Ghp348135

Marginal voices in the media coverage of controversial health interventions: how
do they contribute to the public understanding of science?
M. Hivon, P. Lehoux, J.-L. Denis and M. Rock 2010 19: 34 originally published online 21 January 2009 The online version of this article can be found at: can be found at:
Public Understanding of Science
Additional services and information for
SAGE PUBLICATIONS (www.sagepublications.com) Public Understand. Sci. 19(1) (2010) 34–51
Marginal voices in the media coverage of
controversial health interventions: how do they
contribute to the public understanding of science?
M. Hivon, P. Lehoux, J.-L. Denis and M. Rock
While the media are a significant source of information for the public onscience and technology, journalists are often accused of providing only a par-tial picture by neglecting the points of view of vulnerable stakeholders. Thispaper analyzes the press coverage of four controversial health interventions inorder to uncover what voices are treated marginally in the media and what therelative contributions of these voices are to the stories being told. Our empiri-cal study shows that: 1) patterns of source utilization vary depending on thehealth intervention and less dominant stakeholders are in fact represented;and 2) the use of marginal voices fills certain information gaps but the overallcontribution of such voices to the controversies remains limited. In order tostrengthen the media coverage of science and technology issues, we suggestthat further research on journalistic practices: 1) move beyond the dichotomybetween journalists and scientists, and 2) explore how different categories ofreaders appraise the meaning and relevance of media content.
Keywords:
media sources, media and sciences, marginal voices, knowledge 1. Introduction
Health technologies affect our lives in many ways. “We pay for their implementation and beartheir social costs. Public understanding of their social implications, their technical justifica-tions and their political and economic foundations is in the interest of an informed andinvolved citizenry” (Nelkin, 1987: 172). Yet, to participate in social and political debate ontechnologies and to think critically about decisions affecting their lives, citizens need to beinformed. In this respect, the media represent a significant source of information for lay audi-ences. Hence it is important that they provide their audiences with comprehensive and accu-rate information on issues related to science and technology.
It is often argued that media culture and organizational constraints prevent journalists from providing comprehensive coverage (Nelkin, 1987; Friedman et al., 1999; Seale, 2003). Attimes, journalists are accused of confusing the audience by giving too much weight to maver-ick ideas (Friedman et al., 1999; Weigold, 2001; May, 2005). At other times, they are accused ISSN 0963-6625 DOI: 10.1177/0963662508088668 Hivon et al.: Marginal voices in the media of neglecting the voices of people who could potentially bring much-needed nuance to thedebate (Conrad, 1999; Williams et al., 2003; Bubela and Caufield, 2004). Such conflictinginterpretations may stem from the fact that most empirical studies of media coverage havefocused on a single case, preventing scholars from understanding how certain scientific, socialand political dimensions may affect the scope of voices being mobilized by journalists. By con-trast, this paper addresses a gap in the literature by emphasizing the spectrum of voices—including those seen as marginal—represented in the media coverage of four controversialhealth innovations, and by examining the respective contributions of these voices to the storiesbeing presented to readers. Using a mixed method analysis (quantitative and qualitative), weexamined a corpus of articles published in Canadian newspapers between 2000 and 2006 ontwo controversial therapeutic interventions—electroconvulsive therapy (ECT) and cyclo-oxygenase-2 drugs (COX-2s)—and two contentious screening tests—first-trimester prenatalscreening (PST) for Down syndrome and prostate-specific antigen (PSA) screening.
We first define the concept of “marginal voice” and summarize the issues raised by how such voices are treated in the media. We then present the marginal voices that were includedin our four empirical cases. The last section discusses the extent to which marginal voicescontribute to the understanding of the controversies. In order to strengthen the media cover-age of science and technology issues, we suggest that further research on journalistic prac-tices: 1) move beyond what we call the journalism/science dichotomy, and 2) explore howdifferent media coverage strategies may modify the understanding of, and judgments aboutscience and technology issues of various categories of readers.
The fact that often only a partial picture of a controversy is presented in the media has beenexplained in terms of media culture and organizational constraints. Journalists are faced with orga-nizational constraints such as the limited time and availability of experts on whom they rely forbackground information and clarification (Nelkin, 1987; Einsiedel, 1992; Conrad, 1999; Stocking,1999; Weigold, 2001; Waddell et al., 2005). Furthermore, following the principle of objectivitydear to their culture, they try to avoid taking a position when controversies arise (Dunwoody,1999). Consequently, journalists often strive to provide readers with a balanced story, one in whichextreme and often opposite points of view are presented and given equal weight, without provid-ing any hint as to how representative of the larger community these viewpoints really are (Nelkin,1987; Conrad, 1999; Dunwoody, 1999; Rowan, 1999; Stocking, 1999; Weigold, 2001; Anderson,2002; Seale, 2003; May, 2005). By emphasizing the “pros” and “cons” of technological innova-tions or scientific findings, this dyadic framing ultimately “contributes to the exclusion of morenuanced debate” (Williams et al., 2003: 810). The points of view of certain stakeholders areignored or marginalized. Such apparently balanced treatments may also leave readers with theimpression that there exists considerable difference of opinion on a topic, while in fact there maybe broad consensus. As a result, readers may be misinformed, misled or confused.
Scientific journals and scientific experts constitute, by in large, the main source of infor- mation for health and science journalists (Nelkin, 1987; Einsiedel, 1992; Conrad, 1999;Weigold, 2001; Nisbet et al., 2003). In general, these are followed by government and indus-try (Einsiedel, 1992; Nisbet et al., 2003). By contrast, patients and their advocates are oftencited as being underrepresented. Conrad (1999) underlines their quasi-absence in the mediacoverage of genetics and behavior, concluding: Since news presentations influence how people conceptualize and evaluate new findings… and as the media play a significant role in agenda setting, it is politically important tobring these voices to the fore. Their articulation, even as quotes, adds important balance to the reporting and introduces neglected viewpoints into the public discourse. Untilthese voices are routinely consulted and quoted, journalists will not have gotten thewhole story. (Conrad, 1999: 301) Williams et al. likewise point out the absence of feminist perspectives and women’s voices in themedia reporting of stem cell research despite the role of women “in producing eggs that mightbe used in stem cell research” (2003: 807). They argue that a greater diversity of viewpoints needsto be included to ensure that an “inclusive democratic debate” takes place (2003: 810). Similarly,Collins et al. (2006) highlight the minimal representation of associations of health care profes-sionals in their study of the media coverage of Canadian health care reform debate.
