@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix skos: . vivo:departmentOrSchool "Arts, Faculty of"@en, "English, Department of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Keay, Aimee Lee"@en ; dcterms:issued "2013-06-04T09:12:18Z"@en, "2013"@en ; vivo:relatedDegree "Master of Arts - MA"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description """Healthcare information differs among the United States, Canada and England. Through a critical discourse analysis of fifteen websites on depression from these countries, this thesis reveals how linguistic differences and differences in the websites’ use of visual features persuade readers of the merits of different treatments. The analysis reveals that the trends are complex. However, England’s websites lean more towards talking therapy, the United States’ websites emphasize pharmaceutical intervention, and Canadian websites endorse the use of antidepressants where they mention treatment. These findings are illustrated through a comparison of vocabulary, grammatical, visual and ordering features (building from Fairclough’s 1989 framework). This thesis also reveals that patients from the United States, Canada and England are portrayed as in possession of differing levels of importance in the treatment-decision-making process with physicians. Of particular significance is the Canadian websites’ portrayal of patient deferral to expert physicians. By frequently referencing “your doctor,” using marked grammar for healthcare experts, and providing a reader with limited information on treatment options, Canada’s websites assert a traditional biomedical model of power relations. The patient is secondary to the physician. The United States’ privatized healthcare system indicates that websites from the United States would portray patient input as especially significant. However, it is England’s websites that suggest a high level of patient influence. Stressing patient involvement on England’s websites is perhaps indicative of the NHS’s 2012 constitution, which emphasizes that decision-making should be based on a model of concordance not compliance (Segal 2007). These findings highlight how the different healthcare models of each country might affect the information provided to patients. Above all, this research raises questions about the role of patient information websites, and about the different discursive strategies that subtly persuade a reader to view depression, treatment and their input in decision-making differently. England, United States and Canada all make use of the DSM IV-TR diagnostic criteria and operate under a biomedical model of medicine, but these websites suggest that potentially depressed patients are approached differently in each of these countries."""@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/44540?expand=metadata"@en ; skos:note "“WESTERN(WORLDS(APART?”(A(COMPARISON(OF(PATIENT(INFORMATION(WEBSITES(ON(DEPRESSION(IN(CANADA,(THE(UNITED(STATES(AND(ENGLAND.(((by(((AIMEE(LEE(KEAY(((A(THESIS(SUBMITTED(IN(PARTIAL(FULFILLMENT(OF(THE(REQUIREMENTS(FOR(THE(DEGREE(OF(((MASTER(OF(ARTS((in(((THE(FACULTY(OF(GRADUATE(STUDIES((((English)((((THE(UNIVERSITY(OF(BRITISH(COLUMBIA(((Vancouver)(((((( May(2013((((((((((( © Aimee Lee Keay, 2013 (( ii( Abstract(( Healthcare information differs among the United States, Canada and England. Through a critical discourse analysis of fifteen websites on depression from these countries, this thesis reveals how linguistic differences and differences in the websites’ use of visual features persuade readers of the merits of different treatments. The analysis reveals that the trends are complex. However, England’s websites lean more towards talking therapy, the United States’ websites emphasize pharmaceutical intervention, and Canadian websites endorse the use of antidepressants where they mention treatment. These findings are illustrated through a comparison of vocabulary, grammatical, visual and ordering features (building from Fairclough’s 1989 framework). This thesis also reveals that patients from the United States, Canada and England are portrayed as in possession of differing levels of importance in the treatment-decision- making process with physicians. Of particular significance is the Canadian websites’ portrayal of patient deferral to expert physicians. By frequently referencing “your doctor,” using marked grammar for healthcare experts, and providing a reader with limited information on treatment options, Canada’s websites assert a traditional biomedical model of power relations. The patient is secondary to the physician. The United States’ privatized healthcare system indicates that websites from the United States would portray patient input as especially significant. However, it is England’s websites that suggest a high level of patient influence. Stressing patient involvement on England’s websites is perhaps indicative of the NHS’s 2012 constitution, which emphasizes that decision-making should be based on a model of concordance not compliance (Segal 2007). These findings highlight how the different healthcare models of each country might affect the information provided to patients. Above all, this research raises questions about the role of patient information websites, and about the different discursive strategies that subtly persuade a reader to view depression, treatment and their input in decision-making differently. England, United States and Canada all make use of the DSM IV-TR diagnostic criteria and operate under a biomedical model of medicine, but these websites suggest that potentially depressed patients are approached differently in each of these countries. (( iii( Preface This dissertation is original, unpublished, independent work by the author, A. Keay. (( iv( Table of Contents Abstract ......................................................................................................................... ii Preface .......................................................................................................................... iii Table of Contents .......................................................................................................... iv List of Tables ................................................................................................................ vi List of Images .............................................................................................................. vii Acknowledgements .................................................................................................... viii Dedication ..................................................................................................................... ix 1. Introduction .............................................................................................................. 1 1.1. Aims and Research Questions ............................................................................. 1 1.2. Healthcare Differences ........................................................................................ 4 1.3. Patient Information Websites .............................................................................. 9 1.4. Project Outline .................................................................................................. 11 2. Methodology ........................................................................................................... 12 2.1. The Websites on Depression ............................................................................. 12 2.2. Critical Discourse Analysis Approach .............................................................. 16 3. Vocabulary .............................................................................................................. 20 3.1. Major Depression? Depressive Illness? Depressive Episodes? What’s in a Name? .................................................................................................................................. 21 3.2. Positive and Negative Descriptions of Treatment ............................................. 25 3.2.1. Pharmaceutically Sponsored Websites ...................................................... 25 3.2.2. Descriptions of Talking Therapy and Antidepressants on Non- pharmaceutically sponsored Websites ................................................................ 29 3.3. Collocations with \"Common” ........................................................................... 31 3.4. Section Summary .............................................................................................. 33 4. Grammar ................................................................................................................ 33 4.1. Pronouns ............................................................................................................ 34 4.2. Focus and Prominence: Cleft Sentences ........................................................... 39 4.3. Modal Verbs ...................................................................................................... 41 4.4. Section Summary .............................................................................................. 45 5. Visual and Ordering Features .............................................................................. 45 5.1. Amount of Information ..................................................................................... 46 5.2. Order of Information ......................................................................................... 50 5.3. Layout: Text Callouts and Testimonials ........................................................... 52 5.4. Visual Images .................................................................................................... 59 5.5. Section Summary .............................................................................................. 64 6. Conclusion ............................................................................................................... 65 (( v( 2.1. Overview and Significant Trends ..................................................................... 68 2.2. Why we should be Especially Mindful of Information on Depression ............. 68 Works Cited ............................................................................................................... 67 (( vi( List of Tables ( Table(A:(Background(on(Websites(.......................................................................................................(15(Table(B:(Amount(of(Words(in(Corpora((.............................................................................................(32(Table(C:(Counts(for(Use(of(Second(Person(.......................................................................................(35(Table(D:(Percentage(for(Use(of(Second(Person((.............................................................................(36(Table(E:(Use(of(First(Person(Plural(......................................................................................................(38(Table(F:(Proportion(of(Cleft(Sentences(..............................................................................................(41!Table(G:(Modal(Verb(Counts(...................................................................................................................(43(Table(H:(Word(Count(Information(for(Antidepressants(and(Therapy(in(the(United(States(............................................................................................................................................................................(47(Table(I:(Word(Count(Information(for(Antidepressants(and(Therapy(in(England(...........(47(Table(J:(Word(Count(Information(for(Antidepressants(and(Therapy(in(Canada(.............(47((Table(K:(Order(of(Information(Subsections(.....................................................................................(53!Table(L:(The(Order(of(Information(in(Sentences(...........................................................................(53(Table(M:(Transactional(and(Non`Transactional(Images((...........................................................(60(Table(N:(Percentage(of(Frame(Styles(..................................................................................................(64(( (( vii( List of Images Image(A:(Layout(for(HelpGuide((.............................................................................................................(54(Image(B:(Layout(for(NetDoctor/(............................................................................................................(58(Image(C:(Female(Direct(Gaze((................................................................................................................(61(Image(D:(Male(Direct(Gaze(......................................................................................................................(61(Image(E:(Transactional(Partner(Image(..............................................................................................(62!Image(F:(Transactional(Pet(Image(.......................................................................................................(62(Image(G:(Head(in(Hands((Transactional(Image((.............................................................................(63( ( (( viii( Acknowledgements There are so many people I want to thank for the support, encouragement and help I received on this thesis. Firstly, thank you to my wonderful supervisor, Judy Segal, whose amazing wealth of knowledge, encouragement and speedy feedback made this thesis possible. Thank you to Jessica de Villiers, whose input has also been invaluable – and for suggesting Kundalini Yoga to keep me sane (I’m not sure about the sanity part, but I enjoy the yoga!). Both Judy and Jessica have been outstanding as members of my committee and I am so grateful for their support and expertise. I would also like to thank Louise Soga for the amazing support she has given me throughout this degree. I will be eternally grateful to her for her faith in me and for her encouragement. She is the reason I am still here to write this acknowledgment! Sandra Tomc has also been fantastic as MA advisor. I am very thankful for the advice she has given me throughout this programme, particularly during my first year. She is simply awesome. I think my friends and family also deserve thanks (and endless gratitude!) for listening to me rant about this project, helping me to tease out ideas and for making sure I kept going. It has been a long road! Special thanks to the wonderful Megan Callahan for moral support during our many thesis writing Starbucks visits! Finally, I thank the creators, owners and contributors to the websites I studied on depression. (( ix( Dedication For my Mum and Dad for standing by me when I needed to use these websites. Thank you. And for my beautiful Godson, Bryn Davies-Baker. May you never have to use this information. I love you. ! ! ! 1! 1.Introduction 1.1. Aims and Research Questions How do the websites accessed by people seeking information on health websites affect the way that patients act in the face of illness? Are we persuaded to seek different treatment because of the different ways depression and its treatments are represented to us in the places in which we live? These are the larger questions I raise in this thesis as I compare patient information on depression on websites emerging from the United States, Canada and England. One might expect patient information to differ among the United States, Canada and England. With three different healthcare systems operating with varying degrees of public and private involvement, certain differences are necessary to enable patients to best navigate their healthcare. However, this comparison of fifteen patient information websites (five from each country) targeting individuals investigating depression reveals that differences in the available information from each country are more significant than they may, at first, appear to be. My thesis uses discourse analysis to reveal how users of websites from different countries are persuaded of the merits of different treatment options and to probe who is portrayed as responsible for making these treatment decisions. Different levels of authority, expertise and control emerge from the websites accessed for this research. Overall, this analysis reveals that the websites targeting users from England portray patients as having more choices available to them, whereas websites from Canada suggest that patients should defer to the expert physician. Indeed, two of the five Canadian websites feature a distinct lack of treatment information, ! ! ! 2! reinforcing how information on the Internet cannot replace consultation with a medical expert. Websites from the United States, by contrast, reveal varying levels of expertise and control by physicians and patients. The United States is a nation whose healthcare system is based on the notion of consumer choice. Yet the influence of pharmaceutical companies on this healthcare system appears to persuade patients to choose medication treatment over talking therapy. Patients from Canada are also persuaded towards a pharmaceutical approach, where Canadian websites reference treatment. Of all the websites, it is those emerging from England that are most in favour of treating depression by talking therapy alone. With patient information in the digital age now considerably influencing the choices that patients make about their healthcare, these differences are highly significant. Bioethicist Rebecca Kulka writes that “[t]he increasingly active role that lay people are taking – and indeed, are asked by our culture to take – as medical inquirers and as participants in collaborative knowledge-building is riddled with both empowering possibilities and the danger for distortion and peer pressure” (34). The dissimilarity within websites’ presentation of treatments for depression, then, raises a pertinent question for the field of medical humanities. If patient information reflects the biases of particular healthcare systems, one must ask how far could this information alter the choices patients make themselves? By analyzing the persuasive language on websites for depressed patients, I discuss the ways in which information from different countries could affect patients’ experiences, decisions and judgment about their condition. Annemarie Mol argues that, rather than foster a “logic of choice,” where patients are able to choose treatments based on their personal assessment of their needs, ! ! ! 3! healthcare should be based around “a logic of care,” a system in which expert physicians make the treatment decisions in collaboration with the patient, based on the physician’s judgment about what constitutes the best care (1). This “logic of care” appears an ideal that healthcare systems do not always attain. Websites providing information illustrate the problems within a logic of choice. These problems are especially apparent for an illness like depression, where choices are available to patients who are not always equipped to make good decisions, partly because of the nature of the illness. Furthermore, Mol states that in this logic of choice, “the moment a choice is being made, is embedded in a sequence: (neutral) facts > (value laden) choice > technical action” (54). Yet this thesis suggests that facts are not neutral. These websites exemplify how describing certain treatments negatively, or emphasizing the benefits of some over others, actually reduces the choices available. I argue that, although the language on these websites appears to operate within the realm of a logic of choice, it may, in fact, be a way to reduce the choices patients make. For depression, a general practitioner is usually patients’ first port of call. Dohrenwent, speaking to physicians, stresses that “[p]atients [suffering from depression] are free to decide which of the treatment options offered to them best fit their preferences, beliefs and values” (15). The way in which information on these websites is presented to web-users may provide patients with a clearer idea of which treatment options they would support and what degree of they will have when deciding on treatment with their GP. Furthermore, Dwight-Johnson et al’s patient survey from the United States reveals that patients who are offered a treatment they prefer are more likely to enter care. The sources, taken together raise the question of whether information, which encouraging ! ! ! 