Framing Women’s Health With a Sense-Making Approach: Magazine Coverage of Breast Cancer and Implants Julie L. Andsager Edward R. Murrow School of Communication Washington State University Angela Powers Department of Communication Northern Illinois University This study examined how women’s magazines framed breast cancer and silicone implants to determine whether they used a sense-making framework. Sense-making calls for existing gaps to be closed between what one group views as real and what another group experiences (Parrott, 1996). Analysis included 86 articles on cancer and implants published in four women’s magazines from 1990 to 1997. Overall findings suggest that women’s magazines used a sense-making approach to cancer coverage, framing the disease in terms of coping with its effects, personal experiences, and risk factors. Themes in implant articles pertained to economic concerns of the medical industry and media. The news media are charged with covering issues in a fair, balanced, and informative manner. However, there is constant tension between the social impact versus the economic health and well-being of society (Parrott, 1996). In news, because resources are scarce, decisions must be made about what stories to cover and how they are presented. News framing refers to this selection and emphasis of certain aspects of issues. For magazines, these issues include elements of drama and timeliness, which some critics argue are intended to increase profits rather than offer balance (Parrott & Condit, 1996). Social scientists increasingly believe that media reporting of women’s health problems must be understood as socially, culturally, and economically influenced. A review of the literature suggests that these influences are not consistent in the way they shape media coverage, however. Ruzek, Olesen, and Clarke (1997) stated that the media often take women’s health out of context, presenting unrealistic views of cures and prematurely reporting progress. In addition, the media pay little attention to the downside of medicine, such as treatments that carry more risks than benefits or that contribute little to health conditions. They contend that in this cultural context beliefs about curing support large private and public investments in biomedicine. On the other hand, Vanderford and Smith (1996) analyzed media coverage of silicone breast implants and noted that these accounts “dramatized harm, trivialized benefit, and emphasized danger over safety” (p. 131). News about toxic shock syndrome also focused on its health harms (Weiner, 1986). Other researchers have suggested that women’s medical problems are described by the media and understood as deviations from the norm (K. Johnson & Hoffman, 1994). Furthermore, Fisher, Gandy, and Janus (1981) argued that the promotion of traditional women’s roles is linked to national health policy and funding priorities. In the United States, magazines have long been a major source of information about cancer (Freimuth, Greenberg, DeWitt, & Romano, 1984), and women have been found to turn to magazines in seeking information about breast cancer because the magazines “provide a critical base of information that [women] may use to evaluate subsequently acquired information … when they are confronted with a medical problem” (J. D. Johnson, 1997, p. 9). In women’s magazines, women’s health is a major concern; a content analysis of six prominent magazines found 694 editorial items published during a 5-year period (Kessler, 1989). Although Vanderford and Smith (1996) found that about half of women they surveyed sought information on silicone implants from magazines, as compared to 94% who used television news, magazines provide a richer context for information. Whereas television news is ephemeral, women can return to magazines to double-check statistics, symptoms, and advice. Moreover, women’s magazines, which combine elements of news and entertainment media, may have greater influence on women’s perceptions of risk of breast cancer than newspapers or newsmagazines because entertainment media exposure is positively related to perceptions of personal risk (Snyder & Rouse, 1995). In this study we examined how selected women’s magazines framed information concerning breast cancer—a disease afflicting women—and silicone breast implants, which can result in afflictions that many people believe women have brought on themselves. The two issues are important to study in tandem because of their related nature and the fact that breast implants were often recommended by physicians to breast cancer patients as a solution for the disfigurement of surgery. Moreover, silicone implants can impair detection of breast lumps during mam- 164 ANDSAGER AND POWERS mography (Cowley & Ramo, 1993). Thus, breast cancer and implants have been intricately entwined as women’s health issues. The silicone implant story came to light in 1990 when the media reported that some women were having problems with their implants. Previous studies analyzing newspaper coverage of breast implants indicate that most stories relied on official sources from implant manufacturers, presenting only one side of the controversy (Powers, 1997), and that women’s voices were marginal in newspaper discourse (Andsager & Smiley, 1998). Also in 1990, estimates of breast