PLOS ONE RESEARCH ARTICLE Microbiome research in general and business newspapers: How many microbiome articles are published and which study designs make the news the most? Andreu Prados-Bo ID1,2*, Gonzalo Casino ID1,3* 1 Department of Communication, Pompeu Fabra University, Barcelona, Spain, 2 Blanquerna School of Health Sciences, Ramon Llull University, Barcelona, Spain, 3 Iberoamerican Cochrane Centre, Biomedical Research Institute Sant Pau (IIB Sant Pau), Barcelona, Spain a1111111111 a1111111111 * andreu@andreuprados.com (APB); gonzalo.casino@upf.edu (GC) a1111111111 a1111111111 a1111111111 Abstract The microbiome is a matter of interest for science, consumers and business. Our objective is to quantify that interest in academic journals and newspapers, both quantitatively and by OPEN ACCESS study design. We calculated the number of articles on the microbiome from the total number Citation: Prados-Bo A, Casino G (2021) of biomedicine articles featured in both PubMed and Spanish science news agency SINC, Microbiome research in general and business from 2008 to 2018. We used the Factiva database to identify news stories on microbiome newspapers: How many microbiome articles are papers in three general newspapers (The New York Times, The Times and El Paı́s) and published and which study designs make the news the most? PLoS ONE 16(4): e0249835. https://doi. three business newspapers (The Wall Street Journal, the Financial Times and Expansión), org/10.1371/journal.pone.0249835 from 2007 to 2019. Then, we compared news stories with microbiome papers in PubMed, Editor: Federico Neresini, University of Padova: while also analyzing the frequencies of five study design types, both in the newspapers and Universita degli Studi di Padova, ITALY in the papers themselves. Microbiome papers represented 0.8% of biomedicine papers in Received: July 28, 2020 PubMed from 2008 to 2018 (increasing from 0.4% to 1.4%), while microbiome news pub- lished by SINC represented 1.6% of total biomedical news stories during the same period Accepted: March 25, 2021 (increasing from 0.2% to 2.2%). The number of news stories on microbiome papers corre- Published: April 9, 2021 lated with the number of microbiome papers (0.91, p < 0.001) featured in general newspa- Peer Review History: PLOS recognizes the pers, but not in business ones. News stories on microbiome papers represented 78.9% and benefits of transparency in the peer review 42.7% of all microbiome articles in general and business newspapers, respectively. Both process; therefore, we enable the publication of all of the content of peer review and author media outlet types tended to over-report observational studies in humans while under- responses alongside final, published articles. The reporting environmental studies, while the representation of systematic reviews of editorial history of this article is available here: randomized controlled trials, randomized controlled trials and animal/laboratory studies https://doi.org/10.1371/journal.pone.0249835 was similar when comparing newspapers and PubMed. The microbiome is receiving Copyright: © 2021 Prados-Bo, Casino. This is an increasing attention in academic journals and newspapers. News stories on the microbiome open access article distributed under the terms of in general and business newspapers are mostly based on research findings and are more the Creative Commons Attribution License, which permits unrestricted use, distribution, and interested in observational studies in humans and less in environmental studies compared reproduction in any medium, provided the original to PubMed. author and source are credited. Data Availability Statement: All relevant data are within the paper and its Supporting Information files. PLOS ONE | https://doi.org/10.1371/journal.pone.0249835 April 9, 2021 1 / 14 PLOS ONE A quantitative content analysis of microbiome research news stories in the press Funding: The authors received no specific funding Introduction for this work. On 19 December 2007, four years after the completion of the Human Genome Project, the Competing interests: have read the journal’s policy Human Microbiome Project (HMP), conceived as a “second human genome project” [1], was and the authors of this manuscript have the following competing interests: APB is a paid launched. HMP focuses on microbial communities and their genomes on and in the human consultant to companies commercially involved in body, collectively known as the microbiome [2]. Research into the microbiome dates back to the gut microbiota and probiotics. GC has declared the early 20th century [3]. The 21st century, however, has witnessed a paradigm shift regarding no competing interests. This does not alter our the crucial role microbes play in the way ecosystems—from the ocean to the human body— adherence to PLOS ONE policies on sharing data function, rather than only being seen as infectious pathogens [4]. and materials. Relationships between microorganisms living in our bodies—especially in the gut—and health and risk of disease are currently a major focus of research, public interest and potential business for the pharmaceutical and health industries [4]. Much research in the field has focused on the link between the microbiome and physical and mental well-being, with current unknowns highlighting the need for advancement [5]. An altered human microbiome has been associated with the development of a wide range of diseases, including inflammatory bowel disease, metabolic syndrome, autoimmune disorders, and brain diseases, although cau- sation has yet to be established [6]. Furthermore, microbiome research also has implications for food production [7] and for achieving an environmentally sustainable future [8]. As scholarship and investment in microbiome research develop, it is also important to address the impact of that research in general and business newspapers, which are relevant sources of information and can influence the decisions of the public, investors, health decision makers and healthcare practitioners [9,10]. Scientific articles mentioned in the lay press receive, on average, more citations in academic journals than comparable publications from the same journal that did not appear in the lay press [11–16]. The factors associated with greater levels of newspaper coverage for scientific papers have also been studied. They include the prestige of the journal [17], the