Assembling the submission package
A submission package is everything a journal asks to receive with a manuscript, in the arrangement it asks for, and what that is varies by journal and by article type. Getting it to the journal intact means reading the journal's own instructions, rehearsing its submission site before the deadline, checking the package before sending it, and settling who carries out the upload.
The components of the package
Unless the journal instructs otherwise, assemble the submission letter, title page, abstract, keywords, body of the text, acknowledgments, references, tables with legends and figures with legends, in that order, with each table and figure on a separate page. The title page should give the article's title, each author's name, academic degrees and institutional affiliation, the departments and institutions where the work was done, any disclaimers, the corresponding author's name and full contact information, the person to whom reprint requests should be addressed or a statement that no reprints will be available, sources of support such as grants, and a running head, consistent with the recommendations of the International Committee of Medical Journal Editors (ICMJE) and the New England Journal of Medicine's instructions. A title page should also give the number of tables and figures, a word count and anything else the target journal's instructions request, and name the corresponding author, who represents the research team in conversations with the editor, with full contact information.
A manuscript sent to a journal may be a research paper, written after a study to answer a specific question; a case report on the condition and treatment of an individual patient; a review article that brings together the findings of a number of studies on a topic; or a letter to the editor expressing opinions with reasonable supporting evidence. Each journal usually has slightly different format requirements, but a publisher generally asks for a title page, author names, the institution the authors represent, contact authors for correspondence and for reprint requests, funding details, a short title, any disclaimers, a conflict of interest notification page, a summary, an abstract usually of 150 to 250 words, the body of the text, acknowledgements, references, tables and illustrations, with the journal's own specifications checked for each.
Some journals ask at submission for the details of the individual authors who participated in the study. Many require an additional individual signed statement from each contributing author, such as the declaration of contributions, final approval and conflicts of interest that The Lancet suggests. Some ask for one author to be nominated as guarantor, responsible for the integrity of the work through every stage up to publication.
Acknowledge the individuals who contributed to the investigation or the manuscript but do not qualify as authors, with their individual contributions, and list the study sponsors. A panel of journal editors in 2009 emphasised that all contributors to a publication's development should be clearly and specifically acknowledged. The acknowledgments can include someone who gave purely technical support or help with the report, and financial or material support not disclosed elsewhere. Readers can infer that those acknowledged support the data and conclusions, so everyone named should be pleased to be acknowledged and agree with the substance of the paper, and written permission is needed from each of them. Some journals have specific online forms for that permission, and others accept a signed note on letterhead.
An author disclosure form to facilitate submission, returned before the kickoff meeting, can collect each author's name, academic degrees and honorifics, academic and hospital affiliations and addresses, contact details including an office assistant's and the best time of day to reach them, their login for the journal's author centre, recommended peer reviewers and an image of their signature.
Submitting to a professional medical organisation is often quite a different process from submitting to a journal, because the work will often be published as official information issued by the organisation, and in a larger organisation it may pass through the editorial department before publication. Obtain the organisation's style guide, if one exists, before starting to write, and check its preferred format for submitted documents, any electronic templates already containing standard wording, its style for headings, its policy on images, its referencing requirements and its preferred method of correspondence, since email may not be favoured for confidential patient or commercial information.
What a journal's instructions ask for
From 2007 to 2009 the AMWA Journal accepted manuscripts for consideration only on the understanding that they had not been published elsewhere and were not under review elsewhere. It asked for the manuscript as a Word document attached to an email to the editor, and the same email had to give the name, address, telephone and fax numbers and email address of the author to whom correspondence should be sent, with written permission from authors and publishers to use any previously published material: figures, tables, or quotations of more than 100 words. Hard copies of figures, if necessary, were to be sent by post with complete documentation of the manuscript they accompanied. In 2009 every manuscript also had to be accompanied by a completed author responsibility disclosure and a signed copyright form, both available in the association's information for contributors on its website.
