Abstract
Although scholars explore how news stories’ framing elements may affect reader responses, they have yet to examine how the topics of health-related articles affect those responses. By content analyzing three US newspapers’ online health content and reader comments, this study finds that certain health topics are idiosyncratic with reader responses. Readers reacted to personal health and obesity news with more episodic and gain-framed comments but relied more on loss frames when discussing chronic health issues. Readers also used more thematic frames in comments about mental illness. Health coverage related to politics and the government was associated with fewer episodic comments.
Keywords
Discourse surrounding health concerns has remained among the top news issues in the United States over the last several years, particularly online where traffic to news sites continues to rise. 1 These outlets serve as important conduits of health content and provide a certain level of public engagement, often offering spaces where readers can comment on reports and discuss related issues. Allowing readers to comment on online news coverage opens up public deliberation, exposes people to novel ideas, and invites a variety of perspectives and opinions that may not be included in the news coverage itself. 2 Several recent studies have examined the association between the ways health news articles are framed and the ways readers respond, finding that readers may pick up on particular frames but may not promote them in online discourse. 3 This is particularly important in light of research that shows news consumers who read online content and the comments that follow may align their beliefs and opinions more with reader responses and less with actual news coverage. 4 Even with balanced reporting, reader comments have the potential to skew news consumers’ perceptions.
While public health officials and scholars have urged journalists for years to improve their health news coverage, especially in ways that can help people connect with health issues on societal rather than individual levels, 5 reader comments have only recently come under fire. Indeed, journalists have taken steps to enhance their health coverage by including more societal, or thematic, frames in their reporting. 6 Readers, however, have not entirely embraced the change, continuing to approach some health issues as individual problems in spite of the thematic frames offered by journalists. 7 This suggests that certain topics, rather than framing devices, may drive reader responses. For example, the topic of obesity may engender episodic thinking no matter how much thematic framing journalists employ. Thus, this study questions how the topic of health news content, rather than the frames, may guide reader responses. If it is the case that health topics influence the ways in which readers react, then improving health news coverage may need to extend beyond simply shifting framing techniques.
Using a content analysis of online health articles and reader responses from three major U.S. news outlets, this study explores the association between the topic of health content and the frames with which readers choose to respond. The findings signify the importance of particular health topics in promoting certain forms of reader responses and can help health communication scholars and journalists design potentially more effective approaches to health news coverage by anticipating certain forms of reader reactions.
Literature Review
Over the last several decades, the Internet has opened up new forms of news media that help to inform the public while also promoting a certain level of dialogue between news creators and consumers. 8 News creators have attempted to harness this new form of news interaction, navigating into social media spaces and developing forums for public conversations. While these spaces may not always provide balanced engagement between members of the news media and their audiences, 9 they do allow individuals an opportunity to extend conversations and offer reactions. News organizations may continue to graft traditional journalism routines into digital spaces, keeping the public out of certain aspects of the news production process, but they do so at the risk of losing touch with what the public values as news, as well as public engagement with the news. 10 At least one study shows that reader commentary following news articles, particularly those about science and health, can have a more profound effect on readers than the content of the journalistically-produced product. 11
Several other studies have approached this issue from a health communication angle, exploring how framing techniques used by health journalists can alter the ways in which readers respond to those articles. 12 These studies indicate that journalists have amended, at least in part, the way they cover certain health issues through their framing choices. However, these adjustments have not always resulted in congruent changes in reader responses. Instead, the findings from these studies call into question the role of specific health topics (e.g., cancer, HIV/AIDS, mental health, sexual health) in reader comments. More specifically, they raise the question of whether health topics may influence reader responses as much as or more than news frames.
Health Frames and Reader Responses
Communication scholars have pointed out the importance of reader responses to the news, in particular, the function of reader comments to online news coverage that can potentially reach and influence anyone on the Internet. 13 A growing number of Internet users continue to turn to the media for health information, making the content they encounter—including reader comments—of particular importance. This study is not the first to call into question the role of health topics in relation to perceptions of or reactions to key health issues, but it is the first to date to empirically explore the association between health topics in news coverage and the ways in which readers respond to those issues.