availability of press releases [18–22], the domestic prefer- ence of newspapers for journals from their own country [23], and the newsworthiness of the topic [24]. Although nowadays many people do not obtain their information directly from newspa- pers, news from traditional media outlets continues to dominate the information repertoire of mobile internet users [25,26]. Social media, which has become an emerging source for keeping up with scientific issues [27], also relies heavily on newspapers to disseminate news among young people and adults [28,29]. Despite the current scientific interest in the microbiome, its social impact in newspapers has not been properly analyzed. One relevant way of studying the subject is by analyzing the number of newspaper articles in which authors, papers or journals are cited (called “press cita- tions”) [18,23,30]. As such, our first objective was to analyze the extent to which the predict- able increase in the number of microbiome papers in recent years has had a parallel impact in the press, in both general and business newspapers, given the potential for harnessing the human microbiome to prevent, diagnose or cure disease. Previous studies showed that study designs based on weak methodology (i.e. observational and animal or laboratory studies) are more likely to be covered in newspapers than those of superior quality (i.e. randomized controlled trials) [18,31,32]. Microbiome research over the past two decades has mainly focused on characterizing microbiome composition across cohorts of clinical patients and matched controls, and on using animal models to understand the causal mechanisms [33]. We hypothesize that studies’ methodological rigor is not a major driver when selecting news stories on the microbiome for coverage by newspapers. Our second objective was to undertake a controlled comparison of the study designs of the microbiome papers featured in newspapers against those of the microbiome papers that appear in PubMed. PLOS ONE | https://doi.org/10.1371/journal.pone.0249835 April 9, 2021 2 / 14 PLOS ONE A quantitative content analysis of microbiome research news stories in the press Such an analysis allows us to build a picture of microbiome science’s current level of maturity, which shapes both society’s perception and the decisions individuals make in relation to their health. Methods Newspaper coverage of microbiome research Based on previous studies of press coverage of biomedical research [34–37], we used the Fac- tiva database to search for news stories on the microbiome in three general newspapers (The New York Times, The Times and El País) and three business newspapers (The Wall Street Jour- nal, the Financial Times and Expansión) from the United States, the United Kingdom and Spain, respectively. Those three countries were selected as they were representative of three previously identified national patterns (American, British and Western World) of biomedical reporting in the press [23]. Print and online editions of each newspaper were analyzed together, after ruling out duplicate news stories. The period analyzed begins in 2007—when the HMP was launched [38]—and ends in 2019. Fig 1 shows how news stories on the microbiome were selected and categorized. First, we identified news stories in the Factiva database from the 6 newspapers that mentioned the microbiome and its hyponyms in any part of the text, no matter how many times. We excluded duplicates, infographics and other non-relevant news stories, as well as those in which the term microbiota referred to Microbiota decussata, commonly known as Siberian cypress, or plant flora. After filtering out opinion articles and editorials [19], we went on to study the newspapers’ interest in the microbiome (first objective), for which we identified news stories that devoted 50% or more of the text length (estimated by word count) to reporting on the microbiome (n = 518) [39]. Then, to analyze which study designs made the news compared to PubMed (second objective), we focused on microbiome news stories that cited at least one Fig 1. Process flow diagram of the categorization of news stories on microbiome papers. https://doi.org/10.1371/journal.pone.0249835.g001 PLOS ONE | https://doi.org/10.1371/journal.pone.0249835 April 9, 2021 3 / 14 PLOS ONE A quantitative content analysis of microbiome research news stories in the press scientific paper (n = 361). For this objective, news stories with the following characteristics were excluded [19]: 1) the microbiome was reported but without reference to a specific study; 2) the microbiome was not a variable of the scientific paper cited in the news story; or 3) the news story was based on premature microbiome research not published in peer-review jour- nals, such as studies presented at scientific or press meetings and ongoing clinical trials regis- tered in the ClinicalTrials.gov database. Additionally, we studied news interest in the microbiome in the context of biomedicine by quantifying news stories citing the microbiome in the headline and published by SINC, which is a Spanish publicly-funded news agency specializing in science and technology. SINC agency (https://www.agenciasinc.es/) was chosen because it publishes under a Creative Commons 4.0 license and most Spanish newspapers and scientific online sources usually pick up SINC arti- cles to inform the public. The microbiome in the context of biomedicine To estimate the relative interest in microbiome research in the context of biomedical research as a whole, we counted both papers on the microbiome and the total number of scientific pub- lications in the PubMed database of biomedical literature from 2008 to 2018. News agency SINC, which classified news about biomedicine separately until 2018, was used to count the number of news stories on the microbiome against the whole of the biomedicine category. We used data from SINC because the number of news stories on biomedicine in newspapers can- not be calculated using Factiva. Thus, we had an estimate of the press interest in the micro- biome from 2008 (when SINC was founded) to 2018 (when it finished categorizing news stories) to compare with the estimate in PubMed. Study designs of microbiome papers in PubMed and research news We adapted the criteria used by Bartlett et al. [18] and Lai and Lane [31] for categorizing the study design of medical research news to microbiome research. As