournals.org/ at Michigan State University on February 16, 2015

fear of infection in a significant segment of the population (Mesch et al., 2013). Also, at this time, many older adults already had a skeptical acquaintance with Swine Flu. In 1976, there was an outbreak of Swine Flu (H1N1) at Fort Dix in New Jersey. President Jerry Ford quickly ordered a nationwide vaccination program to include 220 million Americans. The vaccination campaign lasted only a few months when it became clear that the epidemic did not materialize. Only one death resulted from Swine Flu although some side effect illnesses were attributed to the injection. The politics accompanying the failure of the campaign became known as the ‘Swine Flu Fiasco’ and ‘the epidemic that never was.’ (Neustadt and Fineberg, 1983).

Page 3 of 9

Page 4 of 9

G. S. Mesch and K. P. Schwirian

RESEARCH QUESTIONS Given confidence in government, political ideology and attention to the media, we ask: When considered together, are the three—trust, identification and attentiveness—associated with willingness to take the vaccination? Furthermore, does trust in government explain the association between political identification and willingness to take vaccine, should it exist? There are two different expectations for answering this question. One argument implies that the news media environment is characterized more by reinforcement seeking than by triangulation, thereby forging and sustaining bipartisan consensus around even non-partisan issues such as flu vaccination programs. Thus, the explanation of the gap between Republicans and Democrats lies in the different information streams to which Republican and Democrats exposed themselves.

Following this argument, one can expect that the effect of political identification will be reduced or becomes non-statistically significant after controlling for exposure to the media (Baum, 2011). An alternative argument implies that party identification effects are the result of differential confidence in the ability of the government to deal with the pandemic crisis (Gauchat, 2012). ADDITIONAL FACTORS Previous studies have shown that influenza vaccination rates differ according to ethnic/racial status. From 1989 to 2004, Latino and AfricanAmerican older persons consistently experienced 10 and 20% lower vaccination coverage, respectively, than did their White contemporaries. Reasons for the substantial racial/ethnic disparity involve many factors (Hutchins et al., 2009). For this reason, we control for race and ethnicity in our study. Gender has been an important factor in health behavior (Gabe et al., 2005). This had been reflected recently in a research that has indicated that men who report hegemonic masculinity attitudes feel less vulnerable to illness and are therefore less likely to receive preventive care, including vaccination than women (Springer and Mouzon, 2011). The association between education and the willingness to receive a vaccine is unclear. Some studies had reported a direct effect, i.e. the propensity to vaccination increases with more education and household income (Linn et al., 2010). Yet, more recent reviews have argued that the association is not direct but mediated by perceived risks. A large review of studies found that the most important reasons for taking the vaccine are personal risk perceptions, including the perceived risk of being infected by the virus, the perceived risk of severe illness from being infected and the perceived risk of negative vaccine side effects. As education is positively associated with health awareness and risk perception, it seems that education might have an effect on the willingness to take the vaccine (Nguyen et al., 2011). In addition to education, we have included two other status measures: income and homeownership. METHODS AND DATA We conducted a secondary data analysis of the Pew Research for the People and Press October

Downloaded from http://heapro.oxfordjournals.org/ at Michigan State University on February 16, 2015

Attention to the media The cultural cognition theory of risk assumes that the perception of risk is also shaped by social groups to which individuals belong. Consequently, the mass media are viewed as being an important influence in people’s beliefs, opinions and attitudes. Attention, coverage and salience given to an issue shape which issues people pay attention to and regularly follow (Cobb and Elder, 1972, 1983; Dearing and Rogers, 1996). In highly contested and politicized issues, alternative or competing frames in news coverage attempt to persuade the public to adopt one perspective over another. Recent health research has shown the influence of the media on (1) the perception of disease—diseases that are covered more heavily in the media are perceived by people as being more severe (Young et al., 2008); (2) the flu vaccination rate—media attention leads to upward spikes in vaccination rates (Yoo et al., 2010). The Swine Flu pandemic was extensively covered by the media. The WHO and the CDC issued several alerts and warnings over 2009 – 2010 that were widely carried in the press, on radio and on television. Research has shown that the media coverage of the Swine Flu outbreak was closely followed by a large segment of the population, and the extent to which people followed the flu news was highly predictive of people’s concern of possible infection (Mesch et al., 2013).

