Open access Protocol BMJ Open: first published as 10.1136/bmjopen-2019-029532 on 24 August 2019. Downloaded from Media Coverage of the Benefits and Harms of Testing the Healthy: a protocol for a descriptive study Mary O'Keeffe,1 Alexandra Barratt,2 Christopher Maher,1 Joshua Zadro, 1 Alice Fabbri,3 Mark Jones,4 Ray Moynihan2,4 To cite: O'Keeffe M, Barratt A, ABSTRACT Strengths and limitations of this study Maher C, et al. Media Introduction Much testing in medicine is aimed at Coverage of the Benefits healthy people to facilitate the early detection of health ► This will be the first study to analyse the media cov- and Harms of Testing the conditions. However, there is growing evidence that early Healthy: a protocol for a erage of the benefits and harms of tests that have detection is a double-edged sword that may cause harm descriptive study. BMJ Open the potential for overdiagnosis in healthy people. in the form of overdiagnosis. The media can be seen as a 2019;9:e029532. doi:10.1136/ ► Media stories will not be restricted by country. The major generator of consumer demand for health services. bmjopen-2019-029532 results could inform interventions to improve the Previous research shows that media coverage tends to quality of medical reporting in the media. ► Prepublication history for overstate the benefits and downplay the harms of medical ► The study will only consider media coverage of five this paper is available online. interventions for the sick, and often fails to cover relevant To view these files, please visit tests. conflicts of interest of those promoting those interventions. the journal online (http:// dx. doi. org/ 10. 1136/ bmjopen- 2019- However, little is known about how the benefits and harms 029532). of testing the healthy are covered by media. This study will examine the media coverage of the benefits and harms of treatment is more effective when it is tailored 3 Received 30 January 2019 testing the healthy, and coverage of potential conflicts of to the individual. Apparently healthy or well Revised 11 July 2019 interest of those promoting the testing. people are increasingly encouraged to proac- Accepted 11 July 2019 Methods and analysis We will examine five tests: 3D tively monitor, and be vigilant about under- mammography for the early detection of breast cancer; standing their health, with testing seen as a blood liquid biopsy for the early detection of cancer; positive step in consumer health empower- blood biomarker tests for the early detection of dementia; ment. However, there is mounting evidence artificial intelligence technology for the early detection of http://bmjopen.bmj.com/ that testing can harm healthy people, and © Author(s) (or their dementia; and the Apple Watch Series 4 electrocardiogram sensor for the early detection of atrial fibrillation. We the quest for ever-earlier detection of disease employer(s)) 2019. Re-use can lead to unnecessary classification of the permitted under CC BY-NC. No will identify media coverage using Google News and the 3–8 commercial re-use. See rights LexisNexis and ProQuest electronic databases. Sets of two healthy as sick: overdiagnosis. and permissions. Published by independent reviewers will conduct story screening and Although an exact definition of overdiag- BMJ. coding. We will include English language media stories nosis remains the subject of debate, particu- 1Institute for Musculoskeletal referring to any of the five tests from January 2016 to May larly in the context of non-cancer conditions, Health, Sydney School of Public 2019. We will include media stories if they refer to any overdiagnosis can be considered to occur Health, Faculty of Medicine and on September 29, 2021 by guest. Protected copyright. benefits or harms of the test for our conditions of interest. when persons are labelled with a technically Health, University of Sydney, Data will be analysed using categorical data analysis and Sydney, New South Wales, correct diagnosis that does not improve health multinomial logistic regression. 9 10 Australia outcomes. Key drivers of overdiagnosis 2 Ethics and dissemination No ethical approval is required Sydney School of Public Health, have been identified.11 One is the use of more for this study. Results will be presented at relevant Faculty of Medicine and Health, sensitive tests which can detect smaller abnor- The University of Sydney, scientific conferences and in peer-reviewed literature. 