Wikipedia is a freely available online encyclopedia that intends to provide “every single person on the planet free access to the sum of all human knowledge” (Wikimedia Foundation, 2004). To meet this mission, thousands of volunteer editors have created almost six million English-language Wikipedia pages. Many of these pages include hyperlinked footnotes to the sources used to assemble and verify the content (Redi et al., 2018; Wikipedia Foundation, 2019a). These linked references do not only serve as sources of authority for Wikipedia content (Fallis, 2008), but also offer readers a gateway to further learning, with this opportunity enhanced by the growing degree of public access to research literature (Piwowar et al., 2018).

Wikipedia is proving to be a leading source of health information (Heilman and West, 2015; Laurent and Vickers, 2009), and reader engagement with a page’s references can provide opportunities to understand a diagnosis or inform a conversation with their physician. Wikipedia’s health-focused pages, which are maintained by WikiProject Medicine (WPM) editors, are thought to meet a high standard of quality and rigor (James, 2016; Maskalyk, 2014; Trevena, 2011). For example, WPM provides a series of resources for its editors, including “guidelines and policies” for reliable sources, and advice on how to avoid conflicts of interest. In its advice for reliable sources, WPM recommends using “review articles (especially systematic reviews) published in reputable medical journals” (Wikipedia Foundation, 2019b). Wikipedia then provides tools to standardize the bibliographic representation and linking to external sources.

These external sources are particularly relevant for students and practitioners in medicine and other health-related professions, who are active Wikipedia readers (Rössler et al., 2015; Scaffidi et al., 2017; Back et al., 2016; Egle et al., 2015; Allahwala et al., 2013). These readers are familiar with the research literature, and expected to engage in evidence-based medicine (Guyatt et al., 2015). To this end, faculty at several health professions schools, such as schools of medicine and pharmacy, teach courses on editing Wikipedia (Azzam et al., 2017; Joshi et al., 2019; Apollonio et al., 2018).

In a previous study, we investigated Wikipedia as a gateway to further reading through external links by tracking referrals from Wikipedia to research article DOIs (Digital Object Identifier) through Crossref, combined with pageview information from Wikimedia (Maggio et al., 2017). However, with the data available at the time, we could not determine if a click to an external reference with a DOI originated from a WPM page.

To learn more about reader engagement with Wikipedia’s external references, this study leverages Wikimedia’s newly created infrastructure for data collection to compare engagement with external links by readers of WPM pages with that of readers of the rest of Wikipedia. The study is guided by two research questions:

• RQ 1: To what extent do WPM pages differ on average from the rest of Wikipedia pages?

• RQ 2: To what extent do the behaviors of WPM readers with external links differ on average from readers’ behaviors with the rest of Wikipedia?

With the approval and support of the Wikimedia Foundation (WMF), we collected the data presented in this study between March 22^nd – April 22^nd, 2019 from Wikimedia’s Event Logging system and the production MediaWiki database. The data remained within that system, as required by WMF, for a year before being deleted (see Acknowledgements). We anonymized the aggregated data by removing IP addresses, any identifying browser information, and reader-sessions associated with page edits. While this data is not publicly available, data may be available upon request from the WMF research team. To gain access to the data, researchers should review the WMF Research team’s current procedures for data requests (https://www.mediawiki.org/wiki/wikimedia_research/research_and_data). The WMF Research Team will evaluate the request based on Wikimedia’s data access criteria. The code utilized to collect and analyze the data, however, is organized and made publicly available in a collated series of Jupyter notebooks in GitHub (Steinberg and Picardi, 2019; copy archived here).

The Wikipedia pages maintained by WPM are longer, possess a greater density of external links (references), and are viewed considerably more often, on average, than the pages on the rest of Wikipedia (Table 6). The popularity of the pages also speaks to the valuable contribution that WPM makes to Wikipedia’s role as a source of health and medical information, as well as the level of trust that WPM readers have in Wikipedia’s coverage of health and medical topics.

