Tim Albert's tips for preparing your manuscript

Before you start writing

1. Be absolutely clear about your message. Too many scientific papers fail because they don't make it absolutely clear what the authors are trying to say. The solution is to spend time -before you start writing - defining what your paper means. Boil it down to one sentence. If you have nothing interesting to say, consider saying nothing.
2. Make sure your co-authors agree about the message -and where you will publish it.
   Do this before you start writing. There is nothing worse than investing time in your first draft, only to find others trying to turn it into something else as soon as you have finished.
3. Check with the methodologists. No amount of cleverness will be able to rescue a piece of research in which your conclusions are simply not supported by the evidence. Again, sort this out before you start writing.
4. Write a brief plan for each section:
   • Introduction: normally two paragraphs on why you started, with the last sentence summarising what you did,
   • Methods: aim for six paragraphs elaborating on what you did
   • Results: six paragraphs describing what you found
   • Discussion: seven paragraphs on what it all means. Start with a sentence summarising what you found, and end with a clear message (preferably avoiding the cliche 'more research is indicated') in the last sentence.

As you write

1. Write each of the four sections in one go. Try to do a quick version in about 10-15 minutes. Resist the temptation to rewrite as you write, or to check any facts that you might not be sure about. You can do all this later, once you are committed to the general shape of the article.

After you have done the first draft

1. Check that your message is still clear. It should be contained in the last sentence of the last paragraph of the Discussion.
2. Check the structure of your article. Use paragraphs as your basic building blocks. The most important sentence of each paragraph should appear at the beginning of that paragraph, and these key sentences should lead on from one to another in a logical way.
3. Check your facts. Make sure that everything is still reported accurately and that errors have not crept in. Double check your calculations. Make sure that, were you to be challenged, you could back up everything you say with concrete evidence.
4. Check your references - Try to use them only to support the statements you are making, and not to show off your knowledge or plug your previous publications. Write them in the required style. Reread them to ensure that you are still quoting them correctly.
5. Get the basics of language right. Obey the basic rules of English grammar and spelling, and if you are not good in these area get help from someone who is. This is particularly important if English is not your first language.
6. Keep your style as simple as possible Don't be afraid to use short words and short sentences - it will make your work accessible to a larger number of people internationally. Don't be afraid to say ' In this study we…' rather than the old fashioned and pompous 'It was discovered in this study that…'

The final stages<

1. Follow the style of your publisher. Read through the Instructions to Authors and make sure you comply. Also, look at previously published work, and make sure your submission is similar in shape and tone.
2. Get the agreement of you co-authors. They should all sign.
3. Deal promptly with any requests from the editor. Don't be offended if editors ask you to make changes; that's what they are there for. Do them, and do them fast.
4. Celebrate success. This is something we do too rarely and too late. Reward yourself as soon as you send off the manuscript.

Tim Albert has been running courses on writing scientific papers for nearly 10 years.
For further assistance, including information on books and courses visit http://www.timalbert.co.uk/courses.htm

Books by Tim Albert:

A-Z of Medical Writing, BMJ Books, 2000 Winning the publications game, Radcliffe Medical Press (2nd edition) 2000