Content-Type: text/html Social Construction of Depression in Newspaper Frames Cynthia-Lou Coleman* Associate Professor Portland State University Jessica A. Corbitt Public Affairs Officer Fulton County, Georgia Presented August, 2001 Association for Education in Journalism and Mass Communication Newspaper Division Please address correspondence to Cynthia-Lou Coleman [log in to unmask] Department of Communication Portland State University P.O. Box 751 Portland, Oregon 97207-0751 ABSTRACT Social Construction of Depression in Newspaper Frames Cynthia-Lou Coleman* Associate Professor Portland State University Jessica A. Corbitt Public Affairs Officer Fulton County, Georgia Presented August, 2001 Association for Education in Journalism and Mass Communication Newspaper Division The content analysis study examined the social construction of depression in United States news stories over a 14-month period. Three key areas, drawn from the literature, are set against the backdrop of stigma: news routines, exogenous influences and ideological frames. The authors found that stigmatization occurred in stories that relied on routines that focused on violence. Use of the stylistic practices personalization and solutions tended to blame the victim while simplifying causes and cures. Market interests paralleled mentions of medicine-therapies, which were twice as likely to be noted as other therapies, such as counseling. Frames, including the medical disease model and war, placed the individual at the center of depression, rather than viewing the individual within the societal or cultural context. The authors discuss the implications of such coverage. Please address correspondence to Cynthia-Lou Coleman [log in to unmask] Department of Communication Portland State University P.O. Box 751 Portland, Oregon 97207-0751 Depression, page SOCIAL CONSTRUCTION OF DEPRESSION IN NEWSPAPER FRAMES Mental health care professionals have long been interested in stigma surrounding mental illness, and during the last decade government and non-governmental agencies alike have redoubled their efforts in the United States to remove stigma from perception of mental illness. The response has been to launch public education campaigns to heighten awareness of the pervasiveness of mental illness, while challenging notions that mental illness results in violent behavior. One target of such campaigns is media portrayals of the mentally ill. Numerous empirical studies have focused on such portrayals, ranging from depiction of the mentally ill in television entertainment programs (as part of the cultural indicators project at the University of Pennsylvania) to a study of suicide rates following news of Kurt Cobain's death in 1994 (Jobes & Berman, 1996). On balance, researchers concur that portrayals of the mentally ill are stereotyped, stigmatized and just plain false. While critics charge that print media also mischaracterize the mentally ill, studies about newspaper coverage of mental health have taken a back seat to research on movies and television, with some notable exceptions--the Glasgow Media Group's analysis of mental illness in British media (1996) and Wahl's examination of obsessive-compulsive behavior in news (2000). The current study attempts to fill some of the gaps in newspaper coverage by examining the stigmatization of mental illness through the social construction of depression over a 14-month period in the United States. We contend that the study of depression offers a singular opportunity to examine elements of stigma against the backdrop of heightened public awareness of clinical depression and its pharmaceutical treatments. In addition to being a timely topic, depression's impact is felt by more than 19 million Americans (one in four women and one in eight men), and most sufferers remain untreated.[1] The study asks whether and how stigma occurs in the news construction of depression by examining aspects of news routines, influence of exogenous factors, and framing. The paper begins with a review of relevant literature concerning stigma, then defines depression. We turn to aspects of news coverage, specifically routines, and then address external forces that drive coverage, such as pharmaceutical market interests. We next turn to framing, focusing on the role of metaphor and interpretive packages that reveal underlying values that may stigmatize coverage of depression. Next, the hypotheses are presented, followed by methodology and findings. Finally we discuss how the study adds to the current understanding of the social construction of depression in newspapers. Literature Review Stigma Stigma is defined as a mark of shame or discredit, and is similar to the construct "credibility," which can be described as an attribute of an individual ("the witness was highly credible") or a construction of individual perception ("I found her testimony quite credible") (for a discussion, see Gunther, 1992). The parallel is that stigma can reside with the individual who is mentally ill, and is also a function of how publics perceive the mentally ill. We argue that mass media play a pivotal role in constructing the social reality surrounding depression, which in turn, helps shape perceptions about people with depression. Stigma was defined by Goffman (1963) as a "spoiled social identity," and one that was deviant-not normal-judged by others' standards. This definition is most widely accepted among researchers today (Harvey, 2001). Stigmatized individuals are rejected by society and are not altogether human (Taylor, 2000). Finally, stereotypes reinforce value systems by portraying deviant individuals as violators of those values (Pitchford, 2001). In order to demonstrate that news coverage stigmatizes those who are depressed, we need to compare an objective or standard definition of depression with the socially constructed version, and ask how these vary from one another. Take the example of schizophrenia. Wahl (1995) argued that schizophrenia is neither split nor multiple personality. Yet mass media "strongly and repeatedly" communicate that schizophrenia and split personality are one in the same. We turn to the leading scientific, professional and government organizations that have a stake in depression to locate definitions. Defining Depression One of the leading professional mental health organizations, the American Psychiatric Association (APA)[2], classifies depression as a "mood disorder," describing it as "changes in the body's chemistry (that) influence mood and thought processes."[3] The Diagnostic Statistical Manual of Mental Disorders (IV) defines depression as "a mood disorder characterized by at least one major depressive episode with a clinical course."[4] The National Alliance for the Mentally Ill (NAMI), an advocacy group, defines mental illness (including depression) as "disorders of the brain that disrupt a person's thinking, feeling, moods, and ability to relate to others. Just as diabetes is a disorder of the pancreas, mental illnesses are brain disorders that often result in a diminished capacity for coping with the ordinary demands of life."[5] The National Institute of Mental Health (NIMH), a branch of the U.S. Department of Health and Human Services, details three types of depression: major depressive disorder, dysthemic disorder and bipolar disorder. Major depressive disorder is defined as a class of symptoms, and "a diagnosis of five or more of the symptoms" reflects major depressive disorder, while dysthemic disorder is characterized by seven of the symptoms that persist for two years. Bipolar disorder, the third type, is characterized by alternating states of mania and depression.