In addition, the readers of WPM pages not only engage more often with external links, but spend longer on the page before engaging with the links, compared to readers of the rest of Wikipedia (Table 6). WPM readers are more likely to hover over references, causing them to appear in a window, and more likely to view the footnotes than other readers. But readers of WPM pages click on external links less often than readers of other parts of Wikipedia.

This behavior pattern suggests that readers are more interested in validating the external references. This interest may arise from readers who are relatively new to these pages and are in the process of reassuring themselves as to the scientific basis of the content they are consulting. But it may also arise from readers who are trained, or are in the process of being trained, in health professions where they are required to critically appraise and integrate health information into the care of their patients. These evidence-based practitioners are able to judge a lot about the quality of information from the bibliographic reference to the sources, which they view by hovering over or going to the footnote. In addition to being able to see at a glance (or a hover) whether the cited work is a review (which is recommend by Wikipedia as being the “ideal source”), readers are able to identify publication date, language, publication venue (i.e., journal name), and availability (via open access) (Figure 3A).

Figure 3

Download asset Open asset

In addition, in some instances it is possible to obtain from a citation details on patient populations, medical interventions (e.g., treatments, diagnostic tests), and outcomes involved in the study or whether the source is a standard medical textbook in the field (Figure 3B). The greater attention that WPM readers are paying to citations could well be a function of their educational backgrounds, reflecting the demonstrated use of WPM by health professionals, students and practitioners. Although not the focus of this study, the educational value of internal links, in which terms on a WPM page are hyperlinked to other Wikipedia pages or to VideoWiki, should also be noted as a further strategy for maintaining the comprehensive informational and educational quality of WPM pages.

This still leaves the question of why WPM readers are less likely to click through to the external source than readers of the rest of Wikipedia. This question is particularly relevant given the interest demonstrated by WPM readers in viewing the citations (by hovering or clicking on footnote numbers), as well as by the studies (cited above) demonstrating that health professionals use WPM pages, which suggests a potential capacity for further learning through the external links. One possibility is that readers are deterred from clicking on external WPM links in light of our finding that only a fifth of the research cited in WPM pages are publicly accessible, with the rest placed by publishers behind a paywall for subscribers or credit card access. Earlier studies have found that it is common for readers, including physicians, who encounter a paywall to be reluctant to explore the literature further (Moorhead et al., 2015; Maggio et al., 2016). It is possible that if all the research literature was available through open access (and this fact was promoted among readers) health professionals may be more likely to pursue the greater educational advantage of the external links. Currently, there are very few links bearing an open access icon, and readers show no additional interest in viewing these links just because they are open. If readers start expecting that external links will be open (and there is some promotion of this fact), then the two studies just cited suggest that the chances of them pursuing an item of interest will become more likely.

Although this study does not provide a means of determining why readers hover rather than click on external links, this is something that could be addressed using smaller-scale research strategies like reader questionnaires or think-aloud protocols. Alternatively, design experiments, which experiment with external link format and context, could be conducted involving, (a) WPM pages in which all external links are clearly indicated as open access, (b) tested readers are given prior training in how to learn from cited works, and (c) how different mobile interfaces impact the amount research studies are clicked on for further reading.

One of the limitations of this study is the way in which the count for “hovering” (defined as a reader’s cursor lingering over a link for 1000 milliseconds or more, revealing a rollover label) captures both intentional and unintentional acts. So, while the reported number of hovers is of limited value, there is no reason to believe that the incidental hovers would differ between WPM and W readers. As well, with the pageview data, various strategies were used to exclude pages visited by bots, but the limited effectiveness of these strategies is a known weakness in Wikipedia’s infrastructure. It, again, suggests a tempering of the pageview counts but not the ratios between WPM and W in this regard. This is similarly the case with the exclusion from this study of interwiki links that reduced external link counts, as described in the methods. Lastly, this study focuses only on the English-language version of Wikipedia and therefore our findings are limited to this version of the encyclopedia.

