The elephant in the library


Kira Payne photoKira Payne MD, FRCPC, is a recently retired psychiatrist and jazz musician living in Toronto, Ontario


You would think it would be easy to be an expert in the information age: all those books and journal articles, indexed in libraries, ready and waiting to be perused; all those digital bits and bytes coursing through the internet, searchable on academic databases or Google; all those archived videos on You Tube providing information on everything from how to calculate Pi to how to fold a t-shirt in only 2 seconds. Wikis abound, democratizing information, enabling it to exist in a continuously amendable form. Information is everywhere and growing exponentially. It is “kid in a candy store” stuff, right? But there is something sinister about the overabundance of information. Reminiscent of a Grimm fairy tale, there is the very real possibility of paradoxical starvation despite the bounty.

The problem has been described in terms like “information overload,” “infobesity” or “info glut” and the daunting challenge of keeping up with the deluge of facts has been likened to attempts to sip from a fire hose, an image variously attributed to Mitchell Kapor, founder of Lotus Corporation.

Nonetheless, keeping up to date is a moral duty of the expert. This dilemma has been written about by authors and physicians such as Fraser and Dunstan in an article entitled “The impossibility of being expert”. No one would disagree with the expectation that experts stay current - unless forcibly confronted with the practical impossibility of its being accomplishable.

Looking back on the history of information, Aristotle has been credited as being among the last scholars able to grasp the extant knowledge of the world. Over the roughly 87 generations since then, information has been accumulating at an ever increasing rate. Information management has become a significant problem of our time. As of October 18, 2015, a Google search for the term “information overload” returned 12 million results. Limiting the search to Google Scholar still returned 751,000. Experts in all areas are confronted with the practical impossibility of fulfilling their declared ethical obligations to keep up to date. Even before the advent of the internet age, in the mid-1960s according to Abbott, 20 million words of technical information were set down daily. Murray Jr agreed, writing in 1966, that it would have taken an individual reading full time, six weeks to get through, during which time they would have lagged behind more than five years in reading.

In 2010 a cardiologist wanting to read the corpus of literature on echocardiography, would take over a decade to do so, and would be woefully out of date at the end of that time. This dilemma has been surprisingly little studied, let alone acknowledged. Yet, regulatory bodies for professionals, like physicians, hold their members accountable for keeping abreast of the literature, despite evidence that this is patently unattainable in many instances. Since “ought” implies “can,” how do we reconcile the mismatch between the surfeit of information and the temporal and cognitive limitations of our experts? Can we legitimately hold them accountable to perform the impossible? Is there a moral obligation to address this predicament?

The problem involves the logarithmic expansion of knowledge. An apocryphal story of the invention of chess by Shenk illustrates this nicely. In this tale, a King offers to buy the game of chess from its clever inventor (who in some versions of the tale is a mathematician). To the king’s surprise, the inventor asks merely for a grain of rice for the first square on the board, then two grains for the second, then four grains for the third, and so on, squaring the number of rice grains of the previous space, until all the squares on the chessboard are accounted for. The King quickly accepts the terms, thinking he has won the bargain; however, to the King’s dismay, by the time the second half of the chessboard is reached it becomes clear that there are not enough grains of rice in the world, let alone his kingdom, to seal the deal (in fact, Shenk wrote, over 18 quintillion grains are required to be able to fill the chessboard as per the bargain). This is precisely our dilemma today with the 24 hours in a day substituting for chess board squares and research papers and clinical practice guidelines standing in for grains of rice. The dutiful clinician is confronted with what has been dubbed the red queen problem, after Lewis Carroll’s book Through the Looking Glass: "It takes all the running you can do to keep in the same place."