cancer among U.S. women jumped to 1 in 10 (Kahi & Lawrence-Bauer, 1996). An earlier study of breast cancer coverage indicated the media were more attentive to shifting relations and priorities within the health care industry than they were to the public interest, a priority reflected in a steadily increasing breast cancer mortality rate (Corbett & Mori, 1997). To extend such research, we analyzed how four women’s magazines covered these two controversies from 1990 to 1997. THEORETICAL PERSPECTIVE Framing occurs as journalists “select some aspect of a perceived reality and make [it] more salient in a communicating text, in such a way as to promote a particular problem definition, causal interpretation, moral evaluation, and/or treatment recommendation for the item described” (Entman, 1993, p. 52). As Entman (1991) noted, frames emerge as the presence or absence of certain key words, sources of information, and sentences that form thematic clusters. Media scholars have argued that it is important to understand the ways in which journalistic framing of issues occurs, because framing influences public understanding and, consequently, policy formation (Gans, 1979, 1983; Gitlin, 1980; Pan&Kosicki, 1993; Tuchman, 1978). In the abortion debate, for example, the media were found to advocate unique policy messages (Terkildsen, Schnell, & Ling, 1998). In the realm of health-policy formation, framing may have a large impact on public health (Walsh-Childers, 1994). How health and risk issues are framed is important because “the media may affect the nature of regulation, the course of litigation, or the direction of research and development” (Nelkin, 1989, p. 54). The media tend to reinforce conventional definitions of health problems and, hence, the legitimacy of various solutions. News stories about medicine “may increase or diminish the willingness of individuals to present themselves for care, and raise expectations, and dash hopes, or may provoke alarm” (Winsten, 1985, p. 7). During the 1990s, several policy changes related to breast cancer and implants were enacted, in part as a result of raised public concern and media coverage of that concern. In 1990, Congress approved a bill that allowed Medicare to cover biannual mammograms for women over age 65, as well as screening for poor women in four states (M. Beck et al., 1990). In 1992, however, Congress passed a can- FRAMING WOMEN’S HEALTH 165 cer-screening law that delegated cancer control efforts to state health departments (Cancer Registries Amendment Act, 1992). In 1994, the National Cancer Institute said it would increase its funding for breast cancer research 34% from 1993, due in part to a grassroots campaign conducted by the National Breast Cancer Coalition and President Clinton’s directives (Gorman, 1993). In January 1992, the Food and Drug Administration implemented a ban on silicone implants until their effects could be better understood (Schwartz & Kaplan, 1992). The notion of story frame comes from Tuchman’s (1978) work on the social construction of reality within media accounts. Certain pieces of information are selected and put together within the genre constraints of a news story. These choices, based on news values and journalists’ interpretations of social responsibility, have consequences. Readers form impressions of the news stories’ central theme–issue and attitudes toward the policy actors (Hornig, 1990). For the most part, the media are the public’s only contact with technical fields. Journalists can be perceived as brokers, “framing social reality and shaping the public consciousness about science” (Nelkin, 1989, p. 54). Thus, the media create boundaries within which debate can take place. Media coverage of controversial technologies has a tendency to highlight competing interests of the policy actors through disputed data and opposing judgments about risk. When scientists are identified as policy actors, they are perceived as “the source of authoritative evidence and definitive solutions” (Nelkin, 1989, p. 55). However, if the media did not explore scientific issues involving health risk disputes, the public would be left with little or no basis for making meaningful judgments about policy actors’ competing allegations. Thus, medical professionals are granted expert status in health risk conflicts, such as the value of mammograms and the danger of silicone implants. Corbett and Mori (1997) found that the media were generally supportive of medical community activities and willing to follow their lead in changing emphases and priorities regarding breast cancer. Source Factors Overrelianceonofficial sources is of particularimportancein light of increasing criticisms of the media being out of touch with women’s issues. According to Tuchman (1977), use of official sources and public relations subsidies helps satisfy the strategic ritual of balance. Too often, sources are used to cite facts without further investigation and to give credibility to the reporter’s point of view (Ericson, Baranek, & Chan, 1989). In this way, sources can influence the issue’s frames. Entman (1993) posited that “the frame in the news is really the imprint of power—it registers the identity of actors or interests that competed to dominate the text” (p. 7). Powers and Fico (1994) found that despite the