The second panel focused on exploring evidence-based platforms that can be used for media/communication efforts, and on the strengths and weaknesses of the various media types in the context of the social norms targeted for change. The workshop moderator, Joanne Silberner, introduced the first panelist, Vicky Rideout of VJR Consulting and member of the Committee on the Science of Changing Behavioral Social Norms, who has worked on media research and public information campaigns and has spent time in Hollywood getting health information on the air in various forms. She presented a case study of embedding a public health message in entertainment television. Rebecca Palpant Shimkets, associate director, Rosalynn Carter Fellowships for Mental Health Journalism of the Carter Center Mental Health Program, is also a committee member. She designs initiatives aimed at reducing the amount of negative information about mental illness and focused her presentation on getting accurate information into the rapidly changing media landscape. Donna Vallone,1 senior vice president, American Legacy Foundation, which is dedicated to reducing or eliminating tobacco use, focuses her work on health disparities. In her presentation, she discussed the truth® campaign.
Rideout focused on delivering public health messages using “content integration,” in which messages are embedded into the storylines of enter-
1 Vallone is currently the chief research officer at the Truth Initiative.
tainment television shows that are already on the air. In previous work at the Kaiser Family Foundation, she worked with writers and producers on the health content of their television shows, looking for opportunities to write about health topics from a dramatic perspective to fill gaps in public understanding and raise public awareness. Research was also conducted to evaluate the impact of those messages. Rideout described one of these cases that involved the popular television show Grey’s Anatomy.
In season four of Grey’s Anatomy, a storyline about mother-to-child HIV transmission was written into the show. The health communication purpose was to inform the audience that the risk of transmission was less than 2 percent with proper medication on the part of the mother. In other words, there was a 98 percent chance that the child would be born healthy, without HIV. The message was also intended to reduce the stigma against HIV-positive women who choose to have children.
In the episode, a young woman and her husband come to the hospital for a pregnancy test. When the doctor tells them that the test is positive, they both have a very negative reaction, especially the woman, who asks for an abortion to be scheduled as soon as possible. It emerges that she is HIV-positive and cannot bear the thought of transmitting the illness to her child. The doctor tries to tell her that the risk of transmission is very low, but the two get into a heated argument, and the doctor is not strong or clear enough for the patient to hear her. The doctor leaves and returns later to speak with the woman. Rideout played a part of the show for the audience in which the writers incorporated the factual message that there was a 98 percent chance that the child would be born healthy, without HIV. She then presented the methods that were used to evaluate this public health message.
Rideout explained that the week before the episode aired, a nationally representative, random digit dial telephone survey was conducted in which respondents were screened to identify frequent viewers of this television show. They were asked a question about the various television shows they watch and a series of questions on health issues, including the likelihood of mother-to-child HIV transmission and attitudes toward HIV-positive women who choose to have children. One week after the episode aired, a postexposure survey was conducted among a separate national random digit dial sample. The researchers screened for respondents who had actually viewed the episode to assess whether exposure to the episode made a difference in the responses. Finally, to assess whether the information in the episode was retained, another postexposure survey was conducted among another sample of respondents 6 weeks after the episode aired. Each sample included 500 respondents.
Rideout summarized the results of the surveys (Kaiser Family Foundation, 2008). The first question was, “What is the likelihood that an HIV-
positive mother with the proper medication will give birth to a healthy baby?” The response options were that there is (1) more than a 90 percent chance that the baby will be born healthy without HIV, (2) a 50 percent chance, (3) a 25 percent chance, or (4) no chance. Before the episode aired, only 15 percent of respondents said the chance of the baby being born healthy was greater than 90 percent (which is the correct answer); 1 week after the episode aired, the correct response rate jumped to 61 percent.
The second question was attitudinal, Rideout said. Respondents were asked whether they agreed or disagreed with the statement, “If a woman knows she is HIV-positive, it is irresponsible for her to have a baby.” Before the episode aired, 61 percent of respondents said it would be irresponsible for a woman with HIV to have a baby; 1 week after the episode aired, that percentage dropped in the desired direction to 34 percent.
Rideout suggested that delivering public health messages on an entertainment television show—a nontraditional platform for communicating about a health issue—can have a large and powerful impact on viewers. She noted that the reach of the platform may be enormous. In this case, for example, there were 17.5 million viewers. In addition to direct viewership, moreover, entertainment television now has a second life online with fan blogs. Conversational blogs also provide an opportunity for researchers to see, in viewers’ own words, how the storyline affected them.