The study has identified a number of distinctions that set WPM pages apart from the rest of Wikipedia. These differences, such as page length and link density, reflect the quality of these pages as information sources. This finding was further supported by differences in reader behavior, such as time spent on page, and amount of external link engagement, particularly the rate readers hover or click on footnote numbers to examine the bibliographic information on the sources contained in the links. This suggests that WPM readers are more engaged in assessing the citation of the external link, which assists in validating and building trust in the content that makes up the WPM pages, than by the opportunity to visit the sources as a gateway for further learning. Subsequent studies of readers engagement with Wikipedia’s external links should investigate the role citations play in what readers learn about a topic and whether there is the potential, especially for readers in health professions, to increase the degree and depth of learning.

The views expressed in this article are those of the authors and do not necessarily reflect the official policy or position of the Uniformed Services University of the Health Sciences, the U.S. Department of Defense, or the U.S. Government.

With the approval and support of the Wikimedia Foundation (WMF), we accessed the data presented in this study via the Wikimedia's Event Logging system and the production MediaWiki database. Due to WMF data policies intended to protect the privacy of readers and editors, all of the data collected and analyzed has had to remain within their protected data environment. Therefore, the data is not publicly available. While this data is not publicly available, data may be available upon request from the WMF research team. To gain access to the data, researchers should review the WMF Research Team's current procedures for data requests (https://www.mediawiki.org/wiki/wikimedia_research/research_and_data). The WMF Research Team will evaluate the request based on Wikimedia's data access criteria. The code utilized to collect and analyze the data, however, is organized and made publicly available in a collated series of Jupyter notebooks at: https://github.com/ryanmax/wiki-citation-usage/ (Steinberg et al., 2019; copy archived at https://github.com/elifesciences-publications/wiki-citation-usage).

1. 1. Allahwala UK
   2. Nadkarni A
   3. Sebaratnam DF

   (2013) Wikipedia use amongst medical students - new insights into the digital revolution

   Medical Teacher 35:337.

   https://doi.org/10.3109/0142159X.2012.737064

   • PubMed
   • Google Scholar
2. 1. Apollonio DE
   2. Broyde K
   3. Azzam A
   4. De Guia M
   5. Heilman J
   6. Brock T

   (2018) Pharmacy students can improve access to quality medicines information by editing Wikipedia articles

   BMC Medical Education 18:265.

   https://doi.org/10.1186/s12909-018-1375-z

   • PubMed
   • Google Scholar
3. 1. Azzam A
   2. Bresler D
   3. Leon A
   4. Maggio L
   5. Whitaker E
   6. Heilman J
   7. Orlowitz J
   8. Swisher V
   9. Rasberry L
   10. Otoide K
   11. Trotter F
   12. Ross W
   13. McCue JD

   (2017) Why medical schools should embrace Wikipedia: final-year medical student contributions to Wikipedia articles for academic credit at one school

   Academic Medicine 92:194–200.

   https://doi.org/10.1097/ACM.0000000000001381

   • PubMed
   • Google Scholar
4. 1. Back DA
   2. Behringer F
   3. Haberstroh N
   4. Ehlers JP
   5. Sostmann K
   6. Peters H

   (2016) Learning management system and e-learning tools: an experience of medical students' usage and expectations

   International Journal of Medical Education 7:267–273.

   https://doi.org/10.5116/ijme.57a5.f0f5

   • PubMed
   • Google Scholar
5. 1. Egle JP
   2. Smeenge DM
   3. Kassem KM
   4. Mittal VK

   (2015) The internet school of medicine: use of electronic resources by medical trainees and the reliability of those resources

   Journal of Surgical Education 72:316–320.