such, we classified the microbiome paper study designs available in PubMed and reported in newspapers into 6 cate- gories: 1) systematic reviews (SRs) with or without meta-analyses of randomized controlled tri- als (RCTs) in humans; 2) RCTs in humans; 3) human observational studies (defined as prospective and retrospective cohort studies, ecological studies, case-control studies, SRs not of RCTs, and case series); 4) environmental & plant studies (agricultural, aquatic, atmospheric, built environment and terrestrial ecosystems) [40]; 5) animal or laboratory studies; and 6) other designs (interventional studies without randomization and/or without a control group, case reports, narrative or nonsystematic reviews, consensus and reports of expert committees). We excluded commentaries, editorials, perspectives and letters, as such articles do not usually contain research evidence and are not always peer-reviewed [31]. Each scientific study cited in the press was identified on PubMed and downloaded for study design characterization. In PubMed, we set the MeSH and natural terms search to title, abstract and keywords, and applied filters for study designs. It should be acknowledged that PubMed counts articles pub- lished in online and print versions separately [41]. The searches were performed by one author (APB) between January and March 2020. Search phrases and filters used in Factiva and PubMed are listed in the supplementary S1 and S2 Files and the data acquired and used for analyses are included in S1 Dataset. Statistical analyses The primary outcome variable was the number of news stories on the microbiome collected by year from 2007 to 2019 that cited at least one scientific paper. That variable was presented as PLOS ONE | https://doi.org/10.1371/journal.pone.0249835 April 9, 2021 4 / 14 PLOS ONE A quantitative content analysis of microbiome research news stories in the press the mean and standard deviation for the overall sample and subinterest groups: individual newspapers, country (United States, United Kingdom and Spain) and newspaper type (general and business). The relationship between two quantitative variables was evaluated using the Pearson corre- lation (linear adjustment): the number of news stories on microbiome papers per year in news- papers and published by SINC and the number of microbiome papers per year in PubMed. The average annual percentage change was evaluated for both the overall sample and the sub- interest groups: individual newspapers, country and press type. Finally, the comparison between microbiome news/biomedicine news published by SINC vs microbiome papers/bio- medicine papers in PubMed was carried out using a Chi Square test. The level of significance was set at 0.05. Version 3.5.2 of software R (SPSS Inc., Chicago, IL, USA) and version 4.7.0.0 of the Joinpoint Regression Program were used for all analysis work. Results Newspaper coverage of microbiome research Overall, 518 news stories with the microbiome as the main topic were published from 2007 to 2019, of which 361 cited at least one journal article (286 in general newspapers and 75 in busi- ness newspapers) (Fig 1). News stories on microbiome papers showed an irregular pattern of evolution compared to microbiome papers in PubMed (Fig 2A). Apart from a peak in 2008, the interest of general newspapers in microbiome research picked up steadily after 2012 with peaks in 2013, 2016 and 2018 (Fig 2B). News stories on microbiome papers represented 77.9% of overall news stories on the microbiome in The New York Times, 74.1% in The Times, and Fig 2. News stories on microbiome papers compared to microbiome papers in PubMed. (A) Individual newspapers; (B) General vs business newspapers; (C) Newspapers grouped by countries (the USA, the UK and Spain); (D) Microbiome news/biomedicine news published by SINC (2008–2018) vs microbiome papers/biomedicine papers in PubMed (2007–2019). Microbiome papers in PubMed are presented as a black curve on each graph. https://doi.org/10.1371/journal.pone.0249835.g002 PLOS ONE | https://doi.org/10.1371/journal.pone.0249835 April 9, 2021 5 / 14 PLOS ONE A quantitative content analysis of microbiome research news stories in the press Table 1. Number of news stories on microbiome papers, microbiome papers in PubMed, microbiome news published by SINC and correlations between them. Annual cites from Cites in Cites in Average annual Correlations with microbiome Correlations with microbiome 2007 to 2019 2007 2019 percentage change papers in PubMed1 (p-value) news published by SINC1,2 (p- value) Microbiome papers in 9297,0 (6063.3) 2600 21292 19.6% - 0.62 (0.023) PubMed Biomedicine papers in 1111673,6 785933 1397557 4.9% - - PubMed (203280.1) Microbiome/ 0.8% 0.4% 1.4% 9.6% - - biomedicine in PubMed Microbiome news in 8,1 (5.9) 0 9 24.8% 0.62 (0.023) - SINC2 Biomedicine news in 582.1 (81.1) 666 447 -3.7% - - SINC2 Microbiome/ 1.6% 0.2% 2.2% 19.5% - - biomedicine in SINC2 Total newspapers 4.6 (4.9) 2.3 (2.2) 7.8 (7.5) 13.9% 0.88 (<0.001) 0.66 (0.014) Individual newspapers The New York Times 10.3 (6.4) 5 20 16.0% 0.83 (0.005) 0.48 (0.095) The Times 6.8 (4.4) 5 13 14.3% 0.82 (0.005) 0.47 (0.102) El País 5.1 (4.0) 1 8 22.7% 0.74 (0.004) 0.71 (0.006) The Wall Street Journal 4.1 (1.8) 2 4 2.9% 0.14 (0.652) 0.35 (0.236) Financial Times 1.5 (1.6) 1 2 11.8% 0.39 (0.177) 0.58 (0.038) Expansión 0.2 (0.6) 0 0 4.3% 0.41 (0.166) 0.11 (0.713) Country USA 7.2 (5.6) 3.5 (2.1) 12.0 12.0% 0.85 (0.002) 0.57 (0.039) (11.3) UK 4.1 (4.2) 3.0 (2.8) 7.5 (7.8) 14.5% 0.81 (0.001) 0.57 (0.042) Spain 2.7 (3.7) 0.5 (0.7) 4.0 (5.7) 23.1% 0.75 (0.003) 0.68 (0.010) Newspaper type General newspaper 7.4 (5.4) 3.7 (2.3) 13.7 (6.0) 15.7% 0.91 (<0.001) 0.61 (0.024) Business newspaper 1.9 (2.1) 1.0 (1.0) 2.0 (2.0) 7.2% 0.39 (0.185) 0.56 (0.043) Mean followed by the standard deviation in parentheses is indicated for microbiome/biomedicine papers in PubMed, microbiome/biomedicine news in SINC and news stories on microbiome papers in newspapers. 1 The numbers showed the Pearson correlation coefficient. 