Not long ago, if you were reading an article of this type, it was safe to assume that you were holding a printed journal. Today, while some readers might have a document in hand, many are sitting in front of a computer monitor, and still others are reading on a smartphone, tablet, or other wireless device. Accordingly, as paper is replaced by pixels, readers around the world can readily access their reading material at the point of need. Yet this foray into the digital domain does not represent unalloyed progress: the Internet has provided everyone with a printing press, allowing experts, quacks, and all those in between the ability to publish medical material with at least a veneer of authority. Thus, the challenge for the consumer is not so much to find some medical information, but to find valid, trusted, and pertinent medical information1.
This challenge is especially daunting, given the size of the World Wide Web. For example, a keyword search of “carpal tunnel syndrome” in PubMed returns less than 8000 entries (as of October 2011), whereas the same search in Google finds more than 12.3 million results. Beredjiklian et al.2 evaluated the quality of information regarding carpal tunnel syndrome on the Internet and found that 23% of web sites offered unconventional or misleading information, and the mean informational value of the web sites was 28.4 of a possible 100 points. Likewise, Labovitch et al.3 examined Internet sources regarding minimally invasive hip arthroplasty and found information that was often “misleading and of poor quality.”
In response, new models for obtaining information have been created. Web sites of well-respected organizations such as the American Academy of Orthopaedic Surgeons (AAOS) aggregate material and post commentaries. Web 2.0 technologies4, including forums, blogs, and social networks, allow surgeons to comment on orthopaedic issues and modulate understanding accordingly. Search engines that not only find but sort information according to unspoken and idiosyncratic algorithms are increasingly important as well.
In short, the task of information management faces new imperatives and may avail itself of new tools. As the Internet grows, we may see a great transformation of how information is produced, gathered, and used. In this review, we outline our forecasts regarding the contours of the emerging medical information landscape.
Although online publications can maximize content with use of videos, links, and supplemental materials (and minimize heft), there are still some advantages to traditional print media. Printed material can be read in venues without access to electric power or wireless signals. Additionally, as noted by Shafer9, “the attention given to typeface, letter-spacing, line-length, leading, page size, and margins, and all of the other tricks [of typography], gives the eyes and the brain an edge [with paper] over copy published for Web browsers.” As such, we expect paper journals to endure. Perhaps the niche for paper lies in short pieces (such as the “In Brief”10 articles introduced in Clinical Orthopaedics and Related Research), which can be printed on a single sheet and saved for later reading in venues less hospitable to electronic media.
We also predict that the use of hyperlinks will change the style of scientific writing, allowing authors to compose more tersely. With these links, an author can expand on a concept selectively (i.e., for only the users who wish to follow the link). Authors will also be able to embed code within their text that will enable a computer to understand the nature of the information. This process of so-called semantic tagging will enable search programs to get “the concepts right instead of relying on the vagaries of wording,” as noted by Kent Anderson11, the CEO/Publisher of JBJS.
Electronic publishing of textbooks will improve them. Hardcopy printing is hindered by its lengthy revision cycle and the ponderous mass of most textbooks. For example, the current printed edition of Campbell’s Operative Orthopaedics, published in 2007, weighs more than thirty-eight pounds. Some of the Internet sources vying to replace traditional reference textbooks are shown in Table I. On the conservative end of the continuum are the sources that follow traditional approaches, but with online features. On the more radical end lies the wiki-based collaborative authoring model, exemplified by Wikipedia: free, crowdsourced content that is open to continuous revision. Neither is perfect, as the first model does not avail itself of crowdsourcing, whereas the second may be insufficiently authoritative. For example, although Wikipedia was found to be reasonably accurate in some orthopaedic surgery domains (e.g., osteosarcoma12), in general, unsigned articles (e.g., Wikipedia pages) should not be the basis of treatment for our patients.
A hybrid of the two models, such as Orthopaedia.com13, may turn out to be most useful. Orthopaedia is a crowdsourced repository of information that is open to continuous revision, but maintained by a closed community of identified contributors. The success of closed-community wikis in orthopaedic surgery is hardly ensured, however, as in most online communities, less than 1% of the users contribute substantially. Applied to the fairly small orthopaedic surgery community, this rule implies that Orthopaedia might not corral enough contributors to succeed. The success of closed-community wikis may thus depend on the development of new forms of social credit, motivating members to contribute at rates higher than currently seen in most online communities.
