When developing messages for health communication initiatives, public health professionals and communications experts rely on the lessons learned from their experiences as well as on their professional judgment about the strategies that are most likely to succeed, said Marthe Gold of the New York Academy of Medicine and the City College of New York, the moderator of the third panel. These professionals often make decisions without empirical evidence of effectiveness, she said, because the availability of relevant research is limited. The lack of a strong evidence base can be attributed in part to the limited funding available for research on population-based health interventions and also to the difficulty of determining causality in multi-faceted health and communication campaigns. During the third panel session, Sarah Gollust, an assistant professor at the University of Minnesota’s School of Public Health, and Jeff Niederdeppe, an associate professor at Cornell University’s Department of Communication, shared findings from the latest research, funded primarily by the Robert Wood Johnson Foundation and the National Institutes of Health, on the impact of different messages about health inequalities and the social determinants of health.
The Current Focus on Individual Behaviors and Personal Responsibility
Health communication research and practice traditionally have focused on changing individual behaviors such as smoking and eating habits, Niederdeppe said, yet research shows that individuals’ health behaviors and outcomes are influenced largely by the social, political, and economic environments in which they occur. Communication strategies that are effective in influencing individual behaviors may not be effective in advancing policies to improve the environment. In fact, he said, sometimes the strategies conflict.
Public opinion and discourse about many of the leading causes of death in the 21st century—including heart disease, cancer, chronic obstructive pulmonary disease, and accidental injury—are driven largely by a sense of personal responsibility for health. When asked in a national public opinion survey about factors that strongly influence health, the vast majority of respondents cited personal health practices such as smoking, diet, exercise, and alcohol use. Fewer people mentioned factors such as affordable health care and health insurance. Respondents were least likely to attribute health to social determinants such as income, education, neighborhood, and race/ethnicity (Robert and Booske, 2011). People who viewed individual behavior as the sole determinant of health outcomes were much less likely than other respondents to support public policy options addressing the social determinants of health.
Developing communication to explain how the social, physical, and policy environments affect individual health behavior is challenging, Niederdeppe said, because the relationships are complex. It is not always obvious how macro-level factors link to individual actions. Researchers in the United Kingdom created a diagram illustrating 105 systems-level variables contributing to obesity.1 The complexity of the diagram underscores the challenge of communicating effectively about the problem, for which there is no single solution.
Goals and Target Audiences
The goals of communication campaigns and other messaging about the social determinants of health can be divided into five categories,
1 An interactive obesity systems map can be viewed at http://www.shiftn.com/obesity/Full-Map.html (accessed February 18, 2015). For more information see http://www.noo.org.uk/NOO_about_obesity/causes (accessed February 18, 2015).
Niederdeppe said. Some aim to increase the public awareness of health disparities and to show that they are determined in part by social factors. Others seek to increase concern about disparities and to build support for action to address them. Some public health messages promote the belief that societal forces contribute to poor health and health disparities, some seek to build support for policies addressing social factors, and some seek to mobilize people to advocate for social change.
Communication about the social determinants of health will need to address at least three audiences, Gollust said: the general public, the medical and public health workforce, and decision makers in the public and private sectors. There are at least two types of publics for campaign messages: those inclined to agree with messages about social determinants of health and those inclined to disagree. Communication about the impact of social determinants should seek to mobilize the former group and persuade the latter.
The medical and public health workforce is an important audience because of its credibility on population health issues, Gollust said. Health care workers may be able to focus public attention on population health, to educate patients and others, and to mobilize people to act. Communication with decision makers, including policy makers and private-sector leaders within and outside of the health care industry, should promote specific actions that can be taken to address the social determinants of health.
Challenges from Industry
Communication initiatives to advance public health face challenges from well-funded industries that promote products—such as cigarettes, alcohol, and junk food—that are detrimental to health, Niederdeppe said. These industries tend to oppose regulations affecting their profits, and, he estimated, they are able to outspend public health advocates by at least 100 to 1 when the marketing budgets promoting their products are taken into account.
Important Lessons Learned
Based on the current body of communication research, Gollust cited three important lessons about the effectiveness of various strategies to communicate about population health. First, communication initiatives designed simply to increase the awareness of social determinants may not be sufficient to effect change. Second, to influence public opinion and policy, communication will need to reflect important societal values that resonate with large audiences. And finally, when formulating com-
munication strategy, it is important to recognize and address challenges from groups opposed to the population health messages and policy being supported by the communication strategy.