Many studies of health news have worked within the theoretical construct of framing, which can be defined as the structural devices or selection of information used by journalists to explain issues or events. 14 Of particular interest, health scholars have given much focus to two dichotomous framing mechanisms often used by health reporters: thematic/episodic framing and gain/loss framing. 15 Thematic frames connect issues and events with society and emphasize shared responsibility for solving problems, while episodic frames emphasize individual roles in society, often attributing responsibility to a single individual. 16
It is also important to note that health experts have a strong partiality toward thematic framing, often arguing that journalists do not incorporate enough thematic framing in their stories. 17 The thesis of this argument is that the inclusion of thematic frames can cue the public to think about issues on a societal rather than individual level, helping overcome the human tendency to blame individuals, known as the fundamental attribution error. These two different types of frames accomplish different purposes. For example, episodic framing puts human faces on stories and attracts attention. Thematic frames provide information about what society can do; without them, readers acquire an incomplete understanding of health issues. A primary function of journalists is to balance their reporting. Thus, the objective of balanced framing should not be to use only episodic framing or only thematic framing, but rather to achieve a balance of both. Still, a large contingent of health scholars has suggested that the news media can help improve public health knowledge by emphasizing thematic frames in order to help individuals understand societal causes, since it is currently lacking in health coverage. 18 Indeed, research demonstrates that the use of such frames encourages people to assign responsibility for health issues, such as obesity, to society rather than to individuals. 19 Other research has cautioned against such optimism, finding that thematic frames in online health news may not be consistent with reader responses. 20
This is congruent with more general research that shows comments from online news consumers frequently chastise others rather than respond with societal information. 21 Health communication scholars have observed that even with balanced news coverage, the public is quick to assume that sick individuals are responsible for their health problems. 22 Certain issues (e.g., obesity, diabetes) tend to trigger comments that are less thematic and more episodic. 23
Gain framing and loss framing, which have been shown to influence perceptions of and reactions to health communication and issues as well as health behaviors, 24 represent another dimension of framing. Gain frames promote the positive outcomes of specific actions, such as cancer screenings or childhood vaccinations, while loss frames focus on the negative outcomes of not partaking in those actions. While gain frames tend to be persuasive for low-risk actions such as taking vitamins or having blood drawn, loss frames are more effective for high-risk actions such as surgery. 25 When it comes to the ways readers respond to the incorporation of these frames into health news, however, the results are mixed. 26 While the frames health journalists select are important and may affect those who read them, they may not necessarily be the guiding forces behind the ways readers respond. Thus, it is important to understand what other factors influence readers’ understandings of health issues.
Health Topics
Health communication researchers are often very topic specific, usually focusing on a single issue, such as autism, 27 obesity, 28 or HIV/AIDS. 29 When studies are more inclusive, they sometimes create an excessive number of topics, oftentimes incorporating similar issues, such as nutrition and obesity, 30 which can be collapsed into one category, or hepatitis and scoliosis, 31 which can be subsumed under the broader category of chronic health issues that include multiple, related subtopics. In their study of two decades’ worth of health coverage in the U.S. media, Manganello and Blake 32 found eighteen health categories: (1) violence, (2) tobacco, (3) alcohol, (4) illegal drugs, (5) prescription drugs, (6) obesity and nutrition, (7) body image and eating disorders, (8) sex, (9) injury, (10) AIDS, (11) cancer, (12) aging, (13) death and disability, (14) mental health, (15) health providers and organizations, (16) genetics, (17) women’s health, and (18) an “other” category. They also noted that these categorizations could be logically collapsed, could evolve over time, and should be further explored by researchers. Comparable studies 33 have presented similar categorizations, often collapsing some health topics together such as obesity and personal health, women’s health and gender-specific and/or sexual health, and chronic illnesses.
These and other studies that have incorporated multiple health topics have found that differences in public responses can be attributed to the topics. For example, Hatley Major 34 analyzed attributions of responsibility for obesity and lung cancer in a group of 229 U.S. adults who read news articles that were framed as either thematic or episodic and gain or loss. Episodic and loss frames were positively associated with individual attributions of responsibility for obesity, but not for lung cancer, indicating the possibility of preconceived notions of responsibility for certain topics. Another study of interventions for negative behaviors, such as smoking, poor diet, poor exercise, and high alcohol consumption, found that different intervention methods worked best with different health issues, again indicating that varying outcomes—communication responses included—were likely to result from different topics. 35
These studies raise questions about the association between health topics in news coverage and reader responses—an association that cannot be generalized from scholarship examining a single issue or event. As public health officials continue urging journalists to adjust the frames they employ to improve responses and behaviors in society, these same officials and journalists should consider that the topics of discourse may guide how individuals respond more so than how journalists frame them. This study represents a first step in determining whether specific health topics drive the ways in which readers frame their responses, asking:
Method
In order to analyze these questions, this study used a content analysis to examine the possible associations between topics in online health news and the frames readers used in their responses via comments. Scholars analyzing similar research questions have indicated that such analyses should include news organizations that facilitate and publish reader feedback alongside health-related content. 36 In order to develop an appropriate sample for this study, a list of non-localized (i.e., not geographically-specific) news media websites offering health-related content was compiled using HealthNewsReview.org, 37 a nonprofit organization that rates health articles on the websites of twenty-one major media outlets for accuracy, balance, and completeness. Given that there is no comprehensive list of health websites, the authors’ extensive search of the Internet found this site to be the most comprehensive available.
