Google medicine

From HLWIKI Canada
Jump to: navigation, search
Googling.png

Are you interested in contributing to HLWIKI International? contact: dean.giustini@ubc.ca To browse other articles on a range of HSL topics, see the A-Z index.

Contents

Last Update

  • Updated.jpg 11 August 2013

Introduction

See also Electronic health records (EHRs) | Google Drive | Google health projects | Google scholar | Research Portal for Academic Librarians

Google medicine is an idea mentioned in a 2005 article from the BMJ entitled How Google is changing medicine: a medical portal is the logical next step. In the years since, several surveys of their search habits prove that physicians regularly search for information via Google. See recent study by the American Medical Association (see Google survey of physicians) and this study by Duran-Nelson et al. Should we Google it? resource use by internal medicine residents for point-of-care clinical decision making. Acad Med. 2013.

Google is not venturing into medicine despite its ongoing development of Google scholar, and at least three health projects: Google Co-op (Health); Google Health (now closed) and Knol (also closed). Even though Google can produce good results, reliable health information can still be difficult to locate. So much medical information is still poorly organized, repackaged for profit and locked away in proprietary databases. Moreover, medicine’s best evidence is inaccessible to most primary care physicians around the world except those affiliated with universities and teaching hospitals. Medical research buried in the deep web has been a problem for years and is said to be 500 times larger than the surface web. Is this best evidence critical to practicing medicine? A growing body of evidence suggests, when found, it leads to better overall clinical decision-making.

Medical portals

Since 2000, medical portals have been touted as the best one-stop access to peer-reviewed articles. Portal examples include the National Library of Medicine’s Gateway, the National Health Service (U.K.) 'Choices', SumSearch (U.S.) and the TRIP Database. All-purpose tools combining research and genome searching have also been developed such as NLM’s Entrez; even Wikipedia has created a Medicine Portal. Medpedia is part of the portal/vortal craze but is just recently branching out from consumer health, much like MedlinePlus. No portal offers the solution librarians and physicians seem to need; the so-called evidence resides in too many places, is expensive to access and a difficult, time-consuming process. In 2010, physicians need a portal for web searching but some experts are pointing to web 3.0 and the semantic web for more durable solutions to the problem.

Easy access to information = better patient care

For many physicians, conducting a literature search is fraught with difficulties. To gain even basic competencies, there are numerous sources and interfaces to navigate, and the evidence is fragmented across free and fee-based databases. Even some health librarians have a hard time keeping track of this information. Health librarians attending McMaster's "How to Teach Evidence-Based Practice" workshop see firsthand how doctors from around the world make clinical decisions – once they find the evidence. Librarians assigned to small groups are experts at searching but many doctors struggle with basic searching. Most participants are engaged in critical appraisal more than finding the evidence. "Why is searching so difficult?" Why can’t search tools be more straightforward and easy to use? In small EBM tutorials at McMaster, search strategies are discussed superficially. Is information retrieval important in evidence-based practice or not? Clearly, physicians want searching to be less onerous to free up their time to evaluate what they find. Many librarians coach users how to search and hear these comments often. Physicians want to practice evidence-based medicine in the information age, but searching takes them out of their natural domain - patient care. That means, as health librarians, we need to devise strategies and alternatives for them.

Search engines & open access

Since their initial development, search engines have been used for all kinds of reasons. Initially, the web appeared to be a recreational, social place mostly. Now it is a two-way communication tool for business, education and health – and, society as a whole. It is an important entry point to digital information and has transcended its early criticisms even in health and especially for patients. In the web’s early days, hospitals in North America installed firewalls to protect patient and human resources information on their Intranets. Information stored in hospital databanks became vulnerable to hacking due to viruses and spyware on the web. As health organizations moved their services onto the web, some hospital IT departments’ policies changed. Health workers began to use the web for high-quality content and Medline (PubMed) to search the medical literature. Access inequities in a digital divide (those without access or poor search skills) did improve markedly in the period after 1997, when Google was first launched.