Here are a few examples of the scope of the problem. A recent study of knowledge resources available in typical general practitioners’ offices came up with 855 guidelines. Hibbles, Kanka, Pencheon, and Pools concluded that in physical terms this weighed in at over 60 pounds and stood almost 30 inches high. The Cochrane Collaboration of systematic reviews, which is 22 years old this year, is well regarded for providing rigorous syntheses of studies into reviews in aid of evidence-based decision-making. Over the first twenty years, it published more than 5,000 reviews, involving the work of over 30,000 contributors worldwide. Yet, Bastian, Glasziou, & Chalmers estimated that fewer than 50% of these reviews have been kept up to date. Incidentally, according to Shojania et al, the median time to expiry for a group of well performed reviews is roughly 5 years. Moreover, as Mallet and Clarke state, it has been estimated that twice the current number of reviews (i.e., 10,000 reviews) are required to adequately cover the range of information pertinent to health care. So, it seems that the Cochrane Collaboration needs to run faster, too. But, even if it were possible to produce the number of needed reviews and to keep them up to date, how can the end user meet the challenge of reading and reflecting on them all?

According to Wikipedia, the term “elephant in the room” is a “metaphorical idiom for an obvious truth that is either being ignored or going unaddressed.” In the case of information overload for the busy clinician, the “elephant in the room” is in the library.

The push to subspecialization perhaps reflects one strategy to address the dilemma of managing information overload. This seems, on the face of it, like an easy way to hone in on a smaller range of knowledge. Perhaps by staking out smaller territory one can manage the knowledge base. Unfortunately, evidence suggests that this is unlikely. Fraser and Dunstan took a hypothetical look at what a diligent echocardiologist, committed to practising evidence-based medicine and determined to stay on top of the literature, would have to do to keep abreast of this subspecialty. They calculated that it would take him or her over a decade to read the extant applicable literature, reading full time, 40 hours a week, taking only two weeks holiday a year. After accomplishing this, however, another eight years of reading would have accumulated. At no time did our expert see a patient. Much too busy reading!

The task of staying current while also actually seeing patients seems Herculean, indeed. Is it time to admit to what Fraser and Dunstan have called “the impossibility of being expert”? After all, it is not suggested that if only the echocardiologist really worked at it, really sacrificed and dedicated himself or herself, then he or she could manage to keep current. Rather, it is implied that within the constraints of the laws of nature the task may simply not be possible anymore.

So the question then becomes not whether experts are morally obligated to keep current--which would seem to be impossible--but rather whether they are merely obliged to inform the public of the limits of their knowledge. Yet, as long as the professional regulatory bodies continue to espouse the requirement that experts remain au courant and write this as an expectation into the codes of conduct for the professions, then the legal system and the public will no doubt see this as the expected standard of care and the practitioner will be judged as a moral failure if his or her lapses in knowledge are detected. Nobody wants to face the notion that an expert cannot know it all, and no expert wants to publicly declare the impossibility.

Fortunately, there are some potential remedies available and in development. Accredited Continuing Medical Education courses are available which streamline approaches to clinical care, establish benchmarks for a minimum knowledge base, and offer the promise of attainability to the expert and accountability to the public. Point of care databases and computerized “clinical decision support” tools, like Up To Date, Inc. provide valuable services by combing the literature to furnish evidence-based answers to clinical questions, saving the busy clinician from having to labour at this alone.

According to the authors of a 2009 Scientific American article, with the advent of the Semantic Web technology and Web 3.0, astonishing amounts of data from disparate locations will be accessible in searchable formats via computer algorithms and automated search engines to be made available to anyone with a smart phone. The capacity for data integration promised by the “intelligent Web” will be able to furnish answers to particular questions as well as potentially discern new information by detecting patterns in the data previously inaccessible to the unaided human eye. The expert of the future will rely on such computer programs to channel the deluge of data into manageable, “sippable” streams and to perhaps run a little faster than previously thought possible. The “elephant in the library” must be acknowledged, accommodated, managed, using all the technology at hand and still to be developed, or it may run amok.


Leave a Reply

Your email address will not be published. Required fields are marked *