From this initial list, websites offering readers the opportunity to comment directly below stories were selected. Websites that offered reader feedback only through Facebook, Twitter, or other social network sites were excluded, along with those that hid comments, did not allow comments, or did not have accessible or significant commenting (e.g., Health.com, The New York Times). In addition, websites that relied heavily on outside content were excluded. For example, CNN.com consistently covers health issues and allows reader comments, but the majority of health articles on the site come second-hand from health sources. Based on these exclusions, three major U.S. news outlets with an online presence were used: The Washington Post, The Huffington Post, and USA Today.
Sample
Constructed week sampling 38 between one and three constructed weeks produces more efficient samples than simple random and consecutive day sampling of newspapers. More recent research shows when sampling health news articles, six constructed weeks may more accurately estimate one- and five-year populations. 39 Thus, this study developed six constructed weeks using forty-two randomly selected dates for the July 1, 2010, to July 1, 2011, period.
A Google search using the website and/or news outlet name (e.g., The Huffington Post, huffingtonpost.com), the keyword “health,” and the randomly selected dates was used to collect an initial sample. At the time of sampling, articles that did not focus on at least one health issue were excluded from the sample (e.g., vehicle safety, recipes for holiday parties). Letters to the editor, opinion and editorial columns, articles without text (e.g., The Huffington Post’s ‘How To’ health videos), and articles from secondary sources were removed. The resulting sample included 129 articles with an average of 137 reader comments each. Given the volume of comments for some articles and noting that many articles did not receive more than ten comments, the first ten original comments 40 of each article were coded in chronological order (i.e., oldest comments first). If there were fewer than ten comments, then all comments were coded. This procedure resulted in 966 total reader comments for the sample. The unit of analysis for comments was the sentence.
Measures
Topics
Media scholars have yet to develop a single, agreed-upon list of health categories, instead developing their own categories arising from specific data. For example, Smith 41 employed thirteen categories, Manganello and Blake 42 had eighteen, and Berry, Wharf-Higgins, and Naylor utilized fourteen. 43 There was some overlap—all three had categories for cancer and HIV/AIDS, for example. But there also were notable differences. Manganello and Blake, along with Smith, used a “women’s health” category, but Berry, Wharf-Higgins, and Naylor did not. For this study, we relied on the established categories as much as possible, but created our own categories when topics in our dataset did not fit into previous categorizations. For example, a new category of politics and government arose because of Obamacare, which was frequently in the news during the time period. We also collapsed other scholars’ separate categories of heart disease, diabetes, scoliosis, and other chronic conditions into one “chronic” category, and collapsed their separate categories of SARS (severe acute respiratory syndrome), West Nile Virus, injury, staph infections, flu, tuberculosis, and similar acute illnesses into one “acute” category. Our ten categories were (1) cancer, (2) drugs and alcohol (includes smoking and tobacco), (3) personal health and obesity (e.g., diet and nutrition, personal wellness), (4) HIV/AIDS, (5) gender-specific health (e.g., women’s health, sexual health), (6) chronic health issues (e.g., cardiovascular disease, diabetes, scoliosis), (7) acute health issues (e.g., flu, common cold, personal injuries), (8) politics/government health issues, (9) mental health, and (10) other (e.g., cellphone risks, sewage safety). Categories were coded dichotomously: 1 = dominant presence of category, 0 = no dominant presence of category. While an article could contain one or several of the health topics, only the dominant topic was selected. This was identified using cues within the article, including the headline, the lead paragraph, and the overall amount of attention given to a topic.
Episodic/thematic frames
Following Iyengar’s 44 research, comments were coded for the number of sentences that contained episodic or thematic frames. Episodic frames focus on a case study, particular instance, or subject to illustrate an issue. For example, one comment from the sample discussing vitamin intake read, “I don’t take vitamins. My baby was fine. I was fine. I eat right and exercise and that’s better than chemicals going into my body.” Thematic frames emphasize broader roles and issues, such as this sentence: “Other recent research has shown that moms who both breastfeed and co-sleep with their babies get better quality sleep, and more of it, than either moms who formula feed or moms who sleep separately from their babies.” A sentence could fall into one or both categories, and some sentences contained neither episodic nor thematic frames.