4! patients to support particular treatments, differs to reflect the availability and preferences of these three healthcare systems. 1.2. Healthcare Differences Differences among these countries’ healthcare systems suggest that the language on patient information websites will necessarily differ. Website information is not designed to be neutral, but instead it is designed to be specifically applicable to web-users from particular healthcare systems. Thus, despite the biomedical model of medicine dominating the US, English and Canadian healthcare systems, there are substantial managerial differences, which affect the choices a patient has access to (Downie and Randall; Moody; Wolpert). Website information reflects this. These healthcare systems require some explanation. The most significant differences among healthcare systems in the United States, Canada and England lie in the division between socialized and privatized medical care. The three countries analyzed can be classified on a continuum: the United States’ emphasis on managing one’s own care (with limited federal Medicaid) places this country firmly towards the privatized medicine side of the continuum; Canada’s largely socialized healthcare system with limited access to “paramedical services” in the absence of an extended insurance plan places this “public-private hybrid” system (Lewis et al 927) in the middle but more towards socialized medicine; and England’s National Health Service, which provides a wider range of coverage, can be situated even closer to the socialized medicine end of the spectrum. Healthcare expenditure statistics reflect this difference: 67.8% and 43.3% of the expenditure on health in the US and Canada respectively is on private healthcare ! ! ! 5! coverage1 versus only 6.5% in the UK (World Health Organization 2012). However, all of these systems appear to be in a state of flux with regard to the continuum of public- private care. As Lewis et al point out for Canada, “increasing privatization has crept into the system,” and people with money can by-pass queues using loopholes in Canada’s law banning private access to treatments that are publically funded. England’s healthcare system (discussed below by Moody) also contains increased privatization. In contrast, the United States’ the Affordable Care Act (2010) reveals how the United States’ healthcare is shifting slightly towards more socialized care. For purposes of this thesis, current conditions are the basis for research. Peter Conrad’s The Medicalization of Society emphasizes the effects of a private insurance-based healthcare system. Conrad critiques the United States’ approach to healthcare, highlighting how pressures from private industry affect the care a patient receives. Conrad writes “[m]anaged care requires preapprovals for medical treatment and sets limits on some type of care. This has given third-party players more leverage and has often constrained both the care given by doctors and the care received by the patient” (140). Conrad’s argument references the “peer pressure” that Kukla fears within patient information, referencing how pharmaceutical companies and insurance companies have a direct impact on the options a patient selects from. Patients from the United States without coverage have very different options from patients with comprehensive insurance plans. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 1 Private healthcare coverage refers to medical insurance taken out by an individual to cover care in addition to the care provided by the government. This can include insurance to access particular types of care, not covered by federal funding (as in Canada) or insurance that provides access to private clinics, allowing users to by-pass long waiting lists for government-funded healthcare (as in England). 2 For further information on the transformation of the NHS, see Moody’s full article. Moody contrasts UK ! ! ! 6! Indeed, Conrad states of America that “[t]he shift [to pharmacology] is reinforced when third party players will pay for drug treatments but severely limit individual and group therapies. The choice available to many doctors and patient consumers is not whether to have talking or pharmaceutical theory, but rather which brand of drug should be prescribed” (15). Conrad’s argument reveals how different priorities affect the information, and subsequent care, that a patient receives. In Canada, healthcare coverage is federally mandated but the budgets for care are managed provincially. While Canada’s provinces all provide a degree of socialized healthcare, additional insurance can still dictate certain types of care, and many individuals supplement their provincial coverage with additional insurance. The Canadian 1968 Medical Care Act (implemented in 1972) featured four main objectives: universality, portability, administration (non-profit) and accessibility (Clarke 262). Yet, as in the United States, one’s supplementary coverage can dictate treatment options. While, as Joanne Clarke points out, “[o]verall, Canadians of all classes are more likely to visit a physician than Americans are” (267), Canadians still have limitations in what they can access. This is particularly relevant for depression as all Canadian provincial coverage includes psychiatry, not psychology (correct as of 2013). This accounts for why this thesis finds that users from Canada are provided with patient information that emphasizes a psychiatric, not psychological, approach. These coverage limitations are significant for treatment as psychiatrists can prescribe medication, unlike psychologists. However, Blendon et al point out that “Canada has a universal public insurance plan, which prohibits the use of private insurance to pay for services covered by the public plan” (182). Therefore, given that the waiting list for physiatrists is long (in part because ! ! ! 7! physiatrists are covered by government insurance), Canadians with an extended plan may choose to see psychologists. Great Britain’s National Health Service (NHS), established in 1948, is a government run organization. Unlike individuals using Canadian Provincial coverage, individuals from England do not have to supplement their care to receive access to “paramedicals,” such as psychologists. Although a number of people do use private health insurance in England to access the second tier of a “two tiered system” (where people with money can receive care outside of the NHS), this number involves less than 10% of the population (Moody 430). Thus the primary constraint on this nationalized system is its internal financial management. Nonetheless, it would be naïve to assume that the English system operates independent of private factors. Kim Moody criticizes the Coalition Government’s plans for further private involvement in the NHS, arguing that “state funding remains a barrier to complete transformation, and what is likely to emerge is a hybrid system in which competition, consolidation, and pressure on the workforce all increase” (415). In the NHS, then, restrictions on patients’ healthcare are due to the limited availability of the care that the pressurized NHS workforce can supply. This limitation includes the funding of this workforce.2 How, then, do these funding and managerial differences alter the information provided to patients from Canada, England and the United States? Despite England’s constraint on cost, this thesis reveals that information from England stresses a therapeutic approach to managing depression. This result does not align with the necessity of a publicly funded institution to minimize cost given that studies, including John !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 2 For further information on the transformation of the NHS, see Moody’s full article. Moody contrasts UK system with the system of care in the United States, questioning the 2010 Coalition Government’s “White Paper” reform proposal. ! ! ! 8! Donoghue’s overview of pharmaeconomics and Lave et al’s research, conclude that drugs are more cost effective than interpersonal therapy.3 It is possible then to speculate that the trend within England to emphasize psychological therapy may be indicative of a Not-For- Profit ideology. Indeed, the NHS’s most recent constitution (2012) promotes a partnership between the patient and their healthcare providers, using the tag line “the NHS belongs to us all,” and the fourth listed point reads: “NHS services must reflect the needs and preferences of patients, their families and their carers” (3). It is the sixth point that emphasizes “value for money” (4), which perhaps points to an ideal order of priority, with patient choice preceding financial concerns. However, given that the NHS is a centralized healthcare system, the National Health Service employs psychologists, and therefore one would expect psychologists’ promotion within England to exceed their promotion within United States or Canada.4 Therefore more patient choice, on a national or federal funding level at least, does exist within the England than in Canada or the United States. It is how this is portrayed, and what choice is prioritized that I am interested in. These discrepancies demonstrate that, while the Diagnostic and Statistic Manual IV is the worldwide source for diagnosis (Watters 2010: 40), there are still significant variables that affect the patient, and physician’s, choice of treatment. Thus, while, as Berkenkotter and McCarthy and Gerring highlight, the DSM IV affirmed the victory of a !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 3 Donoghue’s summary of pharmaceonomics cites numerous research, including Lave et al, to conclude that medication appears more cost effective than interpersonal therapy (51). Lave et al use Schulberg et al’s medical trial data to measure the effectiveness of treatment by the number of depression free days reported by patients in a year. They conclude that patients on Nortriptyline hydrochloride had significantly more of these than patients receiving interpersonal therapy. 4 There are also numerous private psychologists working in England. The waiting list for Psychological services provided by the NHS is long, and therefore patients may also need to access this “two tiered” system. ! ! ! 9! biomedical perspective on mental health with the dominance of psychiatry over psychology, this psychiatry-psychology hierarchy may not be as ubiquitous as they assume. 5 As outlined, numerous factors interplay to affect the decisions about patients’ treatment, among them the country in which the patient seeks care.6 Ethan Watters, in his overview of worldwide discrepancies in mental illnesses, reveals that “Cross-cultural psychiatrists have pointed out that the mental-health ideas we export to the world are rarely unadulterated scientific facts and never culturally neutral” (40). Depressed patients in these three western countries, then, have the potential to be addressed very differently, and to receive dissimilar information reflective of differing opinions about what is “best” for treating depression. 1.3. Patient Information Websites The role of patient information in web-users’ inquiry about the “best” treatment option has been much discussed in medical humanities literature recently. Indeed, the influence of the web-informed patient on medical practice is subject to much criticism. As Kulka argues, “Autonomous inquiry depends upon the capacity to critically interrogate the legitimacy of the information and advice we are given by others, including by experts.” While she points out that “studies show that Internet users exhibit well-developed and complex strategies for sorting through health information and judging its legitimacy” (31), Kulka also raises questions about how the information is presented. Although the internet-user is often capable of deducing the reliability of a source, there are dangers in representing a medical intervention as “routine.” If medical treatment is presented in this !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 5 Furthermore, the response to the recent release of the DSM5 reflects that numerous concerns exist both inside and outside of the medical world about this biological approach to mental illness. 6 This paper does not claim to account for all these factors. Indeed, gender, social class, education, ethnicity and other factors all play a role on healthcare. Even within nationalized and provincial healthcare systems, access to care is unequal (see Clarke: 262). ! ! ! 10! way, Kulka argues, the patient will rarely question the reasons behind the treatment. Therefore patient information websites presenting medication, or talking therapy, as routine, or “common” are significantly more persuasive on the positive effects of this form of treatment. Angela Coulter, however, is less optimistic than Kukla about the average Internet user’s ability to deduce accurate information, stating of the Internet’s “inaccurate or misleading” information that “it is difficult for non-specialists to sort out the wheat from the chaff” (225). When writing on England’s healthcare information, Coulter argues that a national strategy to ensure the accuracy of information is required, and that “[i]f patients are to be active participants in decisions about their care the information they are given must accord with available evidence and be presented in a form that is acceptable and useful” (225). Many researchers agree with this sentiment; for example, Giménez- Pérez et al’s examination of websites for Type 2 diabetes concludes that “[i]f patients are to be involved in decision-making processes, efforts should be made to update the Internet contents to meet this challenge” (688). Silence et al also argue that “few sites provide sufficient information to support patient decision-making” (663). Furthermore, although the websites examined by this research present themselves as informational, they are, at the same time, persuasive.7 As these websites are not marked as persuasive texts, readers may not be aware of the potential for bias. Therefore web-users may not be as critical as they would be if they were viewing these websites through the frame of persuasion. As a result, this information may particularly influence patients – as persuasion disguised as information is the most persuasive of all. Joseph !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 7 Even websites operated by pharmaceutical companies present themselves as informational, and it is not immediately obvious that they are sponsored websites. DepressionHurts, for example, does not even mention that the website is run by Eli Lilly. ! ! ! 11! Gusfield, in his research on scientific writing, claims that the setting in which a paper occurs “establishes a claim for the paper for be taken as authoritative fact and not as faction or imaginative writing” (18-19). As some of websites are created by sources – such as the National Health Service – which are associated with factual information, these websites on depression do not immediately appear to be potentially biased sources. Even though Kulka points out that Internet users are often able to sort biased and unbiased information, webpages issued by the government and not-for-profit sources do not immediately appear problematic. It is not only the potential impact of these websites on their users that I am interested in; web-users also impact the information they access. Segal (2009) emphasizes that health information is bidirectional, where the patient is informed and constructed by the Internet, and the Internet, in turn, is constructed by the patient (though top result searches, for example). Such dual-construction emphasizes that these websites are even more indicative of the nation’s healthcare preferences: the process is cyclical. Norman Fairclough’s use of the term reproduction is apt here. Fairclough uses reproduction to refer to the intertwined nature of text production. He states that “[d]iscourse, and practice in general, in this sense are both the products of structures and the producers of structures. It is this process of discourse being produced anew (re- produced) through being drawn upon that I refer to as reproduction” (39). National information also has an international audience to influence such reproduction. The Internet is, by nature, global. Conrad notes the following: Cyberspace knows no national boundaries, expediting the dissemination of medical knowledge, commercial promotion, and consumer desires. […] We have ! ! ! 12! no idea yet what the Internet’s impact is on the local and global nature of medical categories and treatments but it is a safe assumption that medicalization will increase with globalization (144). Globalization means that patients need not, and probably do not, have access to only local information. Although Google preferences, and a potential patient’s desire to receive the guidance provided by his or her national healthcare system, may well restrict what a patient views, it is unlikely that a patient will not – at some point at least – come into contact with healthcare information from another country. Therefore, the influence of national patient information websites on depression may well be mitigated by users’ access to other websites. 1.4. Project Outline To probe the issues of patient information and differences among these countries, the analysis will be structured as follows: Section two of this thesis addresses the methodology this project has taken to tackle issues of patient information in the United States, England and Canada. Section three focuses on vocabulary (following Fairlough 1989), where I identify how the different descriptions of patients, depression, and treatments for depression persuade a reader. Section four focuses on grammatical features through an analysis of the syntax, pronouns and modal verbs. In Section five, I examine visual images and ordering features, analyzing what Fairclough deems the “larger textual features” that impact the interpretation of these texts (136). Separating these features, as this project does through Fairclough’s methodology (see below), permits a discussion of how these features individually affect the interpetation, as well as how these features work in conjunction with other elements of the websites. ! ! ! 13! 2. Methodology 2.1. The Websites on Depression Before discussing my means of analysis, I will describe the websites themselves. Corpora of fifteen websites in total was created for this analysis, with five websites selected from each country. All of these websites specifically target patients with depression or symptoms of depression and provide advice on treatment. All fifteen websites were taken from the top results from each countries’ respective Google search (.com, .co.uk and .ca) for “depression” to accurately represent information a web-user is likely to access. Many of these websites also refer to others from the same country, highlighting the connected nature of this information. This emphasizes Segal’s (2009) observations on the bidirectional nature of the Internet, which is created by both readers and texts. By referring to these texts, readers and their search histories encourage other Internet users to refer to the same texts, further increasing the texts’ representativeness of that nation’s online resources on depression. For comparability, four of the five websites selected for each location are run by not-for-profit organizations, including, for example, the national government (US Public Library of Medicine; NHS Choices; Canadian Public Health Association), and one website from each country has a corporate sponsor (Ifred; Netdoctor.co.uk; DepressionHurts.ca). Table A describes the organization responsible for each website used in this analysis. ! ! ! 