Intended for healthcare professionals


Clinical Evidence

BMJ 1999; 318 doi: (Published 12 June 1999) Cite this as: BMJ 1999;318:1570

This month sees the publication of a new resource for clinicians

  1. Fiona Godlee, Editor, Clinical Evidence,
  2. Richard Smith, Editor,
  3. David Goldmann, North American editor, Clinical Evidence
  1. BMJ
  2. ACP-ASIM, Philadelphia, PA 19106-1572, USA

    Clinical review p 1600 See advertisement opposite p 1598 (CR edition), p 1621 (GP), p 1567 (Compact and International)

    This week's BMJ carries a sample of information from a new resource for clinicians, Clinical Evidence, which will be launched later this month (p 1600). The inspiration for Clinical Evidence came in a phone call in 1995. Tom Mann and his colleagues at the NHS Executive asked the BMJ to explore the possibility of developing an evidence “formulary” along the lines of the British National Formulary. They recognised that clinicians were under increasing pressure to keep up to date and to base their practice more firmly on evidence but that few had the necessary time or skills to do this. Their idea was to provide a pocket-size book containing concise and regularly updated summaries of the best available evidence on clinical interventions.

    A small team at the BMJ set to work. In partnership with the American College of Physicians we convened an international advisory board, held focus groups of clinicians, talked to patient support groups, and adopted countless good ideas from early drafts by our contributors. Throughout we kept in mind an equation set out by Shaughnessey et al.1 This states that the usefulness of any source of information is equal to its relevance multiplied by its validity, divided by the work required to extract the information. To be as useful as possible we aimed for high relevance, high validity, and low work in terms of the reader's time and effort. We also kept in mind principles of transparency and explicitness. Readers needed to understand where our information came from and how it was assembled.

    The first issue of Clinical Evidence will contain summaries on the prevention and treatment of about 70 common conditions. Each summary is based on a thorough search and appraisal of the literature, looking for good systematic reviews and, where these are lacking, individual randomised controlled trials. The summaries are written by clinicians with skills in epidemiology and are extensively peer reviewed.

    Clinical Evidence joins a growing number of sources of evidence based information for clinicians. But it has several features that, we think, make it unique.

    Firstly, its contents are driven by questions rather than by the availability of research evidence. Rather than start with the evidence and summarise what is there, we have tried to identify important clinical questions and then to search for and summarise the best available evidence to answer them.

    Secondly, it identifies but does not try to fill important gaps in the evidence. As Jerry Osheroff at the American College of Physicians puts it, Clinical Evidence presents the dark as well as the light side of the moon. We feel that it will be helpful for clinicians to know when their uncertainty stems from gaps in the evidence rather than gaps in their own knowledge.

    Thirdly, it will be updated every six months. This means that clinicians can rely on it to keep them up to date in the topics that are covered.

    Finally, and importantly, it specifically aims not to make recommendations. This is because we feel that simply summarising the evidence will make it more widely applicable. The experience of the clinical practice guideline movement has shown that it is nearly impossible to make recommendations that are appropriate in every situation. Differences in individual patients' baseline risks and preferences, and in the local availability of interventions, will always mean that the evidence must be individually interpreted rather than applied across the board. Clinical Evidence provides the raw material for developing locally applicable clinical practice guidelines and for clinicians and patients to make up their own minds on the best course of action. We supply the evidence, you make the decisions.

    Our expectation is that Clinical Evidence will evolve rapidly in its early years, just as the British National Formulary did when it first appeared. Indeed, Clinical Evidence may well become a family of products, appearing in different formats (including electronic) and languages for different audiences. In particular, it will evolve in response to the needs of clinicians. We have tried hard to anticipate those needs (not least by involving clinicians at every stage), but it is only when people begin to use Clinical Evidence in daily practice that we can know how best to develop it. We hope you will let us know what you think of the sample in this week's journal, and of the first issue of Clinical Evidence when it appears later this month.