2 News stories published by SINC were available from 2008 to 2018. Significant p-values are highlighted in bold. https://doi.org/10.1371/journal.pone.0249835.t001 78.6% in El País. The New York Times showed the most intense microbiome research coverage (10.3 news stories on microbiome papers annually), followed by The Times (6.8 news stories on microbiome papers annually) and, lastly, El País (5.1 news stories on microbiome papers annually) (Table 1). In business newspapers, news stories on microbiome papers represented 56.4% of the over- all number of microbiome news stories for The Wall Street Journal, 52.8% for the Financial Times, and 18.8% for Expansión. The Wall Street Journal was the business newspaper that fea- tured microbiome research the most, followed by the Financial Times (4.1 and 1.5 news stories on microbiome papers annually, respectively). In contrast, news stories on microbiome papers in Expansión were almost null (Table 1). The strong presence of research in news stories about the microbiome in the press was also supported by significant correlations between the number of news stories on microbiome PLOS ONE | https://doi.org/10.1371/journal.pone.0249835 April 9, 2021 6 / 14 PLOS ONE A quantitative content analysis of microbiome research news stories in the press papers in general newspapers and the number of microbiome papers in PubMed (r = 0.91, p < 0.001). The magnitude of that association was greater for the American and British general newspapers (r = 0.85, p = 0.002 and r = 0.81, p = 0.001, respectively) than for their Spanish counterpart (r = 0.75, p = 0.003) (Table 1). The strong interest in microbiome research shown by American newspapers compared to British and Spanish newspapers is also apparent in Fig 2C. As shown in the PubMed curve (Fig 2A–2C), scientific interest in the microbiome gradually grew after 2007 and then picked up speed around 2011. The percentage of microbiome papers available in PubMed and microbiome news published by SINC against the total number of biomedicine publications increased significantly year on year, with a positive annual percent- age change of 9.6% and 19.5%, respectively (Table 1). Of the total biomedical literature avail- able in PubMed from 2008 to 2018, the number of microbiome papers increased from 0.4% to 1.4%. Of all the health and biomedical news stories published by SINC, articles citing the microbiome in the headline went from 0.2% to 2.2% from 2008 to 2018, thus doubling the trends shown by PubMed (Fig 2D and Table 1). The comparison between microbiome news/ biomedicine news published by SINC vs microbiome papers/biomedicine papers in PubMed almost reached statistical significance (p = 0.052). Papers covered in the news and number of press citations In the 361 news stories on microbiome papers, 700 different papers published in scientific journals were cited. Each of those 700 papers was covered in at least one news story in one of the six newspapers. Some papers were covered several times in different news stories from dif- ferent newspapers or from the same newspaper, resulting in a total of 825 press citations of the 700 papers. The most cited papers (8 press citations) were “Artificial sweeteners induce glucose intoler- ance by altering the gut microbiota” (the only paper that was cited in all six newspapers) [42] and “Gut microbiota from twins discordant for obesity modulate metabolism in mice” [43]. Two papers had 7 press citations; 1 paper had 6 citations; 2 papers had 4 citations; 8 papers had 3 citations; 72 papers had 2 citations; and the remaining 613 papers had 1 press citation. The papers were cited in the newspapers generally within three months of publication (64.0% for general newspapers and 48.6% for business newspapers). Study designs of microbiome papers in PubMed and research news The study design of microbiome papers available in PubMed between 2007 and 2019 were as follows: 1.8% were SRs of RCTs in humans; 10.9% were RCTs in humans; 8.5% were human observational studies; 46.5% were environmental & plant studies; 30.4% were animal or labo- ratory studies; and 1.9% had other designs (see methods). Fig 3 illustrates the over-representation (the percentage of microbiome study design in the press was higher than in PubMed) or under-representation (the percentage of microbiome study design in the press was lower than in PubMed) of microbiome study designs in newspa- pers vs PubMed. A common pattern was observed among the general and business newspapers that was characterized by an over-representation of observational studies in humans and an under-representation of environmental & plant studies. In contrast, SRs of RCTs in humans, RCTs in humans and animal or laboratory studies tended to be represented to the same degree in newspapers as in PubMed (Fig 3A and 3B). In terms of study design, no major differences were found between countries (Fig 3B). However, particular features were observed when analyzing newspapers at an individual level. The New York Times showed an over-representation (five-fold increase) of SRs of RCTs in PLOS ONE | https://doi.org/10.1371/journal.pone.0249835 April 9, 2021 7 / 14 PLOS ONE A quantitative content analysis of microbiome research news stories in the press Fig 3. Over-representation and under-representation of microbiome study designs in the press vs PubMed. https://doi.org/10.1371/journal.pone.0249835.g003 humans, while The Times over-represented (seven-fold increase) RCTs compared to PubMed. In contrast, business newspapers covered a similar number of observational studies in humans and animal/laboratory studies (58/181 and 54/181), with the latter over-represented in The Wall Street Journal (Fig 3A). Discussion This is the first study to explore how high-circulation newspapers cover microbiome research in terms of the number of news stories and study design compared to PubMed. Our analysis shows some patterns across newspapers. First, it should be noted that the proportion of papers on the microbiome in relation to all papers in PubMed has increased steadily from 2007 to 2019, with significant year-to-year growth. That proportion is difficult to estimate for newspapers as they do not classify biomedi- cine news separately from other news, but a similar increase is shown in publications by SINC, which categorized biomedical news separately until 2018. Second, an increase in the number of microbiome papers from 2007 to 2019 has resulted in newspapers paying considerable attention to microbiome research, albeit not in a balanced proportion between newspapers. The first peak in press coverage of microbiome research in general newspapers, which took place in 2008, can be explained by the fact that that was the year when large human microbiome research initiatives were