14! Table A reveals that volunteer corporate, governmental and pharmaceutical organizations sponsor and provide the health information I examine on depression. Questions about the credibility of volunteer websites, therefore, are raised, and such biases suggest that attention is needed on the different motivations of people providing information that is not funded by the government. For example, HelpGuide is run by two members of the public in memory of their daughter, who committed suicide when on SSRIs – a significant factor behind their emphasis on treatments other than antidepressant medications. However, while I note these biases, I am interested in the effects on the patients of these different websites, which reflect patients’ from these countries’ highest Google search results. These websites represent the information that web-users locate. The corporate websites are included in the sample to note the heavy pharmaceutical company influence on the information given to patients, particularly within the United States. Indeed, when discussing this worrying conflict of interest, the psychiatrist David Healy states “many of those who advise regulators on the entry of drugs into the market also advise pharmaceutical companies” (2004 xiv). The United States’ Public Library of Medicine, a government website, even permits advertisements for antidepressants. In his 2009 summary of the influence of pharmaceutical companies of psychiatry, Healy writes: Little Pharma made profits by making novel compounds; Big Pharma does it by marketing. Doctors say they consume (prescribe) medication according to the evidence, so marketeers design and run trials to increase a drug’s use. They select the trials, data and authors that suit, publish in quality journals, facilitate incorporation in guidelines, then exhort doctors to practise evidence- based medicine. Because ‘they’re worth it’, doctors consume branded high- ! ! ! 15! Table A: Background to Websites Websites Country Government / Voluntary/ Private Sponsors / Owners Writers Additional Information NHS Choices England Government NHS. Funded by the Department of Health. A team of journalists and editors write on the website, and are required to declare any outside interests or connections in third party healthcare companies. States \"The website helps you make choices about your health, from decisions about your lifestyle, such as smoking, drinking and exercise, to finding and using NHS services in the UK.\" Mind England Voluntary Mind is a registered UK Charity. They accept corporate sponsorship, except pharmaceutical companies. Author policy states that \"Authors who write information products for Mind are expected to have a sound knowledge (through experience or as a professional) about the topic they are covering\" Name change from 'The National Association for Mental Health (NAMH).The organization is run by a board of 16 trustee of which celebrity Stephen Fry is the president. Half of the trustees have experience in mental health. Netdoctor England Private (pharmaceutical sponsored) Pharmaceutical sponsors and, corporate sponsors fund Netdoctor. Owned by Hearst Magazines UK. The content is commissioned directly from numerous doctors, pharmacists and healthcare professionals in England. Pharmaceutical companies are said to have no influence on content. States that \"We want to break down the \"medical language barrier\" between the doctor and the patient\" and \"Ultimately, it is the patient who must make the critical health choices.\" Patient.co.uk England Voluntary Collaboration of two companies by family doctors with interest in public health information, Patient Information Systems and Egton Medical Information Symptoms. Doctors working for the NHS. Members of the editorial team must confirm that they have no outside conflicts of interest. Often printed and given to patients in doctor's surgeries. No drug company funding. Not an NHS organisation. Routinely reviewed. Depression Alliance England Voluntary Registered UK Charily. Accepts Advertisements( (unless they are not in the interests of the public) A selection of staff with experience in the mental health field, and the voluntary sector. Advertisers said to have no control over the editorial content of the site. Run by an executive committee with experience in the mental health field. US National Library of Medicine (Pub Med USA) USA Government USA National Institute of Health. Peer reviewed journal articles by physicians and other health professionals. Antidepressant advertisements are allowed on this website. Psych Central USA Voluntary Owned by Dr John Grohol. Affiliated with the Psych Central Community Connection charity. Written by physicians and other mental health professionals. Dr John Grohol and a team of editors manage the content. Largest Independent Internet Social Network for mental health concerns. Began as Dr John Grohol's webpage for information on mental health, but rapidly grew in size. National Alliance for the Mentally Ill (NAMI) USA Voluntary Registered USA Charity, relies on donations and sponsors, including corporate. The Depression information was reviewed by Ken Duckworth, M.D., April 2013. Editorial Policy not available. Many different state divisions of NAMI across the United States. The website seeks to promote research, information, advocacy awareness and support for mental health conditions. Claims not to endorse any particular treatment or service. Ifred. USA Private (pharmaceutical sponsored) Not for profit company, reliant upon sponsors, including corporate and pharmaceutical companies. The board of directors includes people who formerly worked for pharmaceutical companies, including the founder. Editorial Policy not available. Products sold via the website. Helpguide.org USA Voluntary Owned and run by Robert and Jeanne Segal in memory of their daughter, who committed suicide before the 1996 FDA regulations required the risk of suicide to be noted on SSRIs. The main writer on the editorial team is Melinda Smith, a freelance journalist with a MA in Psychology. Claims to be committed to providing unbiased and reliable information. No advertising or corporate sponsorship. Canadian Mental Health Association Canada Voluntary Reliant upon public and corporate sponsors. Partnered with other mental health groups. Bell is particularly large sponsor. Editorial policy is not available. Regulated by The National Consumer Advisory Council. Public Health Agency of Canada Canada Government Public Health Agency of Canada. Run by the National Government. List of references follows the information provided. Editorial policy not available. Advised by External Advisory Bodies (EABs) are established by the Minister of Health, Chief Public Health Officer or the Public Health Agency of Canada (PHAC). Mood Disorders Society of Canada Canada Voluntary Relies on individual donors and corporate sponsors. Launched by \"mental health consumer leaders.\" Corporate sponsors do not influence editorial content. Editorial policy is not available. Citations to numerous other sources, including other websites, such as CHMA. A key aim of this website is to provide mental health service users with a voice. Primarily a virtual organization. No corporate sponsors are permitted who have \"The potential to cause harm\" to the membership or reputation of the website. Depression Hurts Canada Private (pharmaceutical sponsored) Eli Lily Canada Inc, a pharmaceutical company. Employees of Eli Lily, who reference other sources in a list of references. Eli Lily is only referred to as \"the company\" in the website's privacy statements. Provides a list of references, including the DSM IV and various other pharmaceutical companies. Canadian Psychological Association Canada Private (psychological sponsored) Canadian Psychological Association, a member organization for Psychologists .Members run this organization. Dr Keith Dobson, from the Psychology department at the University of Calgary, wrote the information fact sheet on depression. Canada's largest association for psychology. Has a member only section. Information for this project comes from the public section of the website. ! ! ! 16! cost but less effective ‘evidence-based’ derivatives of older compounds making these drugs worth more than their weight in gold. Posted parcels meanwhile are tracked far more accurately than adverse treatment effects on patients. (85) Healy’s assertions are especially worrying for patient information. Regardless of the websites accessed, Healy’s argument about pharmaceutical companies’ influence on physicians leads to the larger question of whether patients can trust their doctors to represent their interests as patients, the healthcare system that employs them, or – to quote Conrad - these “third party players.” 2.2.Critical Discourse Analysis Approach To discuss the relationship between linguistic features and their wider implications, my thesis adapts Ruth Wodak and Norman Fairclough’s Critical Discourse Analysis framework (henceforth CDA). Combining Fairclough’s methodology with the additional analytical tools of focus and prominence (Gregory; Asp and de Villiers) and visual multimodal analysis (O’Halloran; Kress and Van Leeuwen) enables me to provide a multilevel characterization of the linguistic strategies persuading potential patients. The CDA methodology is designed to allow one to “analyse[…] social interactions in a way which focuses upon their linguistic elements, and which sets out to show up their generally hidden determinants in the system of social relationships, as well as hidden effects they may have upon that system” (Fairclough 5). This approach requires two levels of analysis: interpretation, which is the relationship between the text and the reader, and explanation, the relationship between the interpretation provided by the reader and the social context (Fairclough 26). Any “hidden effects,” are revealed through microanalysis (focusing on interpretation) and explained through social relationships. As ! ! ! 17! discussed, the social factors that this thesis considers are the healthcare system differences among the United States, England and Canada. Therefore, as well as noting any overt emphases these patient information websites place on particular treatment plans for depression, I focus on potentially “hidden” effects and how linguistic differences may affect the interpretive level. Wodak describes this approach as mediating “[t]he link between text and society, between the micro and the macro” (19). Berkenkotter, speaking on combining rhetorical and discursive analysis, which the CDA framework does, advocates for this dual approach to texts. Speaking of her rhetorical-discursive practice, she argues that “[t]his kind of text/context approach is needed, I would contend, to capture the complex interactions between socio-historical, technological, economic, and epistemological factors, especially in disciplines such as psychiatry, a profession that traverses the boundaries between the natural and human sciences” (19). When examining vocabulary, I focus on what Fairclough labels “the experimental value of words” (the importance of individual words when making value judgments), collocations and the use of euphemism (113). Fairclough argues that “[i]n some cases, what is ideologically significant about a text is its vocabulary items per se: for instance, subversive and solidarity belong respectively to ‘right’ and ‘left’ ideological frameworks, and the occurrence of either one will tend to ideologically ‘place a text’” (113). For this analysis, how the vocabulary “places” treatment options is significant, and indicative of the way a patient is persuaded to perceive them. Noting evaluative lexis, then, and the way in which a patient is encouraged to view particular treatments, reveals which treatment the website prefers. Indeed, Fairclough accentuates the significance of lexical ! ! ! 18! items, stating that “[t]he expressive value of words has always been a central concern for those interested in persuasive language. […] It is not so much the mobilization of expressive values for particular persuasive ends that is of interest here, as the fact that these expressive values can be referred to ideologically contrastive classification schemes” (119). In other words, what is significant about the language is the ideological frame set up for a reader to interpret words’ connotations. Therefore the expressive value of words is, as Fairclough argues, especially important for considering the stance a particular website may adopt on treatments for depression. When focusing on grammatical features, I compare the websites’ use of pronouns to uncover the way in which a text asks a reader to respond to the text. For example, by claiming “you should ask your GP,” a website forces a reader to interact very differently with the text than the alternative sentence “they should ask their GP.” CDA presents the language used as an option selected from a series of possible language choices, and therefore even subtle differences are significant. The grammatical analysis I undertake in this research involves a comparison of the websites’ use of modal verbs, another language choice that CDA deems especially relevant. Here, I compare epistemic and deontic modality. Epistemic modality reveals certainty and probability (e.g. “may”), which is particularly relevant for assessing how a text portrays a particular treatment’s outcome. Deontic modality reveals obligation (e.g. “should”), which is significant for examining the sense of responsibility placed on each patient. Syntactic structures also reveal emphases on the options available to a potential patient. By using focus and prominence to approach the websites’ use of cleft and ! ! ! 19! pseudo-cleft sentences, I am able to reveal the way in which certain constituents are marked, and how this affects the way the text is interpreted. For example, the sentence “it is not weak to cry or admit that you are struggling. (Patient.co.uk, n.p.) uses a cleft construction with a dummy subject (‘it’) that makes prominent an evaluation (not weak) and an admission of struggling. The two are therefore more strongly bonded and emphasized. Using a cleft construction is a choice made in place of sentences like “To cry or admit that you are struggling is not weak.” Using clefts and pseudo clefts in this way, as I will discuss later, is a tactic adopted by websites from England to remove personal responsibility from the patient. Instead of labeling the final section of this methodology “Larger Textual Features,” as Fairclough’s methodological framework does, my heading of “Visual and Ordering Features” more aptly summarizes the approach I have taken to the texts. To account for the way in which the texts’ structure affects their readers, this research analyzes of the order of the information provided. I argue that the treatment that a website prefers is suggested when the website lists this first. This is shown in statements such as “[t]here are three well-established types of treatment for depression: medications, psychotherapy and electroconvulsive therapy (ECT)” (NAMI, n.p.). Here, “medications” is listed prior to “psychotherapy and electroconvulsive therapy,” which is indicative of this American website’s leaning towards that treatment. In conjunction with the order of information, and to further assess the differences in the choice and variety provided to patients, I also measure the amount of the information provided on pharmaceuticals and talking therapy respectively. These comparisons further reveal website treatment ! ! ! 20! preferences; common sense dictates that patients seeking information are more likely to ask for treatments about which they are more informed. Finally, this analysis focuses on how different visual elements and the multimodal nature of the websites affect a reader’s interpretation of website information. Indeed, speaking on visual images Kress and Van Leeuwen state that “the placement of the elements (or the participants and of the syntagms that connect them to each other and to the viewer) endows them with specific information values relative to each other” (176). To account for the way visual elements are placed, I focus on the gaze, zoom, and focus of images as well as the text that these websites choose to call out through larger, more eye-catching font. Multimodal analysis also accounts for the ways in which the website structure may lead patients to particular types of treatment information (as noted by O’Halloran 1).8 For example, NAMI’s website structure makes it more likely patients will be able to access information on antidepressants than psychological treatments. From the “Treatment” page, NAMI provides five links to resources on medication, but only two hyperlinks reference psychological treatments (and both of these are linked to the same page on the NAMI website). Thus to consider the use of hyperlinks and where a website persuades it reader to click next is revealing of treatment preference patterns. 3. Vocabulary This section is the first of three analysis sections and focuses on the vocabulary used by the different countries' websites. Fairclough states that “[i]n some cases, what is !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!8!O’Halloran advocates multimodality, which she describes as analyzing and contextualizing language use “in conjunction with other semiotic resources which are simultaneously used for the construction of meaning,” including visual images – as I will discuss in section five (1). ! ! ! 21! ideologically significant about a text is its vocabulary items per se” (113). Here, I highlight the differences in the vocabulary used when describing depression and treatment by medication and talking therapy, focusing on naming, positive and negative descriptors and the words they occur with. Throughout this section, subtle vocabulary differences are shown to persuade readers of the merits of different treatment options. As expected, the three websites affiliated with pharmaceutical companies’ language reflect a preference for medication. Canadian websites also appear to favour medication, whereas England and the United State’s websites’ vocabulary appears more ambiguous. 3.1. Major Depression? Depressive Illness? Depressive Episodes? What’s in a Name? There are numerous different names and titles used to describe depression. Leslie Jeffries’ Critical Stylistics highlights the potential significance of such naming. Building on Fairclough’s vocabulary category, Jeffries emphasizes how “the packaging up of ideas” within a nominal unit can significantly affect how a reader is asked to respond (22). On patient information websites on depression, the terms that the websites use to refer to the illness point to different treatment options for the patient. This is especially the case when considering the different implications using “episode of depression” versus “depressive disorder.” The former suggests short-term treatment is necessary, whereas the latter implies that a longer-term solution is required. In Criteria A for Major Depressive Disorder, the DSM-IV-TR makes two significant distinctions that affect how a patient’s illness will be labeled. Major Depressive Disorder can either include the “[p]resence of a single Major Depressive Episode (296.2x Major Depressive Disorder, Single Episode) or two or more Major ! ! ! 