Confidence in government and vaccination willingness in the USA

DEMOGRAPHIC MEASURES The demographic variables in the past have been shown to be related to both health behaviors and politics. Age was measured by 13 age categories from 18 –20 years old to 75 and older. Gender is a dummy variable indicating 1 for men and 0 for women. Marital status was coded as a dummy variable when 1 indicated married or cohabitation and 0 other. Education was a series of three dummy variables coded as high school or less, college education and graduate education. Income was measured using 5 categories, from low to higher income. Homeownership was a dummy variable with 1 as owner and 0 for renter. Race/ Ethnicity was introduced as series of three dummy variables: white/Caucasian, African-American and Hispanic. The number of Asian-Americans was too small and was excluded from the analysis. FINDINGS The results of the survey indicate that the US public was divided in its willingness to take the Swine flu vaccine. Only 50.4% indicated that they would take the vaccine. (The table is available upon request.) A greater percentage of African-Americans (60.4) and Hispanics (63.6) were willing than Whites (47.6) to take the vaccine. A larger percentage of renters (56.0) were willing than owners (47.7). A larger percentage of those with confidence in government (59.6) were willing than were those with less confidence (32.2), and a larger percentage of Democrats (63.7) were willing than Republicans (43.0) and Independents (43.1), and this party identification finding provides a measure of support for expectations of an association between party identification and willingness to take the vaccine. Those who follow the media (57.9) were more willing than those who did not (30.9). No significant differences were found for age, gender, education marital status. Next, we conducted a logistic regression analysis predicting the willingness to take the vaccination. The model estimated includes endogenous variables ( party identification, following the news on the flu and confidence in the government). We used a two-step logistic regression employing instrumental variables (Rassen et al., 2009; Bollen, 2012). This approach helps to solve the problem of correlated error terms among the variables on the estimates of the relationships of the key independent variables to the dependent

Downloaded from http://heapro.oxfordjournals.org/ at Michigan State University on February 16, 2015

2009. The institute conducts regular surveys of public opinion, and the survey conducted in October 2009 inquired on the willingness of US residents to take the Swine Flu. The survey is based on landline completed telephone interviews among a nationwide sample of 1000 adults, 18 years of age or older, conducted under the direction of ORC (Opinion Research Corporation). Interviews were conducted from September 30 to 4 October 2009. As many as seven attempts were made to contact every sampled telephone number. Calls were staggered over times of day and days of the week to maximize the chance of making contact with potential respondents. At least one daytime call was made to each phone number in an attempt to find someone at home if necessary. Interviewing was spread as evenly as possible across the 5 days in field. The response rate for the landline sample was 20.3%. Weighting was used to compensate for sample designs and patterns of non-response that might bias results. Weights were trimmed to prevent individual interviews from having too much influence on the final results. The use of these weights in statistical analysis ensures that the demographic characteristics of the sample closely approximate the demographic characteristics of the national population. The dependent variable, willingness to take the Swine Flu vaccine was measured by the question: ‘Now thinking about the swine flu that’s been in the news lately. . . . If the swine flu vaccine was available to you, would you get it or not?’ Answers were dichotomized, 1 ¼ yes, 0 ¼ no. Independent variable, confidence in government, was measured by the answer to the question, ‘How confident are you in the government’s ability to deal with the swine flu?’ Answers were 0 ¼ not at all to 4 ¼ very confident. Independent variable political ideology, conservative, independent or liberal was measured by the question, ‘In politics, as of today, do you consider yourself a Republican, a Democrat, or an Independent?’ The variables were introduced as a series of dummy variables; that is, numerical variables used to represent sample subgroups. In order to measure exposure to the news on swine flu, the survey provided a list of topics covered by the media in the last month, and respondents were asked to indicate which topic they had followed more closely. Responses indicating following swine flu news were coded as 1 and responses to other topics were coded as 0.

