How should you decide who will be an author?

When deciding who should be an author, you might feel pressure to include people who have not directly contributed to the work being written about. For example, people who secured funding for the work or who are important in the department that you work in, might say that they should be authors.

As this might lead to questions about ghostwriting and honorary authorship, the International Committee of Medical Journal Editors produced a uniform set of criteria for authorship. These are:

Authors “should have participated sufficiently in the work to take public responsibility for relevant portions of the content” and should meet all three conditions below:

  • Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; AND
  • Drafting the work or revising it critically for important intellectual content; AND
  • Final approval of the version to be published; AND
  • Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

These criteria are also included in GPP2 produced by the International Society for Medical Publication Professionals, which was published in BMJ.

If a person has contributed to some steps but not others then they should be included in the acknowledgements. This might include people who have helped perform an experiment, but not helped in the design, or who have proof read the article prior to submission.

The guidelines also suggest that if the study is very large there could be an authoring committee that takes full responsibility for the content of the manuscript. The full list of study investigators can then be included as an appendix. The article can then be published on behalf of everyone who is involved and if a person who is not on the authoring committee wants to include the manuscript on their CV they can do so.

Hopefully, by having articles published by authority figures on this topic you can feel better able to push back on any demands from people who you feel should not be an author.


ICMJE authorship criteria –

GPP2 –

Don’t aim to write scientifically, aim to write well

When we write something we are often trying to convey who we are, or more specifically who we want to be perceived to be, to the writer. This means that we will alter the tone of our writing according to the audience, an email to a friend will be different from a job application letter. When writing a scientific manuscript it is, therefore, often the case that people write it in a manner they perceive to be ‘scientific’.

What is scientific writing?

When I say scientific writing I mean writing with many technical words intended to convey the idea that you know what you are talking about. These technical words will often be jargon and in some cases might be better conveyed using ‘everyday’ language. This will limit the audience of your work to others who understand the jargon that you are using, as well as potentially making the manuscript more difficult to read.

How can scientific writing be avoided?

Once you have written your manuscript you should get other people to review it, at least one of whom should be reviewing the language used. This should be someone you trust as the feedback might not be what you expect from your ‘perfect’ draft. When you ask this person to review the manuscript you should specifically ask them to keep an eye out for overly ‘scientific’ language.

A second thing that you can do is to read the draft out loud once you have written it. By reading out loud, even if it seems odd, you can identify any issues with ‘pacing’ or where jargon might interfere with the ‘flow’ of the manuscript.

A word of caution

Do not remove scientific words if they are necessary, you should not dumb down just to improve readability. This is a tightrope that you will have to walk, but hopefully by asking others for their opinions this will help navigate the issue.

What are the next steps after the first draft is written?

Once you have produced your first draft and your co-authors have reviewed it, your aim should be to address the comments and get the manuscript submitted to your chosen journal as soon as possible. How you manage this will depend upon the comments you receive.

If they are minimal

  • Make the suggested changes, or ignore comments if you have a reason to do so and can explain it to the person who made the comment
  • Ensure the manuscript is in the style requested by the journal
  • Confirm that all authors are happy to submit
  • Submit to the journal

If there are major or conflicting comments

This will necessitate the development of a second draft and a further round of review. If the comments are conflicting then it is often best to have a teleconference to discuss them and ensure that everyone is in agreement before making any changes. If agreement cannot be reached then the lead author (or the guarantor for the manuscript) should decide what to do.

Depending on the comments on the next draft the cycle will then continue. However, it is important to remember that the manuscript is being written to be published and you should not have innumerable review rounds, demanding perfection if this is at the cost of submitting.

What is an extended outline and why might it be beneficial?

Once an outline has been written, discussed with co-authors, revised as appropriate in line with any comments and been agreed upon, the next step is to bulk out the outline. I believe that producing an extended outline in bulleted form, rather than a first draft written in prose is the best next step. This ‘extended’ outline should be the first draft in bullet form and contain everything that is expected to go into the first full draft.

Why an additional step?

In my experience, it is much easier to make revisions to the order information is presented in, when it is in bullet form. It might be that as you write you realise that the logical flow from one point to the next is wrong and that it is might be improved by shifting bits around. Or it could be that when your co-authors read the manuscript they highlight places where more information is needed, which could again change the flow of the document. If the document is written in full sentences, it can be hard to make these changes as it is difficult to see where it might fit. It is much easier to revise a ‘work in progress’ that a ‘fait accompli’. Overall, I have found that adding in this step saves time in the long run and often results in a better manuscript.

Why write an outline?

I, like many other people I expect, have previously tried to write a novel. I attempted it as part of NaNoWriMo, in which people endeavor to write a 50,000 word ‘novel’ during the month of November. I sadly failed in this attempt and looking back I can see two reasons why I failed. Firstly, although I liked the idea I was sadly not that dedicated to spending all my spare time writing and so only managed about 25,000 words. Secondly, I ‘pantsed’ rather than planned – that is I thought that the great novel would flow fully formed without any real forethought or planning. The resultant work was a complete mess, with what amounted to three discrete short stories linked by events, none of which were that good. This experience has shown me the value of planning or outlining at least some aspects of a novel before starting and in future this is what I will be doing.

Why is this relevant to writing about science?