launched. They included the first phase of the HMP, the International Human Microbiome Consortium, the EU’s PLOS ONE | https://doi.org/10.1371/journal.pone.0249835 April 9, 2021 8 / 14 PLOS ONE A quantitative content analysis of microbiome research news stories in the press Metagenomics of the Human Intestinal Tract (METAHIT) project and the Canadian Micro- biome Initiative, among others [40]. The subsequent publication of milestone papers in June 2012, reporting on five years of research, triggered an increase in the newspapers’ interest in the microbiome [44]. A news peak in 2013 was observed in The New York Times, with half of the stories covering research into microbes’ role in obesity and cardiovascular disease. That sudden increase also reflects the launch of the second wave of human microbiome projects, including the second phase of the HMP, the European Commission-funded My New Gut pro- gram and the French Government’s MetaGenoPolis program [40]. Third, business newspapers are not as sensitive to the two waves of microbiome research projects as general newspapers. That may have two explanations. On the one hand, micro- biome-related patents have increased less markedly than scientific publications [45]. On the other hand, no microbiome therapeutics requiring US Food and Drug Administration and European Food Safety Authority or European Medicines Agency scrutiny have been approved for human use as yet [46]. Fourth, the American and British newspapers were the ones to mention microbiome papers the most. That is not surprising because the US-based National Institutes of Health (NIH) has provided nearly two-thirds of funding for microbiome research [45]. The findings also reflect American and British journalism’s longer scientific tradition and the countries’ dominant position in the scientific literature [23]. British and American newspapers also echo the top medical journals the most, with The New York Times standing out [23]. Similarly, The New York Times covered the Human Genome Project the most [1]. Fifth, the abundance of observational studies in humans in newspapers may be rooted in the over-representation of this kind of study design in press releases from journals and institu- tions [18,21], which can influence the content of subsequent microbiome news stories [22]. Our results are in agreement with those of Lai and Lane, who found that English-language gen- eral and business newspapers were more likely to cover observational studies and less likely to feature SRs of RCTs, RCTs and animal/laboratory studies [31]. Similar to our findings in gen- eral newspapers, the authors identified a similar percentage of systematic reviews of RCTs and animal/laboratory studies in the press (3% and 17%, respectively), but did not provide a com- parison group of study designs available in PubMed [31]. Previous research showed that, when choosing observational studies, the press covers study designs of a lower quality (such as those with smaller sample sizes) compared to those published in high impact medical journals. That, in turn, might contribute to distorting the end image of medical advances [32]. The remark- able under-representation of environmental & plant studies in the newspapers under analysis might be rooted in the fact that these studies may be less newsworthy because they do not have a direct impact on human health. Indeed, surveys have shown that the topics of greatest inter- est to society are those of medicine and health, with scientific and technological discoveries and the environment and ecology generating far less interest [9,47]. Second, the level of knowl- edge required for journalists to understand and communicate the findings of these kinds of studies in layman’s terms can be higher than that required for observational study types, which could lead to their under-representation in the press. Third, environmental & plant studies are not usually published in top journals, which generally issue press releases, and that can have a negative effect on their impact in newspapers [18–22]. It is also important to acknowledge that the over-representation of observational study designs can distort the public’s perceived image of the microbiome, as those study design types are often reported inaccurately in newspapers and usually do not mention any associated cave- ats and limitations [48]. For instance, while an altered microbiome has been reported in a wide variety of health conditions, such as irritable bowel syndrome, obesity and depression, often, it cannot be determined whether differences in the microbiome are causing the disease PLOS ONE | https://doi.org/10.1371/journal.pone.0249835 April 9, 2021 9 / 14 PLOS ONE A quantitative content analysis of microbiome research news stories in the press or, conversely, if the disease itself is causing the differences [49]. The fact that SRs of RCTs, RCTs and animal or laboratory studies tended to be represented to the same degree in newspa- pers compared to PubMed, with animal or laboratory studies being the second type of study design most cited in PubMed, reflects how microbiome research is still mainly based on basic science. It is worth highlighting that the reporting of SRs of RCTs and RCTs in The New York Times and The Times, respectively, may be seen as an indicator of good quality journalism. It should also be noted that the USA and the UK are the countries that produce the most Cochrane SRs [50] and that newspapers tend to report more on domestically produced science [23]. On the other hand, the over-representation of animal/laboratory studies in the business newspapers, which is mainly down to The Wall Street Journal, is expected as preclinical micro- biome research (representing 30.4% of microbiome studies in PubMed) is the first step towards developing microbiome therapies, and that is where most companies’ initial efforts begin. Only a handful of microbiome-related products have entered the end phase of clinical trials [46] and that is reflected in reduced coverage of RCTs in the business press. Beyond the study design of microbiome papers, the fact that we focused on influential newspapers in terms of readership and circulation might explain the intense coverage of scien- tific articles about the microbiome, as these newspapers usually have large science and medi- cine sections with specialist reporters [23]. Other factors might also explain why some