Page 5 of 9

Page 6 of 9

G. S. Mesch and K. P. Schwirian

The model 2 in Table 1 presents the results for the second stage of the logistic regression. With the other variables controlled, following the swine flu news is significantly associated with willingness to take the vaccine. Likewise, party identification is statistically significant with Independents and Republicans less willing than Democrats to take the vaccine. In our search for an explanation for the association of partisan identification and willingness to take the vaccine, in Table 1, in model 3 we introduce confidence in government into the equation. It is statistically significant, thereby indicating that those confident in government are more willing to take the vaccine than are those lacking confidence in government. Interestingly, the effect of identification with the Republican Party is now no longer significant when the measure of confidence in the government was introduced. Also the effect of media reports remains significant—those following the media are more

Table 1: Two-step logistic regression predicting the willingness to take the Swine Flu vaccine Variable name

Step 1

Step 2

Model 1

Age Hispanic Black White Gender (1 ¼ male) Marital Status (1 ¼ Married) Number of Children Education Less than college College Graduate Income Less than 30 K From 30– 49 K From 50– 74 K More than 75 K Homeowner

b

S.E.

Odds

0.01 ---20.06 20.57 0.10 0.06 20.02 --20.27 20.19

0.02 ---0.29 0.22 0.14 0.17 0.05 --1.74 1.74

1.01 ----0.93 0.56** 1.11 1.06 0.97 --0.82 0.91

0.12 20.21 0.19 20.44

0.19 0.15 0.17 0.17

0.88 0.80 1.21 0.63** Model 2

Predicted Step One Model 1 Following media reports Republican Democrat Independent Confidence in the Government Constant Neglerke Pseudo-R 2

0.69 0.035

0.33

2.07*

Model 3

b

S.E.

Odds

b

S.E.

Odds

4.11 1.31 20.48 --20.67

0.94 0.17 0.19 --0.17

61.15** 3.71** 0.61* --0.51**

22.69 0.15

0.52

0.06**

3.92 1.26 20.31 --20.56 1.04 23.32 0.20

0.97 0.17 0.19 ---0.17 0.16 0.55

50.54** 3.52** 0.73 ---0.56* 2.82** .03**

**p , 0.01, *p , 0.05, Contrast categories: Hispanic, Less than College, and Democrats, Less than 30 K

Downloaded from http://heapro.oxfordjournals.org/ at Michigan State University on February 16, 2015

variable. Accordingly, we created two sequential logistic regressions. Using maximum likelihood estimation, the first stage predicts the willingness of vaccination as a function of instrumental variables and other covariates. The second stage predicts the outcome as a function of the firststage predicted outcome and the hypothesized endogenous variables (Rassen et al., 2009). As given in Table 1, model 1, with other variables controlled, there are ethnic differences in willingness to take the swine flu vaccination. For ethnicity, white respondents indicated a smaller likelihood of taking the vaccine than did Hispanics and African-Americans. The effect of AfricanAmericans did not differ significantly from Hispanic respondents in their willingness. Homeownership has been indicated in the past as a potential measure of conservative attitudes. In our study, we found that homeowners reported a lower likelihood of being vaccinated than renters. This finding is consistent with the results regarding political affiliation.

Confidence in government and vaccination willingness in the USA

indicate that the confidence in the government’s ability to deal with the flu crisis is predicted by party identification and exposure to the news. DISCUSSION Influenza is a public health seasonal disease that puts population members at risk of infection. During the 2009 outbreak of the additional and newly appearing swine flu, the press and the public health officials were very active in making the public aware of both the extent of spread of the disease and of the potential negative consequences of infection (Baum, 2011; Mesch et al., 2013). What reasons accounted for the resistance to taking the vaccine? The starting view of this study was culture cognition theory from which we derived the expectation that the willingness of individuals to take the swine flu vaccine was affected independently by three central factors: exposure to the media, political party identification and confidence in the government. The findings both support this view and at the same time clearly expand our understanding of the underlying process of hesitance and rejection to taking the swine flu vaccine.