For the purposes of this paper, we define marginal voices as categories of sources that are given limited or no space and voice in media coverage and that therefore remain mar-ginal or peripheral to the stories being told. Sources are individuals whom journalists con-tact for background information, clarification or comments. We should note that for thepurpose of our analysis, the views expressed by these marginal voices are not necessarily mar-ginal in society. It is the limited treatment they receive in media coverage that makes themmarginal. Through our empirical analysis, we wish to examine the extent to which certainviewpoints have been brought to the readers’ attention as being relevant to understanding andforming a judgment about four different controversial technologies that have generated sci-entific controversy, public controversy, or both.
2. Multiple case study design
In our previous work, we examined the way six Canadian Health Technology Assessment(HTA) agencies interacted with professional and lay stakeholders (Lehoux et al., 2005). Thiswork highlighted the need to better understand how medical innovations that generate a sig-nificant level of scientific and social controversy attract (or do not attract) the attention ofthe media. Did journalists actually use the scientific literature reviews produced by HTAagencies? Did they examine the ethical, social, economic and political issues related to themedical interventions that HTA agencies addressed in their reports? This led us to focus ontwo Canadian agencies (in Ontario and Quebec) and, as a starting point, we selected fourreports that were moderately or significantly controversial from among their recent publica-tions: 1) The Use of Electroconvulsive Therapy in Québec (Banken, 2002); 2) First-trimesterPrenatal Screening for Down Syndrome and Other Aneuploidies (Framarin, 2003); 3)Prostate-specific Antigen (PSA) Screening in Asymptomatic Men (Slaughter et al., 2002);and 4) What Effects do Provincial Drug Plan Coverage Policies for New Drugs Have onPatterns of Use and Cost? (Paterson et al., 2003). The drugs concerned in the last report arecyclo-oxygenase-2 drugs (COX-2s), which are anti-inflammatories used in the treatment ofarthritis and chronic pain.1 We used four Canadian electronic databases of newspaper and magazine articles (Bibliobranché, Repère, CPI and CBCA) to build our sample. We selected articles related to the inter-vention discussed in each HTA report, using keywords such as electroshock (used solely as atherapy), prenatal screening tests, Down syndrome, PSA, prostate cancer, arthritis and COX-2.
The intervention had to be mentioned either in the title or in the text. We looked at articles from Hivon et al.: Marginal voices in the media 2000 to 2006, the period during which the four HTA reports were published. We selectedarticles from both English- and French-language Canadian newspapers and magazines aimed atthe lay audience. We excluded all newspapers and magazines targeting professional audiences,such as the Canadian Medical Association Journal, Médecin du Québec and Medical Post.
Since we wanted to examine the media space our chosen topics occupied, we included dupli-cates (that is, identical articles published concurrently). This method yielded a large spectrumof newspaper articles, including a few articles in which the main focus was not necessarily thehealth intervention per se.2 Overall, our sample comprised 23 articles on prenatal screening testsfor Down syndrome, 24 on electroshock, 139 on PSA screening, and 224 on COX-2s.
The analysis was performed by the first author (MH) and discussed by the whole team on sev-eral occasions. We created one Excel document for each intervention. We developed and pre-tested an analytical extraction form based on a literature review and discussions with colleagueswho had done similar work (Davidson et al., 2003; Kroll-Smith, 2003; Prior, 2003). Articleswere read in full two or three times and their content indexed according to several pre-definedcodes. We first compiled basic information about the sources (e.g., name, title). We then cate-gorized the sources as: 1) scientific expertise (researchers, scientists, studies and reports, scien-tific journals); 2) clinical expertise (associations of health care professionals and professionals);3) lay expertise (associations of patients, patients and their families); 4) management expertise(health care administrations, governmental institutions); 5) industry expertise (industry repre-sentatives and scientists); or 6) other (analysts, lawyers, etc.). These categories were definedempirically as they appeared to encompass all of the cited sources and permitted comparisonsacross the four technologies. They are also very similar to those used in other studies on mediasources (Einsiedel, 1992; Conrad, 1999). Finally, we transcribed comments made by the sourcesinto our Excel file in order to substantiate our findings.
By performing simple descriptive statistics on the source categories used in the articles, we were able to identify which voices were treated marginally by the media for each contro-versial intervention (Table 1). Then, we created comparative tables to assist with a qualitativeanalysis of the sources’ position toward the interventions, across source categories and acrossthe four medical interventions.
We present our results separately for each health intervention. First, we summarize the main storyline of the media coverage for the period 2000–6. Then, we identify which sourceswere treated marginally and examine their views.
3. The media coverage of electroconvulsive therapy (ECT)
Electroconvulsive therapy involves applying brief electrical pulses to the patient’s brain. It isperformed under general anesthesia and requires the administration of a muscle relaxant, oxy-genation of the patient and constant monitoring (Banken, 2002). Our corpus included 9 arti-cles on the subject of lawsuits dating back to use of ECT in the 1950s. The remaining 15articles covered new studies on ECT or discussed the pros and cons of this intervention.
In general, these articles presented ECT as an effective technology, but one that must be usedwith caution (6 articles). ECT was reported to be effective in treating certain mental illnesses, Table 1. Breakdown of source categories cited in the newspaper articles by intervention
in certain individuals, and in cases where medication has failed (4).3 However, it was associ-ated with serious side effects, including various cognitive troubles and memory loss (8).
Seven articles reported that a great deal of uncertainty surrounding ECT remains because noone is able to explain exactly how or why it works on some patients and not on others. It wasalso emphasized that the administration of ECT has improved significantly in the past thirtyyears and that its side effects are now limited (7). Some articles mentioned the need for bettertraining of future psychiatrists and the implementation of quality control programs (4).
Finally, 9 articles mentioned one woman’s lifelong battle for compensation for having under-gone ECT in the 1950s at a Montreal hospital.
Scientific expertise constitutes the most often cited source in the coverage of ECT (31 per-cent). Interestingly, this is closely followed by lay expertise (26 percent) and clinical exper-tise (19 percent). Administrators and industry are almost absent from the coverage,representing only 13 percent and 1 percent, respectively, of all sources cited in the articles.
With respect to industry, we learn from a scholar that the increase of ECT in the province of Quebec may be linked to the pharmaceutical lobby. This assertion, however, is not furtherdeveloped (Capital Santé, 10 September 2003). With respect to administrators, we learn thatthe Quebec government requested an HTA report on ECT following an increase in its use inthe province, suggesting there was cause for concern, but nowhere was it explained why (2).
Health care professionals (19 percent) were mainly cited by journalists to convey the pos- itive and clinical aspects of ECT. According to the remaining clinicians cited, ECT is the besttechnology currently available for treating depression, “eliminating the symptoms in 50% to90% of the cases” (The Globe and Mail, 1 March 2003). They also recognized the uncertain-ties surrounding ECT, but argued there is not enough evidence to conclude that the interven-tion causes brain damage or is responsible for memory loss (1). Some minimized theimportance of side effects, arguing there are consequences to every treatment (5). Others min-imized the perceived violence surrounding this treatment (2) and accused opponents of pre-venting them from treating more patients who would benefit from the technology (4). Forexample, one article noted: “Dr. Vanier claims that if it were not for ‘the big fuss made byhuman-rights groups’ and the taboo surrounding ECT, he would be able to save many moredepressed people” (our translation from the French (TFF), Nouvelliste, 10 December 2001).
