2
The Science of Health Communication: Guiding Principles for Population Health Campaigns
In his keynote address, Robert Hornik, Wilbur Schramm Professor of Communication, Annenberg School of Communication, University of Pennsylvania, discussed four guiding principles for health communication initiatives: (1) identify the targeted behavior; (2) develop an effective strategy for exposing people to the message; (3) take a comprehensive, “all but the kitchen sink” approach; and (4) seek routine media exposure.
The first step in developing an effective health communication strategy, Hornik said, is to understand the targeted behavior. Health communication should focus not on population outcomes (such as increased life expectancy) or on categories of behaviors (such as limiting environmental toxins, reducing exposure to tobacco smoke, or safer sex), but rather on individual behaviors. Health communication can influence people to test for radon and thus help reduce environmental toxins. Health communication can convince policy makers to outlaw smoking in public places and thus help reduce the population’s exposure to tobacco smoke. Health communication can persuade people to use condoms and thus advance the goal of safer sex.
Identifying Determinants of Behavior
Many factors influence individuals’ behavior. For example, Hornik said, a member of Congress’s decision to support funding for National
Institutes of Health research on population health may be influenced by lobbying, campaign contributions, the member’s personal beliefs, or a combination of factors. Once researchers identify the most important determinants of the member’s beliefs, they can tailor communication strategies effectively.
In some cases—specifically, when behavior is driven by forces other than communication—the right intervention will require more than communication alone. For example, in some situations community factors, not individual beliefs, may be the primary influence on behavior. In such cases it may be necessary to tailor communication efforts toward outcomes that were not initially the goal.
The process of developing a population health initiative should begin with investigating the hypothesized determinants of the targeted behavior. This will influence whether the campaign is directly focused on a particular population in order to influence change or whether it should be redirected toward different outcomes or people than initially planned. For example, if fresh produce is not readily available in a community, communicating with residents about the importance of eating fruits and vegetables may have little or no impact, and a more effective approach would be to take action to increase the accessibility and affordability of these foods in the community, possibly in conjunction with communication designed to increase demand. In other words, Hornik said, “If you are trying to influence a behavior in a particular population, then the focus needs to be on what influences them, not what influences you.”
Identifying Effective Paths to Change
Even when communication is the most effective path to change, Hornik said, different behaviors will still require different paths to change, and research will be needed to identify the communication paths with the greatest potential to motivate behavior.
For example, in an anti-smoking campaign sponsored by the Centers for Disease Control and Prevention and the Philadelphia Department of Health, researchers tested 29 messages to determine which had the greatest potential to convince smokers already interested in quitting to seek help. The research found that three messages—“It would be easier to quit if I used help,” “I would set a good example,” and “I would have more energy”—were most likely to convince smokers to seek help with quitting. However, when researchers tested the messages’ potential to persuade smokers to quit (rather than seek help with quitting), they found that two other messages were the most promising: “I would respect myself more” and “I am confident that I could manage intense cravings.” Messages with the most potential to influence the first behavior—seeking
help with quitting—were found to have little or no potential to influence the second behavior, actually quitting. The lesson, Hornik said, is that “depending on the behavior you are going after, even if you choose a communication path, what matters [to people] is going to vary sharply.” What actually matters to people, he concluded, can only be learned by doing systematic, empirical work.
THE NEED FOR AN EFFECTIVE EXPOSURE STRATEGY
A Long-Term Process
Even when population health improvement communication campaigns have promising messages for the target audience, Hornik said, they often fail because they do not have an effective strategy for obtaining the needed exposure strategy. If addressing a particular behavior requires exposing people to the appropriate messages multiple times over an extended period, then a campaign needs to have a realistic strategy for assuring such exposure. For example, if behaviors are rooted in complex social norms, then people are unlikely to change their behaviors in response to a one-time exposure to a message. Changing social norms is likely to require a slow, incremental process of communication over time, with repetition from multiple sources.
Routine Exposure Through Paid Media
Without paid advertising, Hornik said, it is difficult for a population health campaign to provide media exposure that is broad and frequent enough to drive change, but large-scale communication initiatives are expensive. The National Youth Anti-Drug Media Campaign spent approximately $100 million per year to buy its advertising time. The original American Legacy truth® campaign, which focused its anti-tobacco efforts on youth, spent a similar amount, Hornik said. Most of the population health improvement campaigns that rely on free or low-cost public relations strategies to shape media messaging do not have access to that level of funding, which limits the exposure they might expect to achieve.
While a small number of campaigns appear to have found success by relying on the Internet to reach their audiences, a closer look at those campaigns may undermine any enthusiasm, Hornik said. It is quite rare for a specific message to diffuse broadly on the Internet (“go viral”), he explained, and even those that do may have limited staying power. Although communication campaigns based on the Internet and social media can be implemented quickly and at minimal cost, even in those few instances where a message has tremendous viral diffusion, the message
is unlikely to have a lasting presence in the public media environment or to have prolonged effects on behavior. “Virality comes and goes,” Hornik said. “It is not equal to useful effects.”