Rideout added a caveat that there are some challenges in relation to information retention effects that can fade over time. In the example she provided on the risk of HIV transmission, the proportion giving the correct answer 6 weeks after the episode aired declined from 61 percent to 45 percent. Rideout pointed out that this was three times as high as the baseline percentage, but it still had decreased. Similarly with the attitude question, the percentage who said it is irresponsible for a woman who knows she is HIV-positive to have a baby rose to a higher proportion 6 weeks after viewing (61 percent agreement before viewing, 34 percent agreement at 1 week postviewing, and 47 percent at 6 weeks postviewing).
Rideout discussed the advantages of this platform in relation to earlier speakers’ presentations. First, she argued, the evidence is compelling that embedding public health messages in entertainment television shows can work. As discussed during the first panel (Chapter 2), she said, storytelling is emotionally engaging; it has a power that other forms of media, such as a 30-second PSA, may lack. She suggested that this platform could be particularly well suited to addressing issues of mental health and substance abuse and associated negative attitudes. Second, as stated earlier, this platform can reach a huge audience quickly. Third, the audience can be targeted very effectively, once identified, and if one knows what television shows or entertainment media they are already viewing, content can be embedded there. Rideout concluded by saying that using this platform to address
issues of behavioral health and stigma is easier said than done. However, she said it could represent a fruitful approach on a topic that is universal, emotionally engaging. and dramatic.
REPORTING MENTAL HEALTH ISSUES IN A RAPIDLY CHANGING MEDIA LANDSCAPE: RESOURCES, NEW DEVELOPMENTS, AND FUTURE DIRECTIONS
Palpant Shimkets discussed the current media landscape with respect to mental health in terms of issues, improvements, and gaps; such tools as the Rosalynn Carter Fellowships for Mental Health Journalism; and steps forward and opportunities.
Issues in the Media Landscape
Palpant Shimkets stated that when the Rosalynn Carter Fellowships for Mental Health Journalism Program started almost 20 years ago, there was very little reporting on mental health, but many myths and misconceptions about mental illness existed. People did not talk about their own experience as they do today, she said. She described several current issues in the media landscape that require focused attention and improvement.
Random Acts of Violence
According to Palpant Shimkets, one of the greatest challenges in reporting on the topic of mental health and mental illness is around random acts of violence, such as the tragedy in Newtown, Connecticut. She described reporting on such incidents as a challenge for the media because to do so requires being timely, which means not having time to process the issue and provide the context, especially given the role of social media. Current coverage of serious mental illnesses often is associated with violence, she asserted, and most of the news coverage of serious mental illness and gun violence has been focused on the individual, without providing context (McGinty et al., 2013, 2014).
Palpant Shimkets noted the emergence of a number of media guides globally on reporting on suicide.2 She argued that more work is needed in
Palpant Shimkets stated that progress is being made on thematic stories, such as those about system failures, policy, and recovery, but that individual stories continue to target people experiencing mental illness, sometimes jumping to conclusions about their diagnosis without interviewing them about their own experiences and including them in the story. She showed a slide indicating that articles focused on individuals are more likely than generic news stories to have discriminatory content and danger as a theme; and in Canada, up to 83 percent of articles covering mental illness did not include the voice of an individual with a mental illness (Whitley and Berry, 2013; Whitley et al., 2015).3
Images, Headline, and Language
Palpant Shimkets shared that she had recently participated in a meeting with the Global Alliance for Stigma Reduction, in which representatives of 12 countries around the world work together on these issues. There are some media guides in Australia, Canada, and the United Kingdom, but there are no guidelines for headlines, which she said are important because of the opportunities for sensationalizing stories with so many online media sources (Tatum et al., 2010). She added that journalists also need guidance in thinking about how to discuss mental illness in alternative ways so the same terms are not used repeatedly in their reporting.
Media Polarization and Special Groups
Palpant Shimkets stated that, because of political polarization, the media landscape now is increasingly polarized as well. The good news, she said, is that there has been an increase in positive stories about mental health and mental illness. However, she noted, there also are fewer neutral stories, especially related to special populations such as children and men, two groups that are being recognized globally as needing more and better reporting (Henderson and Thornicroft, 2013; Henson et al., 2010).
Palpant Shimkets reiterated that positive media stories about mental health and mental illness are increasing, and the thematic stories are improving. Journalists also are showing a better understanding of diagnosis, she suggested, and the field is acknowledging that social contact with people who experience mental illness is extremely important.