An article in the journal's first 2009 issue closes with an author disclosure certifying that the author participated sufficiently in the intellectual content, the writing of the manuscript and the analysis of data to take public responsibility for it, believes the manuscript represents valid work, and has reviewed the final version and approved it for publication, and noting no commercial associations, such as consultancies, stock ownership, equity interests or patent-licensing arrangements, that may pose a conflict of interest.
The European Association of Science Editors (EASE) Quick-Check Table for Submissions is meant for journals to put at the head of their instructions for authors, so that the basic information needed for submission is quick to find, making life easier for authors and editors, and editors preparing one can delete or add rows. The table was promoted during the EASE campaign Help Scientists Save Time, launched in October 2020 to simplify requirements for initial submission, and by March 2022 a short guide to building one (version 3.1) had been translated into 15 languages. The example table groups what an author needs: general guidelines (complete, concise, clear; the relevant reporting guideline); word limits for body text, a structured abstract, keywords, lay-language highlights and tables and figures; title-page information including running title, affiliations, corresponding author and ORCID identifiers; body structure, required wording such as ethics approval and a statistical analysis subsection, CRediT contributions and funding, and a reference style required only after acceptance; formatting such as consistent UK or US spelling; submission notes on whether a cover letter is needed, which forms are signed by whom and when, proposed reviewers with attention to diversity, file formats and fees; and journal policies on publication model, preprints, data sharing, peer review type, acceptance rate and time to decision.
Rehearsing the submission site
Most scholarly journals require online submission, which means preparing all the files and disclosures needed, with files, figures and tables in the correct format, and working through every step of the journal's submission software, of which there are different types, until the submission succeeds.
Compose everything offline and paste it into the journal's site, because composing online risks being timed out and losing the work. One prolific author allows two to three hours for an electronic submission even to a familiar journal, because there is always something: some journals want the parts bundled and some separate, coauthors' addresses may need looking up, or the journal requests an extra document, such as the British Medical Journal's 'What this paper adds' box for research reports.
Submitting for the first time is nerve-racking, and trepidation before submitting is likely to stay. With preparation and a plan, that fear may lessen. Rehearse with a dummy submission. Create an account on the submission site early, because some congresses give no instructions until you are inside it; once the target journal is known, start a dummy submission in its portal under the submitting author's login. Walk through the steps, with a dummy abstract or manuscript if necessary and stopping short of the final submit button, to learn what questions are asked, who is allowed to submit, and the limits on word count and number of authors. Read every prompt for surprises, such as a required cover letter nobody has written, and look for questions you may not be able to answer or that are unusual, such as one portal's request for the lead author's birthdate. Gathering all the information and materials from the authors ahead of time avoids delays when it is time to submit. The information commonly requested is every author's email address and biography; suggested peer reviewers with their email addresses, which some journals require, up to three names; the conflict-of-interest statements, including the wording for none; whether the authors know of and want to pay an open access fee; and a summary box or key messages.
What the rehearsal finds goes into a template that can be shared with colleagues and reused, headed with the essentials such as the target congress and the abstract deadline. A template can also help authors stay within the guidelines: showing the maximum word count beside the abstract's current count tells authors that any change must stay within the limit. Clients who like to write the initial draft of the abstract themselves can be given the template to keep them on track. Where a rule needs clarifying, it is usually acceptable to ask the journal or congress. One team that submits to a congress whose instructions allow five authors writes to it whenever they have ten, and is always told yes. Colleagues who have submitted there before often have information that can help resolve a dilemma. Having others review the work can help with a hard problem such as getting under the word count or making sure the data are presented correctly: what looks like a roadblock may not be one.
Checking the package before sending it
Repeat the literature review just before submitting, so that no recent key report or breakthrough study is overlooked. A similar paper published first should not discourage submission, but the playing field has changed. Check that all references are accurate, complete and cited in the text, that abbreviations are identified at first use and drugs given nonproprietary names, that any possible conflict of interest is disclosed, that borrowed material has a consent form signed by the copyright holder, and that there is informed consent to use images that may identify human subjects.
