The following is a guest post by Ellen O’Donnell, Technical Writer, National Center for Complementary and Alternative Medicine, who recently spent a year on detail in OSI.
If you’re like me, you’re a fan of online resources for health and medical information. Typically, I use them for things like seeking science-based information on a complementary therapy (such as meditation or an herbal remedy), or deciding what I can do about my sore back from a recent move.
Lots of people share my interest. In a 2011 survey from the Pew Research Center, among the 74 percent of adults who used the Internet, 80 percent had sought information there on any of the 15 broad health topics posed by Pew. One-third had read someone else’s online commentary or experience about health or medical issues. It’s important for consumers, health care providers, and others to cast a critical eye on resources in this area, and guides from sources such as the National Institutes of Health (NIH) and the Library of Congress can help.
A pillar in the field of health information, the NIH’s National Library of Medicine, is celebrating its 175th anniversary. You may have seen its PubMed and MedlinePlus databases, for example; the former offers citations to (and, in many cases, the full text of) peer-reviewed journal articles, and the latter is a kind of one-stop shop for easy-to-understand health information.
Recently, I attended an NLM anniversary event featuring Clifford A. Lynch, Ph.D., director of the Coalition for Networked Information. Lynch, a computer and information scientist, spoke on“Biomedical Libraries in the Next Decades: Open, Diffuse, and Very Personal.”
Lynch envisions “a big, reusable knowledge base of medicine” evolving over the next decade. This will be a distributed, networked, interlinked collection of resources and incorporate three major components.
The first component is the “staggering amounts of data” being generated, thanks to information technologies. This is the case across a wide range of biological and life sciences as well as medicine and health care. The methods include imaging; genomic sequencing; and combinatorial chemistry, which rapidly generates and tests large libraries of compounds and is applied in new drug development.
Researchers and their institutions are struggling with best practices and infrastructure to aggregate, manage, and curate the vast amounts of information generated by such technologies, and to work out what must be kept. Funding long-term data preservation is also a major challenge receiving a lot of attention in policy circles. Data gets tossed out that might be re-purposed usefully, “but in many cases it’s getting to be cheaper to regenerate data than to save it,” said Lynch. There are also ethical considerations; records of clinical trials involving humans probably must be kept, for example.
A second component is scholarly journal articles and related content. The nature of such publications is starting to change. “You’ll be able in some ways to use the journal article as a portal into its data set,” Lynch said. “Audiences will include not only human beings, but computer programs mining, and computing upon, these collections. This, in turn, will affect how we structure content.”
A third component, unique to the medical and health-science world, will be “all the records of health and medical data we have about human beings, including genomic data.” People are already sharing elements of their health history, medical records, and lifestyles in online communities. Interest in genealogy sites is also booming and will continue to grow.
The potential benefits of adding medical records to the knowledge base could be tremendous. Applying large-scale computational techniques could help us identify, for example, “complex, multidimensional interdependencies,” as between genes, medical conditions, and environmental and lifestyle factors in diseases. Getting the largest possible sample sizes will be critical.
Of course, uses of medical records also raise important ethical, regulatory, public policy, social, and economic concerns, “which we overlook at our peril,” Lynch said. “Resolving them will be essential to progress in medicine over the next couple of decades. What’s happening here has to be a negotiation between the broad public and the scientific and medical research worlds.”
Underlying all the questions, he concluded, is one of trust. Whom can the public trust to maintain data, use it wisely, use it in the ways that were promised, and be transparent? “Libraries are still pretty high on that list,” Lynch responded. “It’s a connection we should not overlook as we think about libraries as stewards for these knowledge bases of the future—integrating data, literature, and records, and providing a computational platform. National libraries such as the NLM and the Library of Congress will have very special roles to play in this future.”