Hivon et al.: Marginal voices in the media Patients and their advocates were well-represented in ECT coverage (26 percent).
Journalists cited them mainly to emphasize the negative side effects of ECT. Their main argu-ments were that ECT is inhuman, dangerous and degrading (2). They emphasized the fact thatno one is able to explain why ECT works and why only for some people, and insisted that theside effects outweigh the benefits. They suggested that patients are not always in a positionto give their informed consent; in effect, some are being forced to accept and undergo thetreatment (4). “Little information is given to people and we wonder whether the patient reallyhas a choice,” denounces Doris Provencher, the coordinator of a Quebec advocacy group forthe mentally ill (Le Soleil, 5 March 2003). These sources called for more research on the sideeffects of ECT (1) and on alternatives to ECT (1). Only four patients of the 38 lay sourcescited had positive stories to tell (3). For instance, Curtis Hartmann, age 47, a Massachusettslawyer who has received about 100 electroshocks since 1976 to help control his bipolar ill-ness claims: “Electroshock has been the only thing that has ever let me feel 100% …Depression is like being a corpse with a pulse. I tried everything else. I had a loving family,thousands of hours of good psychotherapy, and none of it ever helped” (Time Canada, 26February 2001). However, the coverage of such success stories definitely took second placeto the horror stories and the opposition of patient associations.
4. The media coverage of prenatal screening tests (PSTs) for Down syndrome (DS)
In this analysis we focus our attention on a prenatal screening test that combines a fetal ultra-sound scan and a biochemical analysis of maternal blood markers. The novelty about this testis that it is performed during the first trimester of pregnancy. It measures nuchal translucency,which is “the subcutaneous space between the fetal cervical spine and the overlying skin”(Framarin, 2003: 11). The test gives the probability that a pregnant woman will be at a higher-than-average risk of having a DS fetus (Vassy, 2006: 2042). Our sample of the press cover-age of prenatal screening tests for DS includes 18 articles covering the pros and cons ofprenatal tests, and 5 articles covering lawsuits launched by parents claiming their physiciansfailed to inform them about PSTs.
The advantages of prenatal screening, and more generally of earlier screening and earlierdiagnosis, were discussed at length based on information provided by scientific experts andclinicians. We read that PSTs make it possible to: a) terminate a pregnancy at a moment whenit is clinically safer and psychologically easier (11); b) avoid intrusive and more risky prena-tal tests (6) or better orient women in the choice of diagnosis methods (5); c) reduce the periodof anxiety for mothers at high risk of having a DS child (1); and d) provide parents with moretime to prepare for the forthcoming birth of a DS child should they decide to continue withthe pregnancy (1). This position in favor of PSTs was also echoed in articles covering law-suits launched by parents against their physicians.
Most cited sources represented scientific expertise (46 percent). This was followed by parentsof DS children or patient associations (28 percent) and clinical experts (13 percent).
Management and industry expertise were totally absent (0 percent) from the coverage.
Here again, the positions of health care professionals emphasize the positive aspects of PSTs (4). Most of them welcomed a PST that could be administered earlier in pregnancy,explaining that earlier screening enables physicians to orient patients toward the best screen-ing and diagnosis tests, thereby avoiding more invasive ones (1). Thus, PSTs are not there toencourage pregnancy termination, but rather to help decide which women should go throughinvasive tests. The cited experts believe the test should be offered to all women at high riskfor having a DS child (1). “‘Prenatal counseling and diagnosis are not a search-and-destroymission,’ insisted Dr. Malone, who has seen many of his patients choose to continue a Downpregnancy. ‘Most of us would not answer the question, Doctor, do you think I should termi-nate? It’s not for us to decide’” (Time Canada, 21 November 2005).
Five articles gave testimonials from parents of DS children, the source that most clearly and directly addressed the ethical and social issues around prenatal screening. It was clear tomany that early PST and diagnosis was aimed at terminating pregnancies (1). The parents didnot oppose a woman’s right to reproductive choice, but expressed serious concerns about theway information is being provided to future parents and what they do with it (1): “… mostCanadians, including many physicians who will be giving the results of tests, have an undulypessimistic view of Down syndrome, and fail to recognize that the vast majority of childrenborn with the condition live full and fulfilling lives” (The Globe and Mail, 21 November2005). Thus, they emphasized the fact that DS fetuses were persons-to-be who are entitled tofull quality of life. “Will people open their eyes to the possibilities of these kinds of kids? …Most of the people who make these decisions don’t know an individual who has Down syn-drome. They don’t know about the advances in recent years” (Time Canada, 21 November2005). Finally, some of these sources said they would prefer to see the money used on PSTsreallocated to services for DS children and to educating society “toward acceptance of dif-ference” (Voices across Boundaries, Winter 2004: 27). This position, however, was not heldby the parents who filed lawsuits against their physicians for having failed to inform themabout PSTs. All of these women said they would have chosen to terminate their pregnancyhad they known they were pregnant with a DS child (5).
5. The media coverage of PSA
The PSA test measures the blood level of a protein released by the prostate, the prostate-specific antigen. A high PSA level can be caused by inflammation, aging-related enlargementor cancer. A positive PSA test has to be followed up with a biopsy or other procedures beforecancer can be confirmed. The media coverage of PSA in newspaper articles included 47 arti-cles on potential screening tests for prostate cancer that could replace or complement PSA,89 articles on PSA and population-based screening for prostate cancer, and 3 articles on alawsuit by the family of a deceased man.
The majority of scholars, health care provider associations and major patient associationscited were against population-based PSA screening. We read about the low accuracy of PSAtesting (51) or its failure to provide any information about the aggressiveness of a tumordetected (5). We learn about the consequences of high levels of false-positive and false-neg-ative results (46), as well as the consequences of radical treatment (e.g., impotence and incon-tinence) (16). The absence of evidence demonstrating a direct link between PSA and a Hivon et al.: Marginal voices in the media reduction in mortality was often stressed (31). On the other hand, we also read about theimportance of early detection of cancer in order to maximize the chances of being cured (48).