microbiome papers are finally echoed by newspapers. They include the impact factor of the journal [16,17], the availability of press releases [18–22], the domestic preference of newspa- pers for journals from their own country [23], and the newsworthiness of the topic [24]. The last factor is especially relevant in the case of the microbiome, due to the ever-increasing inter- est among both researchers and the lay public in targeting the microbiome to maintain health and quality of life [4]. An overall analysis considering all of these factors is needed to better elu- cidate how microbiome research is echoed in the media. In the light of the ever-increasing amount of research about the link between the micro- biome and human health and disease [4], one of the field’s urgent needs is precisely that of ensuring unbiased communication of microbiome research to the general public. In that regard, some journals published by the BMJ group indicate the evidence type and subjects studied to journalists when sending embargoed press releases, which may help inform report- ing on microbiome findings [51,52]. Keeping up with the huge amount of research and publi- cations on the microbiome and receiving training in science communication skills are also necessary for communicating microbiome research with caution and free from misinterpreta- tion [4]. The European Society of Neurogastroenterology & Motility’s Gut Microbiota for Health platform (https://www.gutmicrobiotaforhealth.com/) is an example of a project that aims to translate the latest research on the rapidly-evolving field of the gut microbiome for both the scientific community and the lay public. Although scientific interest in microbiome research has driven an increase in news stories based on research findings, the patterns observed in study design coverage need to be tracked in order to inform on the evolution of the science behind the current microbial momentum being experienced by society. For microbiome scientists, the coverage of their research in newspapers and its dissemination in social media can improve their visibility and scientific citations. It is also important to acknowledge the potential role of media coverage in obtaining research funding, without forgetting that, although mentions of scholarly outputs on social media and news sites are becoming increasingly present in policy papers and research calls, it is too early to consider whether they contribute to the awarding of research funding [53]. One strength of our study is that we focused on both general and business newspapers ranked high in circulation from different countries over a long period of 12 years. Rather than gauging only news stories considered immediately newsworthy (that is, generated in response PLOS ONE | https://doi.org/10.1371/journal.pone.0249835 April 9, 2021 10 / 14 PLOS ONE A quantitative content analysis of microbiome research news stories in the press to a paper within 2 months of its publication), we analyzed all news stories on microbiome papers regardless of paper publication date. In addition, we analyzed the number of news sto- ries on microbiome papers and study designs reported in newspapers vs patterns in PubMed. Our study also has limitations. We did not focus on studying the impact of microbiome research in other mass media such as low-circulation newspapers, magazines, radio, television or the Internet. Moreover, our selection of international general and business newspapers is not representative of the general and business press around the world, even though our selec- tion includes some of the most widely read and best quality international newspapers. Further- more, our analysis only focuses on quantitative aspects. As a result, the study of all representations of the microbiome in the selected newspapers, regardless of whether they cite a scientific study, is limited in scope and deserves the application of a qualitative methodology that is outside the scope of our research objectives. Other authors have previously addressed newspaper coverage of the microbiome based on qualitative aspects, such as the tone of the discourse [54] and language employed to discuss advances in the microbiome [55], highlight- ing the need for microbiology literacy in society due to the role of microbes in the health of our planet [56]. Finally, analyzing the press citations of authors, papers or journals also has its limitations, given that their mentioning in newspaper articles does not provide any informa- tion about the context of the citation or the quality of the journalistic text. Conclusions Our results show that the microbiome is receiving increasing attention in both research and the press. News stories on the microbiome in both the general and business press during the period under study were mostly based on research findings. While the volume of microbiome- based scientific studies in the press mirrors the number of scientific papers in PubMed, the choice of studies covered by general and business newspapers over-represent observational studies and under-represent environmental & plant studies, while showing a similar degree of representation for SRs of RCTs, RCTs and animal or laboratory studies. Supporting information S1 File. Factiva search filters and phrases. (DOC) S2 File. PubMed search filters and phrases. (DOC) S1 Dataset. Data acquired and used for analyses. (XLSX) Acknowledgments We wish to thank Mireia Bosch, Juan Carlos Martı́n and Queralt Miró for their support with data management and statistical analyses, and Marı́a Garcı́a-Puente, Alicia Jarillo and Marta Diaz for their assistance with searches in PubMed. We also appreciate the critical insights received from Prof. Paul Enck while preparing the manuscript. Author Contributions Conceptualization: Andreu Prados-Bo, Gonzalo Casino. Data curation: Andreu Prados-Bo. PLOS ONE | https://doi.org/10.1371/journal.pone.0249835 April 9, 2021 11 / 14 PLOS ONE A quantitative content analysis of microbiome research news stories in the press Formal analysis: Andreu Prados-Bo, Gonzalo Casino. Investigation: Andreu Prados-Bo. Methodology: Andreu Prados-Bo, Gonzalo Casino. Project administration: Andreu Prados-Bo. Supervision: Gonzalo Casino. Validation: Andreu Prados-Bo, Gonzalo Casino. Visualization: Andreu Prados-Bo. Writing – original draft: Andreu Prados-Bo. Writing – review & editing: Gonzalo Casino. References 1. Costa T. The Human Genome Project and the media. Case study: the relation between genetics and the media. J Sci Commun. 