Rheumatoid Arthritis Portrayal by UK National Newspapers 2011–2016: A Service User - Led Thematic Analysis of Language Used

The representation of illness and disease by the mass media

Lay beliefs about the nature, causes, ramifications, time frame, and terminology of a given illness not only originate from personal experience and the medical care system, but also the mass media and wider culture [1, 2]. The mass media can be defined as “outlets of information and entertainment” [3] that includes television, cinema, newspapers, magazines, and the internet. Recent statistics from the United Kingdom (UK) media regulator, ‘The Office of Communications’ (Ofcom) [4], showed that 94% of the British adult population (defined as aged 16+) followed the news and 29% of them still read newspapers in printed form. Given its important influence, various academic studies have explored the media’s portrayal of long-term health conditions as diverse as eating disorders [5], fibromyalgia [6], chronic fatigue syndrome [7, 8], epilepsy [9, 10], diabetes [11, 12], chronic kidney disease [13], oral cancer [14], gout [15], obesity [16], mental illness [17], and neurological disorders [18].

The language utilised by the mass media in the communication of health issues has the potential to affect the public’s attitudes and perceptions [11]. This recognition has given rise to research, which has examined the media’s usage of language in the representation of long-term conditions. For instance, the media’s consistent application of pejorative terminology to describe mental illness has been demonstrated in numerous studies [17, 19, 20]. An analysis [21] of newspaper clippings found that almost half of British tabloid articles applied derogatory terms, such as ‘nutter’ and ‘looney’ when writing about people with mental health problems. There also has been an overwhelming media fixation on the language of violence in the depiction of schizophrenia, often playing on stereotypes of ‘dangerousness’ and ‘criminality’ [22, 23]. The ‘helpless’ or ‘incompetent’ victim are other demeaning terms written by journalists to characterise people with mental health conditions [24]. Corrigan et al. [25] suggested that this wilful adoption of negative language constituted a form of structural discrimination. Graham Thornicroft [26] asked the question of why the media has tended to trade in the use of such negative stereotypes? One of the main reasons he argues is that as most of the media industry is commercially and profit driven, it has had clear incentives to play on the prejudices of its consumers [26].

Diabetes mellitus also has received negative press coverage [11]. An Australian study [11] showed that terms, such as ‘sufferer’ and ‘diabetic’, which should be avoided when referring to people with diabetes, were still commonly used in news articles. Similarly, research [10] in the popular press in the United States (US) found that despite the protestations of epilepsy organisations, the term ‘epileptic’ featured in nearly half of stories analysed. In addition, some of the articles used mystical or supernatural references to describe seizures.

Mass media portrayals of rheumatic conditions

Only a few studies have been conducted on the depiction of rheumatological conditions in the mass media. A telephone survey [27] in the US indicated that the media was the most important source of information for the general public about RA. Television (75%), magazines (62%), and newspapers (52%) were all cited by respondents. Other studies have focused on the amount of coverage and/or accuracy of media portrayals of rheumatological conditions in general, rather than RA in particular. For example, early North American research [28] on television news reporting of long-term conditions over an 11 year period, found that there were only 23 news segments dedicated to rheumatological illnesses. This compared to 215 news features on heart disease and 925 on cancer. The study authors’ concluded that the coverage was biased towards diseases with the greatest mortality rates, rather than highly prevalent, but underreported long-term illnesses. A later Dutch study [29] of newspapers, magazines, and medical related television programmes showed that rheumatological conditions were underrepresented in these media outlets. Journalists also emphasised lifestyle factors and alternative treatments. Such a portrayal does not only downplay the seriousness of these conditions, but can lead to a misperception that they could be mitigated or ‘cured’ by will and effort.

Research in the US [30], Mexico [31], and Argentina [32] have confirmed the earlier findings [28, 29] that despite musculoskeletal conditions being commonly prevalent in middle and later life [33], because of their low mortality, they have not inspired much journalistic coverage. Shaw [34] suggested this demonstrates that long-term illnesses (such as RA) do not have media appeal; or to put it more colloquially, “death makes news” [30]. None of the above studies were conducted in the UK, or investigated the usage of language in media depictions of RA or its treatment.

Three recent studies have gone beyond reporting the frequency and/or accuracy of media accounts of rheumatological diseases. The first study [35], an unpublished PhD, used multiple methods of qualitative analyses to investigate how large circulation UK newspapers and magazines conceptualised osteoarthritis (OA). OA was constructed as a multiple phenomenon, as a ‘disease’, ‘condition’, or ‘ailment’. This in turn, was influenced by cultural conceptions of disability-'saints’ or ‘scroungers’- and ageing-'peril’ or ‘promise’, and their media representation was determined by the targeting of readership, financial priorities, and newsworthiness. The study author concluded that media reporting of OA could de-prioritise the condition, and affect societal perceptions about its legitimateness. The second study [15], involved a content analysis of UK and US newspaper articles about gout. It highlighted that humorous comments were often made about gout, linking it to causal beliefs about lifestyle and the consumption of alcohol. The third study by Hanson et al. [36] concentrated on a specific area of media reporting, namely, local and national UK newspaper stories about research on new treatments for rheumatoid arthritis. With very few exceptions, the media narrative was overwhelmingly positive. Three fundamental aspects framed newspaper stories about new medical treatments for RA: “an ‘innovation’ offers ‘hope’ in the context of ‘burden’” [36].