   https://doi.org/10.1016/j.jsurg.2014.08.005

   • Google Scholar
6. 1. Fallis D

   (2008) Toward an epistemology of Wikipedia

   Journal of the American Society for Information Science and Technology 59:1662–1674.

   https://doi.org/10.1002/asi.20870

   • Google Scholar
7. 1. Guyatt G
   2. Cairns J
   3. Churchill D
   4. Cook D
   5. Haynes B
   6. Hirsh J
   7. Irvine J
   8. Levine M
   9. Nishikawa J
   10. Sackett D

   (2015) Evidence-based medicine: a new approach to teaching the practice of medicine

   JAMA 268:2420–2425.

   https://doi.org/10.1001/jama.1992.03490170092032

   • Google Scholar
8. 1. Heilman JM
   2. West AG

   (2015) Wikipedia and medicine: quantifying readership, editors, and the significance of natural language

   Journal of Medical Internet Research 17:e62.

   https://doi.org/10.2196/jmir.4069

   • PubMed
   • Google Scholar
9. 1. James R

   (2016) WikiProject Medicine: creating credibility in consumer health

   Journal of Hospital Librarianship 16:344–351.

   https://doi.org/10.1080/15323269.2016.1221284

   • Google Scholar
10. 1. Joshi M
    2. Verduzco R
    3. Yogi S
    4. Garcia M
    5. Saxena S
    6. Tackett S
    7. Dexter N
    8. Dempsey A
    9. Whitaker E
    10. Azzam A

    (2019) Wikipedia editing courses at three US medical schools in the 2017-2018 academic year

    MedEdPublish 8:8.

    https://doi.org/10.15694/mep.2019.000146.1

    • Google Scholar
11. 1. Laurent MR
    2. Vickers TJ

    (2009) Seeking health information online: does Wikipedia matter?

    Journal of the American Medical Informatics Association 16:471–479.

    https://doi.org/10.1197/jamia.M3059

    • PubMed
    • Google Scholar
12. 1. Maggio LA
    2. Moorhead LL
    3. Willinsky JM

    (2016) Qualitative study of physicians' varied uses of biomedical research in the USA

    BMJ Open 6:e012846.

    https://doi.org/10.1136/bmjopen-2016-012846

    • PubMed
    • Google Scholar
13. 1. Maggio LA
    2. Willinsky JM
    3. Steinberg RM
    4. Mietchen D
    5. Wass JL
    6. Dong T

    (2017) Wikipedia as a gateway to biomedical research: the relative distribution and use of citations in the English Wikipedia

    PLOS ONE 12:e0190046.

    https://doi.org/10.1371/journal.pone.0190046

    • PubMed
    • Google Scholar
14. Website

    1. Maggio LA
    2. Moorhead LL
    3. Willinsky JM

    (2018) Investigating Wikipedia’s roles as a gateway to medical content

    Accessed September 29, 2019.

    https://meta.wikimedia.org/wiki/Research:Investigating_Wikipedia%27s_role_as_a_gateway_to_medical_content
15. 1. Maskalyk J

    (2014)

    Modern medicine comes online: how putting Wikipedia articles through a medical journal's traditional process can put free, reliable information into as many hands as possible

    Open Medicine 8:e116–e119.

    • PubMed
    • Google Scholar
16. 1. Moorhead LL
    2. Holzmeyer C
    3. Maggio LA
    4. Steinberg RM
    5. Willinsky J

    (2015) In an age of open access to research policies: physician and public health NGO staff research use and policy awareness

    PLOS ONE 10:e0129708.

    https://doi.org/10.1371/journal.pone.0129708

    • PubMed
    • Google Scholar
17. 1. Piwowar H
    2. Priem J
    3. Larivière V
    4. Alperin JP
    5. Matthias L
    6. Norlander B
    7. Farley A
    8. West J
    9. Haustein S

    (2018) The state of OA: a large-scale analysis of the prevalence and impact of open access articles

    PeerJ 6:e4375.