2003; 2(1): 1–20. https://doi.org/10.22323/2.02010203 2. Marchesi JR, Ravel J. The vocabulary of microbiome research: a proposal. Microbiome. 2015; 3: 31. https://doi.org/10.1186/s40168-015-0094-5 PMID: 26229597 3. Prescott SL. History of medicine: origin of the term microbiome and why it matters. Hum Microbiome J. 2017; 4: 24–25. https://doi.org/10.1016/j.humic.2017.05.004 4. Shan Y, Segre JA, Chang EB. Responsible stewardship for communicating microbiome research to the press and public. Nat Med. 2019; 25(6): 872–874. https://doi.org/10.1038/s41591-019-0470-y PMID: 31133694 5. Thomas AM, Segata N. Multiple levels of the unknown in microbiome research. BMC Biol. 2019; 17(1): 48. https://doi.org/10.1186/s12915-019-0667-z PMID: 31189463 6. Zmora N, Soffer E, Elinav E. Transforming medicine with the microbiome. Sci Transl Med. 2019; 11 (477): eaaw1815. https://doi.org/10.1126/scitranslmed.aaw1815 PMID: 30700573 7. Singh BK, Trivedi P. Microbiome and the future for food and nutrient security. Microb Biotechnol. 2017; 10(1): 50–53. https://doi.org/10.1111/1751-7915.12592 PMID: 28074557 8. Cavicchioli R, Ripple WJ, Timmis KN, Azam F, Bakken LR, Baylis M, et al. Scientists’ warning to humanity: Microorganisms and climate change. Nat Rev Microbiol. 2019; 17(9): 569–586. https://doi. org/10.1038/s41579-019-0222-5 PMID: 31213707 9. National Science Board. Science & engineering indicators 2018. In: Science and technology: Public atti- tudes and understanding. Alexandria, VA: National Science Foundation; 2018. pp. 7–1–7–92. Retrieved from: https://www.nsf.gov/statistics/2018/nsb20181/report/sections/science-and-technology- public-attitudes-and-understanding/highlights. 10. Grilli R, Ramsay C, Minozzi S. Mass media interventions: effects on health services utilisation. Cochrane Database Syst Rev. 2002; 1: CD000389. https://doi.org/10.1002/14651858.CD000389 11. Baethge C, Engels M. Citations count-even in the lay press: it is far from true that German science jour- nalists only cite English language medical journals. An evaluation of the citation habits of the FAZ, the Spiegel, the SZ, the Welt, and the Zeit. Dtsch Arztebl Int. 2009; 106(25): 413–415. https://doi.org/10. 3238/arztebl.2009.0413 PMID: 19623309 12. Phillips DP, Kanter EJ, Bednarczyk B, Tastad PL. Importance of the lay press in the transmission of medical knowledge to the scientific community. N Eng J Med. 1991; 325(16): 1180–1183. https://doi. org/10.1056/NEJM199110173251620 PMID: 1891034 13. Kiernan V. Diffusion of news about research. Sci Commun. 2003; 25(1): 3–13. https://doi.org/10.1177/ 1075547003255297 14. Fanelli D. Any publicity is better than none: newspaper coverage increases citations, in the UK more than in Italy. Scientometrics. 2013; 95: 1167–1177. https://doi.org/10.1007/s11192-012-0925-0 15. Mathelus S, Pittman G, Yablonski-Crepeau J. Promotion of research articles to the lay press: a sum- mary of a three-year project. Learned Publishing. 2012; 25: 207–212. https://doi.org/10.1087/20120307 16. Dumas-Mallet E, Garenne A, Boraud T, Gonon F. Does newspaper coverage influence the citations count of scientific publications? An analysis of biomedical studies. Scientometrics. 2020; 123: 413–427. https://doi.org/10.1007/s11192-020-03380-1 PLOS ONE | https://doi.org/10.1371/journal.pone.0249835 April 9, 2021 12 / 14 PLOS ONE A quantitative content analysis of microbiome research news stories in the press 17. Stephenson J. Genetics and journalism. In: Thompson AK, Chadwick RF, editors. Genetic information. Boston: Springer; 1999. pp. 201–205. 18. Bartlett C, Sterne J, Egger M. What is newsworthy? Longitudinal study of the reporting of medical research in two British newspapers. BMJ. 2002; 325(7355): 81–84. https://doi.org/10.1136/bmj.325. 7355.81 PMID: 12114239 19. Casino G. Información de las revistas de biomedicina mediada por comunicados de prensa. El caso del diario El Paı́s. Panace@. 2015; 16(42): 151–157. 20. De Semir V, Ribas C, Revuelta G. Press releases of science journal articles and subsequent newspaper stories on the same topic. JAMA. 1998; 280(3): 294–295. https://doi.org/10.1001/jama.280.3.294 PMID: 9676688 21. Wang MTM, Bolland MJ, Gamble G, Grey A. Media coverage, journal press releases and editorials associated with randomized and observational studies in high-impact medical journals: a cohort study. PLoS One. 2015; 10(12): e0145294. https://doi.org/10.1371/journal.pone.0145294 PMID: 26701758 22. Schwartz LM, Woloshin S, Andrews A, Stukel TA. Influence of medical journal press releases on the quality of associated newspaper coverage: retrospective cohort study. BMJ. 2012; 344: d8164. https:// doi.org/10.1136/bmj.d8164 PMID: 22286507 23. Casino G, Rius R, Cobo E. National citation patterns of NEJM, The Lancet, JAMA and The BMJ in the lay press: A quantitative content analysis. BMJ Open. 2017; 7(11): e018705. https://doi.org/10.1136/ bmjopen-2017-018705 PMID: 29133334 24. Shoemaker PJ. News and newsworthiness: a commentary. Communications. 2006; 31: 105–111. https://doi.org/10.1515/commun.2006.007 25. Wolf C, Schnauber A. News consumption in the mobile era. Digital Journalism. 2015; 3(5): 759–776. https://doi.org/10.1080/21670811.2014.942497 26. Pew Research Center. Newspapers fact sheet. 2019 July 9 [cited 30 September 2020]. In: State of the news media [Internet]. Washington, DC: Pew Research Center. [about 10 screens]. Available from: https://www.journalism.org/fact-sheet/newspapers/. 27. Brossard D. New media landscapes and the science information consumer. Proc Natl Acad Sci USA. 2013; 110(Suppl 3):14096–14101. https://doi.org/10.1073/pnas.1212744110 PMID: 23940316 28. Welbers K, Opgenhaffen M. Presenting news on social media. Digital Journalism. 2018; 7(1): 45–62. https://doi.org/10.1080/21670811.2018.1493939 29. Bergström A, Jervelycke Belfrage M. News in social media. Digital Journalism. 2018; 6(5): 583–598. https://doi.org/10.1080/21670811.2018.1423625 30. Casino G. Cita periodı́stica: impacto de las revistas y los artı́culos cientı́ficos en la prensa generalista. El profesional de la información. 