22! Depressive episodes (296.3x Major Depressive Disorder, Recurrent)” (199). “Major Depressive Episode” also allows a physician to distinguish between depression originating from a unipolar depressive illness and depression as part of bipolar disorder. However, despite heavy use in the DSM, the terms “Major Depressive Disorder” and “Major Depressive Episode” are not consistently applied on the websites of England, Canada or the United States. Although the nuances of the terms for diagnosing depression are not necessarily important to patients seeking advice on how to improve their functioning, these terms do have implications for how a reader is encouraged to frame the illness: “episode” and “disorder” have different connotations. Such differences in naming exemplify Ruth Wodak’s argument that “[a]t all points, intertextuality is important; […] we should include other information which relates to our problem, such as other discourses of the same speakers, other events in the same institution, etc” (21). It is through comparing and contrasting that the significance of these language choices can be assessed. As one might predict, “depression” on its own is the most regularly used term, a shortening of the official diagnosis. However, these countries’ websites differ in their application of other terms for the illness. Numerous websites begin their introduction to depression by providing a list of alternative names for the illness, and then proceed to favour one or two variants of this. For example, NAMI (from the US) states that “Major depression is also known as clinical depression, major depressive illness, major affective disorder and unipolar mood disorder” (n.p.). By opening with “Major Depression,” and using this term more frequently than others, this website – in line with others from the United States – uses terminology that aligns more closely with the DSM IV. By contrast, ! ! ! 23! England’s websites tend to treat “depression” as a hyponym, specifying degrees of depression to create compound nouns (“mild,” “severe,” and “moderate” – and “moderate” is not found on other websites).9 Canadian websites tend to use the hyponym “mood disorder” to reference depression, emphasizing depression’s membership among groups of other illnesses. Although this is the superordinate term in the DSM IV-TR, Canada is the only country whose websites use “mood disorder” for depression. This brings us to the key question: how do different terms for the disorder affect the different treatments recommended? England’s websites’ system of differentiating the severity of depression clearly outlines that for mild or moderate depression antidepressants are rarely an option. Similarly, England uses “episode of depression” more frequently than the other countries, specifying the relative brevity of the illness – again, perhaps an indication to turn to treatments other than medication: certainly short term treatments appear more suited for “episodes.” Here, as Wodak specified, it is through intertextuality, and international comparisons, that this choice is rendered significant; using this term is indicative of a social practice where depression is treated as something that is not constant. Using “episode,” then, perhaps suggests that medication may not be necessary, and that healing may occur without pharmacological treatment. Canadian websites use “mood disorder” to describe depression. “Mood Disorder” is a category inclusive of illnesses that almost always require medication – like bipolar disorder. By grouping these illnesses, by extension, these webpages group treatment, especially as the information frequently discusses the two illnesses together. Furthermore, the term “disorder” suggests a more severe and lasting illness (due to its association with !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 9 These are considered compound nouns, not adjectival phrases, as it would not be possible to retain the same semantic meaning if an additional adjective were to separate the head noun from the compound adjective. E.g.”*moderate regular depression does not work as well as “regular moderate depression.” ! ! ! 24! bipolar and other psychiatric illnesses like borderline personality disorder). Thus, by labeling depressive illnesses with this superordinate term, the vocabulary indicates that a Canadian patient may be more likely to accept medication as a necessary treatment for his or her depression. The United States’ websites, given that they are created for the country from which the DSM originates, use “major depression,” a term from the manual, more frequently than other options. This reflects the United States’ websites adherence to the diagnosis and treatment standards set by the DSM. Using the evaluative term “major” describes the severity of the illness. Using “Major Depression” therefore may predispose a patient to opt for pharmacological treatment. However, as the DSM also recommends other treatment methods, using such a standard term posits the United States in the most neutral vocabulary zone of the three nations. Peter Conrad writes that “[m]edicalization is prevalent in the United States but it is increasingly an international phenomenon. The prevalence of medicalization is partly the result of the expanding hegemony of Western biomedicine, but it is facilitated by multinational drug companies and the global reach of mass media and the Internet” (144). 10 Thus, while the United States’ websites in using DSM terminology adopt supposedly neutral vocabulary, one must not forget that the treatment of depression is significantly influenced by work within the United States. Conrad argues that Western biomedicine is subject to an “expanding hegemony.” Yet deviations in the Canadian and English websites from the terms used by the DSM reveals that this ‘Americanization’ is not complete – and the deviating vocabulary choices point to potentially different approaches !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 10 Conrad describes medicalization as treating more and more human experiences as medical conditions. Peter Lane’s discussion on shyness exemplifies this as he argues that shyness has been medicalized into the condition Social Anxiety Disorder. ! ! ! 25! to healthcare. Ethan Watters argues that “we are engaged in the grand project of Americanizing the world’s understanding of the human mind” (1). Watters notes, in a cross-cultural examination of mental illness, that “[i]n addition, a particularly Americanized version of depression is on the rise in countries across the world” (2). The dispersal of knowledge therefore may reduce variation not only in treatment but also in the illness itself. Part of the way in which the “expanding hegemony” grows is through the language used to describe the illness to patients worldwide. 3.2. Positive and Negative Descriptions of Treatment The discussion of positive and negative language – as central to persuasion - will be tackled in two parts. First, I will examine descriptions on third party, sponsored webpages, and secondly, I will focus on positive and negative descriptions provided by the main selection of data, the voluntary and government-run websites. 3.2.1. Pharmaceutically Sponsored Websites When examining the lexis used in the descriptions of treatments, I must first take note of third party interests. The corporate and pharmaceutically sponsored websites, Netdoctor.co.uk, Ifred.org, and DepressionHurts, each reveals a bias towards pharmacological treatment in their vocabulary – regardless of their country of origin. Indeed, Netdoctor.co.uk lists the following warnings for therapy: • therapy is not usually rapid in its effects, and you need to know that it will be challenging • you need to have a strong desire to find out more about yourself and to see the therapy through • sometimes you will need to face difficult aspects of yourself, which may include hostile feelings about your therapist • therapy can result in changes in the nature of your relationships with other people. The words “challenging,” “strong,” “difficult aspects of yourself,” and “hostile” all paint a negative image, illuminating the difficulties a person might encounter in therapy for ! ! ! 26! depression. The website’s use second person address reinforces the potential impact of these effects; this synthetic personalization (see section four on grammar) ensures that readers identify themselves as the target, the people for whom the “need to face difficult aspects about yourself” is applicable. In describing antidepressant treatment, the webpage couches the negative in positive terms: “[m]any people are wary of taking antidepressants, although they need not be.” Using “many people” posits the reader as someone who is not in this category. Furthermore, negative descriptors of side effects do not feature in the main list of antidepressant information, apart from a statement on SSRI’s that “[t]he main advantage of these new drugs is that they have fewer side effects than older drugs and so are more pleasant to take.” Here, the word “pleasant” denotes a positive and enjoyable experience – a word that speaks to recovery. Hence, the vocabulary of Netdoctor points towards antidepressants as the most desirable, or “pleasant” treatment option. The American pharmaceutical company-sponsored website Ifred similarly promotes antidepressants. This is exemplified through the website’s use of vocabulary that indicates of the success of particular medications and the way in which they work. Ifred’s information asserts serotonin theory and that depression is a chemical illness, without admitting that the precise structure of the brain and the ways in which medications work are still unknown. However, on a non-pharmaceutical-sponsored website, the relationship between serotonin and depression is less explicit, highlighting a gap in researcher’s knowledge. This is seen, for example, on Psych Central (from the US, but not pharmaceutically sponsored): ! ! ! 27! SSRIs work on increasing the amount of serotonin in the brain. Researchers are not sure why an increase in serotonin helps relieve depression (there’s little evidence of support for the serotonin theory of depression), but decades’ worth of studies suggest such medications nonetheless help improve mood. (n.p.) The hyperlink connects to research on the uncertain link between depression and serotonin.In contrast, Ifred has the following description for Cymbalta: This medication works by restoring the balance of two kinds of natural substances (neurotransmitters known as serotonin and norepinephrine) in the brain, thereby improving mood and feelings of well-being” (n.p.). The difference is significant for the web-user seeking accurate information. A web-user reading the tentative language from Psych Central (“suggests,” “not sure,” and “little evidence of support”) would be more cautious about antidepressants. However, someone reading Ifred’s definite assertion of how the medications “work[..],” and how the chemical changes “thereby improve mood and feelings of well-being,” is less likely to be hesitant. Furthermore, Ifred’s introduction asserts that Ifred provides a “variety of information about the latest and greatest medications available to help treat depression.” Such positive vocabulary, particularly with a catchy internal rhyme, lingers with readers as they peruse the following text. The Canadian pharmaceutical sponsored website, DepressionHurts, also uses vocabulary that asserts the validity of the “serotonin theory.” As seen below, DepressionHurts asserts “the science behind depression,” and emphasizes that “many researchers” support the theory that chemicals in the brain are responsible for this illness: The Science Behind Depression Many researchers share the view that depression is caused by an imbalance of naturally occurring chemicals, known as neurotransmitters, found throughout the brain and the body. Neurotransmitters help transport messages between nerve ! ! ! 28! cells. Serotonin and norepinephrine and dopamine are examples of these neurotransmitters. In the brain, serotonin and norepinephrine are thought to be associated with mood as well as regulating and reducing feelings of pain that come from the body. Dopamine is thought to be associated with appetite, loss of pleasure and energy or drive. Here are a few options your doctor may recommend: Medication If your doctor recommends medication, make sure to talk to your doctor about the treatment options available to you. Talk Therapy There are many forms of talk therapy. Cognitive behavioral therapy is the most studied and is an effective treatment for depression Cognitive behavioural therapy helps people correct negative thought patterns and better adapt to the world around them. Talk therapy has few known risks and is generally considered to be safe. This website’s decision to list “Medication” ahead of “Therapy” in the “options your doctor might recommend” relies upon the lexical cohesion of medication with “dopamine,” “neurotransmitters,” and “serotonin.” As DepressionHurts describes depression as chemical, this naturally leads a reader towards medication as a solution. Although DepressionHurts does use positive evaluative terms for talking therapy, such as “an effective treatment,” “helps people,” and “safe,” it is to medication that a reader is naturally directed; the problem of “chemicals” is not explicitly solved in talking therapy. The order of information (discussed in section five) reinforces this evaluation. Finally, before this analysis turns away from the corporate world, I must address the impact of discourse outside of these websites– notably Direct-to-Consumer advertising. The non-sponsored American website HelpGuide joins a range of websites (e.g. RxISK.org) which critique the way in which Direct-to-Consumer advertising affects the information a patient receives, stating “Depression medication may be the most advertised treatment for depression, but that doesn’t mean it is the most effective.” DTC ! ! ! 29! advertising of prescription medication has been legal in the USA since 1985, although in 1997 the stipulation of a detailed list of side effects was removed, which encouraged such advertising to flourish. By contrast, Canada requires that only one of the condition or the product is mentioned, and England bans advertising prescription medicines outright.11 These different advertisement strategies are significant in influencing patient choices (as discussed by Healy and Conrad). Yet, Helpguide also demonstrates how a plethora of different influences affect patients’ choices. By recognizing these influences, the website not only critiques DTC advertising, but also raises the larger question of whose recommendations might be biased –bias which a reader may extend to the Helpguide itself. This website both hinders and extends its credibility through such reasoning; by criticizing sources in its argument, the webpage appears less biased and therefore more trustworthy. 3.3.2. Descriptions of Talking Therapy and Antidepressants on Non-pharmaceutically sponsored Websites The effects of talking therapy and antidepressants are described differently by both English and American websites. Both of these countries’ websites frequently use the verb “helps” in reference to therapy, and this is the most commonly used evaluative term on the English websites. However, “treats” is used by both websites to describe the effect of !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 11 The United States is one of only two countries (the other is New Zealand) to permit the Direct to Consumer advertisement of pharmaceuticals. In 1997, the USA’s Food and Drug Administration (FDA) issued the Draft Guidance for Consumer-Directed, Broadcast Advertisements. The final draft was issued in 1999. See: http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/UCM12282 5.pdf This Guidance requires that “adequate provision” of information the “most important” risks of the drug are provided to the consumer (8). The FDA states that it has made DTC advertisements “more feasible” (7) with these revised guidelines. In 1978, The Food and Drugs Act in Canada states that when advertising a prescription drug (class F), the advertiser cannot reference anything other than the brand name, general name or price and quality of the drug. Therefore, the condition and drug advertisements do not occur together. The UK’s Medication and Healthcare Regulatory Agency bans DTC advertisements for prescription medications outright. ! ! ! 30! antidepressants. The distinction between these two verbs is significant: “helps” is vague, and does not suggest an entirely effective tool, whereas “treats” implies medical intervention: if an illness has been treated, the indication is that the treatment is at least partially successful. While both websites also use adjectives like “effective” to describe talking therapy, and England’s websites use positive phrases like “works well” and “good treatments for depression” (Patient.co.uk, n.p.), the disparity between “treats” for antidepressants and “helps” for therapy should not be underestimated.12 Indeed, it appears that both the American and English websites’ vocabulary persuades patients that drugs are more effective than psychotherapy for treating depression. Canada’s websites, in contrast, tend to emphasize the evidence for a treatment, with terms such as “extensively researched” and “consistent evidence” prevalent in their description of treatments. However, Canada’s discussion of treatment is difficult to measure against websites from England and the USA because, as noted in section four, Canadian websites provide minimal information on antidepressants. However, NAMI’s health brochure claims that “severe depression requires antidepressants” (13). Similarly statements such as “you will find the one that works for you” are persuasive towards antidepressant usage. Websites in all three nations speak to the benefits of combining talking therapy and antidepressants. However, the statements that emerge from the United States and Canada on this combination emphasize more strongly a medication and therapy approach !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 12 Treating talking therapy (which consists of a spectrum of approaches) as one entity was the most practical way to approach this comparison. However, statements about different types of therapy also alter the persuasion. Generally, Cognitive Behavioural Therapy (CBT) is described in positive terms, and is the most common form of therapy available. For example, NAMI describes CBT as one of “several types of psychotherapy that have been shown to be effective for depression” (n.p.). However, NAMI expresses doubt about the effectiveness of psychodynamic therapy, for example, stating that “researchers recommend it less often” and that one study showed that it was “no more effective than a placebo” (n.p.). ! ! ! 31! – the take home message is that therapy alone is not enough. For example, Psych Central (USA) states that a “treatment approach that focuses exclusively on one of these factors is not likely to be as beneficial as a treatment approach that addresses both psychological and biological aspects (through, for example, psychotherapy and medication)” (“Psychotherapy”, n.p.). Furthermore, Psych Central adds “Treatment for depression, like for most mental disorders, usually relies on psychotherapy and medication for the quickest, strongest effects” (n.p.). Again, the positive vocabulary – “quickest and strongest effects” – coupled with the adverb “usually” acts as a persuasive device for medication alongside talking therapy. 3.4. Collocations with “Common” The words that collocate with “common” on these websites also point to the websites’ preferred treatment. To show the different uses of this word in the three countries, I counted occurrences of the word and classified the words it co-occurred with in the corpora. Canadian websites tend to avoid the word altogether. In contrast, those from England use “common” frequently to refer to depression itself, whereas websites from the United States often use “common” in relation to medication and antidepressants. This trend is revealing. Fairclough stresses that in some cases what is significant about a text is the way in which the words co-occur or collocate (113). By frequently positing antidepressants as “common,” as the United States’ websites do in examples such as “common types of antidepressants are…” (Public Library of Medicine, n.p.) or in the adjectival phrase “common SSRI” (Psych Central, n.p.), websites from the United States may persuade their users to ask their physicians for this treatment. The expressive value of the word ! ! ! 