    https://doi.org/10.7717/peerj.4375

    • PubMed
    • Google Scholar
18. Website

    1. Redi M
    2. Taraborelli D
    3. Orlowitz J

    (2018) How many Wikipedia references are available to read? We measured the proportion of open access sources across languages and topics

    Accessed September 29, 2019.

    https://wikimediafoundation.org/news/2018/08/20/how-many-wikipedia-references-are-available-to-read/
19. Website

    1. Redi M

    (2019) New results and third experiment to understand wikipedia citation usage

    Accessed September 29, 2019.

    https://en.wikipedia.org/wiki/Wikipedia:Village_pump_(miscellaneous)/Archive_61#New_Results_and_third_experiment_to_understand_Wikipedia_citation_usage
20. 1. Rössler B
    2. Holldack H
    3. Schebesta K

    (2015) Influence of Wikipedia and other web resources on acute and critical care decisions. A web-based survey

    Intensive Care Medicine Experimental 3:A867.

    https://doi.org/10.1186/2197-425X-3-S1-A867

    • Google Scholar
21. 1. Scaffidi MA
    2. Khan R
    3. Wang C
    4. Keren D
    5. Tsui C
    6. Garg A
    7. Brar S
    8. Valoo K
    9. Bonert M
    10. de Wolff JF
    11. Heilman J
    12. Grover SC

    (2017) Comparison of the impact of Wikipedia, UpToDate, and a digital textbook on short-term knowledge acquisition among medical students: randomized controlled trial of three web-based resources

    JMIR Medical Education 3:e20.

    https://doi.org/10.2196/mededu.8188

    • PubMed
    • Google Scholar
22. Software

    1. Steinberg RM
    2. Picardi T

    (2019) Wiki_citation_usage project

    Github.

    https://github.com/ryanmax/wiki-citation-usage/
23. 1. Trevena L

    (2011) WikiProject Medicine

    BMJ 342:d3387.

    https://doi.org/10.1136/bmj.d3387

    • PubMed
    • Google Scholar
24. Website

    1. Wikimedia Foundation

    (2004) Wikimedia Foundation Vision Statement

    Accessed September 29, 2019.

    https://wikimediafoundation.org/wiki/Vision
25. Website

    1. Wikipedia Foundation

    (2019a) Wikimedia statistics

    Accessed September 29, 2019.

    https://en.wikipedia.org/wiki/Wikipedia:Statistics
26. Website

    1. Wikipedia Foundation

    (2019b) Wikipedia: identifying and using primary statements

    Accessed September 29, 2019.

    https://en.wikipedia.org/wiki/Wikipedia:Identifying_and_using_primary_sources

Thank you for submitting your article "Reader engagement with Wikipedia's medical content" for consideration by eLife. Your article has been reviewed by three peer reviewers, and the evaluation has been overseen by two members of the eLife Features team (Julia Deathridge and Peter Rodgers). The following individuals involved in review of your submission have agreed to reveal their identity: James Heilman (Reviewer #1); Besnik Fetahu (Reviewer #2); Shani Evenstein (Reviewer #3).

The reviewers have discussed the reviews with one another and we have drafted this decision to help you prepare a revised submission. We hope you will be able to submit the revised version within two months.

Summary:

This work enriches our understanding of how readers of Wikipedia's medical pages engage with external links compared to readers of standard Wikipedia pages. However, there are a few major comments that need to be addressed.