[6] Depression is highly treatable, according to the Websites above. Recommended treatments for depression include a combination of "talk therapy" and drug therapy. To summarize, depression is defined by traditional stakeholder groups as changes in body chemistry, brain and mood, but is also defined as a collection of symptoms. Thus, in order to discover if stigma occurs in media coverage, we would expect to see depression characterized in opposition to definitions stated by leading professional organizations. And while critics might argue that such definitions are far from objective, they are nevertheless the standards used to explicate depression. Stigmatized News Coverage Stigmatized coverage may be the result of numerous factors that influence how news is created, and many theories address news construction: framing, filtering, agenda setting and building, reporter bias, organizational bias, channel characteristics, advertisers' influence, hegemony, ideology, space and time constraints, gatekeeping, and more. Our focus is on how the news routines of deviance coverage, personalized bias, the generalizing portrayal, attribution of blame, normalizing bias and source use influence the construction of depression. News Routines. Characteristics that make individuals and events newsworthy include such attributes as timeliness, proximity and extraordinary occurrences. Violent behavior and deviant acts engender news, and stigma occurs, in part, when individuals or groups are described as deviant (Dornan, 1990; Gitlin, 1980; Shoemaker, 1984). One way in which the mentally ill are described as deviant is to characterize them as violent and as outsiders, although practitioners and scholars alike have argued strenuously that violent behavior is rare among the mentally ill (Glasgow Media Group, 1996). Still, violence makes news and we would therefore expect to find mass media reports about the mentally ill focused-in part--on violence, while depressed persons are delegitimized as abnormal. Individualism. Routine reporting also focuses on individual-level explanations of phenomena, rather than social system explanations. Rucinski (1992) contended that emphasis in news content on individual actors rather than institutional forces "is implicitly to argue that other factors like cognitive orientations or cultural values are less potent explanatory factors" (p. 94). Individual-based explanations of phenomena, called "personalized bias," are embedded within the traditions of news coverage (Rucinksi; Price & Czillia, 1996). Bennett (1988) noted that if journalists were truly critical they would, "minimize attention to the individual power struggles, the personal motives and fortunes of leaders, and_instead_report on the evolution of issues, the powerful institutions involved and the factions-like lobby groups, special interests and powerful bureaucracies" (p. 26). In her book on science coverage, Dorothy Nelkin (1987) noted that, "Even journalists who cover numerous risk c ontroversies for the daily papers seldom question the 'system': they prefer to conceptualize each problem separately in terms of aberrant individuals, bad companies, or ignorance" (pp. 104-105). Scheff (1984) has argued that social and cultural settings are overlooked when theorists concentrate on the individual aspects of mental illness. When neurotic behavior is constructed as a disease-the medical model--it unfolds "relentlessly out of a psychological system that is entirely contained within the body" (p. 7). Scheff suggested researchers should instead examine the social system in which mental illness arises, while asking, "What are the conditions in a culture under which diverse kinds of rule-breaking become stable and uniform_[and when] are symptoms of mental illness the result of conforming behavior?" (p. 12). Similarly, Foucault (1965) asserted that the definition of madness is anchored in its cultural ethos, ranging from a function of idleness to romance, immorality to creativity, and violence to art. Foucault challenged madness as defined within a positivistic framework, asking, "was it madness, or a work of art? Inspiration, or hallucination_must its truth, even before its birth, be taken from the wretched truth of men, or discovered far beyond its origin, in the being that it presumes?" (p. 286). Two researchers studying television coverage in Germany labeled one aspect of personalization as the "generalizing portrayal." Here the reporter takes an individual case, and then broadens the story by providing statistical or base-rate information. Such base-rate information is combined with illustrative cases, or exemplars: a "stylistic device used to exemplify an issue or its importance" (Daschmann & Brosius, 1999). In other words, one individual is used to illustrate all such cases, which may lead to misperception of-in this case-depression. Thus, in the context of depression, we can surmise that journalists tend to focus on personal and individual attributes and causes, rather than the social system in which mental illness occurs. Blame. While such coverage obscures the social context of disease, personalized bias most likely frames victims as culpable. Susan Sontag (1978) pointed out that, although the illness is the culprit, "theories of disease assign to the luckless ill the ultimate responsibility both for falling ill and for getting well" (p. 57). Stigma, therefore, may be fueled additionally by blame or causes attributed to depression. The literature suggests that, when individuals characterize the mentally ill as personally responsible for their affliction, then they are less compassionate and less accepting of them (Valkenburg, Semetko & De Vreese, 1999). Stigmatized coverage would therefore "blame the victim." Solutions. News stories that embrace controversy or threats often contain a solution. Some suggested this is to reassure publics by "normalizing" threats in order to protect communities (Donohue, Olien & Tichenor, 1989; Fredin & Kosicki, 1989). Bennett (1980) described this process as normalizing the news, meaning, no matter what the crisis, the journalist will "nearly always return the situation to normal by the end of the story." Solutions, Nelkin asserted, reside under the purview of scientists, and their perspectives largely go unchallenged. Similarly, we would expect to find in news coverage of mental health that scientific experts are legitimate but unquestioned voices in setting the parameters for the discussion of depression, and that they are empowered to chart the solutions to mental illness. Source use. Choosing which sources to use in news coverage is an accoutrement of the news package, and some argue that sources help shape the news agenda (Gitlin, 1980; Herman & Chomsky, 1988). The underlying assumption is that readers adopt the slant delivered by the source-and that sources have their own biases. In science coverage, bias leans toward scientific, rather than social system perspectives, and scientists typically are quoted without challenge (Nelkin, 1987). In coverage of depression, we would expect members of the medical and scientific communities to comprise the bulk of sources used. Exogenous Influences Advertising The role of advertisers in shaping and censoring news has been the subject of much discussion. In their book Manufacturing Consent, Edward S. Herman and Noam Chomsky (1988) argued that advertisers bring pressure to bear on editors and reporters, whether direct or subtle, resulting in self-censorship of news or retooling of news to assuage the advertiser. Others have argued that newspaper editors face pressures from advertisers on an ongoing basis (Donohue, Olien & Tichenor, 1989; Soley & Craig, 1992). Concrete examples included: advertisers pulling their accounts from media organizations after reading unfavorable coverage (Lacher & Rotfeld, 1994); advertisers requiring media to eschew editorial (news) content that ad agencies find offensive; and requests to review editorial copy before publication (Shelton & Curtis, 1998). In one survey of newspaper editors, 84% said advertisers had attempted to influence news, and 70% said advertisers had tried to kill stories (Soley & Craig, 199 2). Indeed, advertisers, marketers and public relations practitioners