32! “common” here relies on the connotation of effectiveness – as well as the denoted frequency. In the medical world, the word “common” tends to collocate with “harmless” (as with the “common cold”) or with a reduced risk; if a treatment is common, then one can assume that the treatment is successful. Associations like these exemplify Fairclough’s argument on the dialectic of structures and practices: “discourse has effects upon social structures, as well as being determined by them, and so contributes to social continuity and social change” (16). Here, the website’s decision to associate medication with something that is “common” may make it so, as more people are encouraged to opt for antidepressants. England’s use of the word “common,” however, relies on different associations. Through linking the word “common” with the illness itself, and not the treatment, these websites promote a message designed to remove stigma. If depression is common, then it is nothing to be ashamed of. Therefore, this collocation is designed to get people who may be suffering from depression to seek help – whatever this may be. It is also worth nothing that websites from England use “common” 49 times, compared with 29 occurrences on the USA websites and eight on Canadian websites. Almost 50% of England’s websites’ use of the word “common” reference the illness itself (24 of 49), compared with under a third of the occurrences from the United States (8 of 29). The difference in the use of “common,” is more significant given that – as Table B reveals – the United States’ corpus is larger than England’s. This difference further supports the idea that England’s websites’ emphasize the removal !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 13 For more comparable corpora, the extensive medication and therapy library section of Netdoctor (England) and Helpguide (USA) is not included in these counts, but discussed in section five. Table B: Amount of Words in Corpora to Nearest 100 Words13 England 19300 USA 23000 Canada 12600 ! ! ! 33! of stigma for the illness, whereas the United States’ websites are more focused on removing stigma for the medication. 3.5. Section Summary: Vocab There are no clear boundaries that divide the countries’ websites’ treatment preferences. However, by naming depression as “episodic” or as a “disorder,” the language used by these websites does appear to persuade potential patients to choose short term (England) or long-term (Canada and the United States) solutions. These solutions may or may not involve medication. Canada’s use of the superordinate term “mood disorder” most strongly suggests that medication is necessary. This section also points to differences in the way that third party websites use persuasion: all three of these websites persuade users to choose pharmacological, above therapeutic, treatments – even if all websites acknowledge that a combination of both treatments is the best predictor of a positive outcome. Finally, this section highlights how the same word – “common” – has a different meaning for different websites. Within England, the word is associated with the illness, and meant to reassure the patient. In general, within the United States, the word is associated with medication; thus it aims to persuade patients that medication is not something to fear. This, of all the vocabulary differences uncovered, is the most striking example of the United States’ subtle use of language to persuade its patients to choose medication. 4. Grammar This section on the grammatical features used by these websites, builds on the discussion of vocabulary, revealing how England’s emphasis on depression as a “common” disorder ! ! ! 34! is conveyed through marked syntax. The marked syntactic structure of cleft sentences further reveals this tendency. I also consider the websites’ application of pronouns and modal verbs to reveal significant differences in the degree of patient control suggested by texts from different countries, and the way in which these websites emphasize particular treatments. While the vocabulary alone does not clearly divide the countries according to treatments, the grammatical features suggest that there are clear differences with regard the portrayal of patients’ responsibility and treatment decisions. Of these differences, the most notable finding is that websites from England convey a higher level of patient input on treatment choices, by using more direct address. This country’s websites also use first person plural pronouns and cleft sentences to emphasize a collective responsibility for the illness and its symptoms. By contrast, Canadian websites accentuate the presence of a medical team, and United States’ websites stress the impact of depression on those closest to the depressed person. 4.1. Pronouns The most significant finding with regard to the websites’ use of pronouns lies in the websites’ use of direct address, or second person pronouns. Table C demonstrates that these pronouns are more frequent in texts from England. Fairclough labels the use of direct address “synthetic personalization,” a technique he describes as “a compensatory tendency to give the impression of treating each of the people ‘handled’ en masse as an individual” (62). England’s websites, therefore, present the reader with what appears to be a more individual approach – in contrast to the mass, or the depersonalized third person (they), the reader is addressed directly, and encouraged to identify as the subject. Althusser describes the device in which a reader is “hailed” and encouraged to respond to ! ! ! 35! the text more personally than they might otherwise do as interpellation (105). The reader is transformed into the addressed subject. For example, by stating that “[t]alking through your feelings can be helpful [and…] you can ask your GP to suggest a local self-help group” (NHS Choices) targets the reader directly: it is “your” feelings and it is “you” who is the agent of the proposed action. The reader is positioned as someone for whom the text is directly relevant, and “hailed” to take up the role of a depressed person who should act. As these websites provide information on treatment for depression, the websites’ decisions to position parts of this information so that it appears especially relevant to the reader is significant. The websites’ use of direct address encourages readers to assess the suitability of particular treatments for themselves, and decide whether they wish to adopt the position of the “you” who “can ask your GP to suggest a local self-help group.” By positioning patients with agency, the distribution trends demonstrated on these websites appear to fit with England’s shift to a concordance (rather than “compliance”) model of care, reflecting the NHS agenda for more patient involvement (NHS 2012; Segal 2007). Furthermore, as these websites address the potential patient in second person, this involvement points to the psychological interactions undergone in talking therapy. Direct address, then, may subtly encourage a patient to consider psychological therapy, a place in which considering “you” is at the forefront. Table C: Counts for Use of Second Person Subject (\"You\") Object (\"You\") Possessive(\"Your\") Reflexive (\"Yourself\") USA 138 58 82 5 Canada 127 62 138 8 England 258 88 117 18 The argument that England’s websites provide readers with more agency is further strengthened by the position of these second person pronouns in the syntax. ! ! ! 36! Tables C and D reveal that over half of the sentences from England’s websites involving direct address use the second person in the subject position. For example, “But as a general rule, if you are depressed, you feel sad, hopeless and lose interest in things you used to enjoy” (“Symptoms” NHS Choices, n.p.). However, only 37.9% of the second person pronouns on Canadian websites are in the subject position, with the United States’ at 48.8%. By addressing the reader in the subject position, the websites place the reader syntactically in a position of power. In contrast, the object position for “you” describes the reader as influenced by other people or items – for example, “Depression can make you feel helpless, hopeless, or empty and numb” (Helpguide, n.p.). Here, the addressee is the experiencer, the one made to feel helpless, with “depression” as the subject, and agent of the verb. While all of the websites aim to motivate the reader to tackle depression, the different power roles point to what it is both permissible and possible to for readers to do for themselves. Table D: Percentage for Use of Second Person Subject (\"You\") Object (\"You\") Possessive(\"Your\") Reflexive (\"Yourself\") USA 48.8 20.5 29 1.8 Canada 37.9 18.5 41.2 2.4 England 53.6 18.3 24.3 3.7 Canada’s relative higher use of second person possessive pronouns also indicates, on the surface, that readers are portrayed as having more control: “your” directly references their ownership. However, the frequent use of “your” is often paired with “doctor” or “healthcare provider” (38 counts). For example, DepressionHurts lists “Questions to ask your healthcare provider” (n.p.). Thus the personal responsibility of the patient is removed: the patient is expected defer to the expert, a phenomenon noted in the amount of information provided (see section five). Patients, then, are the recipients of ! ! ! 37! direct address more in relation to their physicians than to their own role in decision- making, suggestive of a system based on the notion of patient compliance. Furthermore, these websites list expert consultation beyond the GP with a psychiatrist prior to further consult with a therapist for talking therapy. This emphasizes that a physician is contacted to diagnose and prescribe, not for talking therapy. Medication as a treatment option is foregrounded. Examining the use of first person plural allows me to further emphasize the impact of address on the level of control a patient has over treatment decisions that are portrayed on these websites. Significantly, Fairclough notes that by using the inclusive “we” (where the reader and the speaker are grouped together) that “in so doing, it [the text] is making an implicit authority claim […] that it has the authority to speak for others” (128). However, such inclusive address can also be used to remove individual responsibility. Replacing individual responsibility with a collective group removes stigma. For example, Mind (from England) uses the inclusive “we” to avoid isolating a particular group of individuals: “some of us are more prone to depression than others. This could also be because we learn behaviour and ways of responding from our relatives, as well as inheriting our genes from them” (n.p.). Therefore, in terms of patients’ control of their illness and/ or treatment decisions, the inclusive “we” removes individual responsibility: the text makes the “implicit authority claim” that it can speak for the reader. The linguist Alastair Pennycook argues that “if ‘we’ claims authority and communality, it also constructs a ‘we/you’ or a ‘we/they’ dichotomy. Thus, these two pronouns must always be understood with reference to other assumptions about who is being defined as the ‘we’ from which the ‘you’ and the ‘they’ differ” (176). Indeed, the ! ! ! 38! ambiguous nature of pronouns means that they should always be understood with reference to context. Here, the context reveals that “you” and “I” (for the inclusive “we”) are the same. There is no single address to the reader: it is “we,” and not “you” who are sufferers of depression. Table E demonstrates that use of first person plural, while rare on all websites, is significantly higher in England. Although the websites analyzed used more direct address (discussed above), by placing more emphasis on the individual audience, the websites also seek to compensate for this emphasis by removing some specific references to sufferers. This is supported in the previous vocabulary section, which revealed that English websites regularly describe depression as “common.” Table E: Use of First Person Plural Countries \"We\" (inclusive) \"We\" (Exclusive) \"Us\" (Inclusive) \"Us\" (Exclusive) \"Our\" (Inclusive) USA 2 1 0 0 1 Canada 1 1 0 1 0 England 15 7 2 3 2 By stressing the minimal knowledge possessed by the experts in England, the exclusive “we” (to refer to writers of the text and the organizations they represent) further helps to mitigate the distinction between expert and patient. Given that the United States’ corpus is larger (see Table B), the higher count of the exclusive “we” from England is revealing. Of the seven English examples, three underscore the minimal knowledge possessed by the medical community about the causes of depression and, significantly, the way in which drug-induced chemical changes treat the illness. Mind demonstrates this: Some doctors may tell you that you have a ‘chemical imbalance’ and need medication to correct it. But the evidence for this, apart from the effects of medication, is very weak, and if changes to brain chemistry occur, we don’t know ! ! ! 39! whether these are the result of the depression or its cause. Although there are physical tests which are occasionally used in research on depression, they are not very accurate or consistent, and there are none that are done routinely to help make a diagnosis. (n.p.) Here, a dichotomy is established between “some doctors” (“they”) and the people who accept that they do not know (the writers of this website, and other, “better,” doctors). This example of exclusive “we” also presents ambiguity: readers can potentially align themselves with this position and interpret the pronoun as inclusive. Yet, readers cannot align themselves as people who agree with the idea of chemical imbalance: as potential patients, readers are not “some doctor.” Therefore, Mind’s use of pronouns demonstrates English cautious approach to the “serotonin theory,” in contrast to the United States’ tendency to endorse it (see section three). Thus, the way in which pronouns stress speaker and readers’ opinions on these websites affects the how patients view treatment options. 4.2.Focus and Prominence: Cleft Sentences Syntactic organization can also affect the treatment a patient is persuaded to request from his or her doctor. Clefts were selected in this analysis of syntactic structure as I noticed the most salient content differences in clefts in the information provided on the websites. The cleft sentences of the website corpora reveal that the patient’s responsibility for his or her treatment decisions is emphasized differently in all three countries. Asp and de Villiers write that “[t]he basic ideas of FOCUS/PROMINENCE is that there is a statistically high probability that certain types of element will occur as the first element in a clause and that any departure from such normative probabilities will be in some degree marked” (80). The use of cleft clauses particularly reveals emphases as “[c]left and pseudo clauses allow prominence to occur on a single marked constituent” (Asp and de Villiers 82). Therefore, this syntax may reveal which messages a reader is ! ! ! 40! most likely to recall as these messages are marked by the websites. For example, in the sentence below the prominence is on “relief,” which emphasizes the possible feelings of the addressee reader (“you”): It may be a relief to know that you are not going mad Focus = bold Prominence = underlined (Patient.co.uk, n.p.) England’s use of cleft sentences often follows this pattern of emphasizing the “common” nature of the illness, and focuses on reassuring sufferers of depression that, as above, “[they] are not going mad.” Further examples are listed below: It is not weak to cry or admit that you are struggling. (Patient.co.uk, n.p.) It is NOT a sign of weakness to get help for your problems (Netdoctor, n.p.). These two examples contain a prominent negative clause. This further emphasizes the negation, stressing that it is not weak to seek help. Thus, these websites accentuate individual experience, and the importance of the individual when making treatment decisions and managing depression. England’s websites’ cleft sentences also tend to be shorter than the clefts used by other countries. Table F reveals that although the English websites contain the highest number of cleft sentences, the average length of these sentences is significantly lower than the length of those from other countries. Indeed, England’s clefts are, on average, almost half the length of those from the United States, and under half the length of those from Canada. This may reflect the choice of topic: as nine of England’s thirty cleft sentences emphasize that depression is a common illness, it appears more appropriate to make the sentences short and memorable. ! ! ! 41! Table F: Proportion of Cleft Sentences Countries Amount of Clefts Words in Clefts England 30 390 Canada 25 945 USA 15 417 In contrast, twelve of the twenty-five cleft sentences used on the Canadian websites stress the importance of a professional recovery team and the importance for the patient to let the professionals know all available information. Such emphases permit more wordy and elaborate statements about how to best use the team’s expertise: It is important not to try and diagnosis yourself or wait in hope that depression will just go away (Mood Disorders Society of Canada, n.p.) A support system is important to anyone’s wellbeing but it is crucial during recovery. (DepressionHurts, n.p.) Here, the emphasis is on professional input, which corresponds to Canadian websites’ distinct lack of information on treatment options; these are instead presented as something that should be discussed only with the guidance of physicians (see section five). By coupling cleft sentences with lexis emphasizing the urgency of the situation (“important” and “crucial”), these Canadian websites further stress the importance of the professional’s role over the individual’s. Six of the fifteen cleft sentences from the American websites emphasize the impact of depression on those close to the depressed person. As seen with Canadian websites, American cleft sentences deemphasize the role of patients. As the following examples demonstrate, cleft sentences on these websites emphasize the needs of a different audience than the depressed patient: It is common for both the person living with the illness and family members to experience grief because of the drastic changes in their lives and the trauma that previous episodes may have caused. (NAMI, n.p.) ! ! ! 42! It’s just as important for you to stay healthy as it is for the depressed person to get treatment, so make your own well-being a priority. (HelpGuide, n.p.) While these cleft sentences do not point to specific treatment options, the websites’ decision to switch the audience from the depressed patient to those seeking information on behalf of the patient removes the sufferer from the information on the website. Explicitly addressing a different audience therefore may affect the way in which a patient interacts with the rest of the information available on the website. By emphasizing the range of people affected by the condition, these websites evoke a sense of patient responsibility to others, and thus subtly emphasize the need for the patient to defer to— and take into account—the opinions of both professionals and those close to them. The marked syntactic structure stresses this. Furthermore, Segal’s essay on the effects of direct-to-consumer advertising of pharmaceuticals reveals that Zoloft advertisements in 2005, for example, emphasized the way in which one’s mental health affects other people. Segal states that “guilt for not trying hard enough (as parent, sister, ex-wife, or worker) is a key part of the motive to medicate” (2011: 23). Here, forcing patients to recognize the impact they may be having on those close to them appears a similar strategy used by these websites. 