Clinical reviews are 1800-2000 word articles (excluding text in boxes, figures
and the references) that provide a clear, up to date account of the topic
aimed at non specialist hospital doctors, general practitioners, and
candidates for postgraduate examinations—from all over the world.

The review should include a broad update of recent developments (from the past
1-2 years) and their likely clinical applications in primary and secondary
care. It should stimulate readers to read further and should indicate other
sources of information, including web based information—for example, Cochrane
reviews. The article should also try to highlight the bridge between primary
and secondary care and offer specific information on what general
practitioners should know about the condition.

You may consider having a colleague to write the piece with you if it will
help to broaden its perspective. The choice of coauthor would be yours, but do
let us know who the coauthor might be. In particular we would like to know his
or her competing interests. If you do have a competing interest that might
rule you out from accepting this commission please do get in touch so that we
can discuss it before you start.

Increasingly many readers prefer to read articles on bmj.com. You therefore
need to visualise how the article will look on the web not just on paper. You
should try to use the many advantages of web publishing such as links to other
sources of information, extra photographs, figures, tables, or sometimes even
a short video.

Please ensure your article includes:
1. INTRODUCTION: This should be about 100-150 words, explaining what the review is about, why it is important to general practitioners and non specialists,
and what it covers. It should specifically answer the question "Why should I
read this review as posed by a GP or a non specialist hospital doctor?"

2. A “METHODS” BOX outlining in 150 words what sources of information were
used to prepare the review and how you selected what to include. We do not
expect you to perform a systematic review yourself, but we do hope you will
consult Cochrane and other systematic reviews. Please say whether you have
done a Medline search, used a personal archive of references, or consulted
other experts. You need to include a statement commenting on the overall
quality of the evidence supporting your article.

3. EVIDENCE BASED WRITING: Throughout the paper, you need to clarify the
evidence on which the key statements in the review are based, and the strength of the evidence (published trials, systematic reviews, observational studies,
expert opinion). We expect to find sentences structured in the following
way. "A large well conducted randomised conducted trial finds..." "The
findings of a small case series suggest..." and so on. Where evidence is
lacking or is of poor quality we expect you say so.

Please avoid specialist jargon and abbreviations. Do remember that any advice
on managing patients may not apply worldwide—please do make clear if your
advice is country specific.

4. PRESENTATION: We attach a lot of importance to presentation. The body of
the text should be broken up under "reader friendly" sub-headings. These
subheadings need to be questions. (Example - Who gets it? How is it diagnosed?
What new treatments can we expect?) Specialist terminology should be
explained, and all abbreviations and acronyms spelt out. Please remember that
many of our readers read bmj.com only—try and visualise how your article will
look on the web.

5. ILLUSTRATIONS: Please provide two or three illustrations, which can be
clinical photographs, line drawings, and flow charts. Send them preferably as
jpeg files. Colour illustrations are welcome. We can offer help with redrawing
line figures. Do remember that we may be able to publish additional figures or
other material suitable for the web. Please remember that we need informed
consent from patients for any material you obtain from patients even if they are not identifiable (including Xrays, histology slides, and so on).

6. Please provide the following BOXES:

i. A METHODS BOX as outlined above.

ii. SUMMARY POINTS: A key messages box should be included containing 4-6 key
messages in the form of succinct, single sentence bullet points. These should
be the most important "take home" messages from the piece. (Some "speed"
readers may read only this box and the introduction).

iii. ONGOING RESEARCH: A box of key ongoing research studies indicating what
questions they aim to answer is welcome. A list of UNANSWERED QUESTIONS would also be welcome.

iv. CONTINUING MEDICAL EDUCATION RESOURCES BOX: Because we are seeking to make these articles as educationally useful as possible, please include a separate box entitled "Additional educational resources" for those who want to pursue the topics in more detail. We suggest this contains up to 2-6 key review articles, particularly Cochrane reviews where these exist and useful URLs and websites. Please look at the BMJ website (bmj.com) and consider including URLs to any useful archival material.

v. INFORMATION RESOURCES FOR PATIENTS: We are trying to include sound, web based information and other patient information resources which doctors can pass on to their patients. When you cite websites please mention
if registration is needed for access.

vi. A BOX CONTAINING A PATIENT’S STORY: If you can, please include a box of up to 200 words containing a personal account by a real patient about their
experience of being ill and getting professional support and treatment, and
perhaps about managing their own disease. The patient should use his or her
full name in the piece (given and family names) or, can remain anonymous. We
always need to see the patient’s signed consent to publication.

vii. A BOX AIMED AT GPs telling them what is expected of them and when they
should refer a patient. ("Tips for non specialists")

7. REFERENCES should be set out in Vancouver style and should not exceed 25
in the paper version of the article. We are happy to put up additional or
supporting references on the web and these should be labelled w1, w2, w3, etc.