Writing for publication

People do not usually read papers from beginning to end. If they are interested in a title they will skim the abstract; if this does not retain their interest, their attention will wane and they will turn the page and browse elsewhere. The abstract should comprise about six short sentences: onethat describes the introduction, perhaps two about the material and methods, two about the results, and one that summarizes the conclusion. it requires rigorous discipline to limit yourself in this way, but it is worth remembering that MEDLINE limits the abstracts that it prints to 250 words, so stick to the rule. There are a couple of other points to remember about the abstract. Firstly, it is likely to be the only part of the paper that many people will read (apart from the title), and secondly, in some parts of the world doctors can afford only abstracting journals and it may be the patients are being treated on the basis of abstracts. The abstract must therefore contain facts, and those facts must be accurate. More and more journals ar enow specifying structured abstracts, which require additional discipline. Journals vary slightly in the headings that they like to see in a structured astract, so do read the Instructions to Authors of the journal to which you are going to submit your paper to find out what the editor wants. Editors (and referees) are likely to be pleasantly disposed towards those who have obviously taken the trouble to do things their way.