Major comments to address:

a) The results reveal that in general the readers of the standard Wikipedia pages are more prone to actually clicking the external links. This is interesting given that one would think that the motivation for visiting such pages may reflect more personal reasons, i.e. checking on symptoms of a disease. Yet, the non-WPM pages seem to have better engagement with the external links. One possibility for this could be that the page visits are not normally distributed across all the pages. You have highly important pages in Wikipedia that gather a lot of attention, thus, their external links are more likely to be visited. I assume that normalizing the Wikipedia articles according to their page visits might paint a different picture here with respect to the page visit.

b) In the second research question, the motivation behind visiting such external links is postulated. While personally I agree with the points in the discussion as to why this may happen, there is no actual evidence for this in the current study. The authors should include a sentence or two discussing this limitation and what further work could be done to validate their hypotheses, such as carrying out a questionnaire with a diverse set of WPM readers that investigates their reasoning for clicking/not-clicking on external links.

c) As the article aims to review WPM content as a gateway to further learning and deeper understanding of a given topic, especially as it is used by both students, health professionals and the general public, one thing that was missing from this effort was an acknowledgement of the importance of both internal links and media files as a means of enhancing understanding, not only that of references and footnotes. Even if this research effort chooses to focus only on external links and footnotes, then an explanation is expected as to why specifically this focus. Especially in a world where imagery is so dominant, and videos have become a major means of education, further research is needed regarding how readers engage with media content on WPM articles, compared with the rest of Wikipedia. If the research team has any data related to this, not only to external links and footnotes, it could be of great contribution to mention it as well.

d) While research till 2016 (Heilman & West) showed that Wikipedia was the most popular site for medical content online, more than sites like WHO, NIH etc, WikiProject Medicine Foundation has shown that this has changed in the last couple of years. It seems that Wikipedia's medical content is no longer the most visited site in the world for medical content and is "losing" to sites that encourage users to ask different questions on the same topic and have multiple answer pages on that same topic. It would be of interest to the public if the research team considers this and discusses the results in this light. It might offer additional conclusions to the ones already presented.

[We repeat the reviewers’ points here in italic, and include our replies point by point, as well as a description of the changes made, in plain text]

Major comments to address:

a) The results reveal that in general the readers of the standard Wikipedia pages are more prone to actually clicking the external links. This is interesting given that one would think that the motivation for visiting such pages may reflect more personal reasons, i.e. checking on symptoms of a disease, yet, the non-WPM pages seem to have better engagement with the external links. One possibility for this could be that the page visits are not normally distributed across all the pages. You have highly important pages in Wikipedia that gather a lot of attention, thus, their external links are more likely to be visited. I assume that normalizing the Wikipedia articles according to their page visits might paint a different picture here with respect to the page visit.

We appreciate this question on our finding that WikiProject Medicine (WPM) readers clicked on external links less often per view than Wikipedia (W) readers. We investigated ways of normalizing for pageviews and page-length, as two factors that might be associated with this result. The most common approach would be to use matching methods, involving creating matched samples or datasets for, in this case, matching WPM and W for pageviews and for page length. This creates, in effect, random samples for these two factors which then allows one to determine the effect of an intervention, or in this case the extent to which a factor is associated with an external link behavior. But since we set out to determine if and how WPM pages and readers differ from the rest of W pages and readers, our goal has been to describe the pattern of the differences as a whole, rather than ask what would happen if pageviews or page lengths were the same for WPM and W. This helped us realize that the focus on the comprehensive pattern was not previously clear enough, and we have now, thanks to this reviewer’s comment, brought it to the fore in revisions to the research questions and discussion. The principal change is to have introduced a summary table (Table 6) in the discussion that presents the three key factors that distinguish WPM pages from the rest of W, and the five key factors that distinguish WPM readers from W readers. As a result of structuring the results this way, we believe it is easier to grasp consistencies that are to be expected with the reading behaviors of health sciences students and professionals (whose use of WPM has been well established, if not the proportion of WPM readers that they make up). We think that this points to the sort of future research helpfully suggested by a reviewer involving questioning readers either directly about their choices or indirectly through think-aloud protocols. All of that said, if the reviewers still feel that matching methods should be applied, we would, of course, be willing to conduct such an analysis. To reflect the above changes, we have also modified the abstract.

b) In the second research question, the motivation behind visiting such external links is postulated. While personally I agree with the points in the discussion as to why this may happen, there is no actual evidence for this in the current study. The authors should include a sentence or two discussing this limitation and what further work could be done to validate their hypotheses, such as carrying out a questionnaire with a diverse set of WPM readers that investigates their reasoning for clicking/not-clicking on external links.