Recognizing the Limitations of Informational Strategies
Raising public awareness about disparities is not sufficient to induce action, Gollust said, and it may even have unintended consequences. Rigby et al. (2009) found that messages about health disparities primed people to focus on group differences. Messages linking health disparities to race generated the least support for public policy intervention, while messages relating disparities to income generated the most support for government action. That study as well as additional research by Lynch and Gollust (2010) found that people may have different beliefs about the reasons for group differences. Some may attribute racial and gender variations in longevity to individual behaviors. Others may believe these differences are related to societal factors. Some people may believe that discrimination affects longevity, whereas others may think that genetics is the primary determining factor (although evidence unequivocally finds that genetics is not an important factor). Attitudes about responsibility and fairness also may shape people’s views about the reasons for group differences in longevity.
People may process messages differently depending on their knowledge or beliefs about the social determinants of health. Therefore, Gollust said, an educational message may not have the same influence on all people, and educational strategies may not be sufficient to stimulate action on health disparities. A study by Gollust and colleagues (2009) found that individuals’ responses to a news article about the social determinants of diabetes varied by party affiliation. People in the control group, who received no information about social determinants, showed little recognition that social factors affect health. When people in the experimental group received information about social determinants, the recognition of social factors’ role increased substantially among Democrats and Independents, but not among Republicans. Moreover, support for public policies to help prevent diabetes increased among Democrats who read the information, while it declined among Republicans.
When Lundell et al. (2013) studied focus groups, they found similar variations in information processing based on people’s values, beliefs, and prior knowledge of issues. The major lesson learned was that in the absence of information explaining how social determinants create disparities, people form opinions based on their biases. They may be suspicious of data on the topic or attempt to refute the statistics. One focus group participant rebutted data on the association between educational attain-
ment and life expectancy by saying, “Maybe somebody didn’t go on to school or even didn’t finish high school, but they might have gotten a good education at home in terms of how to be a healthy person.” Population research has shown, however, that the link between education and longevity is attributable not only to education about health specifically but also to the skills, abilities, and resources that people gain through higher education.
A study by Gollust et al. (2014) showed that even when policy makers have access to the best information about health disparities and social determinants, they may not consider the data in their decision-making process. For example, despite the abundance of evidence on disparities in obesity, the researchers found that none of the of oral testimony, written reports, fact sheets, bills, or news articles discussed by the Minnesota legislature over 5 years of discussion on this issue actually cited this evidence to support government action on disparities in obesity.
Linking Population Health Messages to Important Societal Values
Population health messages that reflect important societal values may be more persuasive than messages that do not, Gollust said. For example, messages acknowledging personal responsibility for health—a strong value in the United States—probably are more effective than messages that do not (Gollust and Cappella, 2014). Gollust offered several important caveats, however. In a nationally representative study on the impact of three messages about the social determinants of health, the message that mentioned personal responsibility generated less anger among both Republicans and Democrats than the message emphasizing social determinants or the message highlighting the relationship between socioeconomic status and health (Gollust and Cappella, 2014). But a message suggesting that personal responsibility was the major or exclusive determinant of health generated much more anger, particularly among Democrats. Moreover, the message emphasizing personal responsibility sparked many counterarguments from both Democrats and Republicans. These findings indicate that audiences are likely to reject messages attributing health outcomes mainly or exclusively to personal responsibility.
A study by Niederdeppe et al. (2014a) found that personal stories about the social and economic determinants of obesity were more persuasive when the protagonist acknowledged some level of personal responsibility for his or her health. However, placing a strong emphasis on personal responsibility undermined the social determinants message, particularly among people with conservative political views. Additional research has suggested that using personal narratives to convey information about the determinants of health can lead people to focus more on
the role of personal responsibility. A study by Barry et al. (2013) found that people who read obesity information in the form of a story about a child blamed children more for the condition than those who read a story without the individual narrative. In another study, Niederdeppe et al. (2015) found that policy makers tended to develop counterarguments against information conveyed through personal narratives, unless it was linked to statistics or a story about community issues.
Because messages focused on personal responsibility sometimes reduce support for population health objectives, researchers are seeking new ways to show how population health issues relate to important American values. For example, to build support for improving school meal programs, a recent report highlighted the devastating impact of childhood obesity on military readiness (Christenson et al., 2012). In a study by Gollust et al. (2013), messages emphasizing the effects of child obesity on military readiness led conservatives to view obesity as a government responsibility and increased their support for obesity prevention policies.
Additional research is needed to identify effective messaging strategies for linking population health objectives with important American values. Messages that relate population health goals to the values of fairness, equal opportunity, and economic security may generate support for public policy interventions, Gollust said. For example, a study by Lynch and Gollust (2010) found that people defined fairness in a variety of ways, and a plurality (38.4 percent) believed that it meant “Everyone has an equal chance to begin with.” This suggests that messages emphasizing equal opportunity may resonate with a wide segment of the American public.