2018; 27(3): 692–697. https://doi.org/10.3145/epi.2018.may.22 31. Lai WYY, Lane T. Characteristics of medical research news reported on front pages of newspapers. PLoS One. 2009; 4(7): e6103. https://doi.org/10.1371/journal.pone.0006103 PMID: 19568422 32. Selvaraj S, Borkar DS, Prasad V. Media coverage of medical journals: Do the best articles make the news? PLoS One. 2014; 9(1): e85355. https://doi.org/10.1371/journal.pone.0085355 PMID: 24465543 33. Lynch SV, Ng SC, Shanahan F, Tilg H. Translating the gut microbiome: Ready for the clinic? Nat Rev Gastroenterol Hepatol. 2019; 16(11): 656–661. https://doi.org/10.1038/s41575-019-0204-0 PMID: 31562390 34. Gonon F, Konsman JP, Cohen D, Boraud T. Why most biomedical findings echoed by newspapers turn out to be false: the case of attention deficit hyperactivity disorder. PLoS One. 2012; 7(9): e44275. https://doi.org/10.1371/journal.pone.0044275 PMID: 22984483 35. Dumas-Mallet E, Smith A, Boraud T, Gonon F. Poor replication validity of biomedical association stud- ies reported by newspapers. PLoS One. 2017; 12(2): e0172650. https://doi.org/10.1371/journal.pone. 0172650 PMID: 28222122 36. Pallari E, Lewison G, Sullivan R. How is chronic non-communicable respiratory conditions research reported in European newspapers? An impact assessment for policy. Clin Respir J. 2017; 11(5): 657– 665. https://doi.org/10.1111/crj.12685 PMID: 28779548 37. Dumas-Mallet E, Smith A, Boraud T, Gonon F. Scientific uncertainty in the press: How newspapers describe initial biomedical findings. Sci Commun. 2018; 40(1): 124–141. https://doi.org/10.1177/ 1075547017752166 38. National Institutes of Health. NIH launches Human Microbiome Project. 2007 Dec 19 [cited 13 May 2020]. In: NIH News & Events [Internet]. Maryland: NIH. [about 2 screens]. Available from: https:// www.nih.gov/news-events/news-releases/nih-launches-human-microbiome-project. 39. Guasch B, Cortiñas S, González M, Justel-Vázquez S, Peña J. The representation of graphene in the online press of the United States, the United Kingdom, and Spain. Int J Commun. 2019; 13: 966–990. PLOS ONE | https://doi.org/10.1371/journal.pone.0249835 April 9, 2021 13 / 14 PLOS ONE A quantitative content analysis of microbiome research news stories in the press 40. Stulberg E, Fravel D, Proctor LM, Murray DM, LoTempio J, Chrisey L, et al. An assessment of US microbiome research. Nat Microbiol. 2016; 1: 15015. https://doi.org/10.1038/nmicrobiol.2015.15 PMID: 27571759 41. Garcı́a-Puente M, Pastor-Ramon E, Oskia A, Morán JM, Herrera-Peco I. Research note. Open letter to the users of the new PubMed: a critical appraisal. El profesional de la información. 2020; 29(3): e290336. https://doi.org/10.3145/epi.2020.may.36 42. Suez J, Korem T, Zeevi D, Zilberman-Schapira G, Thaiss CA, Maza O, et al. Artificial sweeteners induce glucose intolerance by altering the gut microbiota. Nature. 2014; 514(7521): 181–186. https:// doi.org/10.1038/nature13793 PMID: 25231862 43. Ridaura VK, Faith JJ, Rey FE, Cheng J, Duncan AE, Kau AL, et al. Gut microbiota from twins discordant for obesity modulate metabolism in mice. Science. 2013; 341(6150): 1241214. https://doi.org/10.1126/ science.1241214 PMID: 24009397 44. National Human Genome Research Institute. Human Microbiome Project (HMP) telebriefing resources. 2014 Mar 20 [cited 15 May 2020]. In: National Human Genome Research Institute News & Events [Internet]. Bethesda: NHGRI. [about 3 screens]. Available from: https://www.genome.gov/27549115/ human-microbiome-project-telebriefing-resources. 45. Jones S. Trends in microbiome research. Nat Biotechnol. 2013; 31: 277. https://doi.org/10.1038/nbt. 2546 46. Taroncher-Oldenburg G, Jones S, Blaser M, Bonneau R, Christey P, Clemente JC, et al. Translating microbiome futures. Nat Biotechnol. 2018; 36(11): 1037–1042. https://doi.org/10.1038/nbt.4287 PMID: 30412201 47. Fundación Española para la Ciencia y la Tecnologı́a (FECYT). Percepción social de la ciencia y la tec- nologı́a 2018. 2019 [Cited 2021 February 28]. Available from: https://icono.fecyt.es/sites/default/files/ filepublicaciones/20/percepcion_social_de_la_ciencia_y_la_tecnologia_2018_completo_0.pdf. 48. Schwitzer G. Observations about today’s observational studies in the news. 2018 Aug 28 [cited 9 Sep- tember 2020]. In: HealthNewsReview.org [Internet]. Minneapolis: 2006–2020. [about 1 screen]. Avail- able from: https://www.healthnewsreview.org/2018/08/observations-about-todays-observational- studies-in-the-news/. 49. Bik EM. The hoops, hopes, and hypes of human microbiome research. Yale J Biol Med. 2016; 89(3): 363–373. PMID: 27698620 50. Groneberg DA, Rolle S, Bendels MHK, Klingelhöfer D, Schöffel N, Bauer J, et al. A world map of evi- dence-based medicine: density equalizing mapping of the Cochrane database of systematic reviews. PLoS One. 2019; 14(12): e0226305. https://doi.org/10.1371/journal.pone.0226305 PMID: 31834918 51. British Medical Journal. Mediterranean diet promotes gut bacteria linked to ‘healthy ageing’ in older peo- ple. 2020 Feb 17 [cited 15 May 2020]. In: BMJ Newsroom [Internet]. London: BMJ. [about 3 screens]. Available from: https://www.bmj.com/company/newsroom/mediterranean-diet-promotes-gut-bacteria- linked-to-healthy-ageing-in-older-people/. 52. British Medical Journal. Probiotics alone or combined with prebiotics may help ease depression. 2020 July 6 [cited 9 July 2020]. In: BMJ Newsroom [Internet]. London: BMJ. [about 2 screens]. Available from: https://www.bmj.com/company/newsroom/probiotics-alone-or-combined-with-prebiotics-may- help-ease-depression/. 53. Fraumann G. Discussing indicators in research funding: what role do altmetrics play? 2018 July 3 [cited 1 October 2020]. In: Oslo Institute for Research on the Impact of Science blog [Internet]. Oslo: Univer- sity of Oslo. [about 5 screens]. Available from: https://www.sv.uio.no/tik/english/research/centre/osiris/ osirisblog/discussing-indicators-in-research-funding%3A—what-r.html. 54. Chuong KH, O’Doherty KC, Secko DM. Media discourse on the social acceptability of fecal transplants. Qual Health Res. 2015; 25(10): 1359–1371. https://doi.org/10.1177/1049732314568199 PMID: 25595150 55. Nerlich B, Hellsten I. Beyond the human genome: microbes, metaphors and what it means to be human in an interconnected post-genomic world. New Genet Soc. 2007; 28: 19–36. https://doi.org/10.1080/ 14636770802670233 56. Timmis K, Cavicchioli R, Garcia JL, Nogales B, Chavarrı́a M, Stein L, et al. The urgent need for microbi- ology literacy in society. Environ Microbiol. 2019; 21(5): 1513–1528. https://doi.org/10.1111/1462-2920. 14611 PMID: 30912268 PLOS ONE | https://doi.org/10.1371/journal.pone.0249835 April 9, 2021 14 / 14