Thus, readers are confronted with articles that emphasize early detection by means of a PSAtest: “Early PSA Screening May Reduce Risk of Prostate Cancer Death: Study” (CanadianPress, 8 July 2005). Or they are alarmed by the consequences of this early detection: “MenMay Be Over Diagnosed with Prostate Cancer: Study” (Toronto Star, 5 July 2002). More thanone-third of newspaper articles (36 percent) ended with recommendations similar to thosemade by the Canadian Cancer Society, which emphasize individual choice: Men over the age of 50 years [should] discuss with their doctor the potential benefits andrisks of early detection of prostate cancer using the Prostate Specific Antigen (PSA) testand digital rectal examinations (DRE) so that they can make informed decisions aboutthe use of these tests. Men at higher risk because of family history or those of Africanancestry should discuss the need for testing at an earlier age. (Canada NewsWire, 16September 2004) Here again, the main source cited was scientific expertise (49 percent). This was followed bylay expertise (20 percent) and clinical expertise (13 percent). Administrators and industrysources were marginal voices, representing 7 percent and 4 percent, respectively.
Industry was mainly quoted for its research and development activities in prostate cancer detection. These articles discussed new tests being developed that could, in combination withPSA, improve screening for prostate cancer, help avoid invasive interventions and treatments,and reduce the costs of health care services (12).
Administrators did not really depart from the main storyline; they were usually cited as being against population-based PSA screening (20). In Canada, most provincial governmentsreject the idea of population-based screening for prostate cancer since there is no clear evi-dence supporting a link between PSA and reduced mortality rate (10). “PSA detects cases ear-lier, but this doesn’t mean that lives will be saved. … the test may sometimes even lead tounnecessary interventions, and this is why no jurisdiction will put forward a population-basedscreening program, as has been done for breast cancer” (consultant in a public health divisionof the Quebec government, TFF, Le Droit, 23 September 2003).
Interestingly, there was a discrepancy between the government’s official position on PSA and the testimonials of individual politicians. Two cases of prominent Canadian politiciansbeing diagnosed with prostate cancer were covered. After successful treatment, both mendeparted from the official line, maintaining that PSA screening had saved their lives and rec-ommending that all men be tested. One article cited the long-time adviser of Reform Partyfounder, Preston Manning, who was diagnosed and successfully treated for prostate cancer:“He’s taking a lesson from that, that other men should also be tested” (Canadian Press, 13December 2000).
We observed a similar difference of opinion among health care providers. While major provincial associations of urologists and the College of Physicians have positioned them-selves against mass screening (21), some individual physicians are in favor of making thePSA test available to all men. Talking about his patient, the former health minister Alan Rock,Dr. Jim Paust states: “Mr. Rock’s situation is an excellent example of how early detection andearly treatment can lead to a total cure. … it is a good reminder to all men over the age of 40to be tested regularly and to keep track of their yearly PSA levels” (Canadian Press, 13February 2001). A study of Ontario urologists conducted by a pro-PSA support group “found that 85 per cent of Ontario urologists surveyed said that screening tests for prostate cancerhelp reduce deaths and should be covered by OHIP [Ontario Health Insurance Plan]”(Canadian Press, 3 November 2005). In fact, most physicians quoted recognize the limita-tions of the PSA test but believe it is the best tool available for early detection of this illness(16): “The logical approach is to carry on with PSA until better means are developed” (TFF,Nouvelliste, 13 September 2004). Some asserted that excessive treatment resulting from thePSA’s lack of accuracy could be prevented by the active surveillance of patients at risk (6).
They also seemed to support findings according to which PSA velocity—the fluctuation ofPSA level within a certain period of time—is a good indicator of the presence of cancer andits aggressiveness: “Awareness of PSA levels allows men to know their prostate health status,essential in determining a logical strategy and best therapy in the event of cancer” (The Globeand Mail, 21 September 2005). This, however, means regular screening tests.
Finally, we find a similar discrepancy between the recommendations of patient represen- tatives and the individual testimonials of cancer survivors. Major patient associations such asthe Canadian Cancer Society and the American Cancer Society have come out against uni-versal prostate cancer screening, but they encourage men over 50 to discuss the potential ben-efits and drawbacks of PSA screening with their doctors before making their own decision(12). They believe, however, that men who decide to undergo a PSA test should have accessto it. However, when the testimonials of smaller groups such as the Early Prostate CancerDiagnosis Ontario (EPCDO) or individual patients were used in the newspaper articles, it wasalways to illustrate a clear position in favor of widespread PSA screening (13). DraytonMcLane, owner of the Houston Astros stated: “I feel very fortunate that the doctors caughtthis so early. It is proof that early detection is the right path for everyone” (Canadian Press,8 October 2002). Thus, the media are conveying two messages: get information about the testin order to make an informed choice, and by all means go and get tested.
6. The media coverage of COX-2s
Cyclo-oxygenase-2 or COX-2 drugs are prescribed for the treatment of inflammation and painassociated with arthritis and other musculoskeletal conditions. Mostly known under the brandnames Vioxx™, Celebrex™ and Bextra™, they were introduced in Canada at the end of 1999on the strength of studies showing they cause fewer upper gastrointestinal side effects than tra-ditional non-steroidal anti-inflammatory drugs (NSAIDs) such as Advil™ or Motrin™. Therewere 224 articles published on COX-2s between 2000 and 2006. Of these, 177 were publishedafter September 2004 when one of the COX-2s, Vioxx™, was withdrawn from the market afterstudies showed it was associated with an increased cardiovascular risk.
At the beginning of 2000, COX-2s were presented in the media as an alternative to NSAIDs,which are associated with serious gastrointestinal side effects (44). However, controversyarose soon after they hit the market. There were studies that challenged their superiority interms of efficacy and reducing gastrointestinal effects (8). Moreover, new studies aimed atexamining the efficacy of COX-2s in the treatment of certain types of cancer and Alzheimer’ssuggested that they were associated with an increased cardiovascular risk (20).
In September 2004, Merck Frosst decided to withdraw Vioxx™ from the market after its own study showed it doubled the risk of heart attack or stroke if taken for 18 months orlonger (90). Following this spectacular withdrawal, debate raged in the scientific and health Hivon et al.: Marginal voices in the media professional communities about whether or not the risks associated with Vioxx™ werecommon to all COX-2s (32). Health administrations in Canada and the United States wereaccused of failing to protect the public (13). Merck Frosst was accused of putting profit aheadof public safety (33). Expert review committees were formed by the FDA (US Food and DrugAdministration) and Health Canada to examine the evidence, assess the benefits and risks ofCOX-2s, and provide recommendations (9). They concluded that the benefits of COX-2s out-weigh the risks, and Vioxx™ was allowed back on the shelf provided specific conditions weremet (3). Celebrex™, which remained on the market throughout the uproar, also came undermore severe restrictions (6). Bextra™ was the only medication that was pulled off the shelvesindefinitely because it was shown to cause a rare but fatal skin problem (19). Although thereputation of the COX-2s had been restored, thousands of lawsuits were launched, mostlyagainst Merck Frosst, Health Canada and the FDA (33). And the public controversy is notover for the media continue to cover these lawsuits.