Anticipating and Responding to the Opposition
Health communication professionals often face challenges from opposing organizations that can outspend them by 100 to 1, or even more, Niederdeppe said. Efforts to anticipate and respond to opponents’ messages may have limited success. In one study by Niederdeppe and colleagues (2014b), researchers divided people into four groups to test the effect of alternative messages about a proposed tax on sugar-sweetened beverages: one group read a strong, research-based argument supporting the tax; the second group read both the pro-tax argument and an opposing argument from the beverage industry; the third read a “refutational” message refuting industry arguments against the tax and explaining why the tax was justified; and the fourth group received no message. The refutational message increased support for the tax relative to the control group, but approval remained far below 50 percent. Moreover, people’s
responses to the message varied by political affiliation. The refutational approach was effective among Democrats and Independents, but it seemed to backfire among Republicans. To replicate the real-world environment for health communication, where industry opposition outspends health policy advocates by a considerable margin, researchers conducted a follow-up study in which participants read an anti-tax message 2 weeks later. After reading the follow-up message, a majority of people in all four of the original groups—including those who had read the refutational message—continued to oppose the tax.
Calling attention to public health issues may in some cases activate the opposition, Niederdeppe said. A study by Harwood et al. (2005) found that when an alcohol prevention campaign received extensive media coverage, prevention legislation was defeated. When the campaign received little or no coverage, several prevention bills were enacted. Based on this analysis, researchers suggested that media advocacy had the unintended effect of mobilizing the alcohol industry. In some cases, Gollust noted, industry representatives disseminate arguments against public health initiatives indirectly through third parties such as interest groups and advocacy coalitions. Therefore, she encouraged population health professionals to identify and track the source of counterarguments.
Finding Effective New Messengers
The traditional news media is an important source of information about population health, but the news tends to focus on individual determinants of health—such as behavior, biology, and genetics—rather than on social, economic, or neighborhood determinants (Gollust and Lantz, 2009). Although the news media is beginning to describe issues from a population health perspective, Gollust said, this approach remains relatively uncommon. In a study by Wallington et al. (2010), journalists cited the lack of specialized training, capacity, and time as barriers to covering health disparities. Moreover, they had difficulty making the issue relevant and interesting to target audiences. One journalist said, “I think that [with] any issue that involves race, such as health disparities, there will be opportunities for people or our audiences to turn a deaf ear or say, ‘Oh! Here we go going on about this [race] again.’”
Because the news media has a limited capacity to deliver population health messages, further research is needed to identify effective messengers outside traditional media outlets, Gollust said. Messages are most likely to be persuasive when the audience perceives the source to be credible; therefore, she suggested, public health professionals should conduct research to identify which potential sources are perceived as most credible. Messages from unexpected sources can be particularly effective.
For example, a study by Bergan (2012) found that a Republican’s endorsement of a policy allowing same-sex marriage generated greater support for the policy than a Democrat’s endorsement. A health communication campaign in San Francisco, The Bigger Picture, features youths speaking to youths about the social determinants of type 2 diabetes. Military leaders affiliated with the Mission Readiness campaign are advocating for healthier school meals to reduce child obesity.
In conclusion, Niederdeppe called for additional research, both on the interaction between message and messenger and on the connection between public opinion and actions conducive to policy change. Public opinion—along with factors such as politics and coalition building—plays a major role in the policy-making process. Gradual shifts in public opinion have set the context for significant policy change on issues such as same-sex marriage and tobacco control. Additional research is needed, Niederdeppe said, to identify the communication strategies with the greatest potential to influence public opinion on population health issues. Just as professionals would not make public health decisions without comprehensive epidemiological evidence, they should not make decisions about health communication strategies without sufficient research and evaluation. Communicating effectively about population health is just as important as understanding its fundamental causes, he said.
Following the presentation, panelists responded to questions about effective messages and strategies for advancing population health.
Michael Mangianello, founding partner of HCM Strategists, asked whether a single event could prompt policy change to improve population health to the same extent that National Coming Out Day changed the policy environment for the gay rights movement. Niederdeppe suggested that a multi-faceted, “everything but the kitchen sink” approach would have greater potential to influence policy than any single event. Strategies that go beyond messaging to include coalition building and efforts to stimulate interpersonal conversation and mobilize people are effective ways to engage the public. Gollust agreed and noted the importance of generating conversation in small groups and social networks. Gold said that bringing together diverse groups of people to share ideas and build mutual understanding can be an effective way to approach public policy challenges.
When David Kindig of the University of Wisconsin School of Medicine and Public Health, asked about the state of the art in targeting specific audiences, Niederdeppe said that the research on effective strategies is limited. His study of the impact of various messages on legislators and
the general public (Niederdeppe et al., 2015) found a similar association in both groups between perceptions of personal responsibility for health and support for public policy intervention. To advance the objectives of population health, public health professionals should make personal connections and build long-term relationships with policy makers, he said. Additional research is needed to identify effective relationship-building strategies.