Carter Center Initiative
Palpant Shimkets explained that in training for journalists at the Carter Center, individuals with lived experience with mental illness are included on every panel. These participants talk about what it is like to have a mental illness and about treatment, recovery, and the importance of peers.
Palpant Shimkets explained that the Rosalynn Carter Fellowships for Mental Health Journalism is a year-long fellowship program for professional journalists. She explained that journalists remain in their newsrooms during their fellowship year, which extends exposure to the project to their editors, colleagues, peers, and people who are engaging with them. In measuring the impact of the program, she said, it was found that more than 60 percent of journalists who complete the fellowship maintain their interest in reporting on mental health and continue to report on the topic after their fellowship year. She added that the program is increasing the accuracy and amount of information on the subject and helping journalists produce high-quality work in this area. She cited as one of the greatest benefits of the program affording journalists the time to develop a relationship with the people on whom they are reporting and to pursue questions at deeper levels.
Palpant Shimkets noted that the program has 150 fellows worldwide, with program sites in New Zealand, South Africa, Romania, and Colombia. She explained that it has an increasing focus on postconflict mental health. She said that one of the best measures of the program is its impact on public policy in terms of influencing more funding for mental health services, exposing abuses in state hospitals, and stimulating change in mental health systems. She cited Jonathan Martin with the Seattle Times, who exposed the issue of boarding in emergency rooms because of the lack of psychiatric beds for people in crisis. As a result, this practice was declared unconstitutional by the state Supreme Court, an action that was directly linked back to the journalist’s work in exposing the issue.
In closing, Palpant Shimkets suggested that more work is needed on media guides, which have evidence of being helpful, especially if training is involved, as seen in Australia and Canada. She noted that results of these efforts are mixed, however, and improvements can be made with the addi-
tion of specific guidance on how to report during isolated and very dramatic events involving violence and suicide.
Vallone presented on the truth campaign, the first national branded youth smoking prevention campaign. Launched in 2000, the campaign has generated many lessons learned over the past 15 years. Vallone’s presentation focused on using national paid advertising as a means of shifting population-level attitudes, beliefs, and behaviors.
Vallone pointed out first that a substantial literature demonstrates the impact of anti-tobacco campaigns in shifting key attitudes, beliefs, and behaviors. However, she asserted, lessons learned from prior campaigns demonstrate the critical need to ensure sufficient exposure to campaign messages among the target audience, particularly given the relatively small effects of campaigns and the need analytically to contrast those who are aware of the campaign and those who are not (Emery et al., 2012; Farrelly et al., 2002, 2005, 2009; Holtgrave et al., 2009; Miller et al., 2003; Messer et al., 2007; Vallone et al., 2011).
According to Vallone, in Bob Hornik’s seminal work on evaluating public health campaigns, researchers concluded that despite the salience of the message, if there is not broad exposure, the efforts to evaluate and demonstrate impact are likely to be unsuccessful. She noted that much of the evidence for the impact of anti-tobacco campaigns comes from states and that some of the most successful campaigns have been conducted in such places as California, Massachusetts, New York, and Florida. In fact, she said, the national truth campaign was based on the Florida truth campaign.
Vallone reported that to help guide other states in the implementation of media campaigns, the Centers for Disease Control and Prevention (CDC) issued recommendations on the best approaches and on how to ensure sufficient dose. The CDC recommended using television in particular for broad mass reach and suggested the need to reach at least 75 to 85 percent of the target audience, airing approximately 1,200 gross rating points (GRPs) (a measure of TV media dose) per quarter at the outset of a campaign and then sustaining the effort over time with approximately 800 GRPs. According to Vallone, the CDC recommendations also refer to the duration of the campaign, citing the need to be on air at least 3-6 months to achieve awareness, 6-12 months to influence attitudes, and 12-24 months to see behavior change.
Vallone said that in the early days of the truth campaign, an effort was made to ensure exposure by using not just television but a broad array of platforms, including earned media (e.g., magazines), integrated entertain-
ment or content-integrated entertainment (to which Rideout had referred), promotional activities, and websites (which were considered an extremely progressive platform in 2000).