Many medical journals will also change significantly. William Curtis, President, Springer Science+Business Media, has noted that the “uptake of Internet versions of long-established print journals has been so widespread that increasingly the print editions can be eliminated…only a handful of print editions will exist a decade from now, if not sooner.”14 The appeal of the web is obvious: although the editorial costs for electronic journals can be comparable to those of traditional paper journals, the marginal distribution cost of electronic copies approaches zero. Furthermore, unlike a traditional journal, an online journal can publish an arbitrarily large number of papers of arbitrary length.
With the migration to the web, the concept of “reading a journal” may change, as readers may create their own composite journals by merging electronically accessed articles into a self-created personal interest publication. For example, a reader may use Really Simple Syndication (RSS) feeds to create an electronic table of contents limited to only articles of interest (Fig. 1) or from an automated PubMed search15.
Web-based publishing will affect the content of what is written. With the volume limit lifted, authors may be requested or compelled to post raw data. For instance, authors wishing to describe a cohort of patients undergoing hip arthroplasty in a paper journal would be forced to select the data of interest. However, in a web-based model, authors may post the complete dataset for every patient. To be sure, unfiltered publication of information alone in the name of research is not desirable, and many readers will be unable to draw reliable conclusions from raw data. Nonetheless, comprehensive data reporting, as a supplement to a fully described study, will be particularly useful for those performing meta-analyses and therefore will ultimately benefit the average reader.
In a journal where space is not a concern, peer reviewers might be instructed to ignore the question of whether an article is of interest to the readership and consider only whether the study is valid and potentially interesting to even one reader. This mirrors the “long-tail” business model in which some Internet-based commercial firms profit from selling small volumes of less popular items to many customers16. When only validity (and not potential reader interest) is considered, it is more likely that studies in which no differences between groups are found (i.e., studies thought to be less interesting to readers) will appear more frequently in medical journals. The increased incidence of “negative studies” also abets the performance of meta-analyses. Growth in pay-per-view reimbursement schemes for journals, in which payment is rendered every time materials are downloaded, will also motivate a journal to produce the maximal amount of source content. Granted, a long-tail approach to peer review necessarily implies that publication in a once-exclusive journal might lose its cachet, but new forms of identifying merit (i.e., supplanting the current mode which rests in large part on impressing editors and reviewers) are apt to emerge.
The strongest countervailing force against unfettered expansion of journals is the effect such growth may have on a journal’s “impact factor.” The impact factor is a rating metric derived by the ratio of citations to the journal in a given year divided by the number of items the journal published in the previous two years. Clearly, publishing items that are not cited frequently will diminish the impact factor. In the coming years, we predict that the impact factor will matter less: “times cited” will give way to more nuanced measures of readership and influence. Additionally, with search engines effectively turning the pages for them, readers will be oblivious to the fact that many uninteresting things were published. In the transition, when publishers heed both impact factor and usage rates, we may see the proliferation of “sister journals” spun off the main publication. For example, the JBJS Case Connector allows The Journal to offer access to case reports that are not apt to be cited often, without diluting the main journal’s impact factor.
The face of medical information is changing and will continue to change. This change is coming chiefly because the new medium of consumption is also a medium of production. When users read on electronic devices, entirely new realms of data will be, quite literally, at hand (namely, the data produced by readers in response to primary material, the responses to those data, supplementary materials, etc.). Within that stream will lie gold, in the form of a flow of new and worthwhile information. For example, if there are 12.3 million web pages on carpal tunnel syndrome, and 77% of them are free of unconventional or misleading material, there are approximately 9.5 million potentially useful pages. More than ever, the challenge to users will be to sift through the stream, keeping the valuable and discarding the rest. To help with that task, new entities will arise, established entities will adapt or die, and new modes of social credit will emerge.
Variations on Wikipedia (exemplifying collaborative composition) and Facebook (representing the world of social networks), modified for the medical domain, may obliterate traditional methods of knowledge dissemination. Thus, the potential looms for a more egalitarian and useful web of knowledge. However, there is also the prospect of squandered opportunity. Because patients are helped by any step that enhances the quality of medical evidence or the process of evaluating it22, the orthopaedic community must strive to create the optimal future for orthopaedic information management.