Sanne Magnan of the Institute for Clinical Systems Improvement asked for suggestions for how to anticipate unintended consequences without conducting formal research and how to engage audiences in discussions about health care costs. Gollust said that results from past experience may help health communications professionals recognize the potential for unintended consequences, but empirical research is probably the best source of guidance. Niederdeppe said that it is not always possible to design evaluations to analyze unintended effects, but social science theory may be instructive in some cases. For example, if the target audience has stereotypes that could interfere with effective communication, researchers should avoid messages with the potential for priming these stereotypical convictions, such as the existence of certain types of group differences. Even if they do not test messages in advance, researchers should consider strategies to offset stereotypes. Research on message processing and motivated reasoning processes has identified metrics that signal potential boomerang effects, that is, effects that produce results that are opposite from those intended, Gollust said. By measuring intermediate emotional and attitudinal reactions to communication—such as anger, negative thoughts, and counter-arguing—researchers can assess whether messages are having the intended effect on public opinion or if they have the potential to produce backlash.
Engaging audiences on the issue of health care costs is challenging, Gollust said, because people associate low-cost care with low-quality or substandard care. Negative reactions to cost discussions are strongly associated with moral concerns about rationing. Additional research is needed to explore the implications of these views.
In light of findings reported by Rigby et al. (2009), Pamela Russo of the Robert Wood Johnson Foundation asked whether population health professionals should focus discussions of health disparities on income rather than on racial and ethnic differences. Discussions of race are challenging for many reasons, Gollust said, but because racial minority status is so strongly associated with other types of social disadvantage, it is impossible to discuss one issue without referencing the other. The study by Rigby et al. (2009) found that people had different attitudes about the reasons for racial and income disparities in health. Many people believe that racial and gender differences in life expectancy are mainly biological,
but they do not believe the same thing about income differences. Therefore they may have different responses to messages about disparities, depending on the group being discussed. Additional education may be necessary to change how people understand the underlying factors leading to group differences.
Focusing discussions of disparities exclusively on income inequality would omit many other factors, such as discrimination, that contribute to racial disparities, Niederdeppe said.
Although avoiding discussion of race may be easier in the short term, he added, in the long term it would have negative implications for social justice, and the costs would outweigh the benefits. Gollust commented that structural factors and discrimination are important and distinct determinants.
Paula Lantz of the Milken Institute School of Public Health, The George Washington University, commented that policy change tends to occur when policy makers and the public view affected populations as deserving. For example, policy makers typically view women as a deserving population in the context of policies related to breast cancer prevention. As a result, legislatures in 19 states have passed bills requiring the issuance of mammogram reports to notify women when the results show dense breast tissue. The greatest challenge facing public health professionals, she said, is to effectively address issues affecting populations who are marginalized and viewed as less deserving and who lack a voice in the policy-making process. Messages should not use the term social justice, she said, because discussions of social justice tend to polarize people based on their political and ideological views.
Gold concluded the session by reviewing major points from the presentation. First, she said, it will be difficult to shift public opinion away from the view that individual behavior determines health, but this shift is necessary to make progress in addressing social determinants. The social and environmental factors affecting health are complex, and efforts to explain them succinctly are challenging. Public health organizations also face challenges from well-funded opponents.
Population health messages should focus on three target audiences: the public, health professionals, and decision makers. In some cases, communication should aim to mobilize audiences; in other instances, it should be designed to persuade. The persuasion process is difficult, and increased awareness is not always a sufficient impetus for action. Public health messages linked with important American values are more likely to be persuasive.
Changing the way that messages are conveyed can alter people’s responses. Narratives suggesting that health outcomes are determined by a moderate amount of personal responsibility combined with other
factors may be the most effective way to convey information about social determinants. However, individuals’ interpretations of messages vary depending on their political beliefs and values.
Counterarguments and opposing messengers can create major challenges for population health initiatives. In some cases, focusing public attention on an issue activates the opposition and leads to defeat. In New York State, when policy makers were considering a tax on sugar-sweetened beverages, the beverage industry worked with the National Association for the Advancement of Colored People to defeat the proposal.
The traditional news media have been slow to engage in discussions about social determinants of health. Reporters may be reluctant to use population health terminology because they expect a negative response among readers, but research findings suggest that a change in journalists’ perspectives may be coming. It is important to use the language of population health and to discuss social determinants, Gold said, in order to help change public perceptions and increase support for investment in population health initiatives.
In a concluding comment, Gollust reiterated Lantz’s call for focusing communication initiatives on the needs of populations often viewed as less deserving, in spite of the challenges this poses because of the undercurrent of negative societal attitudes toward some marginalized groups.