find methods to massage the message in news. This also holds true for pharmaceutical organizations promoting drugs for mental illness. News copy is "more compelling" than an ad, according to a 1998 article in Pharmaceutical Executive that gives pointers on how marketers can best promote their products (Levins & Rodgers, 1998). The authors suggest placing stories in local press in order to "drive sales." One way to attract news is to "promote a problem your product can solve." In other words, a marketer who sells antifungal cream should be able to engender news stories about the heartbreak of toe fungus. And a good marketer should also provide the solution to the problem with mention of the brand name and (positive) value statements about the product. Pharmaceutical interests. Clearly pharmaceutical companies have a vested interest in the social construction of depression. This is particularly true in light of the introduction of new antidepressant medications, specifically the class of SSRI drugs--selective serotonin reuptake inhibitors-which include heavily advertised brands such as Prozac and Paxil. The SSRI medications have been regaled because side effects are reportedly much less severe than earlier antidepressant medications. Coincident to the advent of new drugs is the explosion of direct-to-consumer (DTC) advertising, which has grown exponentially since the FDA lifted the moratorium on advertising to consumers in 1985. Overall spending on DTC ads grew nine-fold from $55 million in 1991 to $516 million in 1996, and then doubled the following year from $516 million to $1 billion. Such advertising now brings prescription drug pitches directly to the consumer, whereas, prior to the mid-1990s, such pitches were channeled directly to healthcare professionals. As a result, consumers are arguably more familiar with antidepressant medications as well as such afflictions as panic disorders, obsessive-compulsive behavior and clinical depression. Moreover, television viewers and magazine readers are led to believe they can find solutions to these problems in pill form. Thus, manufacturers of such drugs are keenly aware of the social construction of mental illness and its solutions, and have a stake in defining depression within a medical disease model. That is, if depression is a chemical imbalance, taking SSRI pills will restore that balance. But not all scientists agree. While some drugs may lift sadness they cannot solve the issues that brought on depression in the first place. Frank (1999) wrote, "although many individual depressive episodes can be temporarily 'fixed' by antidepressants, the drugs are not curative, no more than insulin cures diabetes...the demon always returns at some point." Our expectation is that coverage would include in its solutions the SSRI medications, which would be seen as groundbreaking new therapies. Framing We turn our attention to the social construction of phenomena in news coverage with an examination of framing and metaphor, borrowing liberally from scholars who have conceptualized framing as central organizing ideas that undergird news accounts. Framing by definition incorporates news routines as well as exogenous variables such as advertisers' influence. In his treatise on coverage of the Vietnam War protests, Todd Gitlin (1980) characterized framing as "core principles of selection, exclusion and valuation" (p. 23). From this perspective, framing includes how news ideas are generated, where information is gleaned, which sources are quoted, which issues get spotlighted, and how explanations and descriptions unfold (Gamson & Modigliani, 1989; Gitlin, 1980; Hartley, 1989; Iyengar, 1990; Kinder & Sanders, 1990; Luke, 1987; Ryan, 1991). Many writers suggested that framing reveals underlying values and ideologies of those with the power to name and define phenomena. Reese (1990) not ed that media "accept the frames imposed on events by officials and marginalize and delegitimate voices that fall outside the dominant elite circles_the media establish what is normal and deviant by the way they portray people and ideas" (pp. 394-395). Metaphors and frames are interlocked, and the use of metaphors in framing can reveal underlying meanings, as well as provide the implied solution to problems. For example, if society is a machine, then you need a trained engineer to run it (Gozzi, 1999). Similarly, if illness is constructed in a medical disease model, you need medical technicians and treatments to cure it. The Medical Disease Model Others have argued that if mental illness is framed as a medical disease construct, then the public will have greater tolerance for the mentally ill: "If only the public could be persuaded mental illness is like any other illness, just as diabetes is a disorder of the pancreas" (Read & Law, 1999). Thomas Szasz (2000) has been an outspoken critic of the medical disease model in mental health, arguing, "diseases of the mind are not diseases of the brain." He also noted that the social construction of mental illness as a disease, like the construction of alcoholism as "chemical dependence," allows the illness to be legitimized as outside the individual's control. And while the model may indeed lift some of the onus of blame from the individual to her or his brain, Wahl maintained that the model is an invented reality, a construction created in part by those with a vested interest in how the terms are defined. Others have argued that the medical disease model emphasizes pathology, which results in concentration on the symptoms of disease--not necessarily the causes-and thus placing the clinician in the role of expert. Davis & Jansen (1998) noted that most illnesses can be defined in terms of biochemistry, and even love has a chemical explanation (Neuhaus, 1993). Such scholars emphasize that, by focusing on disease, we lose sight of its social context. Thus we would expect the medical disease model to permeate coverage of depression. Metaphors Metaphors effectively convey meaning by loading multiple meanings into one phrase. For example, in a study of the coverage of an environmental controversy Coleman (1994) found that the metaphors used to describe protesters included "outsiders" who would "lead us back to the Stone Age," while proponents in the controversy were labeled "partners in progress." Each metaphor deftly conveyed myriad meanings with negative and positive valences. And such metaphors equip readers with the cues necessary to make sense of narratives (Lakoff & Johnson, 1980). War. Nelkin noted that metaphors in science include the imagery of war. By equating science and medicine with war, experts are not "questioned, new technologies must go forward, and limits are inappropriate" (p. 81). Thus, the spectrum of meaning for the word "war" covers a vast territory, including such concepts as victors and losers, heroes and villains, gods and devils. The war metaphor frequently appears in medical constructions. Doctors, scientists and researchers are described as militia whose job is to wipe out, destroy and kill invading viruses, bacteria and diseases. For example, in media coverage of the Ebola virus, war metaphors pervaded the discourse, with the bug described as stalker and predictor, sufferers as victims, and physicians as heroes (Ungar, 1998; Weldon, 2001). In a similar vein, we would expect coverage of depression to include war metaphors, and that solution frames would emphasize the efficacy of medical treatments over all others. To summarize, we have asserted that the social construction of depression is determined (in part) by news routines, exogenous variables such as advertising influence, and by framing. We now turn to our hypotheses. Research Question and Hypotheses To recap, we believe that examining the social construction of depression in news will reveal the shape of stigma.