Vallone showed a slide of a map illustrating the GRP variation across 210 media markets in the United States for the truth campaign’s message delivery. The natural variation in GRPs was used to link exposure analytically to reductions in youth smoking prevalence. Two outcome studies were published. One linked the GRP data to data from the Monitoring the Future study and demonstrated that the campaign could be credited with 22 percent of the overall decline in youth smoking in the first 3 years of its delivery. The second used data from the National Longitudinal Survey of Youth (NLSY), replicating those findings over a longer period of time, 2000 to 2004, and demonstrating a change in smoking prevalence.
Over the past decade, Vallone noted, the campaign has evolved with the media landscape. It has moved from using traditional media, either on or off air, to using all of the new digital platforms, including those that require steady streams of content, engagement, and discourse. Today, Vallone reported, truth campaign messages are disseminated through an integrated approach, utilizing the interplay among television, digital, and social media platforms.
Vallone discussed the opportunities and challenges of using national paid advertising. Through this platform, she asserted, effective reach to the identified target audience can be ensured. Whether via television or digital media, she said, a certain number of viewers will have exposure to the messages. She noted further that the dose of the intervention can also be quantified, which is extremely important when trying to shift and measure population-level change in attitude and behavior. Paid advertising also ensures optimal placement in television programming, where at-risk audiences will be exposed to messages they can comprehend.
At the same time, Vallone explained that national paid advertising, particularly on television, does require a significant investment. However, analysis of the cost-effectiveness of this approach showed during 2000 and 2002 that the campaign recovered $324 million in costs associated with developing, delivering, and evaluating as well as litigating the campaign. Additionally, $1.9 to $5.4 billion in medical costs to society was saved. Vallone concluded that although the cost of such efforts to society is substantial, these costs can be defrayed by changing attitudes, beliefs, and behaviors.
Vallone added that paid digital advertising is a very important tool, particularly with respect to frequency of exposure. Their campaign’s most recent messaging includes a 2.5-minute video on YouTube. A 60-second short spot leads into a longer video, which has had more than 21 million views. The audience also can be targeted in digital advertising, Vallone
noted, and it is significantly less costly than television advertising. She remarked, however, that its reach can be limited. Therefore, she said, marketers recommend the use of both television and digital components to drive change. She also cautioned that significant resources are required to promote digital properties and digital engagement that may not be anticipated.
Bernice Pescosolido introduced the discussion session by pointing to the cost-effectiveness of some of the communication methods described by the panelists. The presentations, she said, illustrated how structural stigma, which is embedded in institutions and the media, can be addressed and what contributions journalists and television writers can make to removing stigma. She underscored that the presentations showed how progress can be made by working at all levels—culturally, creatively, scientifically, and technologically—to get effective messaging where it will be seen and heard.
The discussion that followed centered primarily on questions related to connecting mental and physical health in messaging; message decay and unintended consequences; violence, mental health, and public attitudes; and challenges in using television and digital media platforms.
Connecting Mental and Physical Health in Messaging
Leslie Richards, University of the District of Columbia, asked why more campaigns connecting mental and physical health have not been undertaken. Vallone cited the difficulty of including multiple messages in a 60-second spot. She noted that the truth campaign has focused on both the health and social consequences of tobacco use in a variety of ways, and that the message content also depends on the audience. Palpant Shimkets commented that efforts currently are focused on addressing this issue with primary care physicians. She reported that Jurgen Unutzer’s work at the University of Washington is helping to identify behavioral health issues in primary care settings and targeting the role of doctors in identifying behavioral health issues. She noted that many people who complete suicide see their primary care physician within months of doing so, a fact that speaks to the importance of reaching primary care physicians.
Message Decay and Unintended Consequences
Beth Angell, committee member, referred to Vallone’s point that to be successful, messaging must be sustained over a long period of time and that there is a trajectory from awareness to behavior. Angell suggested that
this raises the question of the need to continue to evolve and roll out new forms of the message. Vallone replied that one of the lessons learned in the truth campaign is that it is impossible to be all things to all people. The campaign chose to target 12- to 17-year-olds with a prevention message. In past years, a cessation campaign targeted 24- to 54-year-olds with messages about treatment. Vallone stressed the need for changes in content, especially with digital platforms. Analyses can be carried out to address issues of decay and wearout.
A related question came from Vanessa Wellbery, who conducts government relations and community outreach for an organization that provides permanent supportive housing to formerly homeless women with a mental health diagnosis. She asked about the effect of education and awareness in sometimes pushing people further away from the facts. She cited the example of parents opposed to vaccination, who may become defensive and be pushed further toward opposition. Other parents who are on the fence when exposed to the facts may also be pushed further away. Wellbery asked the panelists whether they had thoughts about that phenomenon.
