|
WHAT'S THIS?
|
BASICS << Ten top tips for new medical writers
|
Ten top tips for new medical writers
Published 26 November 2008, written by Jane Fraser
Writing your first (or even your fifteenth) clinical paper on behalf of an investigator can be a daunting task. Often, it’s the preparation that makes it difficult, rather than the writing itself. Here are ten top tips to help you to draft papers that help pharmaceutical companies present clinical studies in a scientifically credible, ethically responsible way. They are based on some of the commonest questions I’ve been asked over the years during courses for new medical writers in medical communications agencies.
Get a good brief
As with any writing project, you need to know exactly what you are trying to achieve. It’s not just a matter of having a copy of the Instructions to Authors, or knowing which facts you need to convey. You also need to be highly sensitive to the perspectives of the author, reviewers and journal editor. Find out about the weaknesses as well as the strengths of the study. If you are trying to get into a top journal, the study will need to have a strong element of originality. Does it have the potential to change clinical practice? Or does it simply confirm previous studies? Get the investigator and client to be frank about the limitations of the study. The CONSORT guidelines specifically ask for this, and some journals now even have a separate subheading for limitations.
Make sure you have all the necessary documents
Your client and the investigator may think they have given you all the information, but you may need to do a little digging. It’s surprising how often people forget to tell you things that might be extremely relevant. Is there an investigator’s brochure that might provide useful references for the Introduction and Discussion? What about any abstracts or posters that have been produced from the study (another agency or investigators may already have done useful work). Are the statistics in the abstract/poster the same as the ones in the study report? What about related papers, perhaps written by other agencies? Your paper needs to complement them and to refer to them where possible.
Be aware of all the relevant guidelines
There are now detailed guidelines for reporting randomised controlled trials, observational studies, meta-analyses…the list keeps growing. Many of the guidelines for reporting studies are now drawn together on a new website. You also need to be familiar with the International Committee of Medical Journal Editors (ICMJE) Uniform Requirements for Manuscripts Submitted to Biomedical Journals and, of course, the Good Publication Practice guidelines.
Have a good model – learn from successful (and unsuccessful) papers
Take a look at similar papers in your target journal for ideas. For example, you might get some hints on how to display data effectively, how much detail to give in the Introduction, or what sort of data could be provided in supplementary information online. You can also learn from the fate of other papers drafted by your agency. If a paper is rejected first time round, why? Was it the wrong target journal? A problem with the study design? Missing data? A disagreement on interpretation? Reviewers’ comments can help improve all future papers in the same publications plan.
Add value with the Introduction – but keep it brief
Writing the Methods and Results sections is largely a matter of carefully extracting the relevant information and putting it in a sensible order – it’s quite a mechanical process. But the Introduction and Discussion are where medical writers can add value to a paper. It’s rare for a paper to be rejected on the basis of the Introduction (unless the reviewers disagree with the objectives of the study). But the Introduction does offer an important opportunity to draw attention to important points that might otherwise be overlooked. For example, is there an unmet clinical need that needs spelling out? Does the study drug have a novel mode of action? What other key studies on your client’s drug should be cited? It’s also important to target the Introduction to the readers of the journal, and to match the length to the journal’s expectations – some allow you much more space than others.
Remember that most clinical papers are rejected on the basis of the Methods
You can’t change the way the study was done. However, you can make sure that you include all the components needed to comply with the CONSORT statement. You can also discuss potential problems with the investigators and the client, and address them in the Discussion. For example, if an unusual rating scale has been used to measure a clinical outcome, why? If the study was not blinded, why not? If the study was stopped early, what was the justification?
Don’t worry if the Results is short – but check and double-check the numbers
There’s no reason for the Results to be padded out – it’s often shorter than the Methods. Discuss with the client and investigator whether you will be reporting all the outcomes in one paper, or reserving some for another paper. This may influence what you need to put in the Methods. Of course, you should never exclude any results just because they don’t fit in with the hypothesis! However, if you are reporting a large study with too many outcomes to report in one paper, it is legitimate to just mention them all in the Methods (a requirement of CONSORT). Then you can state that some of them will be reported in full in subsequent papers. Be aware that discrepancies often arise between data presented in abstracts, presentation slides, posters and study reports. Surprisingly often, the data in the abstract don’t match those in the body of the paper. You may have to refer to the client’s statistician to resolve such discrepancies.
Add value with the Discussion – but keep it balanced
Reviewers and editors recognize that the Discussion is where authors defend their conclusions, so it’s legitimate to draw attention to the good points of the study. It’s also legitimate to interpret the results of the study in the light of other findings – which means that you can speculate on the clinical, scientific or economic consequences of the study. But it’s also important not to over-interpret the data or to ignore opposing evidence. You must mention alternative interpretations and highlight the limitations of the study, or the reviewers will consider the paper to be biased. This is an especially critical issue with the leading clinical journals, the editors of which are acutely aware of the potential for pharmaceutical industry bias. Don’t irritate reviewers and editors by over-selling the conclusions – they must be soundly based on the data presented. On the other hand, don’t miss out on the opportunity to draw attention to what’s new and clinically relevant – it may not be immediately obvious to the reader.
Make sure you are up-to-date with developments in the therapeutic area
When you are immersed in a single study, or a single drug, it’s easy to forget what’s going on in the wider world. The way you write the Introduction and Discussion will need to take account of recent developments in the field. Are you, the investigator or your client aware of any recent guidelines, papers, meta-analyses or editorials that have been published? Double-check before submission that you have not missed a critical reference, or the paper could look sloppy to the reviewers.
Write formally but clearly and concisely
We’re nearly at the end of the article, and we haven’t mentioned the writing yet. Writing style is yet another area where professional medical writers can add value. Poor writing is a common reviewers’ complaint (though it is not a common reason for outright rejection). The BMJ website tells authors: “Please write in a clear, direct, and active style. The BMJ is an international journal, and many readers do not have English as their first language.” Thus you should keep sentences and paragraphs to a manageable length, choose familiar words, and not be afraid of using ‘we measured’ rather than ‘a measurement was performed’. You should be formal, but there is no need to be old-fashioned or wordy.
For a reading list see www.janefraser.com.
About the author:
Jane is a former Editorial Director whose company Jane Fraser Associates Ltd provides training in scientific writing and editing skills to medical communications agencies, pharmaceutical companies and universities worldwide. (E: jane@janefraser.com)
|
Feedback: Do you have any comments about this article? Contact the Publisher, Peter Llewellyn.
|
|
|