[7] News Routines Hypothesis 1. News stories on depression will vary from official constructions of depression by focusing a substantively significant portion of coverage on violence. Hypothesis 2. On balance, depressed individuals will be deligitimized with derogatory mentions or labels, thus indicating they are abnormal. Hypothesis 3. News reports about depression will significantly reflect the personalization bias by focusing on individuals. Hypothesis 4. The generalizing portrayal will be incorporated into personalized news stories about depression. Hypothesis 5. When causes for depression are noted, individuals will be mentioned as responsible for their illness significantly more frequently than other causal factors. Hypothesis 6. News stories about depressed individuals will incorporate solution frames. Hypothesis 7. The majority of sources quoted in stories about depression will be significantly more likely to include medical and scientific personnel. Exogenous Variables Hypothesis 8. When treatments or cures for depression are discussed, prescription medication will be significantly more likely than other treatments to be recommended. Hypothesis 9. When medication for depression is mentioned, best-selling SSRI brands will be most likely to be named. Hypothesis 10. Depression will be defined as a chemical imbalance or brain-related problem. Framing Hypothesis 11. Coverage of depression will employ the medical disease model frame when describing the illness. Hypothesis 12. Coverage of depression will employ war frames when describing the illness. Methods A 14-month time frame of August 1, 1999 to October 1, 2000 was chosen to sample news coverage concerning depression, using the Lexis-Nexis database of daily newspapers in the United States. The first foray into Lexis-Nexis revealed thousands of articles that contained the word "depression." A close examination revealed that many of the stories focused on the economy or weather, and others were not relevant to clinical depression. By limiting the pool of stories to the term "depression" and the root "depress" we reasoned that the stories would be more likely to have depression as its central theme.[8] A total of 640 news stories with the term "depression" and "depress" were found in the time frame. In the spirit of parsimony, and in order to achieve a confidence level of 95%, a sample size of 70 to 85 would be necessary for sufficient reliability. To ensure an adequate sample size we selected 110 stories using a random, blind method[9]. Of the 110 stories chosen, 37 were deemed irrelevant: some simply did not focus on clinical depression, while others were columns, letters or opinion pieces. As a result, 72 news stories passed muster. Using a confidence level of 95%, the sample size of 72 yielded a confidence interval of 11. Two researchers devised the coding scheme in order to answer basic questions about the stories-the newspaper name and size, date, story length, etc.-and to address the research questions and hypotheses. Questions explored both manifest and latent content. Whenever possible coding was designed to create independent, rather than overlapping categories. Two individuals met twice after separate coding trials to compare answers and discuss areas that resulted in disagreement. Each researcher independently read and coded the news articles. Intercoder Reliability A random sample of three stories was chosen to check intercoder reliability. Coding fell into five categories: dichotomous variables (66% of all coding), three-way variables (11%), four-way variables (6%), seven-way variables (5%), and variables with 15 categories (11%). Observed and expected agreements were calculated for each category. Intercoder reliability, which was averaged and weighted across the five categories, was .83, and judged as acceptable. Operationalizations of Terms[10] Coders first decided whether or not depression was central to each story. Coders counted whether and how often the following terms were included in news stories and headlines: depression (and related terms such as melancholy, mania, sadness, etc.); technical classifications of mental disorders (bipolar, schizophrenia, etc.); mentions of sane, normal or healthy, and derogatory mentions of mental illness (crazy, nutty, wacko, etc.). Coders noted whether and how often mentions were made about violence, suicide, alcohol drinking, illicit drug abuse, stress, hormones, chemical imbalance and sexual abuse in relation to depression. Treatment categories included whether prescription drugs were mentioned as therapy for depression; whether specific antidepressant brand names were mentioned; whether therapeutic counseling was mentioned; whether a combination of medication and counseling was mentioned; and whether alternative treatments were mentioned (exercise, herbal remedies, etc.). Fifteen categories were selected based on information gleaned from government publications and the literature as primary causes of depression, and coders noted which cause (or "no cause") best fit each story.[11] Coders also noted whether a secondary blame was mentioned and categorized it. Stories were coded to reflect whether the primary subjects were men, women, both or neither. Coders also noted whether the stories focused on adults, elderly, children or teens, some combination of the above, or no group. A note was made if the story featured celebrities and whether stories referred to any individual or groups being "at risk" for depression. Story Issues Coders noted whether and how often references about depression were made to public policy (legislation, political agendas, regulations, etc.), and whether references were made to pharmaceutical issues, such as marketing, advertising, drug trends, etc. Mentions of specific scientific studies were noted, along with whether quantitative data were mentioned. Coders noted total number of personalized accounts was noted. Name, title, gender and affiliation were noted for all individual sources (not institutions) named in stories. Frames and Metaphors Coders noted the type of frames and metaphors that described depression or mental illness. Coders looked for the following specific frames: solution-resolution; anti-solution or hopeless; war; peace; scientific progress; anti-progress; medical-disease model; non-disease or anti-disease; pill culture; and anti-pill or alternative health care. A notation was made if the frame incorporated another frame and thus overlapped. Findings Although efforts were made to obtain a representative regional sample of newspapers, a greater number of stories was selected (randomly) from the west: 38%; with 25% from the south; 19% from the north; and 15% from the Midwest. Circulation size of newspapers ranged from 18,000 to 1,086,000. About 17% of stories were attributed to a wire service, with the remainder written by staff or unidentified writers. The majority of stories were from the news section of the paper (42%). Stories also appeared in the health (13%) and lifestyle or entertainment areas (11%). Other articles appeared in miscellaneous sections (10%) or in sports (6%). The remaining were unidentified. While the lion's share of stories was carried in section A (21%) and B (17%), stories also appeared in sections D, E, F, I, L and S. The majority of stories were on page 1 (29%) and page 2 (10%) of the section, with the remaining on pages 3 through 48. The average story length was 783 words (s.d.