4.3. Modal Verbs Fairclough states that “interest is in the authenticity claims, or claims to knowledge […] are evidenced by modality forms” (129). It is these “claims to knowledge” that can significantly alter the treatment a patient is persuaded to choose; degrees of certainty of a treatment’s success or availability are conveyed through modal verbs. To recap, epistemic modality reveals certainty and probability, and deontic modality reveals obligation. ! ! ! 43! Table G: Modal Verb Counts Epistemic Deontic Countries Can Could May Might Must Should Will Would Can May Must Should England 178 9 200 19 2 2 41 18 32 1 1 29 Canada 100 4 77 4 3 0 52 0 8 2 5 15 USA 150 2 98 39 8 2 71 7 32 1 3 21 As Table G shows, the most significant difference in the websites’ use of modal verbs lies with the epistemic modal verb “may,” which occurs 200 times in the English corpus. Proportionally, this modal makes up 1.04% of the English corpus, compared with 0.42% of the USA corpus and 0.61% of the Canadian corpus. This verb is frequently used in relation to what a physician may or may not recommend or prescribe, and highlights that decisions about the best course of treatment are not universal: Remember that even when you feel better, your doctor may ask you to continue taking your medication for a while – this is just to make sure that the depression does not return. (Depression Alliance, n.p.) Your GP may refer you to a qualified fitness trainer for an exercise scheme (NHS Choices, n.p.) The websites’ use of “may” affirms that that different treatment plans are advisable for different patients – a fact that websites, such as NHS Choices and Patient.co.uk, emphasize by sub-dividing treatment information into categories dependent upon how severe the depression is (see section five). Yet “may” also emphasizes that the power is with the doctor: it is up to the physician which treatment they recommend – perhaps this is indicative of a system of patient compliance masked by concordance. The websites from Canada, which avoid detailed discussion of antidepressants, also affirm that control is with the physician. This is supported by Canada’s lower counts of the deontic “can,” where things that a patient can do are not emphasized through modal auxiliaries as frequently as they are by other websites. ! ! ! 44! Examining the use of modal verbs on websites from the United States reveals a higher count for the epistemic “will.” The increase use of this modal verb reflects the greater certainty with which the outcome of treatment is described, particularly for antidepressants. To reflect variations in the effects of medications, the modal is accompanied by adverbs pertaining to time, such as “usually,” “sometimes” or “often” to allow room for error. Nonetheless, even with these adverbs, using the definite modal “will” suggests a greater degree of certainty than the modal “may,” as the examples below illustrate: The effects of medications will usually be felt within 6 to 8 weeks of taking an antidepressant (Psych Central, n.p.). !! Most people will feel a positive impact in one to two weeks, but they won’t experience the full impact for one to two months, said Dr. Dietrich. (Psych Central, n.p.) Your doctor will often prescribe higher (but still safe) doses of an antidepressant, or a combination of medications. (US Library of Pub Med, n.p.) Sometimes people will need to stay on medications for long periods of time. (US library of Pub Med, n.p.) These sentences also suggest that medication will be effective. Through stating that people “need” to stay on medication, the Pub Med emphasizes the utility of this treatment. Similarly, the first example couples the possibility of a positive impact in one to two weeks with the certainty of a full impact in one to two months. This certainty is masked by the use of the conjunction as it is the first clause that readers are encouraged to focus on. Such certainty stands in contrast to England’s use of “may.” Even England’s third party sponsored website, Netdoctor, states that “[a]ntidepressants may help to reverse these changes” (n.p.). ! ! ! 45! 4.4. Section Summary In this section, the linguistic patterns discussed reveal that websites from England suggest that a patient has a higher level of control over his or her treatment choices. Such use reflects the NHS agenda for more patient involvement (NHS 2012). Yet to mitigate personal responsibility on the patient, websites from England also make use of the inclusive “we” function when describing symptoms or possible causes for the illness. Collective responsibility and the high frequency of the illness are also stressed by England’s corpus in cleft sentences. The United States’ websites recognize the multifaceted nature of their audience, often using cleft sentences to emphasize the impact of assisting a depressed person, and do not address patients themselves. Canadian websites also deemphasize the role a patient plays themselves, by emphasizing the importance of a medical team, and not discussing certain treatment options (this is discussed further in the next section). When writing on the treatments for depression, England’s language is more hesitant about the effects of antidepressants than the United States. Even the English pharmaceutically sponsored website, Netdoctor, qualifies the effectiveness of antidepressants with the modal “may,” whereas the United States’ websites uses modals such as “will” to convey a higher degree of certainty, qualifying them only with temporal adverbs such as “usually.” 5. Visual and Ordering Features In this final section of analysis, I build on the issues raised by the vocabulary and grammar sections to discuss the larger discursive features of the websites on depression. Here, I consider the significance of website organization, including the order and amount ! ! ! 46! of information, and use Kress and Ven Leuween’s visual grammar to uncover the way in which these websites use images to engage with their audience. The United States’ images place a higher emotional demand on their readers, adding to the previous section’s discussion on making the patient feel guilty about their effect on others. The amount and order of information also adds to prior discussion where the United States emphasizes antidepressants more strongly than England. By minimizing the information provided, Canadian websites emphasize the significance of one’s healthcare provider, making it necessary for readers to visit their physicians. 5.1. Amount of information The amount of information provided on each treatment option hints at the treatment a website prefers. I carried out a simple word count analysis on each website to reveal whether patients are provided with more information about therapy or antidepressants. This is considered in proportion to the amount of information provided by these webpages overall, where the United States provides the most information overall and Canada provides the least (see Table B14). The percentage of the information each treatment comprises was calculated to reveal the distribution of information. To create the counts, I isolated the sections on the websites that spoke directly to either antidepressants or talking therapy. If a website spoke about both treatments in a single sentence, I only counted the words “medication,” “antidepressants,” or “therapy”15 towards the overall count (this explains the low counts in Table J). !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 14 In addition to the counts on Table B, the amount of information includes the extended information Netdoctor and Psych Central provide in their library of antidepressants. 15 When referencing talking therapy, and excluding any mentions of pharmacological therapy. ! ! ! 47! Table H: Word Count for Information on Antidepressants and Therapy in the United States Amount of Information AMERICAN Ifreg.org NAMI Helpguide.org Psych Central Pub Med Total Percentage Antidepressants 1376 926 2566 87486 265 92619 94.6 Therapy 80 414 3923 771 102 5290 5.4 Table I: Word Count for Information on Antidepressants and Therapy in England Amount of Information ENGLISH Patient.co.uk Mind.org Depression Alliance Netdoctor.co.uk NHS Choices Total Percentage Antidepressants 356 230 398 64143 40547 105674 87.7 Therapy 390 310 264 5520 8380 14864 12.3 Table J: Word Count for Information on Antidepressants and Therapy in Canada Amount of Information CANADA Mood Disorders Public Health Agency of Canada Depressionhurts.ca Canadian Mental Health Association Canadian Psychological Association Total Percentage Antidepressants 638 1 249 1 0 888 39.3 Therapy 591 2 53 2 728 1374 60.7 On Tables H and I, the word count reveals that antidepressants (partly because there are so many of them to discuss) constitute a larger percentage of the information than therapy for websites from the United States and England.16 Both of the samples from these countries contain a webpage that lists each antidepressant and its side effects. Significantly, only 5.4% of the information on these two treatments from the United States addresses therapy. However, 12.7% of the information from England addresses !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 16 As indicated, there is more information to include about antidepressant medication due to the variety of drugs available and their various side effects. Therefore, one would expect an unequal distribution. The amount of information can also be influenced by Internet users search patterns (for example see section one and Segal 2007, below). ! ! ! 48! therapy, a small, but significant difference.17 Given the increased influence of pharmaceutical companies on healthcare in the United States (Conrad), this difference is, perhaps, not surprising. However, it must be noted that the high proportion of information from England reflecting medication treatment may also be traced to the influence of pharmaceuticals. The British Pharmaceutical association plays a large role in the public’s understanding of health and illness, funding research and promoting information.18 This may explain England’s websites’ larger information on medication than Canada’s websites. Canadian websites provide more information on therapy: 60.7% of the information they provide on antidepressants and therapy treatment options focuses on therapy, and only 39.7% on antidepressants. While the inclusion of the Canadian Psychological Association does somewhat bias this measure, it accurately reflects a Canadian enquirer’s Google results. As the order of search results depends in part upon the rate at which web users access them, this further supports the interpretation that Canadian Internet users, searching for depression, receive more information on therapy than antidepressants. This difference in amount is even more significant as the Public Health Agency of Canada and the Canadian Mental Health Association both provide no detailed information on either treatment option. For example, the following information !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 17 These figures do reveal some disparity even within the countries. For example, Healthguide.org, an American site, leans towards therapy, whereas Netdoctor.co.England and NHS Choices (ENGLAND websites) provide more information on antidepressants. Nonetheless, the overall trend is suggestive of geographic preferences. Indeed, the pharmaceutical sponsored webpage from England – Netdoctor.co.England – is expected to show this bias (see Section Three). NHS Choices also provides information on available antidepressants and their side effects, which necessitate a larger word-count than different therapeutic approaches. 18!For more information see: http://www.bps.ac.uk/details/aboutPage/841591/Aims__objectives.html?cat=bps12a5c851a17#743539,806 839 ! ! ! 49! from the Public Agency of Canada is the only time that this website broaches the possibility of therapy or antidepressants: What Can I Do to Help Myself if I Feel Depressed? You should tell your doctor that you think it might be depression, so that he or she can ask you the right questions and come to the most accurate diagnosis. Therapy is also very important. Many studies have shown that the most effective way to fight depression includes a combination of medication, self-care and psychotherapy. (Public Health Agency of Canada, n.p.) Here, there is a clear deferral of authority to “your doctor,” which implies that information from the Internet, or other non-medical sources, may not be trustworthy. Indeed, while the United States’ information offers disclaimers, such as “[t]his reprint is for information only and NOT a substitute for professional diagnosis and treatment” (NAMI, n.p.), the Canadian Public Health Agency’s refusal to provide the information in the first place contains a stronger message.19 The lack of information, then, leads one to doubt, not the legitimacy and truthfulness of Internet information (as discussed by Kulka), but its appropriateness. The question here is not on the issue of the autonomy of “lay people” that Kulka raises when considering the effect of patient information websites (34), but the risk of patients attempting to diagnose and select treatment by themselves. This is an issue Segal addresses, when speaking of the fear within a web-user researching illnesses: “In the case of the anxious Web user, the emotion most rhetorically powerful is not the emotion that, for example, a pharmaceutical company brings to the rhetorical situation; it is the emotion already in the user when she begins to click” (2009 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 19 By contrast, such disclaimers are difficult to locate on websites from England. This is reflective of England’s centralized system, where the patient’s choice is already limited by the availability of treatment. It is, therefore, inevitable that a doctor’s opinion will be sought as there are fewer possibilities to obtain treatment without consulting a GP. ! ! ! 50! 363). The information (or lack thereof) provided by these websites, then, suggests that the creators of these websites may be particularly aware of this problem. Indeed, by asking “What can I do to help myself?,” the Canadian Public Health Agency also asserts a hierarchy of information: “I” references the reader, who is reminded that he or she is seeking information, and therefore he or she is referred to an authoritative source (“your doctor”). By providing readers with practical (non-medical) advice in the period before they do contact their doctors, this website reinforces the authority of the physician, asserting a traditional biomedical system, and suggesting that readers should not act on their treatment preferences prior to consultation with doctors. 5.3. Order of Information The order in which these websites list their information further hints at psychopharmacological or therapeutic leanings. Eva Laeurbach indicates that “emphasis in language is a sequential phenomenon: as a local stylistic surface effect, it works linearly against the backdrop of the immediately preceding co-text, and on an underlying level against the background of the norms of culture, language, genre and the text so-far” (135). The data here are not straightforward. By counting both the order of the distinct paragraphs on therapy and antidepressants, and the order in which these treatments are listed together in sentences, I had hoped to obtain a clear overview of the websites emphases: what is foregrounded helps establish what Laeurbach labels the norms of the text. However, there are some disparities between the order in which the paragraphs on treatments are provided, and the order in which these treatments are grouped in single ! ! ! 51! sentences.20 For example, Pub Med (from the USA) presents a section on antidepressants prior to a section on therapy, yet lists therapy three times ahead of antidepressants in the same list. While these are low counts and typify the difficulties of deducing general trends from a small sample, they are indicative of the complex issue of persuasion and healthcare. Both the United States and England have three websites that list antidepressants ahead of therapy and two websites that list therapy first. While I will discuss below how subtle layout differences do suggest that the United States’ websites emphasizes antidepressants, these figures highlight that categorizing overall emphases of these countries is complex. Yet, with regard to Canadian websites, there is a significant emphasis on antidepressants, which precede talking therapy. Tables K and L illustrate this distribution of preferences. Canadian websites both emphasize antidepressants first in their lists and in the websites that contain sections on both talking therapy and medication. Indeed, despite having less influence by pharmaceutical companies than the United States, pharmacological treatment appears to receive more emphasis in patient information. By listing antidepressants before therapy, these websites signal that antidepressants are a “norm,” a route one should approach first. Listing antidepressants first is almost unanimous by these websites – even the Canadian Psychological Association (which does not have a section discussing medication alone) lists antidepressants ahead of therapy in one example advocating a dual treatment option: “However, because some of !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 20 All of the paragraphs provided on treatment discuss antidepressants and talking therapy separately, apart from those that mention them in a general summary of treatment or in a paragraph about the benefits of combining psychological and pharmacological therapy. In these instances, the order in which treatments are listed in a sentence were counted, as indicated in Table K. ! ! ! 52! the drug therapies may lead to a quick response they may be effectively combined with psychological treatments which provide for longer term change” (Dobson, “Psychology Works Fact Sheet” from The Canadian Psychological Association, 3). Furthermore, the Canadian Mental Health Alliance references The Mood Disorder Society’s leaflet for depression, an intra-national connection that demonstrates that this data is both recommended by and representative of information in Canada. The Canadian Society for Mood Disorders even lists the “last resort” treatment (Patient.co.uk, n.p) Electroconvulsive Therapy (ECT) prior to therapy. Such placement indicates a clear distinction between what the medical field offers, and therapy, which is posited as an additional – or paramedical – option. This stands in contrast to England’s almost entirely public healthcare system and the United States’ almost entirely privatized system, where both psychological and medication routes require one to engage with the same medical system and/or insurance company.21 5.4. Layout: Text Callouts and Testimonials Unlike Canada’s order of information, the order of information from the United States or England does not show a clear bias towards one treatment or the other. However, when I adopt a multimodal analysis approach, the information from the US does reveal a higher !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 21 This raises a question for further research about how this partially socialized healthcare system affects the choices available to patients. The WHO (2012) statistics show that 49.7% of Canada’s private expenditure on health is out of pocket; therefore healthcare plans are not always adequate for patients’ needs. ! ! ! 53! ! ! ! 54! preference for antidepressants than England. This can be shown through examining the effects of layout and text call outs,22 which illuminate emphases in these texts. Kress and van Leeuwen note that “[l]ike linguistic structures, visual structures point to particular interpretations of experience and forms of social interaction. […] For instance, what is expressed in language through the choice between different word classes and clause structures, may, in visual communication, be expressed through the choice between different uses of colour or different compositional structures. And this will affect meaning” (2). Below I use HelpGuide (from the USA) to exemplify the effect of layout. This website lists therapy prior to antidepressants. Yet, when discussing types of therapy, HelpGuide features a side box, shadowed in blue, with the hailing question “Are antidepressants right for you?” See Image A: Image A: Layout for HelpGuide (HealthGuide, n.p.) The blue box attracts a reader’s attention through its colour, but its position is also significant. Kress and Van Leuween describe left to right layout structures as representative of given (left) and new (right) information. Here, the box is on the right, !