Google changed the landscape

In 1998, Google was launched. In head-to-head comparisons with AltaVista, Lycos and other search engines, Google scored higher in usability and relevancy. Over the next few years, Google has been adopted by health professionals with no search skills and many adopted it by choice for all kinds of reasons. For those working in hospitals, information retrieval is increasingly important because patients use the web to find evidence. Google’s popularity hit a high point in mid-2004 after its initial public offering. By 2005, it had announced Google Print (now Book Search) and Google scholar, the latter giving searchers a second option in locating medical literature. Almost overnight, health workers began to favour Google for literature reviews; PubMed citations began to appear in Google and Google scholar and by 2006 requests from library users to teach Google scholar became common. Users asked why PubMed was so difficult to use so health librarians began to teach it alongside Yahoo and Google. Open Access (OA) journals and search engines have an undeniable synergy in the digital age and are central to searching in the open access era. Accessing free content – and learning how to use all manner of search tools (not just Google) - is a core information skill for health professionals and librarians alike.

What new information technologies mean

Over the past few years, we have seen a number of trends and counter-trends on the web. Two of the most obvious are the rise of open access and mass digitization of print books. In medicine, it also seems obvious that user-generated content is on the rise. To see how much information and discussion is generated every day in the blogosphere take a look at some of the activity on Twitter and on blogs. Some of the traditional hierarchies and boundaries in publishing are shifting quickly in the web 2.0 universe and there is a sense that physicians are using social media in higher numbers than ever. How does collaborative writing and editing affect medical publishing in web 2.0? Do open peer-review processes change how research is produced? With ever-increasing numbers of documents produced it also seems self-evident that information overload is a constant problem.

Information overload, monopolies, monetization

Traditional search tools are not immune from the problems associated with information overload. PubMed contains over nineteen million articles. Even though widely regarded as the gold standard in medical databases, PubMed is more difficult to use and unreliable. Some doctors view NLM’s journal selection as elitist and American-biased. Some librarians question NLM's indexing practices as they seem idiosyncratic and slow to respond to changes in research. NLM is also considering machine-driven indexing which librarians need to watch in terms of upholding standards and quality control. Scirus is one of a few specialized science and medical search engines and indexes 500 million items but is criticized for being too European and Elsevier-focused. As for Google scholar, it will eventually index the most content due to its crawling of all journal sites, bibliographies, library catalogues and open access repositories. However, no one knows for sure what is indexed because Google isn’t saying. If that were not enough, the digital age has introduced new political and logistical problems for searching. For instance, many librarians fear search engines’ monopolies. For better or worse, librarians are custodians of the human record. Once Google’s digitization project is complete, it stands to hold the record of human history. Is this what we want for society? Do we want private interests to control our culture? Does it make sense for librarians to be the stewards of this information? Little wonder the National Librarian of France has vowed to challenge Google. Google’s monetization of search is its bread and butter. Well-placed ads and “click-through” increase profits. But free web content is easy to index. In medicine, much information is inaccessible and evidence-based content “locked down” by publishers. ACP Pier, Clinical evidence, DynaMed and UpToDate are but a few examples. Access is available only to those who can afford to pay, like rich physicians and universities. This closed access is detrimental to human health, and true evidence based practice.

Conclusion

Medical librarians should continue to monitor trends that affect searching, especially OA and physician-created content in web 2.0. A variety of medical wikis (such as Google's Knol - http://knol.google.com/k) and blogs provide access to a range of scientific content but we are still experimenting with these tools. Newer models of scholarly communication strive to wrest back control of the knowledge-base in medicine. For example, physician-created content built on open-source platforms have enormous potential for disseminating research. Physicians can submit papers to OA journals and self-archive so that this material will appear in search results. Quick, easy access to evidence-based studies is the ultimate goal but we are not there yet. Major search companies like Google should be working more closely with OA publishers and vendors to free up access to medical evidence. Pay-as-you-go models could be devised to obtain full-text, for example. Rich countries could pay; developing countries could be granted free access. Liberating evidence for all presents the best future for medicine in a more open evidence-based world.

References

Personal tools
Namespaces

Variants
Actions
Navigation
Toolbox