Clifford Lynch Presentation: “Biomedical Libraries in the Next Decades”
On December 16, 2011, in celebration of the end of NLM’s 175th anniversary, Clifford Lynch, Ph.D., Executive Director of the Coalition for Networked Information, gave the presentation Biomedical Libraries in the Next Decades: Open, Diffuse, and Very Personal, at the National Library of Medicine. Overriding themes of his presentation included the implications surrounding the convergence of the increasingly data intensive nature of the biological sciences with traditional journal publications, and the possibilities for large scale computational scholarship. Another key topic was how developments in the broader economic, social, and regulatory climate will affect the evolution of medical knowledgebases. He emphasized the importance to the life sciences community of having an institution such as NLM, as a focal point for aggregating the staggering amount of information and data generated in the life sciences. There’s nothing analogous to NLM in other areas of scholarship.
Dr. Lynch talked about rethinking the nature of scientific journal articles. From roughly 1990-2005 scientific journal literature largely transitioned from print to electronic format, with access through publisher-based servers. However, the underlying essence of journal articles did not change. He noted that in the future there will be a trend towards the scientific article as a portal to reported data. Dr. Lynch projects that in 10-15 years journal articles will not be mere digitizations of their former paper versions, but instead they will become more interactive vehicles acting as collections of text and data sets to serve computers and human users. The difference will be quite striking! Dr. Lynch noted that journal publishers generally do not want the archival responsibility for data, making this an ideal opportunity for libraries.
Moving from biological to medical data, Dr. Lynch noted the incredibly large potential collection of data related to individuals. He predicted an age of personalized medicine, where it might be realistic to sequence the genome of every individual. The coming decade will see a linkage of medical records, genetic data, and biological data, creating a massive knowledgebase. The power of big data sets lies in the capacity to do revealing statistical analyses and pattern determination, as well as the ability to sort out a complex of factors, such as genes, the environment, etc. There is already great interest and budding efforts in both the public and private sectors to amass this data. Dr. Lynch stated that medical libraries may have a substantial role to play in working with medical records and data, perhaps more than people expect.
Personalized medicine will raise the importance of certain economic, social, and regulatory issues. Dr. Lynch mentioned the recent announcement by British Prime Minister David Cameron that essentially everyone in the UK will be a research subject, unless they opt out. The rationale behind this decision is the enormous centralized data pool held by the National Health Service (NHS), with the accompanying possibility of reducing medical care costs and generating revenue. At this point, it is hard to gauge public opinion regarding this newly announced initiative. The US does not have the equivalent of the NHS, with data widely dispersed among hospitals and other institutions. Also, there is a pervasive feeling in the US that medical records data can be used against an individual by insurance companies, employers, and other groups. Resolution of these types of concerns is essential to making progress in building the medical knowledgebase.
Dr. Lynch acknowledged the wealth of data that has been made publicly available to those who take the time and trouble to use it, allowing many people to discover research level data. A good percentage of the public has the aptitude to successfully interpret this data. However, a big sector of the public is not sophisticated enough to interpret data, due to a lack of statistical acumen, reading comprehension ability, etc. The popular press helps with the translation process, but this is a major area for NLM and similar institutions to have an impact. The stakes will rise even higher in the next 20 years, as data becomes even more complex and personal. One suggestion to address this issue is the development of computer-based tools to help the public work with and interpret data.
Another issue arises with the concept of consumer genetics and companies that provide genotyping services. These services may eventually include sequencing of the entire genome of an individual. Since this process occurs outside the formal health care system, the data will not be included in a medical record unless it is deliberately added. Some states have tried to shut down these private services for that reason, which could lead to a political fight. Dr. Lynch also noted the rising popularity of medical data sharing in closed communities or publicly through social media such as Facebook and Twitter, and he posed the question of how this information could fit into medical records. People with rare diseases who get genotyping services will provide an important data collection, which will increase in future decades. Much of that data currently resides on unstable platforms, such as Google Groups. We need to think about integrating it into medical records. This process is about making genomic data available for the public good. People will be more inclined to make personal data available if they trust that the information will not be used against them.
Dr. Lynch then raised the issue of medical records for deceased persons. With data increasingly stored on electronic medical records, data for the deceased could easily be added to the knowledgebase. But who owns this data? Who decides if these records are destroyed or saved and made publicly available? He suggested the possibility of allowing individuals to donate their medical record information in the same way that they can donate their organs upon death.
Dr. Lynch then addressed the international nature of medical knowledgebases. He noted that many countries are amassing data pools, which are increasingly viewed as global resources. The communication of scientific knowledge knows no boundaries, and data sharing across nations will increase the resolution of data computations. Dr. Lynch also noted the ethical issues raised by human experimentation. Protections will remain very important through review boards, etc.
In closing, Dr. Lynch observed that the shape of the future medical knowledgebase, including the integration of medical records, involves more than technological issues. With the issue of public trust, it is important to determine which agencies are believed to be competent and capable of handling data transparently. Libraries are high on that list of institutions to act as stewards of data and providers of computational platforms. Dr. Lynch believes that medical knowledge will split off from biological knowledge, due to the inclusion of medical record data. The medical library community is very fortunate to have NLM as a focal point for these efforts as we move forward into the next decade!