availability of numerous information sources, journalists’ decisions to use them may be influenced by concerns 166 ANDSAGER AND POWERS other than audience needs or adherence to professional standards. Results from a survey of journalists from top-circulation newspapers indicated that news content was most powerfully shaped by journalists’ own orientation toward source qualities. The journalists’ judgments had the most powerful and numerous influences on source selection in both routine and conflict situations. According to Sigal (1973), such beliefs have resulted in a lack of source diversity. Furthermore, research indicates that men far outnumber women as newsmakers and sources. Because women do not hold as many prominent positions in society as do men, their lack of status as experts makes it more difficult for the news media to include them in coverage. Although news organizations claim objectivity in their reporting, they do not claim to balance the viewpoints of men and women (Rakow & Kranich, 1991). Others have suggested that news is written for the political and economic interests of men. For example, distinction has been drawn between hard news, which is serious, important, and masculine in nature, and soft news, which is likely to be written by female reporters for female readers (Rakow & Kranich, 1991). Hartley (1982) stated that the majority of news stories are written by men about men’s issues. Holland (1987) added that two prominent themes in television news—excitement and seriousness—are both based on masculine values. This is why news is not one of women’s favorite categories of television programming (Morley, 1986). Both television news and newspapers, the media outlets analyzed in these studies, are directed at general audiences; it appears that male interests tend to dominate those presumed of the general audiences. However, women’s magazines should not be as subject to this criticism because their writers may be more likely to be female, and their audiences are clearly dominated by female readers. Although little research has examined news framing of breast cancer, early studies on implant coverage suggest that the media were biased against their use (Anderson & Larson, 1995). Palcheff-Wiemer, Concannon, Conn, and Puckett (1993) surveyed women who had breast implants. Respondents described the reporting as biased, sensational, and overgeneralized. Vanderford and Smith (1996) found that the media relied on dramatic narratives that exploited an audience’s identification with victims. The researchers concluded that sensational narrative forms portrayed dangers, trivialized medical benefits, and portrayed rare problems that created negative perceptions of the silicone implants. A Sense-Making Approach The key assumption of a sense-making approach states that gaps exist between what one group views as real and what another group experiences (Parrott, 1996).A sense-making approach requires that these gaps close. The media contribute to gaps of knowledge by omitting information, such as overlooking the importance of so- FRAMING WOMEN’S HEALTH 167 cial support networks (Parrott, 1996). According to Parrott, these social support networks are discounted because they do not involve the use of medical technology or drug therapy and may reduce the need for such treatment. Such a contention is supported by previous research indicating that journalists tend to rely on official or authoritative sources, such as doctors, scientists, and corporate spokespeople (Gans, 1979; Gitlin, 1980; Shoemaker & Reese, 1996; Tuchman, 1978). Another gap in messages communicated includes overlooking what women may have already learned about their health. A lack of understanding of facts and procedures will keep women from making informed decisions on their health. In the case of breast implants, for example, many studies, some funded in part by implant manufacturers, found no conclusive link between implants and autoimmune-related diseases. Other studies indicated that such links existed. Reports of studies of silicone implants, like other medical news, were “characterized by ambiguity, uncertainty, and contradiction” (Vanderford & Smith, 1996, p. 110). The conflicting studies may have resulted in confusion, preventing some women from seeking help. Unless media reports question the validity of manufacturer-financed studies in particular, the social structure linking science and corporate interests as a major influence on women’s health will not be elucidated. Examination of groups’ and institutions’ responsibilities for the factors affecting health care availability and funding is part of a sense-making approach to covering women’s health (Dervin, 1989; Parrott, 1996). To develop a sense-making approach, creators of health messages need to determine how the consumers of those messages make sense of their lives, how they connect health messages to their personal behaviors and social structures (Dervin, 1989). More promotion of self-efficacy in the media is also important in a sense-making approach to communicating about women’s health. Self-efficacy is the belief that women can exert control over their motivation, behavior, and social environment. If women know they have control over their environment and that agencies are available to help, there is a greater