The most striking observation here is the relative absence of clinical and lay points of view. Whileclinical expertise represented 13 percent to 19 percent of all the sources quoted in the three othercases, it represented only 4 percent in the case of COX-2s. The same is true of lay expertise,which represented 20 percent to 28 percent of the cited sources in the three other cases, comparedto only 7 percent in the case of COX-2s. A different pattern emerged for administrator and indus-try expertise: while almost absent from the coverage of the other three health technologies, theyrepresented 21 percent and 17 percent, respectively, of all the sources cited in this present case.
Scientific expertise is also dominant, representing 39 percent of the sources cited.
Before 2004, there were very few references to clinical expertise in the media. The media refer to this expertise in general terms such as “doctors.” We read that “doctors” continue toprescribe Vioxx™ and other COX-2s despite warnings against cardiovascular risks becauseof the medication’s effectiveness at controlling gastrointestinal problems (1). After 2004, theirposition toward Vioxx™ and COX-2s as presented in the media was more variable. Somepositioned themselves as completely against COX-2s and called for a ban (5). Others arguedthat COX-2s constitute an important alternative to traditional medications and that, when usedwith caution, they can benefit many people suffering from excruciating chronic pain (1). Afew stated that the withdrawal of Vioxx™ from the market was not a “big deal” since otheralternatives exist and the medication is expensive for the little additional benefit it provides(1). Representatives of rheumatologist associations argued that studies showing increased car-diovascular risks were incomplete. Those cited contended that COX-2s are valuable as theysignificantly reduce gastrointestinal problems and show promise as a cure for certain types ofcancer. Besides, they say, any treatment entails both benefits and risks. Dr. François Couture,president of the Quebec association of rheumatologists stated: We condemn a class of medication that is very promising for preventing cancer and that hasimportant benefits in the treatment of arthritis on the basis of far less rigorous studies thanthose necessary for the approval of new drugs. I don’t want data to be hidden, but medicinemust be based on evidence, not speculation. (TFF, La Presse, 23 December 2004) They disagreed with the stance that all COX-2s present similar risks (3), but advised theirmembers to be careful in prescribing the drug until additional studies have clarified the situ-ation. According to some, the Vioxx™ controversy led physicians to re-examine their pre-scription habits (2). They underlined that these medications have often been prescribed topeople for whom they were not designed, and they called for greater caution in the future (2).
How the media used lay expertise was very similar to how it used the expertise of health care providers. The media presented the testimonials of people for whom COX-2s had notbeen very beneficial: “It had no reaction whatsoever on me except on my pocketbook”(National Post, 27 August 2001). It also presented patients that felt that the risks associatedwith these medications did not outweigh the benefits: “I appreciate Health Canada’s concernfor my health, but in the final analysis, I should, as an adult, be able to weigh all the risks anddecide my own course of action. As adults, we have choices to make. My choices are livingwith risks and a quality of life or living in a long-term facility with no quality of life”(Canadian Press, 9 June 2005). Throughout the controversy, the position of the ArthritisSociety of Canada favored the cautious use of COX-2s.
Each patient is unique. There are over 100 types of arthritis. The severity of the diseasevaries from patient to patient as does the stage of their disease, and the presence of otherdiseases. Physicians are in the best position to assess the relative risks and benefits ofeach medication prescribed to a patient. … We need to make sure that we strike a bal-ance between giving people access to the best available drugs and ensuring that both doc-tors and patients know about all the risk factors when they are making treatmentdecisions. (Canadian NewsWire, 22 December 2004) The Society recognized that COX-2s are effective and that the associated risks are generallyminimal (2). However, as in the case of prenatal tests for DS, this cautious position was notshared by the thousands of patients who launched lawsuits against Merck Frosst for havingput their safety at risk (33).
Throughout the controversy, contrary to the relatively nuanced position of most stake- holders in the coverage of COX-2s, the industry was always presented as being thoroughly infavor of these drugs. Before 2004, both Merck Frosst and Pfizer maintained their productswere safe and that studies suggesting the contrary were incomplete (13). Commenting on theVIGOR study that showed Vioxx™ users were twice as likely to suffer serious cardiac prob-lems as those taking naproxen, Merck’s Senior Director of Cardiovascular Clinical Researchstated: “The Vioxx™ study involved patients with rheumatoid arthritis, a disease that raisesthe risk of heart trouble. The results may reflect naproxen’s potential heart-friendly benefitsrather than any negative effects from Vioxx™” (Canadian Press, 21 August 2001). Merckdenied the risks associated with Vioxx™ up until its own study confirmed it. Once the riskbecame obvious, the company reacted immediately by withdrawing the product from theshelves. Merck Frosst’s competitor, Pfizer, also denied similar risks for its own products,Celebrex™ and Bextra™. Pfizer’s President of Worldwide Development stated: “the resultsrevealed in that study [increased cardiovascular risks] are not consistent with a ‘large body ofdata’ that the company had collected” (Canadian Press, 1 February 2005).
Finally, the media mainly referred to the government in three specific contexts: a) when it announced the decision to refund a specific drug (2); b) when it approved the commercial-ization of a new drug (6); and c) when new risks associated with a drug required taking action(10). The controversy surrounding the Vioxx™ withdrawal put this last function of the gov-ernment to the test. Both Health Canada and the FDA managed the risks by asking the com-panies for additional data on the drugs (11), by compelling them to modify their productmonographs (14) and by warning the public about the new risks (9). The official position ofboth administrations in the controversy around Vioxx™ and COX-2s was one of caution.
They suggested that until more complete data on the safety of COX-2s became available,patients should discuss with their physicians their particular risks for cardiovascular or gas-trointestinal problems before using the drugs (14). Thus, they emphasized the need for addi-tional studies (6) and underlined the need to balance the risks and benefits of the drugs (8).
Hivon et al.: Marginal voices in the media They assembled a review panel of experts to examine the evidence, assess the benefits andrisks of COX-2s, and provide recommendations (9). Following these recommendations, theyallowed Vioxx™ back on the market subject to specific conditions, maintained the saleof Celebrex™ subject to severe restrictions, and withdrew Bextra™ from the market (28).
The controversy also highlighted the limited power of government to compel the industry toprovide additional data once a medication has been approved for sale (2). In response, thegovernment re-examined the review process leading to the approval of new medications forsale (11). Thus, in the media coverage of COX-2s, health care administration was an impor-tant actor in the controversy.