=500) with a range from 97 to 2,436 words. The majority of depression stories ran on Sunday (22%), followed by Tuesday (18%), Monday (17%) and Wednesday (15%) with the remaining on Thursday and Saturday (8% each), Friday (7%), or unknown. Nearly three-quarters (71%) of the stories focused on specific age groups, with adults being the most frequently described group (50%). About 10% of the stories concentrated on teens or children, while 11% of the stories concerned the elderly. Another 10% included a combination of age-types, and the remaining 19% concerned no particular group. Many stories focused on women--about 24%--while 17% concerned men only. Another quarter (26%) focused on men and women (in the same story), while the remaining 33% concerned neither. About 11% of the stories contained a depression checklist for readers, 25% had a referral telephone number, 13% gave an address, and 11% had an email address or website listed. Routines Stigma and Definitions. Recall our contention that stigmatized coverage of depression would vary from the definitions of conventional mental health stakeholders, and that stigmatized coverage would equate depression with violence while delegitimating those with depression. Suicide was mentioned in relationship to depression in 47%[12] of all stories while 21% of all stories mentioned violence. Another 21% mentioned stress, and 21% mentioned brain or chemical imbalance as relevant to depression. Alcohol was mentioned in 19% of stories, as was drug abuse (18%). Hormones were mentioned in 14% of stories and a relationship was mentioned in 13%. Violence was a common theme, occurring in one story out of every five, and suicide-which occurred in almost half of all stories-and is also arguably a violent act, and thus Hypothesis 1 is supported. Violence mentions did covary with suicide mentions: about 21% of stories that carried such mentions contained both suicide and violence. Some violence mentions concerned harming others (italics added): Kathleen Amper's remoteness got so bad that hospital personnel thought she was purposely trying to hurt her infant son...at the extreme, some patients neglect their baby, some deliberately harm it. Others commit suicide. (Whitely, 2000) Other mentions linked depression with violent thoughts, and in this example, the subject threatened to kill his classmates: He spoke of being depressed. "I don't see me having a very happy life," he said. "I'm not happy now"_eight pages [of a diary] of rambling teen angst adorned with swastikas_"I hate people so much I would be willing to kill them." (Pacenti, 2000) Other references to violence showed its causal relationship to depression: The study's findings offer further confirmation that traumatic experiences can have a profound effect on brain chemistry_A history of childhood abuse, studies have shown, puts people at higher risk for developing depression. (Goode, 2000). Deviant or derogatory mentions were indeed mentioned in 18% of stories, in support Hypothesis 2. Such mentions illustrated deviance by highlighting the abnormal qualities of depression: Karyn Bates is not your ordinary board chairwoman. But this is not your ordinary board. (Ventura County (California) Mental Health Board)_During severe bouts of mania, she rarely slept and did bizarre things. She recalls once walking into heavy traffic, stopping a shocked driver and asking for a cigarette. She stole food and cigarettes and ended up in jail. (Johnson, 1999, October 10, p. B-1) Many of those interviewed talked about disparaging attitudes: "It's like mental illnesses are inherently wrong," she said. "You'll never find a get-well card that says, 'Glad you're on anti-depressants. Hope things are going well'_you don't want something like that to be wrong with you because it's so taboo_you don't want people to say you're a nutjob"_some people think emotional problems are just signs that they need to quit being a whiner. (Forster, 1999) Personalization. Recall we also hypothesized that stories would incorporate the personalization bias by focusing on individual attributes of depression. About 45% of all stories included personalization, where at least one depressed individual was described at length, which supports Hypothesis 3. We also predicted that the generalizing portrayal would be incorporated into personalized stories about depression. About 40% of such stories specifically detailed quantitative data as background to the individual featured, in support of Hypothesis 4. In the following example, the article begins with one individual, and then fans out to incorporate statistics about the disease's effects--an illustration of the generalizing portrayal: Fourteen-year-old Stephanie (not her real name) used colored markers and white paper to draw a stick-figure image of herself being pushed off a cliff. (Hernandez, 2000) Later the story takes on general facts: As many as one in 33 children suffer from depression, according to the U.S. Department of Health and Human Services. Up to 3 million young people are estimated to suffer from a form of mental illness. Blame. Turning to causes of depression, stories were more likely to attribute depression to personal-level factors rather than societal factors. Indeed, societal factors were mentioned as the primary cause of depression in only 4% of stories. Turning to the list[13] of depression's primary cause (not mere mentions), "no blame" was found in 25% of stories, followed by blame to "self" (22%). The third most mentioned cause was chemical-brain imbalance (11%). The following categories all gleaned less than 5% of causal attributions: pregnancy or hormones, family, other disease or disability, someone else (not family or spouse), aging, work, or genetics. When comparing "no blame" to "self" blame, the difference was statistically significant when examining the Pearson chi-square statistic: p=.009. Using the Spearman correlation statistic for ordinal by ordinal measure was also significant: p=.008. This supports Hypothesis 5, that causal attributions to self are significantly different from "no blame." Following is an example of a story that placed blame on "self": Struggling with a sense of hopelessness and loss, Massing had grown far from the sturdy survivor that he had always been. Massing nearly did shut himself off from others_finally realizing he could not extricate himself from his depression on his own. (Oppermann, 2000) Societal values, while rarely mentioned, demonstrated coverage from a cultural rather than personal bias: In the survey, obese women were 37% more likely than women of average weight to have clinical depression in the past year. But obese men were 37% less likely than normal-weight men to suffer depression or to attempt suicide. "An overweight woman is so at odds with what's acceptable in this society, but our culture values muscular, powerful-looking males. It's the frail man who's at odds with society's ideal." ("Obese women more likely," 2000) Solutions. Solutions to depression were best captured as story frames, in view that mentions alone may not adequately depict a solution. Solution frames occurred in half (51%) of all stories, followed by war (29%) then disease frames (15%). This finding supports Hypothesis 6. The following examples illustrate Bennett's (1988) normalizing routine in which solution packages neatly wrapped up depression's cures: The problem got so bad in 1995, he agreed to meet with doctors suggested by the Sonics, was subsequently diagnosed with clinical depression, took his prescribed medicine, quit playing for two weeks and got better. (Newnham, 2000) A positive spin shows another solution frame: The happy part of the postpartum depression story is that, like any other depressive episode, it is largely treatable. And generally the treatment is the same: drugs and talk therapy_Seven months after the turmoil began, Ann thinks she has largely regained her equilibrium_She has stopped all of her medication and feels very much like her old self. (Uhlenhutch, 1999) Sources. We also hypothesized that the majority of sources quoted in depression stories would be medical and scientific personnel. We found that of all the sources[14] (182 total) quoted in all stories, 22% were counselors, psychologists or psychiatrists. Scientists, researchers and academics represented 16% of all sources, while health care workers, physicians and nurses comprised 13%. In all, 51 percent of the sources were from the medical or scientific community, in support of Hypothesis 7. Other sources included the depressed individual (22%) while family members made up 8%. Friends, neighbors and witnesses comprised 7%, law enforcement personnel comprised another 7%, elected or appointed officials 6% and legal representatives such as attorneys, 4%. Exogenous Variables The next set of hypotheses concerned drug therapy. Some type of therapy was noted in 65% of all stories, and among those stories, prescription drugs were mentioned twice as often as any other therapy--72% of the time, in support of Hypothesis 8. Other therapy mentions included counseling (36%), alternative treatments (28%) and drug-counseling therapy combined (12%).