We interpreted this important point as partly a valuable critique of our second research question. We have reframed the research questions in the introduction so that they follow the pattern of the study, dealing with differences between WPM and W pages and readers. This had led us to extend the analysis of the greater hovering and footnote-number clicking in the Discussion, in which WPM readers must be gaining additional reassurances from how the title and publication venue of the references can provide scientific characteristics of the study, using the “ideal sources” of systematic reviews and medical textbooks as examples. That said, we fully embrace and now include the reviewer’s point about follow-up qualitative studies among a diversity of users to determine what influences their choices around engagement with external links.

c) As the article aims to review WPM content as a gateway to further learning and deeper understanding of a given topic, especially as it is used by both students, health professionals and the general public, one thing that was missing from this effort - an acknowledgement of the importance of both internal links and media files as a means of enhancing understanding, not only that of references and footnotes. Even if this research effort chooses to focus only on external links and footnotes, then an explanation is expected as to why specifically this focus. Especially in a world where imagery is so dominant, and videos have become a major means of education, further research is needed regarding how readers engage with media content on WPM articles, compared with the rest of Wikipedia. If the research team has any data related to this, not only to external links and footnotes, it could be of great contribution to mention it as well.

We did fail to properly acknowledge the contribution of internal links for reader learning. And while our focus remains on the association between Wikipedia and the body of external research, that has now been made clear (in the Introduction), in the context of how richly cross-referenced Wikipedia is and how much of that lateral reading can contribute to the informational and educational quality of WPM pages.

d) While research till 2016 (Heilman & West) showed that Wikipedia was the most popular site for medical content online, more than sites like WHO, NIH etc, WikiProject Medicine Foundation has shown that this has changed in the last couple of years. It seems that Wikipedia's medical content is no longer the most visited site in the world for medical content and is "losing" to sites that encourage users to ask different questions on the same topic and have multiple answer pages on that same topic. It would be of interest to the public if the research team considers this and discusses the results in this light. It might offer additional conclusions to the ones already presented.

This comment raises interesting points about changes to the questions readers may be asking and how that is affecting the popularity of WPM pages. After some discussion among us, as well as searches for relevant literature on it, we wish to respectfully suggest that this comment is out of scope for this paper, given our focus on reader engagement with external links on WPM pages. Based on recent studies on medical and health provider use of Wikipedia (Scaffidi, 2017), we see the relevance of this site for health information is still substantial, and a particularly relevant source for the initial and continuing education of health professionals. We appreciate the reviewers’ encouragement to consider broader issues around such topics as reader interests and questions.

Author details

1. Lauren A Maggio

   Lauren A Maggio is in the Department of Medicine, Uniformed Services University of the Health Sciences, Bethesda, United States

   Contribution

   Conceptualization, Formal analysis, Investigation, Methodology, Project administration

   For correspondence

   lauren.maggio@usuhs.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-2997-6133

2. Ryan M Steinberg

   Ryan M Steinberg is in the Lane Medical Library, Stanford University School of Medicine, Stanford, United States

   Contribution

   Conceptualization, Data curation, Software, Formal analysis, Investigation, Methodology

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-0101-4490

3. Tiziano Piccardi

   Tiziano Piccardi is in the École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland

   Contribution

   Data curation, Formal analysis, Investigation, Visualization, Methodology

   Competing interests

   No competing interests declared

4. John M Willinsky

   John M Willinsky is in the Graduate School of Education, Stanford University, Stanford, United States

   Contribution

   Conceptualization, Formal analysis, Methodology

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-6192-8687

• 1,056

  views

• 80

  downloads

• 22

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.