7. Discussion
Several studies on the media coverage of health technology, when examining a single case,often end up criticizing the fact that specific groups are silenced by journalists (Conrad, 1999;Williams et al., 2003; Collins et al., 2006). They also argue that a more inclusive treatment ofstakeholders in the media would make the debate around science and technology issues moredemocratic. Our study contributes to current knowledge by suggesting that the above criti-cism is perhaps not fully justified, and that the above suggestion, while valid, needs qualifi-cation. Our results show that: 1) the pattern of source utilization varies between healthinterventions and does indeed include stakeholders typically seen as vulnerable; and 2) theviews expressed by the marginal sources fill information gaps and bring nuance to the storiesbeing told, mostly by emphasizing aspects not addressed by the dominant voices (Box 1).
However, as we argue below, journalistic practices tend to reproduce social assumptions about whose views matter and why, and about what kinds of technologies are desirable orthreatening. As a result, if the goal of scholars in the area of the public understanding ofscience is to strengthen the democratic foundations of the perspectives brought forward byjournalists, two assumptions need to be revisited. Further research should: 1) move beyondthe journalism/science dichotomy; and 2) explore more systematically how different journal-istic strategies may modify the understanding of, and judgments about, science and technol-ogy issues on the part of different categories of readers.
Moving beyond the journalism/science dichotomy The current body of research on the media coverage of science and technology issues addressesjournalists’ use and misuse of scientific expertise and highlights the distortion of scientific factsthat may ensue (Conrad, 1999; Dunwoody, 1999; Nelkin, 1987; Waddell et al., 2005). Severalcritics have suggested cross-fertilizing the worldviews of scientists and journalists by improv-ing the scientific training of journalists and the communication skills of scientists (Dunwoody,1999; Nelkin, 1987; Ransohoff and Ransohoff, 2001; Waddell et al., 2005; Weigold, 2001). Theassumption is that these “two worlds” need to be bridged. Nevertheless, not much research hastried to understand how non-scientific sources are identified, selected and mobilized by jour-nalists and what contribution they can make to the public understanding of science and tech-nology. Conrad (1999) and Williams et al. (2003) suggest including the voices of sociallymarginal stakeholders in order to ensure a more nuanced and democratic debate. But, whatsources are most likely to provide thorough information on the scientific, political, social andethical issues raised by science and technology matters? And how can journalists ascertainwhether the views they gather are meaningful to the story being told? MARGINAL VOICES IN THE MEDIA COVERAGE OF CONTROVERSIAL
HEALTH INTERVENTIONS:
Industry may be linked to increased ECT use.
Administrators are concerned by the increase in ECT use.
Health care professionals are mainly in favor of ECT.
Lay experts are mainly against ECT, although a few successstories are given.
DS prenatal screening
Industry and administrators are completely absent from the For health care professionals, PST helps orient patients toward For lay experts, PST leads to the desirable/undesirable termina- Industry is involved in R&D activities to find tests that can be All major institutions in all source categories are generally against populationbased PSA screening.
Individual testimonials in all source categories are in favor of Administrative, clinical and lay experts are divided on the risks Industry is clearly in favor of COX-2s and minimizes their Box 1. Summary of the views expressed by marginal voices.
Our analysis of the coverage of four health technologies that are controversial for different reasons has highlighted various patterns of source utilization. Contrary to what has been sug-gested by several authors (Conrad, 1999; Nisbet et al., 2003; Williams et al., 2003), our analy-sis shows that sources treated marginally in the media are not necessarily socially marginalstakeholders. While affected groups—the vulnerable or voiceless and their advocates—have Hivon et al.: Marginal voices in the media often been pinpointed as the key “silences” in media coverage, our study draws a signifi-cantly different picture. Patient associations, patients and their families were, in fact, signif-icantly represented in the media coverage of three of the four health interventions studied.
In the case of ECT, they came close to scientific experts in terms of representation (26 per-cent versus 31 percent). While other studies have suggested that the main cited sources areusually scientific experts, industry and government (Einsiedel, 1992; Nisbet et al., 2003), wefound that the latter two sources were almost absent in the coverage of three of the four inter-ventions. These voices were thus treated marginally when in fact they are significant stake-holders whose views matter from a public health policy perspective. Our point is not thatall the categories of sources we examined should be equally represented in the media. Butunderstanding why journalists deem certain groups to be relevant stakeholders for the pur-poses of their stories calls for further research in order to build on the current conflicting andinconsistent findings.
For instance, in his study of the press coverage of “behaviors” associated, either rightly or wrongly, with genetics (alcoholism, mental illness and homosexuality), Conrad (1999)suggests that the level of politicization of the socially marginalized group may favor its inclu-sion by journalists. When he asked journalists why they sought the points of views of homo-sexuals, but not those of mentally ill people or alcoholics, the response was that “it simplyhad not occurred to them” (1999: 300). Conrad concludes: “… the rise of the gay liberationmovement and the struggle for gay rights has produced organizations that have acquired a seatat the press’s table. Whenever a new genetic finding is reported, science writers seek outviews of gay spokespersons and some of the ‘affected’” (1999: 300). This interpretation partlyconverges with our own findings. Most patient associations concerned with ECT or Downsyndrome are politicized since they militate for human rights (Heitman, 1996). This maycompel journalists to seek their views when reporting new scientific findings in these areas.
This is less obvious, however, for patient associations dealing with prostate cancer or arthri-tis, where the main mission is to obtain better diagnosis and treatment as opposed to defend-ing patient rights. This apparent lack of any stake to be defended may explain why the voiceof laypeople is less represented in the press coverage of PSA and COX-2s. But it does notexplain why the government or industry is largely absent from debates around ECT, PST andPSA. The responsibility and accountability of these groups are, in principle, key to the storiesbeing told. These groups are, by definition, highly politicized and usually occupy an impor-tant place in the media. Moreover, in all the cases we examined, their formal authority andresponsibility in the development and use of health technologies could have been used by themedia to invigorate and democratize the debate. In the case of COX-2s, where journalistsclearly addressed the failure to protect the public, the larger presence of government andindustry is noticeable but happens somewhat late in the unfolding of the controversy. Thistreatment may simply reflect journalists’ (sometimes) uncritical social assumptions aboutwhose views matter. However, it may weaken the democratic process by which members ofthe public form judgments about new technologies.
The contribution of marginal voices to the debate around health interventions Our study shows that marginal voices do fill certain information gaps and bring some inter-esting nuances. For example, in the case of ECT and prenatal screening tests, patient associ-ations put forward ethical issues around informed consent and the consequences of thesetechnologies on patients and society—issues that were not necessarily developed by the sci-entific experts cited by the media. Are severely depressed people truly informed about therisks and benefits of ECT? Will pregnant women be adequately informed about the pros and cons of living with a DS child? Does this lead to an informed choice about whether or not toterminate their pregnancy? In the case of ECT and COX-2s, health care professionals alsobrought new perspectives. They emphasized the side effects of these technologies, remindingreaders that any treatment bears consequences, and that these consequences vary betweenindividuals. Reflecting on their own prescription practices, they pointed out that the problemwith a medication like Vioxx™ might not be with the drug itself, but rather in the way inwhich it is used or overused.