[15] Turning to brand names, specific drugs-like Prozac-were mentioned in only 15% of stories that included some type of drug treatment. However, when brands were mentioned, they tended to be the best sellers (in order of mentions): Prozac, Wellbutrin and Buproprion (the chemical name for Wellbutrin), Zoloft and Serzone (the chemical name for Zoloft), Paxil and Paroxetin (the chemical name for Paxil), Risperdal, Celexa, Xanax and Lithium (the chemical name for Eskalith). The most frequently mentioned brand-Prozac-is the top-selling antidepressant and ranked the fourth overall as the best-selling drug in retail sales in 2000[16]. The other brand names mentioned (in order) are ranked among the top 100 in all drugs sales as follows: Wellbutrin (ranked 22), Zoloft (7), Serzone (72), Paxil (8), Risperdal (21), Celexa (22), and Xanax (159). Lithium was ranked in another category-the top 200 best-selling generic drugs (rank of 115)[17]. This finding offers support for Hypothesis 9: of the nine drugs mentioned, six were in the top 25 of all drug sales, and two were in the top 160. Part of the pharmaceutical message concerning antidepressants is that the drugs offset the chemical imbalance of depression, and as such, we expected that depression would be defined in news copy as a chemical imbalance. Overall we examined both mentions, and the medical disease model frame. Mere mentions of chemical imbalances occurred in 21% of all stories, contrary to Hypothesis 10. Typical brain chemistry mentions were described as follows: Researchers now know that the disorder reflects a chemical imbalance in the brain as well as a deficiency of certain salts in the blood. Treatment involves restoring the balance. (Groves, 2000) A mutant suicide gene that affects brain chemistry in unknown ways may drive some depressed people to kill themselves, report Canadian psychiatric researchers who have studied the brain for 10 years. (Ferraro, 2000) Similarly we examined how the medical disease model-which incorporates the chemical imbalance definition--was framed. Frames Like solutions, the medical disease model was best captured as a story frame since mere mentions failed to explicate the disease model's attributes. Disease frames were located in 15% of stories, lending support to Hypothesis 11. Disease frames tended to describe depression "like any other disease." Some similes described depression as akin to diabetes, high blood pressure, heart disease or a broken leg. One patient noted: She says the key to her improvement in the last couple of years has been her medication. Without it, "I would have been dead years ago," she says. "My meds are definitely what keeps me alive, like insulin helps a diabetic_Depression's really rotten_It's an actual disease and it strikes and you don't know why it strikes." (Borsuk, 1999) And: Rebecca was depressed, not feeling down for a day or two every once in awhile, but suffering from a chemical imbalance in her brain. She is feeling better, now that she is taking medication. (Forster, 1999) War frames were found in 19% of all stories, lending credence to the notion that medicine is often described in militia-type terms, and supporting Hypothesis 12. Such stories were laced with battle metaphors, and articles used such words as arsenal, bout, fight, cavalry, combat and war. For example: Treatment includes a lot of medication and treatment-round after round, just about everything in the arsenal. (Borsuk, 1999) And: But depression, it seems is a foe that can be beaten or at least knocked into the ropes_the problem_is to recognize the opponent and take the enemy seriously. (Stevens, 2000) Note that some metaphors did overlap. In the following example, a medical disease model frame meshes with a war ("fight") and policy metaphor: Many people don't understand that mental illnesses have a biological basis and can be treated with medication, just as, say, someone with high blood pressure takes pills_advocates said they would use the report to fight for equal health insurance coverage ("Mentally ill mostly untreated," 1999). Discussion Our contribution Why study depression coverage? One reason is that people often learn about health issues from mass media, whether or not they are purposefully seeking information, and we should therefore pay attention to the construction of mental illness in media accounts. You don't need to be depressed to know Prozac has been heralded as a groundbreaking drug. Similarly, you don't need to be depressed to believe there is a sense of shame and deviance associated with depression, as there certainly is with any mental illness. While some scholars might blanch at the analogy, we argue that if mental illness were a crime, then depression would be the misdemeanor to schizophrenia's felony[18]. And when depression is framed as a type of mental illness, the stigma is carried along with the definition. Our analysis of news coverage reveals that references to depression are not overtly biased-individuals are saved from labels like lunatic or wacko in print. But mentions of violence-harm to self and harm to others-did occur in tandem with stories about depression. Such stories may reinforce mental health advocates' long-standing criticism wrought by entertainment media and tabloid newspaper portrayals that equate mental illness with violence. The link we found with violence may be a function of the newsworthiness of the story-such as the woman who drove her car into a lake, killing her daughter. The story is not so much about depression as it is coverage of infanticide. What makes stories newsworthy-like crime and violence-may be the very driving force that results in newspapers' delegitimization of depression. We also argue that the frames used in coverage of depression may enable stigma and stereotypes to occur. For example, personalization of news, by its very nature, places the individual at the heart of the story, directing the reader's attention to the person, rather than the community. And while the prevailing assumption is that the personalization technique-putting a human face on a story-makes the story more compelling to readers, some studies reveal the opposite--that personalization reduces, rather than increases sympathy for the mentally ill (Valkenburg, Semetko & De Vreese, 1999). Moreover, by focusing on individuals rather than social systems, both cause and cure take on individual features. And we argue that there is indeed a cultural context of depression that plays an underreported role. The notion of the brain as the site of depression implies, by default, that the cure is a pill that restores chemical balance. Moreover, readers and viewers are reminded of the cure wit h every antidepressant advertisement. Yet there are myriad critics who argue that depression's causes occupy a continuum, and that only some depression is the result of brain chemistry, and therefore antidepressant drugs are not always the most prudent therapy choice. The construction of depression in news stories includes solution frames, another common journalistic routine. The solution frame suggests that depression is best solved by taking medication. Yet organizations like the APA, NAMI and NIMH agree that drug therapy should be combined with counseling therapy, and point to clinical trials that demonstrate the efficacy of the combined therapies. One explanation-which has yet to