Publishing research in England

Gordon B. Drummond,
Edinburgh University, Edinburgh Scotland

How to go about it

The main aim of this article is to advise writers who do not have English as their native language. My experience of this subject comes from having read many submissions and articles written by such authors. I have read even more by native English speaking authors (many of which are less well written).

"The worth of a thing is best known by the want of it"
English saying

Unfortunately there are no scientific rules on how to write,
so I shall not cover complex grammar, construction, or style. Such things often need a very good command of English to understand. For those whose English is this good. the best book is Gowers' "Plain Words" (see the reading list). This book was written to improve the writing of British government bureaucrats, whose complex language often obscured the meaning of their writing. I am certain that other governments have the same type of person. Of course, politicians do this deliberately all the time, to discuss unpleasant facts like murder, unemployment, and telling lies.
Instead, I shall cover more practical aspects of the writing, submission, and revision process emphasising the parts where non-British people find difficulty. Much of this may appear straightforward or even trivial. Sadly, this "idiot's guide" approach appears to be needed (although there are now many good simple articles and books on the subject; the list at the end gives two that are in print). I am afraid that I have no references to articles in other languages that may be helpful: I would appreciate any suggestions and comments from readers. However a list of instructions, as in a food recipe, may not always be applicable because scientific studies differ. For example, an account of a double blind randomised trial is not the same as an evaluation of a new laryngoscope or a new method of teaching.
Nevertheless, I shall emphasise the common ideas and themes in the presentation of any scientific work.
Try to lay out your ideas logically and clearly. State the aim, the means you used to answer the question the results and the conclusion then you will have achieved your purpose both scientifically and as an author. Editors see a lot of papers and have to reject many: why should they choose to publish those that are unclear? An editor I know uses the term "easy way out" when referring to a rejection, and I understand completely what he means: he is making it easy for the readers of his journal as well as himself.

Basic consideration

You must have some good scientific data to report. You started with a clearly stated hypothesis, preferable as a question, such as "does anaesthesia in the upright subject cause hypotension?". The possibility that this proposition is not in fact true is the e null hypothesis (such as, for example "anaesthesia in the upright subject causes no important change in blood pressure") and your statistical aim is to show that this possibility is small enough to neglect. You used an adequate method to investigate and test this hypothesis, and now' have a result that is statistically sound. (Please remember that if you present a "negative" result, you must provide an assessment of the study power. This will allow the reader to see that your "absence of proof" can also be taken as "proof of absence".) Your data are either new, or they support previous observations, or they contradict previous findings. You wrote a short discussion to consider these results in terms of the technical validity, and their implications, and also wrote a summary (abstract).
Even if it isn't required, you would do well to write the abstract in a "structured" way: it will be easier to follow, be more likely to contain all the necessary information, more likely to be accepted, and much more likely to be read and understood after it is published (Table 1)

Remember, it is all that many people will ever read of your work. If this is to be all they will see of your work, make sure that it is the best you can do, and don't be upset by the fact that the rest of your carefully written paper will be ignored. Humility is one of the key characteristics of a successful passage through the authorship process!
Another very useful check-list is one for reporting controlled clinical trials. This should make sure that your data can be properly assessed, used by others, and perhaps included in a meta-analysis.
The essential features are in Table 2. In a large study, a diagram of the flow of patients from eligibility to trial completion, indicating dropouts, withdrawals, lost to follow-up, and so on, can be very helpful.
Decide which journal you will send the article to. This has an important bearing on both the content and style of what you write. This decision will depend on the importance, complexity, and area of interest of the study. An article of wide anaesthetic interest might go to the BJA whereas a more specialised one could be aimed at Regional Anaesthesia.

If you have difficulty in writing English. I suggest:

Avoid too much "cut and paste" from other articles even if they have been accepted for publication or published. Writing is not automatically good, just because it is in print; a published article may merely perpetuate poor English.
If you must "cut and paste" remember that English styles differ between North America and Britain.
Find a colleague with a good command of English to help you. Give your colleague a double spaced copy and ask him/her to be as severe as possible with what you have written. It can only help!

If you can't find a good English writer,
get a good (large) dictionary. Small dictionaries are dangerous: if you use a large one you will see how many things even a simple word can mean (for example, look up the word "mean" and see how a small dictionary could get you into trouble).
Get a good book on medical writing: the present article can't contain all the help you are likely to need. I have suggested two useful short books but there are many more, almost all good. When the writing is done, give it to two colleagues to read. One of them should know what the paper is about: an informed subject for your trial write-up. The other should not know about the design and execution of the study before reading your draft: a naive subject. They will give you different but equally useful sorts of feedback. You will have to revise the article on the basis of what they say about the layout, sequence of the argument, conclusions, and so on. Then, PUT THE PAPER AWAY for two weeks. Then take it out and read it again:

I can guarantee that you will see things to change.