While these perspectives bring important insights to the debates, the problem is they remain marginal in the media coverage that purports to be balanced. Thus, the extent to which theycatch the attention of readers and inform their judgment about the various issues at stakeremains unclear. Indeed, despite health care professionals’ efforts to make ECT look less alarm-ing, the media coverage of this technology remains fairly negative and dramatized. Despitepatient associations’ concerns over the consequences of prenatal screening and the positiveaspects of raising children with DS, the coverage of PSTs is overwhelmingly positive.4 Our study suggests that the use of marginal voices to create a balanced story may also increase confusion. For instance, the current dominant, scientific view is that there is no solidevidence that the PSA test saves lives and that the benefits of the test outweigh the risks andinconveniences. However, the marginal voices brought in—those of a few health care profes-sionals and cancer survivors—encourage men to get tested. These voices thus convey a radi-cally different message. Contrary to the nuance brought by marginal voices in the otherstories, here the position of the various types of marginally treated stakeholders converge andmake a compelling case in favor of the test. Readers are told two opposing stories that arenonetheless portrayed as equally valid: the test is worthwhile and cancer over-detection dueto the test is a serious issue. How does such a dual message affect the reader’s judgment aboutPSA testing? As noted above, researchers who have found that journalists tend to misrepresent scien- tific facts have more or less implicitly suggested that if journalists were scientifically moreliterate, their news stories would be more valid. Nevertheless, while scientific experts couldhelp journalists understand “the processes of research, dead ends, and wrong turns” (Nelkin,1987: 173), or the benefits, risks and costs of new medications (Cassels et al., 2003;Moynihan et al., 2000), they are not necessarily in the best position to clarify the “ideologiesor social priorities that guide science policy decisions” (Nelkin, 1987: 173). This wouldrequire, on the part of journalists, telling news stories through the lenses of an explicit socio-political framework and seeking sources that can offer a solid scaffolding for public under-standing of science to take place.
8. Conclusion
The broad objectives of our research program were to understand how the media shape thesocio-political environment within which health technology assessments are disseminatedand used. As suggested by one reviewer, the opposite exercise would have been to examinehow the socio-political environment shapes journalistic practices. Such an analysis mayindeed help explain why “industry remains hidden from view” in several news stories and“appreciate the rise of social movements, including their links with the industry, many ofwhich are also hidden.” Despite its limitations, one of the values of our study lies in the fact that it shows the respective contribution of marginal voices at the same time as it shows their limited role in theunfolding of science and technology-related controversies. Different groups of stakeholders Hivon et al.: Marginal voices in the media 49 tend to shed light on different issues, which can in principle contribute to readers’ understand-ing of science and technology. The key question for journalists thus becomes: how can theyidentify, select and organize a meaningful set of views and develop more than a seemingly neu-tral, balanced treatment of advantages and disadvantages? We therefore believe that the call by some researchers to bridge the two worlds of journal- ism and science and to include other categories of sources, while important in that it helps ensure amore democratic debate, is not sufficient. How do different categories of readers form a judgmentaround scientific controversies? Do they assess the credibility of non-scientific and scientificsources in the same way? Do they give the same weight to the various voices being presented bythe press? How do they assess the validity of the knowledge claims presented in the media? Howdo these processes differ between newspaper articles and other sources of information?Interestingly, while studies of Internet users have proliferated in recent years, very few scholarshave examined how various publics form a judgment about what they read in the newspapers(Rogers, 1999). Further research could examine how publics appraise the cited sources and howthese sources influence their views on scientific news and controversial health technologies.
Acknowledgments
This research was funded by an operating grant from the Canadian Institutes of Health Research(CIHR; #MOP-64200). P. Lehoux holds a Canada Research Chair on Innovations in Health(2005–10). Jean-Louis Denis holds a Chair jointly sponsored by the Canadian Foundation forHealth Services Research (CFHSR) and CIHR. Our research group infrastructure is supportedby the Fond de la recherche en santé du Québec (FRSQ). We would like to thank Héléna Urferand Stéphanie Tailliez for providing useful feedback and comments on previous versions. Weare also grateful to Geneviève Daudelin for her valuable insights.
1 The report on COX-2s is slightly different from the other three in that it emphasizes policies rather than tech- nologies. However, the HTA agency was about to follow up on this report by publishing several papers on theseantiinflammatory drugs, and extensive media coverage of these reports was expected. We therefore selected thisreport as a starting point for examining the media coverage of COX-2s.
2 For instance, many were about lawsuits in which certain aspects of the technologies were contested. This had an impact on our results since it contributed to an overall feeling of negative coverage for ECT, PST and COX-2s.
3 Owing to space limitations, for the remainder of the paper, we give the number of articles mentioning the given 4 The positive or negative character of the coverage of each technology was measured in two ways. We examined whether the intervention was presented in the title mainly as a problem or as a solution (Smith, 1987), and whetheradvantages and disadvantages were mentioned (Einsiedel, 1992; Cassels et al., 2003). The detailed results of thisanalysis are available upon request.
References
Anderson, A. (2002) “In Search of the Holy Grail: Media Discourse and the New Human Genetics,” New Genetics Banken, R. (2002) The Use of Electroconvulsive Therapy in Québec. AETMIS (Agence d’Evaluation des Technologies et des Modes d’Intervention en Sante) 02-05 RE. xvii + 96 pp. Montréal: AETMIS.
Bubela, T.M. and Caufield, T.A. (2004) “Do Print Media ‘Hype’ Genetic Research? A Comparison of Newspaper Stories and Peer-reviewed Research Papers,” CMAJ 170(9): 1399–407.
Cassels, A.H., Merrilee, A., Cole, C., Mintzes, B., Lexchin, J. and McCormack, J.P. (2003) “Drugs in the News: An Analysis of Canadian Newspaper Coverage of New Prescription Drugs,” CMAJ 168: 1133–7.
Collins, P.A., Abelson, J., Pyman, H. and Lavis, J.N. (2006) “Are We Expecting Too Much from Print Media? An Analysis of Newspaper Coverage of the 2002 Canadian Healthcare Reform Debate,” Social Science andMedicine 63: 89–102.
Conrad, P. (1999) “Uses of Expertise: Sources, Quotes, and Voice in the Reporting of Genetics in the News,” Public Understanding of Science 8: 285–302.
Davidson, R., Hunt, K. and Kitzinger, J. (2003) “‘Radical Blueprint for Social Change’? Media Representations of New Labour’s Policies on Public Health,” Sociology of Health & Illness 25(6): 532–52.