Gain/loss frames
A gain frame emphasizes the benefits of a certain activity, such as the positive outcomes of screening behaviors or healthy dieting. One comment in the sample stated, “Regular check ups will keep your heart healthy and keep you off the operating table.” A loss frame does the opposite, focusing on the negative outcomes of avoiding screening behaviors or unhealthy dieting. 45 As an example, one of the responses read, “You eat fast food, you get fat, you die.” The number of sentences that contained either gain frames or loss frames was recorded. Again, sentences could have one, several, or no instances of either frame.
Coding Reliability and Analysis
The researchers developed the initial coding scheme and trained two independent coders in multiple practice sessions using articles and comments not included in the final sample. Slight adjustments were made to the coding scheme to improve reliability. Acceptable Krippendorf’s alpha scores 46 of .81 or higher were achieved for all comment variables during the final training session. The sample was then randomly split between both coders, and reliability was measured for 20% of the coded sample. 47 Krippendorf’s alpha scores were .88 for coded health topics in articles and .84 or higher for all coded comment variables.
Data were analyzed with point-biserial correlations 48 given the comparison between the dichotomous topics variable and the interval frames variables.
Results
The majority of the 129 articles included in the sample came from USA Today (42.7%). Articles from The Huffington Post (36.4%) and The Washington Post (20.9%) comprised the remainder of the sample. Similarly, more comments came from USA Today (48.1%) than from The Huffington Post (34.8%) and The Washington Post (17.1%). There were approximately 137 comments per article (M = 137.4, SD = 308.23). As indicated in Table 1, four topics dominated health news coverage within the sample: (1) obesity and personal health (34.1%), (2) politics/government health issues (17.8%), (3) gender-specific health (10.9%), and (4) chronic health issues (10.9%).
Topics in Online Health News Articles.
To examine RQ1, the association between the health topics of online articles and thematic and episodic frames in reader responses to those articles, point-biserial correlation analyses revealed several significant associations (see Table 2). Issues of personal health and obesity were associated with significantly more episodically-framed comments (rpbi i = .057, p < .01), politics and government issues were associated with significantly fewer episodic comments (rpbi = −.069, p < .05), and mental health topics were associated with significantly more thematic comments (rpbi = .074, p < .05).
Point-Biserial Correlations between Health Topics in Online Articles and Frames Used in Reader Responses to Those Articles.
p < .05. **p < .01. ***p < .001.
To examine
Discussion
Public health professionals have implored journalists and news producers to rethink the ways they frame certain health issues, suggesting that changes toward more thematic and gain-framed stories could help improve public knowledge and connectedness with those issues. Researchers have recently observed a shift toward thematic coverage of health news but have duly noted that these changes have not necessarily improved the way society responds. 49 The findings from this study suggest that those responses may be linked with some topics of health news, indicating that the frames news producers choose to cover may not be the only significant elements driving public reactions. Yet, it should also be noted that despite the statistical significance of the findings here, only one health topic was associated with more thematic comments: mental health. At a time when health scholars and practitioners continue to implore health journalists to frame their stories more thematically—a call that journalists appear to be heeding—readers may not be as responsive to such frames as hoped. This would suggest that more consideration, both scholarly and in practice, should be given to the ways in which readers consider and respond to health topics in news media coverage.
In terms of the individual findings, some of the relationships were expected. The correlation between episodic responses and topics of personal health and obesity, which included prevention, screenings, diet, and exercise, is shown consistently in studies. People see diet and exercise, as well as obesity and taking care of oneself, as something individuals have complete control over. They are typically not inclined to recognize the role of health insurance or access to affordable health care (e.g., safe, free places to exercise, availability and cost of healthy foods, high concentration of fast-food restaurants in certain neighborhoods) as contributing to these issues. Despite the work of some journalists who have increasingly included this type of thematic information in their articles, the public overwhelmingly continues to believe that ill individuals are responsible for their conditions. Perhaps only time and decades of consistent, thematic framing will correct this—similar to what has occurred with smoking.
In seeking ways to help the public balance its views about personal health and obesity between episodic and thematic lenses, other means of altering attitudes must be explored. Journalists may be doing their part, but more of the same does not appear to be moving the responsibility needle sufficiently. Among the constructs that may be worth exploring, involvement, personal relevance, and emotion are shown to be important for many different outcomes. Perhaps this is the case with health topics. Just as personal relevance inoculates one against agenda-setting effects, it may also ingrain attitudes of responsibility toward certain health issues. Emotion is an important construct that acknowledges how humans are not always rational, and this is one aspect of health beliefs that is understudied. Stereotypes may also play a role. Perhaps stories aimed at dispelling various stereotypes would be more effective in altering whom people hold responsible for certain health problems than would continuing to hammer home societal responsibility. Moreover, health scholars should speak with health professionals to determine what they believe works to change attitudes and test those insights in future research.
