Editorials

Britain's gift: a "Medline" of synthesised evidence

Worldwide free access to evidence based resources could transform health care

America's two greatest gifts to the world are jazz and Medline. Now the British government has the chance to match Medline by funding universal free access to what might be described as "a Medline of synthesised, reliable, and up to date evidence." This could be even more useful to clinicians and patients and has the potential to change health care across the world.

Medline is an electronic index of nearly 4500 journals from over 70 countries compiled by Washington's National Library of Medicine. It has been available online since 1971. Later, Hilary Clinton, then America's first lady, announced worldwide free access to Medline through the internet (www.ncbi.nlm.nih.gov/entrez/). Since then the number of people using it has increased exponentially, and many of them are patients. There is no better free starting point for finding high quality medical information.

But a search of Medline may be frustrating. Although Medline often includes abstracts and free access to the full text of some articles (including those in the BMJ ), clinicians and patients may be overwhelmed by an avalanche of references and abstracts. They only rarely have the time and resources to sift through the output of a search, let alone obtain the full texts of all the articles that may answer their questions. Clinicians and patients need ready access to syntheses of valid, up to date information relevant to their questions.

Recent years have seen several initiatives to serve these needs more effectively. A consensus is growing that the most valid answers to their questions will come from systematic reviews based on rigorous research methods. The most obvious manifestation of this trend is the international Cochrane Collaboration. ^{1 2} The Cochrane Library (www.update-software.com/cochrane/), brings together an unequalled collection of reviews of research about the effects of healthcare interventions.

Cochrane reviews tend to address fairly specific questionsfor example, is echinacea helpful for a cold? Reviews published in other web based sources, such as Clinical Evidence (www.clinicalevidence.org, published by the BMJ Publishing Groupsee competing interest) draw on the evidence in these specific reviews to address broader questionsfor example, what's good for a cold?³ And because Clinical Evidence is based on questions that clinicians and patients want answered, this may be the right starting place for a search for relevant evidence.

Whether a question implies the need for a specific or a broad systematic review, a mountain of evidence remains to be synthesised before it will become clear just which questions can be answered using existing research evidence. But what are clinicians and patients to do if their search of these new resources shows genuine uncertainty about the relative merits of alternative forms of care? Their most imaginative step would be to consult the metaRegister of Controlled Trials (www.controlled-trials.com) to assess whether a relevant controlled trial was open to participants,⁴ and, if so, to agree that "the trial would be the treatment."⁵

Imagine the benefits of linking these sources of information electronicallyand of making this linked resource freely available. A whole new way of practising medicine opens up. A clinician and a patient trying to solve a problem together would start by searching Clinical Evidence, which might provide a helpful summary of the evidence. If they wanted to check the pedigree of the summary they could "drill down into" the Cochrane and other systematic reviews on which it had been based. This evidence could then inform decisions about treatment, which would take account of the patient's preferences as well as the availability of the preferred treatment.

If their search showed uncertainty about the best course of action then they might look to see if a relevant clinical trial was underway. The patient might choose to enter such a trial, particularly since patients tend to do better when they take part in trials.⁶ The wider benefit might be that we would more quickly know the answer to many important questions. For example, we still do not know which treatments are useful for acute stroke, but if every patient in the world experiencing a stroke were admitted to trials we would have enough patients within 24 hours to answer many of these questions. If there were no trials underway addressing the patient's question then the patient and doctor would send a signal to a central database that the question needed answering. This would allow trials to be designed to answer the questions that mattered most to patients.

Is this scenario of electronically linked resources serving the interests of patients and doctors unrealistically fanciful? We believe it is essential. We need to take advantage of the possibility of designing intelligent software that will flag the arrival of new data or substantive changes in the evidence. Without such help it will be increasingly difficult for people to keep information up to date and trustworthy.

A way should be found to make this informationlike the information in Medlinefree to anyone who has access to the world wide web. Current Controlled Trials has already undertaken to provide worldwide free access to the metaRegister of Controlled Trials. Many clinicians and people in higher education already have free access to both the Cochrane Library and Clinical Evidence, and both these resources are either already (or very soon will be) provided free to everybody in the 100 poorest countries in the world. Why stop there? These resources could be free to everyone at the point of use. Wide access would also ensure that errors would be spotted and quickly corrected.

Is there a rolepossibly a responsibilityfor Britain here? Britain has given the world Shakespeare, newtonian physics, the theory of evolution, parliamentary governmentand the randomised controlled trial. Tony Blair's speech at the Labour party conference suggested that the response to the attacks of 11 September must be not just war but also to build a new world that ultimately destroys extreme inequities. Universal free access to an integrated information resource built from the Cochrane Library, Clinical Evidence, and the metaRegister of Controlled Trials would go some way to reducing the inequities in access to information for improving health care. For a cost which might be as little as 10p for each Briton, the British government has the chance to match Medline by funding universal free access to the system we have outlined. It would provide a lasting memorial of the Queen's jubilee next yearand, in her honour, perhaps it could be called "Lizzie."

Richard Smith, editor. 

BMJ

Iain Chalmers, director. 

UK Cochrane Centre, Oxford OX2 7LG

Footnotes

   RS is chief executive of the BMJ Publishing Group, which publishes Clinical Evidence. He is, however, paid a fixed salary and would not personally gain financially from extra sales of Clinical Evidence. IC is director of the UK Cochrane Centre, but has no financial involvement in the Cochrane Library. RS and IC are also members of the international advisory group for Current Controlled Trials but have no financial interest in it.