Table 2: Two-step logistic regression predicting confidence in the government’s ability to cope with the swine crisis Variable name

Age Hispanic Black White Gender (1 ¼ male) Marital Status (1 ¼ Married) Number of Children Education Less than college College Graduate Income Less than 30 K From 30–49 K From 50–74 K More than 75 K Homeowner Predicted Step One Following media reports Republican Democrat Independent Constant Neglerke Pseudo-R 2

Step 1

Step 2

b

SE

Odds

20.05

0.02

0.94*

0.06 20.17 0.03 0.02 20.02 --20.02 20.01

0.30 0.23 0.14 0.17 0.05 --0.02 0.01

1.06 0.84 1.03 1.02 0.97 --0.97 0.95

0.06 20.46 0.47 0.10

0.19 0.15 0.18 0.18

1.06 0.63** 1.60* 1.10

1.15 0.13

0.35

3.16**

b

SE

Odds

4.44 0.62 20.87 ---20.51 22.30 0.10

0.88 0.16 0.18 --0.17 0.60

85.4** 1.86** 0.41** ---0.59** 1.86**

**p , 0.01, *p , 0.05, Contrast categories: Hispanic, Less than College, and Democrat, Less than 30 K.

Downloaded from http://heapro.oxfordjournals.org/ at Michigan State University on February 16, 2015

willing to take the vaccine than those who do not follow the media. As we have seen, confidence in the government is a strong predictor of the likelihood of taking the vaccine, and with other variables controlled the people who expressed confidence in government are still more likely to take the vaccine. For this reason, we turn now to investigate the factors associated with confidence in the government’s ability to deal with the swine flu crisis (Table 2). We conducted a two-step logistic regression analysis to deal with the potential endogeneity of party identification and confidence in the government. Using maximum likelihood estimation, the first stage predicts confidence in government’s ability to deal with the flu crisis as a function of instrument variables and covariates. Age is now statistically significant; that is, with the other variables statistically controlled, older people have less confidence in the government’s ability to deal with the Swine Flu outbreak. This in part might reflect their direct experience with or knowledge of the 1976 ‘Swine Flu Fiasco.’ In the second stage, the predicted odds of the first stage are introduced in the model together with party identification and exposure to the media. The findings

Page 7 of 9

Page 8 of 9

G. S. Mesch and K. P. Schwirian

LIMITATIONS OF THE STUDY Among the limitations of this study are: the measurement of confidence in government, the effect of the method of data collection, measurement of news exposure and vaccination status verification. First, confidence in government was measured by one question. It would have been preferable to measure it with a several item scale; however, it was not possible with this data set. Second, the data were collected through telephone interviews; the interaction situation of face-to-face interviews may have elicited

different responses to the question of confidence in government. Third, news exposure was measured by one question. Exposure may be a multiple dimensional phenomenon better measured by a multiple item scale that gets at the nature of the content of media exposure. Furthermore, no distinction has been made here as to the actual content of the media reports. This sampling was underway before the vaccine was in wide distribution, so the assumption here has been that the media content focused on the dangers of the infection and the degree of spread rather than response to possible dangers of the vaccine. Fourth, the interview did not request a verification of vaccination status, so the responses were taken as an accurate depiction of the situation. CONCLUDING REMARKS Finally, overall our results provide support for a moderate version of the politicization hypothesis. As we had shown, party identification is not directly associated with vaccination, as previous studies had argued (Baum, 2011). The link is dependent on the extent of the respondents’ confidence in the government; that is, the perception of the public that the government is able to deal with a public health crisis. Furthermore, it also appears to depend on the segmentation of the media along with a selective exposure or a worldview that rejects modern medicine and science, and understates the government’s ability to deal with major public health events. Future studies need to move a step forward and test the simultaneous effect of political partisanship, confidence in the government as well as other factors that had been tested in various studies (such as concern with risks of being infected and risks of the disease) and not separately as is the current situation. Health issues are complex and in this paper, we deal with only one flu vaccination. It would be interesting for future research to determine which, if any, public health programs or health issues conservatives and liberals are more similar in program confidence.

REFERENCES Bandura, A. (2001) Social cognition theory: an agentic perspective. Annual Review of Psychology, 52, 521– 526. Bandura, A. (2002) Social cognitive theory in cultural context. Applied Psychology: An International Review, 31, 269– 290.