The personal health and obesity stories in this study also were significantly correlated with gain-framed comments, showing that readers were likely to have positive attitudes toward such articles and talk about them in terms of benefits to be had rather than risks. This finding is a notable one and suggests that readers may be decreasing the level of vitriol associated with this topic in exchange for more encouraging words. This is not all that surprising, given the effectiveness of positive persuasion on individual health. 50 As Cooper 51 noted, positive messages not only help people feel encouraged but also help create a sense of belonging and community. Perhaps when it comes to health, individuals are recognizing the power of positivity and its effectiveness as a bonding agent. While journalists may work toward thematically framing issues such as obesity in the United States, they continue to write personal health and obesity articles in a sort of “news-you-can-use” format, implying an individual, how-to-help-yourself approach. 52 While this may prompt more episodic responses, such as those seen in this study, it may also have the unexpected benefit of promoting gainful conversations. People may be more inclined to provide well-meaning comments as well as share personal success stories to encourage other readers.
Topics of politics and government were significantly associated with fewer episodic comments, which was not unexpected, given the government’s role in health care is a society-wide issue. Nor was the chronic illness association with more loss frames surprising, given that heart disease, diabetes, and the like are long-term challenges that people must cope with and for which there are no cures.
The positive correlation between mental health topics and thematically-framed comments also is encouraging. Perhaps this is one topic that journalists have covered as a societal issue for so long that people are beginning to incorporate that into their discourse. This may also be a positive sign for reducing the stigma associated with mental health issues, an area where the news media play an important role. While popular-culture products, such as movies and television programs, often crystallize mental illness through one event or individual, the news media have the opportunity to connect society to issues and concerns it might otherwise be unaware of. 53 This is not to suggest that the news media play a more important role than popular culture in improving the discourse and knowledge surrounding mental illness, but rather to promote the news media’s role as a possible outlet for the reduction of the stigmatization of mental illnesses. While mental illnesses have been episodically portrayed as violent, dangerous, and disparaging conditions in the past, 54 they may now be cast less episodically and more accurately by the news media—a source of public discourse historically critiqued for its focus on individuals with mental illnesses rather than society’s perception of them. 55 While the news media most likely are not the sole source of society’s shift toward thematic thoughts of mental illness, this is an area that deserves, at the very least, more consideration by scholars researching communication and disabilities. Research to determine why this shift has occurred could also help inform efforts on other topics that are still mired in individual responsibility.
Conclusion
While scholars have begun exploring the impact of online news stories on reader comments, this is the first study to date that analyzes how readers respond to specific health topics. This study focused on reader comments in online outlets, providing a foundation for future studies to investigate other platforms, such as Twitter and Facebook, where people can interact with the news and each other. These studies could help provide a wider understanding of how people converse about health issues in mediated venues, as well as interpersonal environments.
Scholars may also code articles for thematic/episodic and gain/loss sentences in order to determine if the framing of the articles reflects the framing of the reader responses. Future research may then empirically assess whether an article “sets the agenda” for reader comments or if the framing of an article instead stirs up explicit counter-framing by commenters.
Instead of focusing solely on the framing of health stories, research should explore other important constructs that may be influential in changing the way people think about health—personal relevance, emotion, and stereotyping are a few that have been important in changing attitudes about other issues. When these constructs are combined with framing elements, journalists may find their work yields quicker and more long-lasting attitude changes.
Moreover, recent research notes the power of personal narratives within health-specific stories. Some research has suggested, for instance, that even thematically-framed stories—those that focus on how societal conditions play the primary role in health outcomes such as obesity—often include “real-person” sources because journalists believe this mechanism makes stories more interesting. Dorfman, Wallack, and others 56 have postulated, however, that the inclusion of these individual-focused anecdotes may draw attention to the individual and lead readers to assign responsibility to, rather than empathize with, the individual even when the rest of the article makes the case that the individual is not responsible. Other researchers have noted the importance of personal narratives in explaining complex health issues, noting that positioning individuals within a story may improve comprehension and positive health behaviors. 57 However, the same research cautions that many variables may factor into how individuals respond to the ways health information is framed, ranging from their emotional states and their prior knowledge to their perceptions of an issue and their media consumption habits.
Although it is important for health journalists to thematically frame their articles on all health topics, something about the topic of coverage remains a powerful force in determining how the reader will feel about the piece. Framing theory, especially the thematic/episodic and gain/loss aspect, which is so important to health communication, should incorporate the role of topic and seek to understand the underlying dimensions that drive readers’ frame selections.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.