Dunwoody, S. (1999) “Scientists, Journalists and the Meaning of Uncertainty,” in S.M. Friedman, S. Dunwoody and C.L. Rogers (eds) Communicating Uncertainty: Media Coverage of New and Controversial Science, pp. 59–79.
Mahwah, NJ: Lawrence Erlbaum Associates.
Einsiedel, E.F. (1992) “Framing Science and Technology in the Canadian Press,” Public Understanding of Science Framarin, A. (2003) First-trimester Prenatal Screening for Down Syndrome and Other Aneuploidies. AETMIS (Agence d’Evaluation des Technologies et des Modes d’Intervention en Sante) 03-01. xxii + 81 pp. Montréal:AETMIS.
Friedman, S.M., Dunwoody, S. and Rogers, C.L., eds (1999) Communicating Uncertainty: Media Coverage of New and Controversial Science. Mahwah, NJ: Lawrence Erlbaum Associates.
Heitman, E. (1996) “The Public’s Role in the Evaluation of Health Care Technology,” International Journal of Technology Assessment in Health Care 12(4): 657–72.
Kroll-Smith, S. (2003) “Popular Media and ‘Excessive Daytime Sleepiness’: A Study of Rhetorical Authority in Medical Sociology,” Sociology of Health & Illness 25(6): 625–43.
Lehoux, P., Denis, J.-L., Tailliez, S. and Hivon, M. (2005) “Disseminating Health Technology Assessment: Identifying the Visions Guiding an Evolving Policy Intervention in Canada,” Journal of Health Politics, Policyand Law 30(4): 603–41. May, R. (2005) “Media Should Campaign on the Basis of Facts,” Nature 7060(43): 814.
Moynihan, R., Bero, L., Ross-Degnan, D., Henry, D., Lee, K., Watkins, J., Mah, C. and Soumerai, S. (2000) “Coverage by the News Media of the Benefits and Risks of Medications,” New England Journal of Medicine342(22): 1645–50.
Nelkin, D. (1987) Selling Science: How Press Covers Science and Technology. New York: W.H. Freeman and Company.
Nisbet, M.C., Brossard, D. and Kroepsch, A. (2003) “Framing Science: The Stem Cell Controversy in an Age of Press/Politics,” Press/Politics 8(2): 36–70.
Paterson, M., Bassett, K., Mamdani, M., Wright, J., Naglie, G., Laupacis, A. and Anderson, G. (2003) What Effects do Provincial Drug Plan Coverage Policies for New Drugs Have on Patterns of Use and Cost? Toronto:Institute for Clinical Evaluative Sciences (ICES).
Prior, L. (2003) Using Documents in Social Research. London: SAGE Publications.
Ransohoff, D.F. and Ransohoff, R.M. (2001) “Sensationalism in the Media: When Scientists and Journalists May be Complicit Collaborators,” Effective Clinical Practice 4(4): 185–8.
Rogers, C.L. (1999) “The Importance of Understanding Audiences,” in S.M. Friedman, S. Dunwoody and C.L.
Rogers (eds) Communicating Uncertainty: Media Coverage of New and Controversial Science, pp. 179–200.
Mahwah, NJ: Lawrence Erlbaum Associates.
Rowan, K.E. (1999) “Effective Explanation of Uncertain and Complex Science,” in S.M. Friedman, S. Dunwoody and C.L. Rogers (eds) Communicating Uncertainty: Media Coverage of New and Controversial Science, pp. 201–23. Mahwah, NJ: Lawrence Erlbaum Associates.
Seale, C. (2003) “Health and Media,” Sociology of Health & Illness 25(6): 513–31.
Slaughter, P.M., Pinfold, S.P. and Laupacis, A. (2002) Prostate-specific Antigen (PSA) Screening in Asymptomatic Men. March, 65 pp. Toronto: Institute for Clinical Evaluative Sciences (ICES).
Smith, D.E. (1987) “The Process of Popularization—Rewriting Medical Research Papers for the Layman: Discussion Paper,” Journal of the Royal Society of Medicine 80: 634–6.
Stocking, S.H. (1999) “How Journalists Deal with Scientific Uncertainty,” in S.M. Friedman, S. Dunwoody and C.L. Rogers (eds) Communicating Uncertainty: Media Coverage of New and Controversial Science, pp. 23–41.
Mahwah, NJ: Lawrence Erlbaum Associates.
Vassy, C. (2006) “From a Genetic Innovation to Mass Health Programmes: The Diffusion of Down’s Syndrome Prenatal Screening and Diagnostic Techniques in France,” Social Science and Medicine 63: 2041–51.
Waddell, C., Lomas, J., Lavis, J.N., Abelson, J., Shepherd, C.A. and Bird-Gayson, T. (2005) “Joining the Conversation: Newspaper Journalists’ View on Working with Researchers,” Healthcare Policy 1(1): 123–39.
Weigold, M.F. (2001) “Communicating Science: A Review of the Literature,” Science Communication 23(2): 164–93.
Williams, C., Kitzinger, J. and Henderson, L. (2003) “Envisaging the Embryo in Stem Cell Research: Rhetorical Strategies and Media Reporting of the Ethical Debates,” Sociology of Health & Illness 25(7): 793–814.
Hivon et al.: Marginal voices in the media 51 Myriam Hivon, Ph.D. is an anthropologist working with the Interdisciplinary Health ResearchGroup (GRIS) at the University of Montreal, Montreal, Quebec, Canada. Her research inter-ests include lay participation in policymaking and knowledge transfer.
Pascale Lehoux is an associate professor with the Department of Health Administration at theUniversity of Montreal, Quebec, Canada. Her primary research interests lie in the sociology ofinnovation and knowledge transfer. She holds a Canada Research Chair on Innovations in Health.
Correspondence: Department of Health Administration, University of Montreal, P.O. Box 6128,Succ. Centre-ville, Montreal, Quebec, H3C 3J7, Canada; e-mail: pascale.lehoux@umontreal.ca Jean-Louis Denis is a full professor with the Department of Health Administration at theUniversity of Montreal, Montreal, Quebec, Canada. His research focuses on innovation andorganizational theory. He holds a CIHR/CFHSR chair on the transformation and governanceof health care systems. Melanie Rock is an assistant professor, Community Health Sciences, University of Calgary,Health Sciences Centre, Calgary, Alberta, Canada. She is also an AHFMR Population HealthInvestigator and a CIHR New Investigator in Societal and Cultural Dimensions of Health.

Source: http://getoss.enap.ca/GETOSS/Publications/Lists/Publications/Attachments/12/Marginal%20voices%20in%20the%20media%20coverage%20of%20controversial%20health%20interventions.pdf