Hornik pointed out yet another concern about campaigns that depend on Internet diffusion. Often such campaigns encourage people to actively seek a message—by visiting a website or by becoming a Twitter follower or a Facebook friend—but in practice the portion of the target population which will actually visit a specific website will be very small, and thus campaigns that require an active seeking of information by audiences typically have minimal exposure and are not effective in changing population behavior. Often, Hornik said, campaigns must assume that most target audience members will not be active seekers, and then programs must choose exposure strategies that assume a more passive audience. Often that requires the use of paid media so that people are exposed to messages as part of their everyday routines, while they are using media in a passive way.1
A COMPREHENSIVE, “ALL BUT THE KITCHEN SINK” APPROACH
The most significant changes in public health behaviors have been associated not with one-time communication efforts, Hornik said, but rather, with multi-faceted, “all but the kitchen sink” campaigns by multiple entities over long periods of time. The substantial reduction in the number of smokers in the United States since 1967 can be attributed to many activities, including but not limited to communication, by many organizations in the past 40 to 50 years. Other successful campaigns—including the National High Blood Pressure Education Program, implemented from 1972-1984 to reduce stroke mortality, as well as long-term efforts to reduce HIV transmission—likewise involved multiple stakeholders and used a variety of strategies.
THE IMPORTANCE OF ROUTINE MEDIA EXPOSURE AND EFFORTS TO SHAPE COVERAGE
Hornik estimated that public health communication campaigns account for less than 1 percent of the total volume of health information conveyed through the media. The vast majority of health information in the media comes from sources such as news programs, documentaries, doctor shows, drug advertisements, health magazines, radio talk shows,
________________
1 Hornik did not address whether social media could be effective components of the “all but the kitchen sink” campaigns discussed in the next section.
and Internet stories that people access in the context of their everyday routines. Because individuals get far more of their exposure to health-related messages through the media than through deliberate messaging campaigns, health communication efforts should consider strategies that seek to shape routine media coverage.
One way to shape media coverage is through media advocacy. Originally developed by Larry Wallack and Lori Dorfman, media advocacy links grassroots organizations and lobbying in order to influence mainstream media coverage and, ultimately, specific policy outcomes.2 Other strategies to affect coverage include standard public relations practices (e.g., issuing press releases and communicating with reporters to shape the content and framing of stories), programs to educate health reporters about population health, and efforts to influence the content of entertainment media.
Media content can affect people’s behavior directly—by modeling certain activities and condemning others (such as alcohol use)—and it can frame how people think about issues (e.g., by presenting health behavior as a matter of individual choice versus presenting it as the result of public policy).
Evaluation is a critical component of health communication, Hornik said. To test the effectiveness of a message, it is necessary not only to count the number of people exposed to it, but also to calculate the proportion of the total audience exposed, to measure audiences’ understanding of the message, and to determine whether the message changed the audiences’ attitudes and behavior, if indeed that was the purpose of the communication program.
Hornik concluded by reiterating the four tenets of health communication: identifying a targeted behavior, not just a population outcome or a category of behaviors; developing an effective exposure strategy; pursuing an “all but the kitchen sink” approach; and seeking routine exposure in the media.
In the discussion following his presentation, Hornik responded to a series of questions about effective communication strategies.
In response to a question about how to maintain focus on a goal in
________________
2 Wallack and Dorfman are the founding director and current director, respectively, of the Berkeley Media Studies Group (BMSG). BMSG defines media advocacy as “the strategic use of mass media to support community organizers’ efforts to advance social or public health policies.” For more on media advocacy, see http://www.bmsg.org/resources/media-advocacy-101 (accessed January 27, 2015).
the context of a multi-faceted campaign, Hornik cited the examples of two political movements aimed at improving population health. Mothers Against Drunk Driving organized people and maintained media attention over long periods of time to reduce drunk driving. Gay men got organized through advocacy groups to focus people’s attention on HIV/AIDS. The World Health Organization’s Global Program on AIDS shaped media coverage in an effort to frame the disease as a universal problem requiring public investment to protect future generations, since everyone was at risk.
Responding to a question about media manipulation of campaign messages, Hornik said that population health professionals cannot prevent the media from reframing messages. For example, a local news show may cover a campaign’s television ad in a way that presents the message as controversial or as particularly relevant in a local area. Such changes are virtually impossible to avoid.
Asked about measuring the relative impact of individual components of a multi-dimensional campaign, Hornik emphasized two points. First, he said, successful population health movements such as the anti-smoking movement were effective because multiple entities focused on the theme. No single organization managed the campaign, and each entity focused on what it did best. Second, in a multi-faceted campaign, it is impossible to determine which component has the greatest impact because there is interaction among strategies. Each strategy may have different effects, with the various effects working together synergistically to achieve the desired outcome. For example, one part of the campaign may drive behavior change, while another may gain public attention and affect public policy.
When an audience member asked about developing messages to influence media coverage rather than individual behavior, Hornik said that such an approach could work if there were a clear path from media coverage to behavior change. For example, media coverage about climate change could reframe beliefs about electricity use, change social norms, and ultimately motivate people to replace standard light bulbs with energy-efficient alternatives. However, Hornik said he would be concerned about developing messages focused on media coverage without clear evidence that such a strategy would change beliefs and drive behavior change.