I feel that the same can be true when writing about science, although often to a lesser degree, as the planning and research for conducting experiments should hopefully lend an idea of structure. Nonetheless, whilst it is often easy to list of the key points in the narrative of the research, these can be lost or forgotten when writing a prose document resulting in a loss of coherence.

I would, therefore, suggest writing an outline when starting to work on a document. This should contain the key points for each section and detail any data to be included. This outline will enable you to see if the manuscript makes sense and if there are any gaps, either data gaps or gaps in reasoning, before more words are in place and the manuscript is more difficult to revise. The outline will also allow all proposed authors a chance to input into the manuscript at an early stage and make suggestions that could potentially strengthen the final piece of work.

How much detail should be included in an outline?

The first outline should not be overly detailed as this allows for the best discussion about the proposed structure and makes people more likely to offer input. I would suggest that key statements are included, without over much detail, in the introduction and discussion sections and more detail in the methods and discussion section as these are fixed and will enable people to make more informed comments on the discussion section.

Could an example be given?

If I were to be writing a manuscript on the epidemiology of type 2 diabetes in the South Asian population of the UK, the first outline for the introduction might look something like this:

  • Type 2 diabetes is a major public health issue and the incidence is increasing
  • Type 2 diabetes is more prevalent in South Asian individuals
  • It has been suggested that there are differences between South Asian and White people in the presentation and outcomes of type 2 diabetes
  • For example lower BMI cut-offs have been suggested for overweight and obesity in South Asian individuals
  • The South Asian population is not in fact a single entity, but rather made up of a number of different sub-populations
  • This database study investigates the clinical characteristics at diagnosis of type 2 diabetes for people from these sub-populations

This (fictional) example hopefully highlights the broad nature of statements made and it can be seen how these build up to the research question, showing why the issue is important. Once all authors have agreed upon the first outline, the points can be expanded upon and fully referenced.

If the message is incorrect, then the document fails…

I have just started perusing The Craft of Editing  by Michael Alley, which is subtitled A guide for managers, scientists, and engineers. Right near the start three different types of editing are highlighted, editing for content, editing for style and editing for form. However, the most important thing emphasized at this point is that whilst a lot of time might be spent discussing editing for style or form, these are relatively unimportant compared with editing for content.

Editing for content is defined as ensuring that the information included is correct, complete, appropriate for the audience, appropriate for the purpose and acceptable for distribution. This is generally managed by a different person to the other types of review as it will often require knowledge of the topic being discussed.

This is extremely important as we might have a beautifully written piece with no grammatical or spelling errors that is completely wrong for its intended use, and does not convey the correct key message, or even worse contains incorrect information and misleads rather than educates. This is very well put in the book where it states:

 If the message is incorrect, then the document fails, no matter how well the message is communicated or what form it is in.

I think this is something we all need to remember as we try to write or edit scientific literature.

In case you want to read this book it can be found on Amazon here: The Craft of Editing: A Guide for Managers, Scientists, and Engineers

What is a key message and why have only one?

When you read a scientific article to gain knowledge of what people have done and what they are thinking, what do you take away? Often when I was working in the lab I would specifically look for reaction conditions that might work for my experiments and ignore the introduction and discussion. However, if you read the entire article you likely will not remember exact experimental conditions once finished, rather you will hopefully leave with the idea that the article answered a question and you will take away an idea. This idea is the key message for the manuscript.

Determining the key message

Often when writing articles you will discuss what the message will be when planning the article and putting together the outline. This message will infuse the article, being proposed or hinted at in the introduction, expanded upon in the discussion and possibly explicitly stated in the concluding remarks. However, if you try to include two separate messages then you will muddy the message and it will be less impactful.

For example, if my message is that the new treatment I am discussing is efficacious and well tolerated I can easily build the article around this message. Whereas if the messages are that the treatment is efficacious and well tolerated, and also that it is extremely cost-effective compared to therapies currently on the market, then there will be two narratives running through the article aimed at different groups of people and this may cause confusion.

What to do if you want to include multiple key messages?

I would suggest that if you want to have two different messages for different audiences that you consider writing two articles. This will improve the ability of the article to reach the appropriate audience, and also allow each message to have maximum impact.

Why should you choose a journal before writing a first draft?

In life it often pays to be prepared. For example, before going for a night out we check that we have enough cash to avoid embarrassment when trying to get drinks. This same level of preparation and preparedness is also essential when you are writing a manuscript for submission to a journal.

Each journal is different and may have different rules on how much you can write and how to structure your manuscript. This might limit the number of figures or references you can include, or request specific details be included in subsections. You may think that you can edit your manuscript to meet these guidelines at a late stage, but this can be hard. If you have cited 100 references and are allowed a maximum of 30 which should you cull?

Maybe you think that you do not need to abide by these rules and your work will be accepted whatever. I have found that this is often not the case. For example, if you submit a manuscript that is overly long the editor will send it back with a request to edit the word count down and resubmit. This is additional work and time that could be better spent moving the research forward or writing grant applications.

I would, therefore, advise you to select your target journal, having written an outline and before you write a first draft. This enables unconstrained creativity when developing the initial flow of the article and deciding what to include, but then allows a targeted approach after this, which can maximize your chance of being considered by your chosen journal. By selecting the journal after the outline you also have more information at your disposal about the proposed manuscript and can reject journals if they do not allow a particular aspect of your manuscript, possibly seven huge tables or a video.

So be prepared and increase your chances of success!