Finally, make absolutely sure that the written form follows EXACTLY the format requested by the journal you are sending the article to, and follow all the instructions about layout and numbering of the pages, margins, number of copies, and so on. I think you should start this as soon as possible in the writing process, rather than drafting the article roughly first, with the intention to "tidy it up", but ending by hastily trying to get it done properly when you are in a hurry to submit it. Some journals have a very useful check-list to use: it is worth making one anyway.

You must follow the rules: this is a serious game.

Use a spell checker if you have written the article on a word processor, but remember that this will not check the context of the word: you can easily get a word correctly spelt, but that is no help if it turns out to be the wrong word!
Very few articles are accepted without modification. If your article is accepted, it is likely to need revision. Follow the assessor's reports carefully. Divide each report up into a series of separate points. Take each point in turn. For each point, you can either

accept the assessor's criticism and modify your text (in your covering letter, tell the editor the page and line where the change is to be found) or
you must answer the criticism in your covering letter to the editor.

This is the next place you have to be humble: most assessor's reports are appropriate and relevant. Even if the report shows that the assessor hasn't read the article, hasn't understood what has been stated, and doesn't know anything about the topic, be very careful how you reply: you may have missed the point of a very relevant criticism.
Occasionally, an assessor may review unfairly or slowly. If you and your colleagues think this has happened, you should first be absolutely sure that you are not mistaken (nobody likes to be criticised harshly). However the Council for Biology Editors has issued "guide-lines for Reviewers"which contain a stringent set of instructions (Table 3). If, after careful thought, you believe that these guide-lines have been ignored I think that you should write a polite but firm letter to the editor indicating your views. This is the only way that this behaviour (fortunately rare) can be dealt with.
If your article is rejected, revise the manuscript anyway and then reconsider submission. If the reviews say that the research is fatally flawed, consider abandoning the paper: far too much poor research is published. If it was always flawed, it was not ethical even to start, but I suppose that you didn't realise that! If you do send it to another journal you MUST rewrite it in the style required: otherwise you are saying "here is a paper rejected by journal A. perhaps journal B won't be as fussy". This is not a good start to getting it accepted by journal B.

Practical points

Use simple and practical words. Abstract ideas are hard to express anyway, and particularly difficult in a foreign language. This may mean that you can't introduce your favourite hypothesis in the discussion I suggest that you stick to practicalities: the reader will have a different hypothesis. Most discussion sections are too long anyway!
Articles such as this one, where the message is ideas rather than facts, are particularly susceptible to mistakes with "fancy words". For example, when I re-read this article I removed the words "These aspects" and replaced them with "This sort" or "This part".
Write short sentences. In some places you may wish to reduce the word count and this can be done by removing little words like "of". This can often make reading very difficult. For example "observed peripheral mean haematocrit reduction" has less words but is much more difficult to understand than "observation of the mean reduction of haematocrit in peripheral vessels". Avoid putting many nouns together: you can often, without meaning to, get absurd or humorous effects.

Avoid jargon: do NOT write as you would speak.

You may speak simply, without a lot of the big words that you may be tempted to put down on paper: but you also speak very loosely.
English jargon includes "line" for cannula, "stick" for arterial sample, and "protocol" for plan. In everyday use, I may say that the blood urea level is rising, but I would prefer to write "the blood urea concentration is increasing". The level of a river may rise, but we rarely deal with "levels" in this sense, in contrast to the many other values that are often called levels (pressure, concentration, tension incidence).
Try to put things in an active way rather than a passive form. For example, say "We measured plasma concentration of morphine.." rather than "Plasma morphine concentrations were measured."
The word "demographic" is often used wrongly. It means "the study of the life conditions of communities of people, as shown by statistics of births, deaths, and diseases". A better word is "characteristics" or "details" when referring to the subjects of a study.
The word "data" is plural, so data are simple, or interesting.

When you summarise data, say what the numbers are (e.g. median, standard deviation).
There are some times when you should use one, and other times when you should use another, form of expressing "variability". Confidence limits are becoming popular, but often with data that are not normallv distributed the quartile values are very useful.

Words commonly misused by non English speakers:

The words "a" and "the": it is not correct, for example, to say "the effect of vertical position"; this should be "the effect of the vertical position".
Plural words such as complications, systems, and their possessive forms. For example, the respiratory and cardiovascular system
(not systems) have complications, but the expression "cardiovascular systems complications" is not used.
Padding words: I dislike particularly the expression "In this respect it should be noted that...". which is often completely unnecessary, or can be replaced by the little word "As..". Other frequent padding- words include "it has been shown that", and "it was noted that".

Final advice

Follow ALL the rules:
keep it simple, and stay humble.

USER NAME

PASSWORD

Publishing research in England

How to go about it

Basic consideration

Practical points

Final advice

Suggested further reading.

News