A panel of journal editors in 2009 encouraged authors and publication planners to make their submissions lucid, succinct, well balanced, and relevant to the reader.
Check for background information in the Results, interpretation in the Results rather than the Discussion, new information introduced in the Discussion, overloaded tables and figures, the same data repeated across tables, figures and text, and conclusions inconsistent with the data.
Have colleagues or peers read the work and give feedback before it goes, and have all contributors review the draft. Input from an experienced person who has published extensively and successfully can be invaluable. The final review should always include checking that the references are correct and complete. Reference formats should conform to the instructions for authors, and the references should be checked for consistency and correctness so that the reader can find them.
Check the manuscript against the target journal's instructions for authors just before submission. Instructions vary considerably, and the target journal is ideally decided at the outlining stage so that drafts are developed against its requirements; that is not always possible, and instructions may be revised and updated even when it is. It is also a good idea at this stage to check the consistency of the data across figures, tables and text and cross-check the values against the source data, because errors may have been introduced during multiple rounds of revision. The final documents should be submission-ready and completely free of errors.
Proofread for meaning as well as spelling: a spell-checker accepts correctly spelled wrong words and misses omitted words, and an omission such as a dropped 'no' can change the meaning. A paper will never be perfect, so avoid multiple trivial revisions and at some point declare it done; editors make allowance for small deviations from minor technical requirements that do not interfere with reading, reviewing or editing, and such minor variations will not cause rejection.
An author who is not very computer literate should have someone who is check that the technical formatting matches the publisher's specifications, to avoid delays from work returned because the editor could not open a file or the images were not suitable for printing. Always follow the publication's specific directions for submitting electronic documents, usually given on its website. Keep a copy of everything submitted, which is later used to check proofs. After submission, the journal should send notification that the manuscript has been received.
At submission, have a coworker take a final look, correct what remains, submit, notify the authors and save the notification that the submission succeeded. Technical corrections may still be needed afterwards and a submission may be rejected; mistakes will happen, and the advice is to fix them and move on.
Submitting on the corresponding author's behalf
Medical writers are frequently asked to submit a manuscript on behalf of the corresponding author, typically with the author's own login, because the author is too busy or lacks the experience or the English to handle the upload and the other tasks of submission. The ICMJE Recommendations make the corresponding author the one individual who takes primary responsibility for communication with the journal during submission, peer review and publication, and who typically ensures that the journal's administrative requirements, such as authorship details, ethics committee approval, clinical trial registration documentation and conflict of interest forms, are properly completed, although those duties may be delegated to one or more co-authors. One medical writer's analysis reads this as meaning that a medical writer completing the submission for the corresponding author is not acceptable, which puts the writer in the awkward position of disregarding the Recommendations or refusing and possibly upsetting the client.
The analysis sets out three options. The best, in its view, is to insist that the corresponding author complete the submission, but corresponding authors are often not chosen for their availability, English or ability to manage the submission and correspondence: the role often goes to the first author as an honour, which makes it hard to shift to a secondary author. The writer should then explain that the role is functional rather than an honour, describe the responsibilities, and insist that failing to perform them will create severe problems for publication, and if that fails, explain very diplomatically that a journal could treat it as an ethical breach and grounds for rejection or later retraction. Many clients will appreciate the advice; others will not, and the relationship can suffer.
The second option is to make the writer a co-author, which would let them serve as the corresponding author or act on the corresponding author's behalf, and it turns on whether the writer meets the ICMJE's four authorship criteria.
The third option is to change the system to reflect practice, since the corresponding author often needs a medical writer's help to upload the files and complete the submission. Good Publication Practice (GPP), the main ethical guidelines for industry, permits a medical writer, or an appropriately supervised delegate, to complete the administrative tasks of submitting a publication with the corresponding author's permission and if the journal or congress allows it, which leaves journals as the only barrier. The analysis concludes that the ICMJE Recommendations say a writer should not submit for the corresponding author and GPP says a writer may, and that journals should update their submission systems to allow writers to upload articles and the ICMJE Recommendations should be brought into line with GPP.