11 Sydney, New South Wales, malities, many of which are benign. There Australia is growing evidence demonstrating the pres- 3Charles Perkins Centre and INTRODUCTION ence of overdiagnosis, often arising through School of Pharmacy, Faculty of Much testing in medicine is aimed at appar- testing healthy people, across different areas Medicine and Health, University ently healthy people to identify those at an of medicine. Examples include screening for of Sydney, Sydney, New South increased risk of a disease or disorder.1 These cancer (eg, breast, prostate, thyroid), cardio- Wales, Australia 7 8 12–16 4Institute for Evidence-Based ‘healthy’ people can subsequently be offered vascular disease and dementia. Inap- Healthcare, Bond University, more tests, treatment(s) or preventive strat- propriate screening in this context is likely to Gold Coast, Queensland, egies (eg, a preventive medicine).1 The lead to higher healthcare spending and worse Australia increasing popularity of testing is indicative of outcomes (eg, psychological effects and 2 17–21 Correspondence to recent enthusiasm for early detection, which unnecessary and harmful treatments). Dr Mary O'Keeffe; is part of the promise of ‘precision medicine’. Sustained promotion to the public, mary. okeeffe@ sydney. edu. au That is, early detection is always better, and patients and clinicians of the importance O'Keeffe M, et al. BMJ Open 2019;9:e029532. doi:10.1136/bmjopen-2019-029532 1 Open access BMJ Open: first published as 10.1136/bmjopen-2019-029532 on 24 August 2019. Downloaded from of early detection and testing, including via the media, promoting the testing, by examining the coverage of five is considered another driver of overdiagnosis.11 Uncrit- tests. ical coverage of new tests, without consideration of their potential downsides, contributes to the general lack of knowledge about the potential harms of getting tested METHODS AND ANALYSIS when healthy. In fact, research has shown that only a small Overview proportion of people are knowledgeable about overdiag- We will conduct a large descriptive cross-sectional study of 22 23 nosis. This includes individuals offered tests where the global English-language media coverage of five tests from 22 23 potential for overdiagnosis is high. As such, patients January 2016 to May 2019. (and clinicians) overestimate the benefits of testing, while underestimating the harms.24 25 Tests and conditions of interest There is also concern about how changing media envi- This study will focus on five tests: ronments, such as the rising influence of social media, 1. 3D mammography for the early detection of breast can lead to ‘junk-food news’.26 Indeed, previous studies cancer on the media have identified evidence of exaggera- 2. Blood liquid biopsy tests for the early detection of can- tion,27 28 inaccurate media coverage of published scien- cer(s) tific papers,29 30 overstating of benefits of treatments, 3. Blood biomarker tests for the early detection of de- downplaying of harms27 29 and failure to report important mentia conflicts of interest of the experts cited in the story.27 4. Artificial intelligence (AI) technology for the early de- Concern has also been expressed about the financial tection of dementia closeness between journalists and industry. For example, 5. Apple Watch Series 4 ECG sensor for the early detec- pharmaceutical industry funding of journalism practice, tion of atrial fibrillation (AF) awards and education, has been documented.31 32 Further, We identified these tests based on the following criteria there appears to be a lack of independent medical 1. Evidence-based concerns about overdiagnosis research commentators in the media. Only one in six 2. Evidence of Food and Drug Administration (FDA) ap- media reports of research published in high-impact proval for certain classes of each test medical journals included comments from people who 3. Concern that the results of these tests may not lead to were independent of the study investigators.33 Moreover, improved health outcomes for some individuals; either one in three of the independent commenters had finan- due to the unavailability of effective treatment options cial conflicts of interest, most of which were not disclosed (eg, dementia) or treatments that may cause more in the media stories.33 harm than benefit (eg, early mammography) Poor media coverage of medicine is a significant issue; it 4. Identifiable groups or companies with a financial in- can influence how the public perceives the risk of health terest in promoting these tests, or maximising the mar- http://bmjopen.bmj.com/ services and how patients make treatment decisions.29 31 kets for downstream treatments For example, media coverage about the celebrity Kylie 5. Notable media coverage Minogue’s self-referral mammogram bookings led to a One reviewer (MO) used Google News to track media 20-fold increase in media coverage about breast cancer coverage of tests for the healthy between April 2018 and and a 40% increase in mammogram bookings during the October 2018. Results were recorded in a Microsoft Excel 2-week peak after the interview.34 Six weeks later, media file. Based on these results, the same reviewer designed a coverage was still up by 30%.34 series of Google Alerts (running between April 2018 and Media coverage of overtreatment has been examined December 2018) with specific keywords related to testing. on September 29, 2021 by guest. Protected copyright. in one study35 which examined the framing of medical The Google Alert results were screened to identify media overtreatment in United States (US) newspapers from coverage on tests for healthy people which met the first January 2007 to December 2010.