Editorials Christmas 2010: Editorial

Strategies for coping with information overload

BMJ 2010; 341 doi: http://dx.doi.org/10.1136/bmj.c7126 (Published 15 December 2010) Cite this as: BMJ 2010;341:c7126
  1. Richard Smith, chair
  1. 1Patients Know Best, Cambridge CB4 0WS, UK
  1. richardswsmith{at}yahoo.co.uk

You need a machine to help you

Fraser and Dunstan (doi:10.1136/bmj.c6815) show that even within a narrow specialty it is impossible to keep up with published medical reports.1 Trainees in cardiac imaging reading 40 papers a day five days a week would take over 11 years to bring themselves up to date with the specialty. But by the time they had completed that task, another 82 000 relevant papers would have been published, requiring another eight years’ reading. And this analysis assumes that trainees need to know about cardiac imaging only, whereas they surely need to keep up with other areas of medicine and healthcare. The authors conclude that it is impossible to be an expert.

This problem is not new. Dave Sackett, the “father” of evidence based medicine, found some 20 years ago that to keep up to date in internal medicine it was necessary to read 17 articles a day 365 days a year.2 He also found that the median time spent reading by newly graduated doctors was zero, while for senior consultants it was 30 minutes, with 40% reading nothing.2

Some 10 years ago I asked around 100 doctors how much of what they should read to do their job better they actually read. About 80% said less than 50%, and 10% said less than 1%.3 More than half felt guilty about this, and when asked to describe in one word how they felt about their information supply it was mostly negative (impossible, overwhelmed, crushed, despairing, depressed), with just a few answering “challenged.”3

One of the best known responses to information overload was the founding of the Cochrane Collaboration, named after the epidemiologist Archie Cochrane who called for a “critical summary . . . adapted periodically, of all relevant randomised controlled trials.”4 He knew that most of the new information was of poor quality, and Brian Haynes showed later that less than 1% of studies in most medical journals reach stringent scientific standards.5 John Ioannidis has argued in the best read paper in PloS Medicine that most research findings are false.6 So it makes no sense for doctors to try and read everything: rather, argued Cochrane, they should rely on critical summaries.

But 20 years after the launch of the Cochrane Collaboration a review has found progress to be poor.7 Around 75 clinical trials and 11 systematic reviews are published every day, with no sign of abating. Yet many clinical topics have no Cochrane Library systematic reviews, and perhaps three quarters of interventions lack a firm evidence base. We have what Muir Gray, once director of the National Library for Health, calls an information paradox—we are overwhelmed by new information yet have many unanswered questions. The average 10 minute consultation between a doctor and patient will throw up at least one question that cannot be answered.8 The box lists the possible strategies for dealing with the problem, but the only one that might bring success is to use a machine.

Strategies for dealing with information overload

The ostrich strategy

With this strategy doctors simply ignore the torrent of new information. If Sackett’s data are right, many doctors adopt this strategy, especially as they get older.9

The pigeon strategy

Perhaps the most common strategy is to hang around with other doctors and pick up titbits of information. You attend grand rounds and the occasional postgraduate meeting, follow some guidelines, and rely on drug company representatives to tell you about new treatments. When you have a tricky question about a patient you consult a colleague—the most common way to get an answer.8 You sometimes flick through journals, but you learn more from the mass media. The most annoying way that you learn new things is from patients who bring newspaper clippings, garbled stories about something on the television, or long printouts from the internet.

The owl strategy

Probably the rarest strategy is that proposed by the originators of evidence based medicine. You build your knowledge patient by patient by identifying questions that arise during interactions. You refine the questions to one that can be answered, search for all relevant evidence, and systematically analyse it, abandoning the large amount that is of poor quality and combining, preferably numerically, that of high quality. The advantage of this strategy is that your information relates directly to your patients. Unfortunately, almost nobody has the time and very few the skills to pursue such a strategy.

The Jackdaw strategy

Doctors who pursue this strategy follow the pigeon strategy but also regularly search for highly refined evidence—from perhaps the Cochrane Library, Clinical Evidence, guidelines, or other sources of evidence based reviews. Unfortunately these sources are full of holes (because the evidence simply doesn’t exist), and the evidence is not useful—and may even be harmful—for patients with comorbidity (who now constitute most patients).10

The inhuman strategy

John Fox, once director of the Advanced Computing Laboratory, said that practising medicine is an inhuman activity, meaning that it’s absurd for doctors to practise without the help of machines. Individual doctors have no chance of keeping up with new research, but teams of people can process new information and feed it into machines that doctors (and patients) can use. The most popular of these machines is UptoDate, which has 400 000 users, but there is also BMJ Point of Care, the Map of Medicine, and more.

Several years ago after conducting a semi-systematic review of the information needs of doctors I tried to identify the characteristics of the machine that would finally solve the seemingly impossible problem of answering all the questions that arise in medicine with the very latest research.8 Here are the characteristics:

  • Part of the information system that doctors use as they see patients

  • Able to answer highly complex questions

  • Connected to a large valid database

  • Electronic

  • Fast (answers within five seconds)

  • Easy to use (as easy as a car)

  • Portable

  • Prompts doctors in a way that is helpful not demeaning

  • Connected to the patient record

  • Gives evidence related to individual patients

  • A servant of patients as well as doctors

  • Provides psychological support.

Some of these characteristics may never be achieved. For example, it is impossible to give evidence related to individual patients because evidence is gathered on populations. It would also be hard for machines to provide psychological support, but many of the questions that doctors ask themselves, such as, “Did I do the right thing by that young woman who died of breast cancer last week?”, are really a request for psychological support.8

And will doctors be willing to use such machines? De Dombal showed that computers are better at diagnosing acute abdominal pain than doctors, but his strategies were never widely adopted.11 My father resented my mother buying a dishwasher because he feared it would replace his role, and perhaps doctors are worried that machines might precipitate the reformation, described so beautifully by Joanne Shaw, where priestly doctors with their Latin bibles will have to give way to plebeians speaking the vernacular.12

“Will we ever solve the problem of information overload?” I imagine myself asking God as I arrive in heaven. “Sure,” he’ll answer, “but not in my lifetime.”

Notes

Cite this as: BMJ 2010;341:c7126

Footnotes

  • Research, doi:10.1136/bmj.c6815
  • Competing interests: The author has completed the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declares: no support from any organisation for the submitted work; RS is on the board of the Public Library of Science and chair of Patients Know Best, a company that uses technology to enhance patient-clinician relationships; RS was the editor of the BMJ and chief executive of the BMJ publishing Group.

  • Provenance and peer review: Commissioned; not externally peer reviewed.

References