Downloaded from http://heapro.oxfordjournals.org/ at Michigan State University on February 16, 2015

Consistent with the cultural cognition theory of risk is our finding that individuals identifying as Democrats are more willing to take the Swine Flu vaccine than are individuals identifying with the Republican Party. This finding is consistent with the studies that had suggested that over time the confidence of the political conservatives in some areas of science has decreased (Gauchat, 2012). It is also consistent with the argument that has suggested that due to media segmentation, willingness to vaccinate has become a contested political issue (Baum, 2011). We also found that respondents expressing more confidence in the ability of the government are more likely to be willing to take the swine flu vaccine than those expressing less confidence. This finding, in addition to the finding on political affiliation, provides additional support for the understanding of the extent of politicization of public health issues. With the introduction of confidence in the government in the equation, the effect of identification with the Republican Party was no longer statistically significant. This finding expands the expectations of the culture cognitive model and underscores the mechanism of association of political identification and vaccination. It is not that Republicans reject vaccination because of their conservative views or excessive exposure to the media, but mainly because of their lack of confidence in the government’s ability, that at the time of the swine flu outbreak was a Democratic administration. In order to further increase the credibility of the findings, we conducted a two-step logistic regression analysis of the correlates of confidence in the government. Again consistent with the cultural cognition of risk perspective, the results show that party identification is a major explanation of the extent of confidence in the government.

Confidence in government and vaccination willingness in the USA

self-selection and partisan gaps in swine flu vaccination. Journal of Health Politics, Policy, and Law, 36, 1021–1059. McNeil, D. G. (2008) A multitude of vaccine benefits, yet controversy persists. New York Times, March 29, 2008. McNeil, D. G., Jr. (2013) W.H.O. estimate of Swine Flu deaths in 2009 rises sharply. New York Times, November 2013. Mesch, G. S., Schwirian, K. P. and Kolobov, T. (2013) Attention to the media and worry over becoming infected: the case of the Swine Flu (H1N1) epidemic of 2009. Sociology of Health and Illness, 35, 325 –331. Mooney, C. (2005) The Republican War on Science. Basic Books, New York. Nelson, K. (2010) Markers of trust: how pro- and antivaccination web sites make their case. http://papers.ssrn. com/sol3/papers.cfm?abstract_id=1579525 (last accessed 31 February 2013). Neustadt, H. V. and Fineberg, H. V. (1983) Flu: The epidemic that never was. Vintage Books, UK. Nguyen, T., Henningsen, K., Brehaut, J., Hoe, E. and Wilson, K. (2011) Acceptance of pandemic influenza vaccine: a systematic review of surveys of the general public. Infection Drug Resistance, 4, 197–207. Novella, S. (2010) Exploring issues and controversies in the relationship between science and medicine. The Lancet retracts Andrew Wakefield’s article. Science Based Medicine. http://www.sciencebasedmedicine.org/index.php/ lancet-retracts-wakefield-article (last accessed 13 February 2013). Offit, P. A. (2011) Deadly Choices: How the Anti-vaccine Movement Threatens us all. Basic Books. Ravnovich, R. (2009) H1N1 Influenza vaccine manufactures ramp up for Oct. 15. Gen: Genetic Engineering & Biotechnology. http:www.genedngnews.com. Rassen, J. A., Schneeweiss, S., Glynn, R. J., Mittleman, M. A. and Brookhart, M. A. (2009) Instrumental variable analysis for estimation of treatment effects with dichotomous outcomes. American Journal of Epidemiology, 169, 273–283. Springer, K. W. and Mouzon, D. M. (2011) “Macho Men” and preventive health care: implications for older men in different social classes. Journal of Health and Social Behavior, 52, 212– 227. Steinhauer, J. (2009) Swine flu shots revive a debate about vaccines. New York Times. http://www.nytimes.com/2009/ 10/16/health/16vaccine.html (last accessed 13 February 2013). Stern, A. M. and Markel, H. (2005) The history of vaccines and immunization: familiar patterns, new challenges. Health Affairs, 24, 611–621. Tavernise, S. (2012) Washington state makes it harder to opt out of immunizations. New York Times, 19 September 2012. Willrich, M. (2011) Pox, an American History. Penguin Press, New York. Wolfe, R. M. and Sharp, L. K. (2002) Anti-vaccinationists past and present. BMJ, 325, 430. Yoo, B., Holland, M. L., Phelps, C. E. and Szilagy, P. G. (2010) Effects of mass media coverage on timing and annual receipt of vaccination among Medicare elderly. Health Services Research, 45, 1287–1309. Young, M. E., Norman, G. R. and Humpreys, K. R. (2008) Medicine in the popular press: the influence of the media on perceptions of disease. PLoS One, 3, e3552.