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 22 For this project, text call outs are defined as sections of text which are separate from the main body of information and marked by different font. These often include testimonials. ! ! ! 55! representing an alternative—a new—avenue for a reader to explore, and thus is especially salient. Furthermore, the box’s prominent position is coupled with the synthetic personalization (see Fairclough and section three), where readers are encouraged to answer the question of whether antidepressants are right for them – even as they read about the benefits of therapy. The information on therapy, like that on antidepressants, is peppered with quantifiers like “often” and “some,” which suggest that an alternative option is available. As this alternative is positioned next to the information, the website’s layout works as a persuasive device for antidepressant use. The website HelpGuide cannot be classified as simply promoting antidepressants, as the section on antidepressants also frequently refers to therapy, and emphasizes that a patient should carefully consider whether antidepressants are appropriate. While no box “is therapy right for you?” is provided to the side of the information on antidepressants to direct a reader to an alternative treatment, the website does make use of a disclaimer prior to presenting information on antidepressants: The bottom line If you have severe depression that’s interfering with your ability to function, medication may be right for you. However, many people use antidepressants when therapy, exercise, or self-help strategies would work just as well or better—minus the side effects. (Helpguide, n.p.) Here, the order of information requires that a reader decide early in the sequence whether medication “may be right” (my emphasis). Kress and Van Leuwen argue that for layout using the vertical axis, the top represents the ideal, with the bottom part indicative of reality. They state that “[t]he upper section tends to make some kind of emotive appeal and to show us ‘what might be’; the lower section tends to be more informative and ! ! ! 56! practical, showing us ‘what is’” (186). It is this question of “what might be” that is raised in the reader’s mind through the main relational clause “medication may be right for you.” Readers are further reminded to assess their options with the section “Deciding Whether Antidepressants are Right for You.” There are two ways to interpret this demand that readers assess treatment options and weigh the risks and benefits themselves. On one hand, HealthGuide’s use of the above disclaimer, and repeated references to the benefits of therapy, appears a persuasive device to discourage readers from requesting antidepressants. Yet, on the other, if a reader is able to answer “yes” to the benefits of antidepressants, perhaps he or she is more likely to preempt his or her doctor’s cautions, and receive medication. Indeed studies, such as Mintzes et al, report that 75% of patients who request a particular treatment are likely to receive it. Thus, by placing the power for this decision with the reader, this decision-making process only strengthens the idea that medication is an ideal option for certain readers, and makes it more likely that these readers will mention medication to— and receive medication from—their treating physicians.23 Therefore, as different patients respond differently to information that repeatedly asks them to question their preferences, it is difficult to determine whether the website is in favour of a particular treatment or not. Other websites from the United States that appear to favour therapy show a similar ambiguity. Although Ifred lists “counseling” first on its drop-down menu, the amount of information this section provides is minimal; it does not describe the benefits of counseling, instead referring a potential patient to a “find a therapist” tool.24 Psych !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 23 It is worth noting here that this information stands in sharp contrast to Canadian information, where a reader is encouraged to ask a doctor about the benefits, not decide “for themselves.” 24 Although the “Find a Therapist” tool shows an awareness of patients’ needs, by not providing information about how therapy can help the patient, the website is requiring that its users already know how ! ! ! 57! Central’s layout provides readers with easy access to information on antidepressants, despite also favouring therapy in the order of information. Unlike Ifred, Psych Central does provide information on the topic of therapy, and complements this with the “find a therapist” search box to the bottom right. Furthermore, the website also lists an “ask a therapist” hyperlink ahead of those on antidepressants. However, below this, the website has three links to antidepressants, which – to a reader scrolling down through information – has a more lasting presence on the page. Therefore, even when a reader is accessing information on psychotherapy, the preceding therapy links quickly disappear as a reader scrolls down, leaving the links to antidepressants available. Given the website’s explicit support of a dual treatment approach, these links appear especially effective: A treatment approach that focuses exclusively on one of these factors is not likely to be as beneficial as a treatment approach that addresses both psychological and biological aspects (through, for example, psychotherapy and medication). (Psych Central, n.p.) The bold font of the conjunction “and” emphasizes the website’s support for medication alongside therapy. The website appears to recognize the availability of the “antidepressants” hyperlink on the left side as the website only provides a further link to “psychotherapy,” not “medication.” In contrast, England’s websites present their preferences with less ambiguity. Links to redirect a reader’s attention, as seen on HelpGuide, are not present. Instead, the layout works to further enhance the treatment they prefer. The NHS Choices website, which orders therapy prior to antidepressants, presents a video on Talking Therapy next !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! therapy can assist them. Therefore, the user must already know that he or she wishes to pursue therapy, and why it could help before using the “Find a Therapist” tool. Describing counseling simply as “important,” as Ifred does, does not inform the reader as to why it might be “important.” ! ! ! 58! to its introduction on Treatment. Mind further persuades on the benefits of therapy with testimonials: (Mind, n.p.) Such testimonials are dotted throughout the page in the bold blue font. In contrast, medication testimonials are noticeably absent. The websites’ coupling of testimonials with the positive emotive language of personal experience, further reinforces the benefits of therapy over antidepressants. For English websites favouring antidepressants, the layout is also clear in persuading patients of their benefits. Depression Alliance lists antidepressants under the heading “treatments,” a label that does not direct readers to the subsequent “talking treatments” link, which is in miniscule font. (Netdoctor, n.p.) Similarly, the website Netdoctor also lists only medication under its heading “Treatments for Depression,” as seen below: Image B: Layout for Netdoctor ! ! ! 59! Here, therapy does not even feature in the website’s overview of depression, indicative of the websites’ preference towards pharmacological treatment. Furthermore, the heading “Depression Therapy on the NHS” hints at difficulties a patient may have when obtaining therapy on the NHS – a topic that dominates this section of the information. 5.5. Visual Images As the extract above demonstrates, part of the persuasion used by Netdoctor rests on visual images. Above, a healthy woman appears relaxed, smiling into the distance. By positioning this image next to information on medication, the website encourages readers to make a causal link between this image and the information on the page. This message is that medication enabled this person to recover. The images on the websites on depression predominately focus on people, who are portrayed in various emotional states. Kress and Van Leuwen emphasize that the way a viewer is encouraged to relate to images of people is deliberately constructed: “The relation between the human participants represented in images and the viewer is once again an imaginary relation. People are portrayed as though they are friends, or as though they are strangers” (original emphasis, 126). How closely the reader is asked to identify with these images is then a persuasive device, calling for different responses to both the pictures and the surrounding texts. This closeness is signaled both by the direction of the person’s gaze, and the proximity of the person in the frame to the viewer. To discuss the closeness conveyed by the direction of the gaze, I use Kress and Van Leeuwen’s categories of transactional relations and non-transactional relations. In a transactional relation, the people in the frame are looking at each other or a visible object, and in a non-transactional relation the person in the picture is gazing outside of the frame, ! ! ! 60! at something the audience cannot see (175). As Table M displays, these countries’ websites use gazes differently. Of the differences uncovered, the most striking is the high tendency of American websites to use non-transactional relations in over 75% of the images of people. Table M: Transactional and Non Transactional Images Percentage of Transactional Percentage of Non- transactional Canadian 56.7 43.3 American 24.1 75.9 English 39.5 60.5 By directing the gaze outwards, instead of within the internal frame of the picture, these websites use visual hailing, a visual form of synthetic personalization, which appears to directly target the viewer. Most images from the United States look directly outward, forcing a viewer to make eye contact with the person featured in the image. Communication expert, Paul Messaris comments on this device in advertising, stating that “[o]ne of the simplest examples of the former [drawing attention to the advertisement] occurs in advertising images in which someone looks directly at the spectator. This ubiquitous device […] draws its attention-getting power from our real-life tendency to look back when we are looked at” (4). Perhaps, in the United States, a society that permits Direct to Consumer Advertising, this technique is favoured because it reflects the norms of the medical genre. By contrast, other gaze directions found in these images show a person looking down, suggesting that the depicted depressed person is seeking to avoid scrutiny. By using images that directly gaze at the audience, the United States’ websites suggest that scrutiny is essential for recovery. As readers are already visiting these websites for information, I argue that the “tendency to look back” that Messaris identifies has less to do with “attracting attention” ! ! ! 61! and more to do with looking internally. When viewing these images, a reader is required to contemplate their own experience of depression: can they identify with the expression on these people’s faces? Have they seen it in a friend or relative? If so, they are persuaded to “look further” and pursue the information that follows the image. It is this image that persuades the reader of the relevance of the information. The level of empathy the viewer expresses with the image creates an environment in which they are acutely aware of the circumstances which lead them to the website, and therefore potentially more open to uncritically accept this information. Indeed, to requote Segal: “in the case of the anxious Web user, the emotion most rhetorically powerful is […] the emotion already in the user when she begins to click” (2009 363). These images require users who may be experiencing depression, or who have knowledge of the illness, to access these emotions to make sense of the images. The following images from NAMI demonstrate this. Both feature expressions that prompt a user to identify with the sad and despondent emotion that they subtly convey: Image C: Female Direct Gaze Image D: Male Direct Gaze (NAMI, “Depression” 1) (NAMI, “Depression” 3) These images also exemplify how such non-transactional relations eschew background ! ! ! 62! information; instead, the single blocks of colour behind the images ensure that the sad expression is the most salient feature. Viewers are therefore encouraged to identify with the image through accessing their own emotions, not by assessing the circumstances of the person featured. Transactional images, by contrast, often make use of surrounding environmental features to convey emotional distress or relief. Doing so allows these images to provide direct prompts to the cause of these emotions, and the possible treatment options available to a reader. Instead of relying on a strong emotional identification with the picture, these images use explicit signals to support their information and direct a reader to the relevant section. See the following images from Canadian Websites: Image E: Transactional Partner Image Image F: Transactional Pet Image In the first image, the partner is portrayed as supportive; both of those pictured use a transactional gaze to make eye contact. Furthermore, the surroundings feature a pet, and hint at exercise, which can play a positive role in recovery from depression. The second image also capitalizes on the importance of pets in helping with recovery. Yet the tag line “Jim talked to his doctor” suggests that medical help is the reason why he is able to look happier when playing with his dog. The image then directs Mood Disorders Society of Canada, “Depression” 20)! !(DepressionHurts, n.p.)! ! ! ! 63! a reader to information on discussing depression with one’s doctor. 25 The most frequent type of transactional image found on Canadian websites is the well-recognized media trope of a figure with their head in their hands. In Writing the Image, Yve Lomax states “[t]o speak figuratively and by the way of tropes, it is said that one speaks indirectly. Indeed, the trope implies and very quickly we assume something that isn’t stated directly” (184). Here, the indirect “speech” is the head in his hands: this implies despair. There are eleven images of a person despairing with their head in their hands on the Mood Disorders Society of Canada and DepressionHurts (the two Canadian websites that use images). Image G: Head in Hands Transactional Image (Mood Disorders Society of Canada 9). For images, like Image G, the gaze is still transactional in that it can be discerned and located within the image, but – as with the non-transactional gaze – the featured person’s internal focus emphasizes their mental state. These images require that the reader recognize the trope to signal despair. However – unlike with the non-transactional direct gaze – these images do not demand the same poignant contemplation of the experiencer; the viewer cannot make eye contact with the experiencer, and is not asked to identify on !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 25 DepressionHurts uses only sketched images of people. This provides an emotional distance from the problem of depression: by removing the situation from real life people, the website appears to persuade its visitors to view depression through a more objective lens. The prevalence of transactional images adds to this, as discussed. The television advertisements make use of both cartooned and filmed people. The cartoon images precede the website URL, linking the website with the advertisement. ! ! ! 64! the personal level demanded by the direct gaze. Table N also illuminates the Canadian trend to make fewer emotional demands on the viewer as a higher percentage of the images used are at a greater distance than those from Canadian or American websites. To explore how frame further affects this relationship between the viewer and the image, I have used Kress and Van Leeuwen’s system for classifying the level of zoom in images of people. Using Hall’s (1964) observations, Kress and Van Leeuwen classify images as intimate distance (face or head only), close personal distance (head and shoulders), far personal (waist up), close social (whole figure), far social (whole figure with space), public (torso of four plus people). Table N: Percentage of Frame Styles Percentage of Intimate Percentage of Close Personal Percentage of Far Personal Percentage of Far Social Percentage of Public Distance Percentage of No Face Canadian 9.4 40.6 28.1 15.6 3.1 3.1 American 24.2 48.5 15.2 0 6.1 6.1 English 25 43.2 11.4 4.5 6.8 9.1 Table N illustrates that Canadian images are rarely at the intimate level. By contrast, intimacy constitutes a quarter of the American and English images. The distance in images from Canada therefore contains a higher proportion of the categories of far personal and far social images. These images make use of the environment within the picture to explain depression and recovery (see Image Five, where the partner aids recovery). 5.6. Section Summary Images from these websites can be summarized as follows: Canadian websites favour transactional relations, and England and the United States’ websites favour non- transactional relations. The images on Canada’s websites make fewer demands on their ! ! ! 65! audience. The United States’ images make the highest demands, calling for a higher level of emotional identification. The amount and order of information require a more complex summary. From word count information, these websites highlight that the websites from the United States provide more information on antidepressants than those from England or Canada. Canadian websites provide a higher percentage of information on therapy than both of these countries. However, two of the five Canadian websites do not provide any treatment information at all, indicative of a higher rate of deferral towards patients’ physicians – an argument that may explain why Canada’s images make fewer emotional demands on the viewer. However, with regard to the order of information, Canadian websites appear strongly biased towards antidepressants, positing medications as the treatment which patients are more likely to discuss with their doctors. 6. Conclusion 6.1. Overview and significant Trends As expected, antidepressant use was emphasized by pharmaceutically sponsored websites from all three countries. Websites from England tend to emphasize the high frequency of depression, and focus on relieving the patient of personal responsibility for the illness. The frequent collocation of “common” and “depression,” and the prominence of statements about the illnesses frequency demonstrate this. Yet, England’s websites also provide the highest rate of direct address, indicative of a higher degree of choice for the patient. This appears to reflect the NHS’ agenda for a collaborative modal of care (NHS 2012). Such direct address implicitly seeks to persuade a patient to collaborate and confide in healthcare professionals. Therefore patients will perhaps choose talking ! ! ! 