Despite the insights of these three studies [15, 35, 36], to our knowledge there have been no publications on the construction of RA in the UK popular press. Given the prevalence of RA in the UK population [37] and the important role of the media as a source of information about the condition [27], this omission is surprising. In contrast to Hanson et al.’s [36] study, our research was led by a researcher with RA, and in addition, was designed to examine the use of language by the printed press in the portrayal of RA. Specifically, the aim of the study was to explore the language used by UK national newspapers in its reporting of rheumatoid arthritis (RA) over a defined five year period between 26th July 2011 and 26th July 2016.

A patient and public involvement (PPI) approach

The study was based on a patient and public involvement (PPI) approach [38] and involved academics and people with RA working together on this project. An inherent advantage of PPI is that it draws on the invaluable perspective of people with RA, who by their daily lived experience of the condition are experts in their own right. PPI can have different degrees of service user involvement in research, including consultation (researchers ask for views/advice from service users and/or carers), collaboration/co-production (researchers and service users and/or carers work together, e.g. to identify research questions), and user-led/controlled (service users make the decisions about the research, e.g. are the principal investigator) [38].

This study was service user-led by the second author, who had primary responsibility for the data collection and analysis of findings. Lempp [38] has argued that PPI is both ethically imperative and important in ensuring research is directly relevant to people with RA and carers. Indeed, the original impetus for the study was the frustration of service users with RA, who we come into daily contact with as a rheumatology department about society’s lack of awareness of RA and its impact on people with the disease. Recent research has shown that members of the public with no direct experience of RA had a limited understanding of the condition [2, 39–41], including a conflation of RA with OA or osteoporosis [39, 41], and a mistaken assumption that RA is a natural consequence of ageing [39, 41].

Data selection/collection

The decision to restrict the study to newspapers is that they are a highly accessible outlet of media, and they are not limited by the subtopics that they can cover about a given health issue [30]. As Adelman and Verbrugge [30] argued, “scientific findings, therapeutic regimens, government policies and programs, personal stories of adaptation, deaths of individuals, and more” have all featured in the popular press.

Tabloids can be defined by their straightforward use of language and “light, bright, and vivid writing, shorter stories, and extensive use of photographs and graphics” [44], that appeal to a wide readership. In addition, the colourful imagery that is typical of a tabloid’s front page is anchored by striking headlines [44]. Another feature of these mass market newspapers is the subject matter of entertainment, sports, crime, and celebrity stories that often invoke scandal or have a sexual element [44]. Although these characteristics are also found in the broadsheet newspapers, they assume greater importance and prominence in the tabloids [44].

By contrast, broadsheets also known as quality newspapers were traditionally defined by their large format and their higher level of language compared to tabloids [45]. Furthermore, they are characterised by an editorial and presentational style where the focus is on the ‘serious news’ of politics, business, economics, and world affairs [46]. The middle market newspapers in contrast, combine the entertainment features of the tabloids with the coverage of ‘serious’ topics associated with the quality press [47].

The inclusion criteria for the analysis were: (i) articles with the phrase ‘rheumatoid arthritis’ in the headline and/or lead paragraph, and (ii) published between 26th July 2011 and 26th July 2016. The term rheumatoid arthritis (RA) refers to "a chronic inflammatory disease characterized by joint swelling, joint tenderness, and destruction of synovial joints, leading to severe disability" [48]. As one person (second author) had primary responsibility for the analysis of the newspaper items, a five year time span for the identification of articles was agreed among the authors as manageable given the amount of data. A 5 year time period has also been deemed appropriate in other research [15, 17] on newspaper portrayals of long term illnesses. The exclusion criteria were: (i) items which referred to RA only once, (ii) duplicate articles, (iii) letters that sought medical advice, and (iv) product advertisements.

Data analysis

All articles were uploaded to NVivo Pro 11 [49] to aid the analytical process. A thematic analysis based on a realist perspective [50] was applied to the dataset of newspaper articles. In this approach, codes and themes were not predetermined deductively by a theoretical framework, but were generated inductively from the data [51]. As the results of the literature search/review discovered little information about the language used in the popular press to portray RA, an inductive orientation [51] to the analysis fitted in with the explorative nature of the research.

In reference to a grounded theory approach, the second author systematically interrogated the data set through ‘open coding’ [52] with the ‘constant comparative’ method [53]. In this way, simultaneous comparison of codes and data were identified, followed by the refinement of codes into themes. A random sample of 25% of the coded data was then cross-checked by the third author and a consensus was reached on final codes and themes. This form of triangulation [54] was designed to limit researcher subjectivity during the analytical process. The results of the thematic analysis were member checked [55] by proxy with a group of clinicians (e.g. nurses and doctors) drawn from the authors’ academic rheumatology department. We used this strategy, because departmental expert service users recommended not to approach service users for the validation of findings, but instead involve clinicians in this process. Service users have at times disclosed negative reactions to media coverage of RA to clinical staff during consultations.

Theme 1: Language used to describe rheumatoid arthritis (RA)

Theme 2: Language used to refer to those who live with rheumatoid arthritis (RA)

Theme 3: Language used to report on potential new treatments for rheumatoid arthritis (RA)