Downloaded from http://heapro.oxfordjournals.org/ at Michigan State University on February 16, 2015

Baum, M. A. (2011) Red state, blue state, flu state: media self-selection and partisan gaps in swine flu vaccination. Journal of Health Politics, Policy, and Law, 36, 1021–1059. Bollen, K. A. (2012) Instrumental variables in sociology and the social sciences. Annual Review of Sociology, 38, 37–72. Cobb, R. W. and Elder, C. D. (1972) Participation in American Politics: the Dynamics of Agenda-Setting. Allyn and Bacon, Boston. Cobb, R. W. and Elder, C. D. (1983) Participation in American Politics: the Dynamics of Agenda-Building. Johns Hopkins University Press, Baltimore, MD. Cohen, D. and Carter, P. (2010) WHO and the pandemic flu “conspiracies”. BMJ, 340, 1274–1279. Dearing, J. and Rogers, E. (1996) Agenda Setting. Sage, Thousand Oaks, CA. Fitzpatrick, M. (2005) Review of Nadia Durbach, the antivaccination movement in England, 1853–1907. Journal of the Royal Society of Medicine, 98, 384– 385. Gabe, J., Bury, M. and Elston, M. A. (2005) Key Concepts in Medical Sociology. Sage Publications, London. Gardner, A. (2011) Fear of Vaccination has a Long History. US News and World Report. 26 January 2011. Gauchat, G. (2008) A test of three theories of anti-science attitudes. Sociological Focus, 41, 337– 357. Gauchat, G. (2012) Politicization of science in the public sphere: a study of public trust in the United States, 1974 to 2010. American Sociological Review, 77, 167 –187. Gidengil, C., Parker, A. M. and Zikmund-Fisher, B. (2012) Trends in risk perceptions and vaccination intentions: a longitudinal study of the first year of the H1N1 pandemic. American Journal of Public Health, 102, 672– 679. Goren, P., Federico, C. M. and Kittilson, M. C. (2009) Source cues, partisan identities, and political value expression. American Journal of Political Science, 53, 805–820. Harvey, C. (2008) Cited in survey confirms parents’ fears, confusion over autism. eScience News. http://esciencenews. com/articles/2008/10/03/survey.confirms.parents.fears. confusion (last accessed 3 October 2008). Hobson-West, P. (2007) Trusting blindly can be the biggest risk of all: organized resistance to childhood vaccination in the UK. Sociology of Health & Illness, 29, 198– 215. Hutchins, S., Fiscella, K., Levine, R. S., Ompad, D. and McDonald, M. (2009) Protection of racial/ethnic minority populations during an Influenza pandemic. American Journal of Public Health, 99, S261– S270. Jacoby, W. G. (2010) Policy attitudes, ideology and voting behavior in the 2008 election. Electoral Studies, 29, 557–568. Kahan, D. M. and Braman, D. (2006) Cultural cognition of public policy. Yale Journal of Law and Public Policy, 24, 147–170. Kahan, D. M., Braman, D., Cohen, G., Gastil, J. and Slovic, P. (2010) Who fears the HPV vaccine, who doesn’t and why? Law & Human Behavior, 34, 501–516. Linn, S., Guralnik, J. and Patel, K. (2010) Disparities in influenza vaccination coverage in the US 2008. Journal of American Geriatrics Society, 58, 1333–1340. Malka, A. and Lelkes, Y. (2010) More than ideology: conservative- liberal identity and receptivity of political cues. Social Justice Research, 23, 156– 188. Maugh, T. H. (2009) Beck and Limbaugh Weigh in on Swine Flu. LA Times: Booster Shot blog. Cited in Baum, M. A. 2011. Red State, blue state, flu state: media

Page 9 of 9