66! treatments over, or in combination with, medication. Of the three countries, England’s websites are the only ones to imply that therapy alone is an adequate treatment. Although England’s National Health Service necessarily restricts patient choice, the language on these websites suggests that potential patients have more choices than seen on Canadian or American websites. Given that the NHS constitutional emphasis on patient involvement is a relatively recent development, the emphasis in language may be due to the healthcare organization’s awareness of a need to shift patients’ perception of their influence. Indeed, Judy Segal’s argument that the United Kingdom’s shift to a system based on “concordance” rather than patient compliance that “in fact, harbors an ideology of compliance” (2007 81); it only the language framing healthcare for the patient that has changed. By portraying talking therapy as a more useful approach than seen in the United States’ and Canadian literature, England’s websites make it more likely that patients will choose – and so comply with – the option of counseling as treatment. Therefore, a concordance model of care does not appear more empowering than a model based on compliance. When examining Canada’s websites, I found, most significantly, a lack of information on treatment. Unlike the websites from England and the United States, Canadian websites tend to imply that treatment information is something that should only be obtained by a practicing physician or other expert. For example, far more references are made to “your doctor,” in the second person possessive on Canadian websites. Such texts therefore show a heightened wariness about the appropriate sourcing of information and echo bioethicists’ concerns over “[t]he increasingly active role that lay people are taking – and indeed, are asked by our culture to take [… and the] empowering ! ! ! 67! possibilities and the danger for distortion and peer pressure” (Kulka 34). These concerns are further emphasized by the websites’ vocabulary. Instead of simply describing the effects of a treatment with the verbs “to help” or “to treat,” Canadian websites tend to cite evidence for these effects, using phrases such as “extensively researched” or “research suggests” throughout the information that they do provide. Furthermore, even the images on Canadian websites make fewer demands on their audience, relying on more transactional relations, where the person featured in a picture is rarely looking outside of the frame to the reader. Through referencing internal parts of the picture and not relying on a reader’s interpretation of the cause for a facial expression, Canadian websites once again emphasize the need for readers to rely on the expert, and information that they are provided with over their own beliefs. Canadian websites do, however, reveal a bias towards antidepressants when both therapy and antidepressants are addressed together. This is perhaps due to the increased amount of research on the effectiveness of drugs worldwide. By emphasizing medication, these websites may affect the treatments for depression that Canadians are persuaded to choose. As psychologists are not covered by basic health insurance in Canada, treatment (prescription) by GP’s or psychiatrists may be most accessible. Still, the Canadian Psychological Association is prominent on a Google.ca search. Information from the United States’ websites, like information on Canadian websites, also appears to emphasize the benefits of treatment via antidepressants. Prior to this research, I hypothesized that this bias would be evident in all of my analysis. I expected the United States’ websites to consistently emphasize medication before discussing therapy, given the increased influence of third parties and pharmaceutical ! ! ! 68! companies but this is not the case, and the order in which therapy and medication are presented vary. Nonetheless, the United States’ websites do use tactics to persuade readers of the merits of medication: “common” frequently collocates with medication, about which there is more information. Furthermore, the United States’ websites stress the personal responsibility of the depressed patient to seek help by addressing those who may be caring for someone with depression. Such tactics emphasize the guilt a depressed patient may experience. Unlike England’s information, which seeks to remove this guilt, the United States’ websites appear to play on guilt to persuade a reader to choose antidepressants. 6.2. Why should we be especially mindful of patient information on depression? Finally, I wish to comment on why these language differences are especially significant for depression. Why do differences in emphases on treatments matter? I have revealed the importance of persuasion on the patient reader, who is encouraged to view these websites as a source of information, not persuasion. Yet, the significance of such differences go beyond websites that subtly encourage patients to support a treatment in line with the preferences of their healthcare system. What I wish to stress is this: that these websites’ emphases on particular treatments are especially relevant for patients with depression because, as there is no universal treatment plan for depression, the decision to medicate or not is largely left to the patient (Dohrenwent 15). Thus, while the effect of such websites on patient decision-making cannot be determined through this thesis’ comparison, the potential effects of these differences should not be underestimated. Websites that persuade potential patients to choose medication may well increase the rate at which physicians prescribe antidepressants. Indeed, Conrad suggests that the role of ! ! ! 69! physicians in treatment decisions is changing due to medicalization: “[d]octors are still the gatekeepers for medical treatment but their role has become more subordinate in the expansion or contraction of medicalization” (142). All of these healthcare systems emphasize that patient preferences need to be strongly considered by the treating physician: England’s NHS stresses patient choice in its policy; the United States’ system is built around patient choice; and the Canadian system is a blend of these two health services. These websites present what Annemarie Mol would label “a logic of choice” system where no choice is prescribed as the “right” one and the patients’ rights to choose their treatment can outweigh the physicians’ initial treatment plan as noted by!Mintzes et al. Therefore, if these websites are providing what appears unbiased information and persuading patients to request different treatments, the effect of patient information websites could be dramatic. The title of this project asks: are these Anglo-American Western countries’ “worlds apart” in their information? Yes, this thesis argues – because subtle linguistic differences can make a world of difference. ! ! ! 70! Works Cited “Advertising for Medicines.” Medication and Healthcare Regularity Agency. MHRA, 2013. Web. 8 May 2013. “Aims and Objectives.” British Pharmacological Society. British Pharmacological Society, 2013. Web. 15 May 2013. Althusser, Louis. “Ideology and Ideological State Apparatuses (Notes Towards an Investigation).” Anthropology of The State: A Reader. Ed. Aradhana Sharma and Akhil Gupta. Oxford: Blackwell Publishing, 2006: 86-112. Print. From “Ideology and Ideological State Apparatuses (Notes Towards an Investigation).” Lenin and Philosophy and Other Essays. Trans. B. Brewster. New York and London: Monthly Review Press, 1971: 127-186. Print. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4th ed., text re. Washington: American Psychiatric Association, 2000. Print. Asp, D. Elissa, and Jessica de Villiers. When Language Breaks Down. Analysing Discourse in Clinical Contexts. Cambridge: Cambridge UP, 2010. Print. Berkenkotter, Carol. “Genre Systems at Work: DSM-IV and Rhetorical Recontextualization in Psychotherapy Paperwork.” Written Communication 18.3 (2001): 326–349. Web. 4 Oct. 2012. ---. “A Case for Historical “Wide-Angle” Genre Analysis: A Personal Retrospective.” Ibérica: Journal of the European Association of Language for Specific Purposes 18 (2009): 9-21. Print. Blendon et al. “Inequities In Health Care” A Five-Country Survey.” Health Affairs 21.3 (2002):182-191. Web. 14 May 2013. Clarke, Juanne Nancarrow. Health, Illness and Medicine in Canada. Toronto: Oxford UP, 1996. Print. Conrad, Peter. The Medicalization of Society: On the Transformation of Human Conditions into Treatable Disorders. Baltimore: John Hopkins, 2007. Print. Coulter, Angela. “Evidence based patient information.” BMJ 317.7153, 1998: 225-226. Web. 8 May 2013. Donoghue, John. “Depression.” Pharmacoeconomics in Psychiatry. Eds. Taylor, David. Martin Knapp and Robert Kerwin. London: Martin Duntz, 2002: 43-55. Print. Dohrenwent, Anne. “11 Patient Misconceptions About Antidepressants: How to Respond.” Fam Pract Manag. (May – June 2009): 15-17. Web. 18th Jan. 2013. ! ! ! 71! Downie, Robin, and Fiona Randall. “Choice and Responsibility in the NHS.” Clinical medicine (London,UK) 8.2 (2008): 182–5. Print. Dwight-Johnson et al. “Can Quality Improvement Programs for Depression in Primary Care Address Patient Preferences for Treatment?” Medical Care 39.9 (Sep. 2011): 934-944. Print. Fairclough, Norman. Language and Power. London: Longman, 1989. Print. First, Michael et al. The Essential Companion to the Diagnosis and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision. DSM-IV-TR Guidebook. Washington: American Psychiatric Publishing, 2004. Print. Gregory, Michael. \"Phasal analysis within communication linguistics: two contrastive discourses.\" Relations and Functions within and around Language. Ed. Peter H. Fries, Michael Cummings, David Lockwood and William Spruiell. London: Continuuim, 2002. 13-31. Print. Giménez-Pérez et al. “Dissemination of ‘patient-oriented evidence that matters’ on the Internet: the case of Type 2 diabetes treatment.” Diabetic Medicine 22 (2005): 688-692. Web. May 15 2013. Gusfield, Joseph. “The Literary Rhetoric of Science: Comedy and Pathos in Drinking Driver Research.” American Sociological Review 41.1, Feb. 1976: 16-34. Web. 8 May 2013. Halliday, Michael and Christian Matthiessen. An Introduction to Functional Grammar. London: Arnold, 2004. Print. Healy, David. Let Them Eat Prozac: The Unhealthy Relationship between the Pharmaceutical Industry and Depression. New York: New York UP, 2004. ---. “Psychiatry and the Pharmaceutical Industry - in 100 Words.” The British Journal of Psychiatry 194.1 (2008): 85–85. Web. 27 Aug. 2012. Jeffries, Leslie. Critical Stylistics: The Power of English. London: Palgrave Macmillan, 2010. Print. Kress, Gunther; Leeuwen, Theo Van. Reading Images: The Grammar of Visual Design. Taylor & Francis, 2006. Print. Kukla, Rebecca. “How Do Patients Know?” Hastings Center Report 37.5 (2007): 27-35. Print. Lane, C. “How Shyness Became an Illness: A Brief History of Social Phobia.” Common Knowledge 12.3 (2006): 388–409. Web. 12 Oct. 2012. ! ! ! 72! Lauerbach, Gerda Eve. “Emphasis.” Pragmatics in Practice. Jan-Ola Ostman and Jef Vershueren (eds). Amsterdam: John Benjamins, 2011.130-148 Lave, Judith R. et al. “Cost-effectiveness of Treatments for Major Depression in Primary Care Practice.” Archives of general psychiatry 55.7 (1998): 645–51. Lewis et al. “The Future of Health Care in Canada.” BMJ 323.7318 (October 2001): 926-929. Web. 15 May 2013. Lomax, Yve. Writing The Image: An Adventure with Art and Theory. London: I.B. Tauris, 2000. Print. McCarthy, Lucille Parkinson, and Joan Page Gerring. \"Revising psychiatry's charter document DSM-IV.\" Written Communication 11.2 (1994): 147-192. Messaris, Paul. Visual Persuasion The Role of Images in Advertising. Thousand Oaks, CA: Sage, 1997. Mintzes, Barbara et al. “How does direct-to-consumer advertising (DTCA) affect prescribing? A survey in primary care environments with and without legal DTCA.” CMAJ 169.5 (2003): 405-412. Web. 15 Oct. 2012. Mintzes, Barbara. “What are the Public Health Implications? Direct-to-Consumer Advertising of Prescription Drugs in Canada.” Health Council of Canada, 2006. Web. 8 May 2013. Mol, Annemarie. The Logic of Care. Health and The Problem of Patient Choice. London: Routledge, 2008. Print. Moody, K. “Capitalist Care: Will the Coalition Government’s ‘Reforms’ Move the NHS Further Toward a US-style Healthcare Market?” Capital & Class 35.3 (2011): 415–434. Web. 17/08/2012. National Health Service. “The NHS Constitution: The NHS Belongs to Us All.” National Health Service, 2012. Web. 24 Sep. 2012. O’Halloran, Kay L. Multimodal Discourse Analysis. Systemic Functional Perspectives. London: Continuum, 2004. Print. Pennycook, Alistair. “The Politics of Pronouns.” ELT 48.2. April 1994: 173-178. Print. RxIsk. N.p., 2013. Web. 15 May 2013. Segal, Judy Z. “‘Compliance’ to ‘Concordance’: A Critical View.” Journal of Medical Humanities 28 (2007): 81-96. ! ! ! 73! ---.“Internet Health and the 21st-Century Patient: A Rhetorical View.” Written Communication 26 (2009): 351-369. ---. “What, in Addition to Drugs, Do Pharmaceutical Ads Sell? The Rhetoric of Pleasure in Direct-to-Consumer Advertising for Prescription Pharmaceuticals.” In Deborah Dysart-Gale and Joan Leach, eds., Rhetorical Questions of Health and Medicine. Lanham, MD: Lexington Press, 2011. 9-32. Silence, Elizabeth et al. “Trust and Mistrust of Online Health sites.” CIHT 6.1., 2004: 663-670. Web. 8 May 2013. Watters, Ethan. “The Americanization of Mental Illness.” The New York Times Magazine. Jan 10 2010: 40. Web. 10 Sept. 2012. Watters, Ethan. The Globalization of the American Psyche. Crazy Like Us. New York: Free Press, 2010. Print. Wodak, Ruth. Disorders of Discourse. London: Longman, 1996. Print. Wolpert, Lewis. Malignant Sadness. The Anatomy of Depression. London: Faber and Faber, 1999. Data “About CHMA.” Canadian Mental Health Association. N.p., n.d. Web 15 May 2013. “About CPA.” Canadian Psychological Association. N.p., 2013. Web. May 15 2013. “About Depression Alliance.” Depression Alliance. N.p., n.d. Web. 15 May 2013. “About NAMI.” NAMI. N.p., n.d. Web. 15 May 2013. “About Psych Central.” Psych Central. N.p., March 2013. Web. 15 May 2013. “About Us.” Mood Disorders Society of Canada. N.p., 2011. Web. 15 May 2013. “About Us.” Patient.co.Uk. N.p., March 2013. Web. 15 May 2013. “About the Agency.” Public Health Agency of Canada. N.p., December 2012. Web. 15 May 2013. ! ! ! 74! “About the National Library of Medicine.” US National Library of Medicine (Pub Med Health). N.p., n.d. Web. 15 May 2013. “Advisory Board.” Ifred. N.p., 2013. Web. 15 May 2013. “About Netdoctor.co.Uk.” Netdoctor. Netdoctor.co.Uk, 2012. Web. 09/11/2012. “Clinical Depression.” NHS Choices. N.p., 2012. Web. 09/11/2012. “Clinical Depression - Diagnosis. NHS Choices. N.p., 2012. Web. 09/11/2012. “Clinical Depression – Introduction.” NHS Choices. N.p., 2012. Web. 09/11/2012. “Clinical Depression – Living With Clinical Depression.” NHS Choices. N.p., 2012. Web. 09/11/2012. “Clinical Depression – Symptoms.” NHS Choices. N.p., 2012. Web. 09/11/2012. “Clinical Depression – Treating Clinical Depression.” NHS Choices. N.p., 2012. Web. 09/11/2012. “Corporate Support.” Mood Disorders Society of Canada. N.p., 2011. Web. 15 May 2013. “Counselling.” Ifred. N.p., 2012. Web. 09/11/2012. “Depression.” (A) Canadian Alliance on Mental Illness and Mental Health. CAMIMH 2012. Web. 09/11/2012. “Depression.” (B) Canadian Mental Health Association. Canadian Mental Health Association, 2012. Web. 09/11/2012. “Depression.” Mood Disorders Society of Canada. Mood Disorders Society of Canada, n.d. Web. 09/11/2012. “Depression.” Patient.co.Uk. Egton Medical Services, n.d. Web. 09/11/2012. “Depression Alliance – Help and Information About Depression.” Depression Alliance. N.p., n.d. Web. 09/11/2012. “Depression Facts.” Netdoctor. Netdoctor.co.Uk, 2012. Web. 09/11/2012. “Depression. Psychology Works Factsheet: Depression.” Canadian Psychological Association. Canadian Psychological Association, 2009. Web. 09/11/2012. “Depression Treatment.” Psych Central. N.p., November 2012. Web. 09/11/2012. ! ! ! 75! “Depression Help Center.” Helpguide.org. N.p., n.d. Web. 09/11/2012. “Depression Treatment. Therapy, Medication and Lifestyle Changes that can help” Helpguide.org. N.p., n.d. Web. 09/11/2012. “Doctoring Depression – talking to your GP.” John Pillinger (ed.). Netdoctor. Netdoctor.co.Uk, 2011. Web. 09/11/2012. Duckworth, Ken. “Depression. Find Help. Find Hope.” National Alliance on Mental Illness. NAMI: 2012. Web. 09/11/2012. “Editorial Policy.” Mind. N.p., n.d. Web. 15 May 2013. “Editorial Policy.” NHS Choices. NHS, 2011. 15 May 2013. “Editorial Policy.” Patient.co.Uk. N.p., January 2012. Web. 15 May 2013. “Governance.” NAMI. N.p., n.d. Web. 15 May 2013. Grohol, John. “An Introduction to Depression.” Psych Central. N.p., 2006. Web. 09/11/2012. ---. “Depression.” Psych Central, N.P., Jan. 2012. Web. 09/11/2012. ---. “The Causes of Depression.” Psych Central. N.p., 2006. Web. 09/11/2012. ---. “Where to Get Help for Depression.” Psych Central. N.p., 2006. Web. 09/11/2012. “Help and Information.” Depression Alliance. N.p., n.d. Web. 09/11/2012. “HelpGuide’s Story.” HelpGuide. N.p., n.d. Web. 15 May 2013. “‘Just the Facts.’ Mental Illness and Addiction in Canada.” Mood Disorders Society of Canada. N.p., November 2009. Web. 15 May 2013. “Major Depression.” Fred K. Berger, La Jolla and David Zieve (eds). U.S. National Library of Medicine. A.D.A.M., 2012. Web. 09/11/2012. “Medication Library.” Psych Central. N.p., 2012. Web. May 15 2013. “Medications.” Ifred. N.p., 2012. Web. 09/11/2012. “Mood Disorders.” Mood Disorders Society of Canada. Mood Disorders Society of Canada, n.d. Web. 09/11/2012. ! ! ! 76! Mulholland, Ciaran. “Depression.” Netdoctor. Netdoctor.co.Uk, 2010. Web. 09/11/2012. ---. “Depression Therapy and the NHS.” Netdoctor. Netdoctor.co.Uk, 2010. Web. 09/11/2012. “Privacy Statement.” DepressionHurts. N.p., n.d. Web. 15 May 2013. Stewart, George. “Understanding Depression.” Mind.org. 2nd Edition. Mind, 2010. Web. 09/11/2012. “Symptoms.” Ifred. N.p., 2012. Web. 09/11/2012. “Symptoms, Causes and Diagnosis.” National Alliance on Mental Illness. NAMI: n.d. Web. 09/11/2012. “Terms of Use.” DepressionHurts. N.p., n.d. Web. 15 May 2013 “Treatment, Services and Support.” National Alliance on Mental Illness. NAMI: n.d. Web. 09/11/2012. “Treatments for Depression.” Depression Alliance. N.p., n.d. Web. 09/11/2012. “Treatments for Depression.” Netdoctor.co.Uk. N.p., n.d. Web. May 15 2013. “Understanding Depression. Depression Symptoms.” DepressionHurts. DepressionHurts, 2012. Web. 09/11/2012. “Understanding Depression. Frequently Asked Questions.” DepressionHurts. DepressionHurts, 2012. Web. 09/11/2012. “Understanding Depression. Important Points to Consider.” DepressionHurts. DepressionHurts, 2012. Web. 09/11/2012. “Understanding Depression: Signs, Symptoms, Causes and Help.” Helpguide.org. N.p., n.d. Web. 09/11/2012. “Understanding Depression. What is Depression?” DepressionHurts. DepressionHurts, 2012. Web. 09/11/2012. Webber, Christine. “Depression – how it affects sex and relationships.” Netdoctor. Netdoctor.co.Uk, 2010. Web. 09/11/2012. “Welcome.” Mood Disorders Society of Canada. N.p., n.d. Web. 09/11/2012. “What can I do to Help Myself if I Feel Depressed?” Public Health Agency of Canada. Public Health Agency of Canada, 2012. Web. 09/11/2012. ! ! ! 77! “What is Depression?” Depression Alliance. N.p., n.d. Web. 09/11/2012. “What is Depression?” Mood Disorders Society of Canada. Mood Disorders Society of Canada, 2009. Web. 09/11/2012. “What is Depression?” National Alliance on Mental Illness. NAMI: n.d. Web. 09/11/2012. “What is Depression?” Public Health Agency of Canada. Public Health Agency of Canada, 2009. Web. 09/11/2012. “Who We Are.” Ifred. N.p., 2013. Web. 15 May 2013. !"@en ; edm:hasType "Thesis/Dissertation"@en ; vivo:dateIssued "2013-11"@en ; edm:isShownAt "10.14288/1.0073879"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "English"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "Attribution 3.0 Unported"@en ; ns0:rightsURI "http://creativecommons.org/licenses/by/3.0/"@en ; ns0:scholarLevel "Graduate"@en ; dcterms:title "\"Western world's apart?\" A comparison of patient information websites on depression in Canada, the United States and England"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/44540"@en .