likelihood that they will benefit from the media coverage concerning their health and will take action (Parrott, 1996). Increased attention to the institutions and agencies that affect women’s health may cause them to “reassess their priorities and agendas” (Parrott, 1996, p. 415). Theuse of a women-centered sense-making approach—including the use of avariety of sources, facts from all sides of an issue, and information promoting self-efficacy— may provide women and the public with a better understanding of their health. Women-centered means that media messages about breast cancer would focusonwomen’sneeds to obtainmammogramsor practice self-examinations, for example, as well as examining barriers in the environment that affect women’s health and efficacy, such as the availability of breast cancer screening forwomenin low-income or remote geographic areas. Messages about silicone implants would discuss howsociety influencessomewomento feel the need to obtain implants orwouldemphasize their personal experiences rather than the economic impact on implant-pro- 168 ANDSAGER AND POWERS ducing corporations—all of which are socially constructed influences on the women’s health care industry. In other words, a women-centered, sense-making approach to coverage of these health issues would allow alternative perspectives to compete against the dominant frame provided by individuals in power and transmitted by means of the media (Dervin, 1989). Such alternative perspectives could focus on women’s own experiences regarding their health or on activist or support groups that challenge the conventionalwisdomandenvironmentaffecting research funding and availability of health providers, for instance. According to Parrott (1996), the identification of linguistic discontinuities or gaps in communication between campaigners’—or, in this case, the medical, scientific, and economic communities’—and audience’s experiences is critical to the success of communicating with women. In this light, we ask the following research questions: What issues do women’s magazines address concerning breast cancer and implants? What types and numbers of sources are used in reporting such stories? What frames emerge from magazine stories on breast cancer and implants? METHOD Four women’s magazines were selected for analysis: Good Housekeeping (GH), Ladies’ Home Journal (LHJ), McCall’s, and Ms. Ms. magazine does not have one of the highest circulation rates, but we included it because it does not accept advertising and thus should have less influence from economic interests. GH, LHJ, and McCall’s each have a circulation of about 5 million, with readers averaging about age 40 (Endres, 1995; Gottlieb, 1995; Zuckerman, 1995). Ms., which is aimed at a feminist audience, has a much smaller circulation of about 250,000 (Prijatel, 1995). The articles analyzed include all of those listed in the Readers’ Guide to Periodicals from 1990 to 1997 under the headings “breast cancer” and “breast implants” for these four magazines. These years were chosen because significant events occurred during this period that related to both implants and breast cancer. Each article was examined in its entirety. We used two content analysis methods. We used traditional content analysis to identify the main issues addressed and the number and type of sources used in the articles. This analysis was important because of claims that too often official sources are relied on by the media when presenting controversial information. In the case of breast cancer and breast implants, a need exists for expanded coverage presenting the viewpoints of not only the medical establishment but also of the patients. The unit of analysis was the magazine article. Two coders analyzed each story. After a pretest, the level of intercoder agreement was 98%. We used chi-squares to identify whether differences in coverage and sources existed among the magazines. FRAMING WOMEN’S HEALTH 169 A computer-assisted content analysis program was used to determine the frames in coverage. Two files—one for breast cancer and one for implants—were formatted for analysis in VBPro (Miller, 1993b), a program that calculates frequency of all terms in the stories. The researchers select key terms based on the articles’ content; the program then groups these terms on the basis of their co-occurrence within articles to illustrate the frames (Miller, 1993a). We selected 120 of the most frequently occurring, unambiguous, and representative terms in the breast cancer and implant articles. Synonyms, such as doctor, doctors, physician, and physicians, were coded as one term. The magazines were also coded so that any differences in their frames would be apparent. The program produced unstandardized eigenvectors for each of the terms based on their co-occurrence within stories. We cluster analyzed these eigenvectors in SPSS (SPSS for Windows, 1999) to group the terms in order to determine frames, using hierarchical clustering with the cosine method. The eigenvectors for each cluster (frame) can be used to plot the cluster in three-dimensional space to illustrate the relations among frames. This process is discussed further later in the article. Selection of terms and subsequent cluster analyses were conducted separately on the breast cancer and implant articles. RESULTS