According to Vallone, the truth campaign knew that a directive approach that involved telling teens not to smoke was going to raise their defenses; teens associated experimentation with rebellion. So, none of the messages put forth ever said specifically not to smoke. Vallone noted that there are ways to gauge receptivity as well as reactions to a message, which goes back to the importance of establishing, knowing, and being committed to the target audience and tailoring the message so it will not prompt dissonance with their experience and what they know to be true. Rideout agreed that this is the reason it is critically important to do formative research with the target audience.
Palpant Shimkets added that education alone is not enough and that one must give the audience something to do about the issues. Pescosolido underscored that unintended consequences are a real problem and a reason to continue to evaluate a campaign while it is in progress so the intervention can evolve as needed.
Violence, Mental Health, and Public Attitudes
A virtual workshop participant asked whether there are any data regarding the impact of shootings, such as the incident in Newtown, Connecticut, and related media stories on public attitudes. Palpant Shimkets cited research done in Germany when tennis star Monica Seles was attacked on the court and stabbed. A study conducted by Matthais Angermeyer previously had shown a certain level of stigma attached to mental health issues in Germany, which peaked dramatically after that incident and took 9 years to return to its prior level. Rideout added that research conducted by Colleen
Barry at Johns Hopkins University specifically addressed public opinions on gun policy and mental illness after Newtown (Barry et al., 2013). She suggested that the more reporting of these events is presented and the more repetitious it is, the greater are the increases in desire for social distance from those with serious mental illness.
Another virtual participant commented that the surgeon general’s report on mental health cites violence and fear as central to stigma. In response, Palpant Shimkets explained that a great deal of work on attitudes and stigma reduction is being done outside of the United States in the United Kingdom, New Zealand, Australia, Scotland, and some Scandinavian countries. From that work, she knows that contact with a person living with mental illness is one of the best antidotes to stigma. If that contact cannot occur in person, there are some promising studies involving use of the media to tell stories of recovery. Pescosolido added that the work of John Monahan and the MacArthur Law and Mental Health Network is critical in this regard.
Marla Hendriksson of SAMHSA asked the panelists about the challenges of dealing with changing attitudes, from negative to neutral and neutral to positive, for example, and of dealing with wholesale behavioral health versus focusing on specific issues, behaviors, or conditions. Rideout reiterated that this is where formative research is essential in the strategic planning for any kind of communication campaign. She suggested that identifying the specific outcome desired among the target audience is key.
Challenges in Using Television and Digital Media Platforms
Patrick Zornow of Vanguard Communications and the George Washington University School of Public Health asked about issues to consider in identifying opportunities to integrate public health messages into entertainment. Rideout replied that this is a difficult undertaking and that one of the key issues in content integration is the investment required. She noted that the Kaiser Family Foundation made a decision to invest heavily in working with entertainment media on the health content of their programs. It funded a substantial, consistent, and sustained effort that involved staff on the ground in Los Angeles who cultivated relationships with the entertainment industry, who followed programming, and who thought about where messages might be integrated in a positive way.
Another issue, said Rideout, is not having creative control. She explained that at the end of the day, the message may come out in a way that unintentionally reinforces negative social attitudes. She shared an example of a show in which the objective was to integrate content on the uninsured, but the storyline showed a bus of people who were homeless coming to an emergency room, which was not the intended message. A
later storyline was about a mother who worked but did not have health insurance. Her son had a chronic illness and was having to be treated in the emergency room. This storyline was a more realistic portrait of the nature of being uninsured and the health consequences.
Marie Dyak of the Entertainment Industries Council asked about how to integrate the necessary research into all of the messaging processes so that the content and outcomes have the texture of intention. Rideout elaborated that the research can be linked to the particular audience, and data will show, for example, who is watching Nashville. One can then see whether the program is a good match for the intended audience of a particular public health message. And when evaluating the effect of the embedded content in a show, Rideout continued, its effectiveness can be assessed by having a control group of people who are frequent viewers of the show but have not seen the particular episodes in which the content is incorporated. Bringing in the research and public health communities early in the development of entertainment programming would be ideal, she suggested. She emphasized that with public health messaging, the task is more difficult than saying “drink our soda instead of the other guy’s soda,” but it can be achieved through parallel investments in research and in applied efforts in the entertainment community and in messaging.