be demonstrated unequivocally-is that the attention to drug therapy in popular press is the result of aggressive marketing on the part of pharmaceutical companies, which place news stories in popular media that extol the virtues of SSRI drugs. Moreover, the solution frame, when combined with the medical disease model, suggests that drugs are the preferred method for treating the disease of depression. We also found that war frames permeate coverage of depression, playing out a narrative where the individual becomes engaged in battle with herself or himself. The war takes on an additional metaphor of the struggle over light and dark: individuals with depression sink and fall into metaphorical abysses, holes and pits. Depression is heavy, dark and burdensome. And our culture equates the dark with evil. It's a short distance, then, from being dark to being evil. And the natural conclusion is that depression is evil, offering another opportunity for delegitimation. But perhaps the greatest disservice is placement of the burden on the individual with relatively rare acknowledgement of the role of the community and culture in mental illness. Moreover, the structure and routine of newsgathering and coverage parallel and reflect the western values of individualism and bootstrap mentality of solving one's own problems. In addition, the characterization of mental illness as a function of brain chemistry within a disease framework places the solution squarely on medication. But medication alone rarely solves the core forces that brought on depression in the first place. Alteration in brain chemistry often results from factors that engendered depression: stress from trauma or profound life changes, or low self-worth created by perceptions based on societal expectations. The act of taking medication does not change the trauma or perceived expectations. But in the journalist's quest to bring about solutions in news stories, the larger social issues te nd to be ignored. On a more positive note, we found that news coverage spotlights some issues that warrant additional attention-the effect of depression on youth, the elderly and on men in general. Many stories concerned youth and the elderly, who are at greater than average risk for both depression and suicide. And a number of stories specifically focused on men, whose depression, critics say, has been underreported and under-discussed. Limitations We examined a rather brief period of news coverage, and a lengthier time frame would undoubtedly expand our understanding of how depression is constructed. Moreover, it would be useful to compare coverage before and after introduction of the SSRI antidepressants. Another limitation is the sheer volume-or lack of volume-of news stories. Our sample may not be representative of the total population of depression news stories. Although we use techniques to ensure reliability and validity, the sample of news stories may not behave like a sample of say, voters. The assumption of probability is that, like randomly chosen voters, each news story represents ten others just like it. Such an assumption may prove problematic in content analysis studies like ours. We argue that substantive-rather than statistical-significance is more appropriate in this case, because data to not behave in the manner assumed for multivariate and linear statistical models. It may be unwise, therefore, to place great confidence when generalizing these stories to the entire population of depression stories. Readers should be mindful that such stories might not necessarily reflect public opinion, and should refrain from making assumptions about audience response to such stories based only on our examination of coverage. We do assert, however, that news construction reflects values and interests driven by a number of factors, from routines embedded in the manufacture of news, to social constructions articulated and managed by special interest groups ranging from professional organizations to pharmaceutical companies. Reference List Bennett, L. (1988). News: The Politics of Illusion. New York: Longman. Borsuk, A.J. (1999, November 21). Darkness lifting: Teen finds inner strength during long struggle with depression. Milwaukee Journal Sentinel, Lifestyle, p. 1. Brown, C. (2000, January 2). The lives they lived: William Woodward III. New York Times, p. 48, section 6. Coleman, C.-L. (1994). Science, technology and risk coverage of a community conflict. Media, Culture and Society, 17, 65-70. Daschmann, G., & Brosius, H.-B. (1999). Can a single incident create an issue? Exemplars in German television magazine shows. Journalism and Mass Communication Quarterly, 76(1), 35-51. Davis, D. R., & Jansen, G. G. (1998). Making meaning of Alcoholics Anonymous for social workers. Social Work, 43(2), 169-183. Donohue, G. A., Olien, C. N., & Tichenor, P. J. (1989). Structure and constraints on community newspaper gatekeepers. Journalism Quarterly, 66(4), 807-812, 845. Ferraro, S. (2000, February 6). Mutant gene eyed in suicides: Canadian research could lead to tests. Daily News (New York), p. 32 Forster, D. (1999, August 1). Eliminating stigma key to curing depression. Wyoming Tribune-Eagle. Page A-1. Foucault, M., & Translated by Richard Howard. (1965-1973). Madness and Civilization. New York : Vintage. Frank, C. (1999). Skirmish or Siege? Psychology Today, 32(2), 35-37. Fredin, E. S., & Kosicki, G. M. (1989). Cognitions and attitudes about community: Compensating for media images. Journalism Quarterly, 66(3), 571-578. Gamson, W. A., & Modigliani, A. (1989). Media Discourse and Public Opinion on Nuclear Power: A Constructionist Approach. American Journal of Sociology, 95(1), 1-37. Gitlin, T. (1980). The Whole World is Watching: Mass Media in the Making and Unmaking of the New Left. Berkeley: University of California Press. Glasgow Media Group. (1996). Media and Mental Distress. London and New York: Longman. Goffman, E. (1963). Stigma: Notes on the Management of Spoiled Identity. Englewood Cliffs, N.J.: Prentice Hall. Goode, E. (2000, August 2). Childhood abuse might affect response to stress, depression. The Commercial Appeal (Memphis), New York Times News Service, p. A-7. Gozzi Jr., R. (1999). The power of metaphor. ETC: A Review of General Semantics, 56(4), 380-405. Granello, D. H., Pauley, P. S., & Carmichael, A. (1999). Relationship of the media to attitudes toward people with mental illness. Journal of Humanistic Counseling, 38(2), 98-111. Groves, S. (2000, June 18). Bernard sheds light on his struggle with bipolar disorder. The Sunday Oregonian, p. T-9 Gunther, A. (1992). What we think others think. Communication Research, 18(3), 355-373. Hannigan, B. (1999). Mental health care in the community: An analysis of contemporary public attitudes toward, and public representations of, mental illness. Journal of Mental Health, 8(5), 431-440. Hartley, J. (1982). Understanding News. London: Routledge. Harvey, R. D. (2001). Individual differences in the phenomenological impact of social stigma. Journal of Social Psychology, 141(2), 174 (16 pages). Online serial, available Ebscohost Hayward, P., & Bright, J. A. (1997). Stigma and mental illness: A review and critique. Journal of Mental Health, 6(4), 345-355. Herman, E., & Chomsky, N. (1988). Manufacturing Consent: The Political Economy of the Mass Media. New York: Pantheon. Hernandez, R. (2000, June 4). "Teen-age depression called common problem." The Press-Enterprise, Riverside, California. Page B-2. Iyengar, S. (1990). Framing responsibility for political issues: The case of poverty. Political Behavior, 12(1), 19-40. Jobes, D.A., & Berman, A.L. (1996). The Kurt Cobain suicide crisis. Suicide & Life-Threatening Behavior, 26(3) 260-265. Johnson, P. J. (October 10, 1999). "Mental Health Board Head Delighted, Petrified by Role," Los Angeles Times, page B-1. Kinder, D. R., & Sanders, L. M. (1990). Mimicking Political Debate with Survey Questions: the Case of White Opinion on Affirmative Action for Blacks. Social Cognition, 8(1), 73-103. Lacher, K. T., & Rotfeld, H. J. (1994). Newspaper policies on the potential merging of advertising and news content. Journal of Public Policy & Marketing, 13(2), 281+. Lakoff, G., & Johnson, M. (1980). Metaphors we live by. Chicago: University of Chicago Press. Levins, I., & Rodgers, C. (1998). Getting good ink: Measuring PR effectiveness. Pharmaceutical Executive, 18(12), 12-15. Luke, T. W. (1987). Chernobyl: The packaging of transnational ecological disaster. Critical Studies in Mass Communication, 4, 351-375. Martin, J. K., Pescosolido, B. A., & Tuch, S. A. (2000). Of fear and loathing: The role of 'disturbing behavior,' labels, and causal attributions in shaping public attitudes toward people with mental illness. Journal of Health and Social Behavior, 41(2), 208-223. Meisol, P. (1999, October 21). "Men and depression: A disease usually associated with women is epidemic among men, and they struggle-literally to hide it." Forth Worth Star-Telegram, Life & Arts, p. 3. "Mentally ill mostly untreated: Report aims to combat stigma, make affordable care available," (1999, December 14). Asbury Park Press (Neptune, N.J.), p. A-1. Meryhew, R. (1999, October 27). Long Prairie mom charged in girl's death: the woman, said to have emotional problems, told police that she drove her car into a lake "to drown," Star Tribune (Minneapolis), p. A-1. Nelkin, D. (1987). Selling Science. New York: W. H. Freeman and Company. Neuhaus Jr., C. (1993). The disease controversy revisited: An ontologic perspective. Journal of Drug Issues, 23(3), 463-477. Newnham, B. (April 18, 2000). "Sonics Have 87 Million Reasons to Help Baker." Seattle Times. Page D-1. "Obese women more likely to be depressed, study says." (2000, February 60. Arizona Republic, p. A-29. Oppermann, J. (2000, May 22). Surviving severe depression: Support groups, poetry helped him deal with the loss of his wife. Chicago Daily Herald, Health & Fitness, p. 4 Pacenti, J. (2000, March 3). Arrested teen's diary details hate, depression. Palm Beach Post, p. A-1. Phelan, J. C., Link, B. G., Steve, A., & Pescosolido, B. A. (2000). Public conceptions of mental illness in 1950 and 1996: What is mental illness and is it to be feared? Journal of Health and Social Behavior, 41(2), 188-207. Pitchford, S. R. (2001). Image-making movements: Welsh nationalism and stereotype transformation. Sociological Perspectives (44) 1, p. 45 (21 pages). Online serial available on Ebscohost Price, V., & Czilli, E. J. (1996). Modeling patterns of news recognition and recall. Journal of Communication, 46(2), 55+. Read, J., & Law, A. (1999). The relationship of causal beliefs and contact with users of mental health services to attitudes to the 'mentally ill'. International Journal of Social Psychiatry, 45(3), 216-229. Reese, S. D. (1990). The News Paradigm and the Ideology of Objectivity: A Socialist at the Wall Street Journal. Critical Studies in Mass Communication, 7, 390-409. Ryan, C. (1991). Prime Time Activism. Boston: South End Press. Scheff, T. J. (1966-1984). Being Mentally Ill: A Sociological Theory. New York: Aldine . Schwartz, M. (August 3, 1999). "The Up Side (To Being Down): Experience of local therapists and patients substantiates long-term research that individual or group exercise benefits people with mild to moderate depression," The Press-Enterprise (Riverside, Calif.), D-1. Shelton, E., & Curtis, J. (1998 April). Advertisers tighten their grip. Marketing, pp. 28-29. Soley, L. C., & Craig, R. L. (1992). Advertising pressures on newspapers: A survey. Journal of Advertising, 21(4), 1-10. Sontag, S. (1978). Illness as Metaphor. New York: Farrar, Straus and Giroux. Stevens, R. (2000, June 28). "There is hope, help for Tapia to ride out his depression." Albuquerque Tribune. Sports, page B 1. Szasz, T. (2000). Mental disorders are not diseases. USA Today Magazine, 128(2656), 30-33. Szasz, T. (1994). Mental illness is still a myth. Society, 31(4), 34-40. Taylor, S.J. (2001). You're not a retard, you're just wise. Journal of Contemporary Ethnography (29)1, p. 58 (35 pages). Online serial, available on Ebscohost Tichenor, P., Donohue, G., & Olien, C. (1980). Communication and Community Conflict. 11-25. same?, Community Conflict and the Press. Beverly Hill: Sage. Ungar, S. (1998). Hot crisis and media reassurance: A comparisons of emerging diseases and Ebola Zaire. British Journal of Sociology, 49(1), 36-57. Valkenburg, P. M., Semetko, H. A., & De Vreese, C. H. (1999). The effects of news frames on readers' thoughts and recall. Communication Research, 26(5), 550-569. Wahl, O. F. (2000). Brief report: Obsessive-compulsive disorder in popular magazines. Community Mental Health Journal, 36(3), 307-312. Wahl, O. F. (1995). Media Madness. New Jersey: Rutgers University Press. Walsh, B. (August 13, 2000). "Surviving the Darkness of Divorce: Teenage sisters endure depression, eating disorders and 'crappy times' in the long process of remaking family." Portland Press Herald (Portland, Maine), page A-9. Uhlenhutch, K. (2000, October 25). Postpartum blues. Kansas City Star, p. E-1. Weldon, R. A. (2001). The rhetorical construction of the predatorial virus: A Burkian analysis of nonfiction accounts of the Ebola virus. Qualitative Health Research, 11(1), 5-26. Whitely, J. (2000, January 4). Beyond the blues. Las Vegas Review-Journal, p. 1-E. [1] http://www.nimh.nih.gov/publicat/depeffect.cfm [2] In our study, APA refers to the American Psychiatric Association [3] http://www.psych.org/public_info/overview.cfm "Mental Illness: An Overview [4] http://www.psyweb.com/Mdisord/DSM_UV/dsm_iv.html [5] http://www.nami.org/illness/whatis.html "What is Mental Illness?" [6] http://www.nimh.nih.gov/publicat/invisible.cfm [7] Significance refers to substantive significance unless indicated otherwise [8] A test of five articles culled randomly with the key word "depression" resulted in only two relevant stories. A separate test was run with the words "depression" and "depress" and all five articles were found to be relevant. [9] Each of the 640 story headlines was printed and cut into strips that were tossed into a straw hat. A blindfolded volunteer selected one piece of paper at a time and the hat was shaken after each selection. A total of 110 headlines slips was pulled from the hat. [10] Unless noted, all categories were independent [11] Categories were independent, and included: no blame, self, family, someone else, disease or disability, trauma, hormones or menopause, puberty, aging, pregnancy or post-partum, work stress, chemical imbalance, society, genetics, religion, and other. [12] Totals exceed 100% because a story may contain more than one mention of terms such as brain, chemistry, violence, stress, etc., as related-not necessarily the cause-of depression. [13] Items were independent [14] Sources, even if quoted repeatedly, were counted in stories once only [15] This is an independent category, meaning; it was not double-coded with the other therapy variables. [16] Top 200 prescription drugs by retail sales in 2000, Scott-Levin's Source Prescription Audit, in http://dt.pdr.net/dt/public.htm?path=content/journals/d/data/2001 /0319/dsrtop20003b.html [17] Lithium, the chemical name for Eskalith, is ranked 115 in the list of generic drugs. Eskalith is not ranked in the top 200 brand name list [18] David Sillars made this analogy Acknowledgements The authors would like to express our heartfelt gratitude to the following students at Georgia State University who assisted in framing this study and providing helpful insights: Cordell Clealand, Mollie Evans, Robin Gilliam Sloan, Vince Stuntebeck, Angela Turk and Nalanie Tyrrell. Acknowledgements The authors would like to express our heartfelt gratitude to the following students at Georgia State University who assisted in framing this study and providing helpful insights: Cordell Clealand, Mollie Evans, Robin Gilliam Sloan, Vince Stuntebeck, Angela Turk and Nalanie Tyrrell.