Follow us on twitter Follow us on X
Expand All
Copy-editor's brief

This section gives a basic outline of what tasks are required for each article. More detailed information on each point is given elsewhere in the style guide.


  • Please acknowledge receipt of articles sent for copy-editing and let us know whether you can take them.
  • We normally request that articles be returned within 3 working days. Please let us know if you need any significant extension to this.


  • The text should be edited for accuracy, clarity and consistency, following the style guide.
  • In general, retain the author’s style but make changes to assist readability.
  • PerfectIt should be used as a pre- or post-editing tool. Let us know if you have any problems with this.
  • Papers for copy-editing will have words or parts of words highlighted in red, blue or green to help you identify words or phrases that may need attention (see guide to pre-editing checks for details).
  • Please complete an author query form for each article with a list of queries to be passed on to the author.
  • Please also complete the copy-editor’s checklist and note any queries for the attention of the Editorial Office.


  • Please cross-check all data in the paper, where possible. In particular, check that results reported in the Abstract are consistent with the rest of the paper and are rounded correctly; similarly, cross-check data reported in the Results to the figures and tables, as appropriate. However, note that some data may be reported in the Results that are not also included in a table or figure; this is fine.
  • Make sure that numbers given in flow charts add up and are consistent with the text.
  • Where possible, data within tables should be checked – for example percentages can be checked where it is clear how they were derived.
  • Check citations and discussion of figures and tables in the text to ensure they match the information given in the figure/table, and that the correct table number or figure part has been referred to.

Tables and figures

  • Check that figures are cited in the main text in consecutive order. Individual figure parts, however, do not need to be cited in the main text, or in consecutive order.
  • See the Tables and figures section for more details.


  • ESM does not need to be copy-edited, but please check the following:
    • that ESM is all present and correctly labelled
    • that the ESM citations match the information given in the ESM (i.e. that the correct number or figure part has been given)
  • If ESM citations are missing or in an incorrect order, this should already have been picked up by the Editorial Office and an author query raised at the top of the paper.


  • Before papers are sent for copy-editing the reference citations are checked for numerical order, the reference list is checked for duplicates and the author will have been asked to change the reference format if it deviates too much from our standard style.
  • Further formatting changes are made by the typesetters as detailed in the References section of the style guide.
  • Scan references for potential problems and, where possible, correct any mistakes by checking on PubMed or raise as an author query.
  • Cross-check all author names given with citations in the text to make sure they match the reference list. Either the first author or the research group leader (usually the last author) may be named; either is acceptable.
  • Check for missing information (see References for details of format and required information).


  • Please check all website URLs in the text and reference list, and raise any problems as an author query and/or note them on the copy-editor’s checklist.
  • Ask authors to provide an access date for websites given in the main text and references, except where the website is for general information purposes only, for the download of software for which a version number has been given, or links to a reference with a clear publication date.

Top of section

General points
  • Language – set to UK English in Word
  • Work with track changes on
  • Spelling: UK English, Oxford English Dictionary, -ise ending
  • In general retain the author’s style, but make changes to assist readability
  • Not Oxford (serial) comma, except for clarity. For example: Diabetes is common in Somerset, Cumbria and Lancashire. The incidence was measured in Somerset, Cumbria, Hereford and Worcester, and Lancashire
  • Papers for copy-editing will have words or parts of words highlighted in pink to help you identify words or phrases that may need attention (see guide to pre-editing checks for details)
  • The typesetters automatically format the following items in the text and tables (but not in table headers and footnotes or figure legends) so there is no need for you to:
    • change x to a multiplication sign or ‘ to a prime symbol
    • italicise p in p value, n, F in F test, U in U test, z in z score or Z test, r or R (printed version will be p value, nF test, U test, rR etc.); however, the typesetters will not change the case, so please ensure that, for example, p is lower case
    • change hyphens to en-rules between pairs of numbers, or close up spaces around en-rules
    • alter spacing around mathematical symbols (e.g. +, ±, =) (printed version will be, for example, 0.38 ± 0.5, p<0.05, with thin spaces); signs abutted by text will have space around them, e.g.mean ± SEM; multiple units and equations will be spaced, e.g. units for blood glucose AUC: mmol/l × h; the AUC for the low-glucose diet was 24 mmol/l × min × 10
    • alter spacing between values and units (printed version will be, for example, 5.6 mm, 8.8 g)
    • change hyphens or en-rules to minus signs in mathematical contexts, although this will still be needed in genetic contexts; you will not be expected to mark up hyphens to be changed to minus signs in figures, as this can result in too many fiddly changes
    • add commas above 9999
Author queries

Use the author query form to list author queries. We have compiled a list of commonly used standard queries that you can adapt for specific circumstances. These should be used as the basis for your queries. We also ask you to complete a copy-editor’s checklist. Please keep language queries to a minimum – they can often be condensed into one or two global queries. Once we receive your copy-edited paper and the queries, we will forward the queries to the author. We will also forward the copy-edited paper to the author, asking him/her to review the paper and make changes in response to the queries.


Proofs are supplied as a pdf together with a Word version of the copy-edited paper and a pdf of the copy-edited figures. Please proofread the pdf proofs and figures against copy. If the metadata include ESM figure legends (rare) please check these; otherwise there is no need to check metadata. We do not ask proofreaders to check ESM

  • Please check that abbreviations have been used consistently and are defined in the abbreviations list and at first mention in the Abstract and text.
  • Please cross-check data where it is possible to do so in case errors have not been picked up during copy-editing. In particular:
    • check that results reported in the Abstract are consistent with the rest of the paper
    • ensure that numbers given in flow charts add up correctly and are consistent with the text
    • cross-check data reported in the Results with figures and tables, as appropriate
    • where possible, check data within tables; for example percentages can be checked where it is clear how they were derived. Sometimes inconsistencies in percentage calculations can be due to data not being available for the whole cohort for certain measurements – this should be explained in a footnote
  • Check that decimals are rounded consistently.
  • Please proofread references and correct any errors; if a reference looks incorrect, check on PubMed or raise as an author query.
  • Check that colours and styles listed in figure legends match the figures e.g. ‘grey bars/black bars’; ‘red dotted line/green dashed line’.

Returning proof corrections

Please either print out the proof and mark up corrections by hand, returning the marked-up proof by email (adding the suffix ‘proofread your initials’ to the file name) or fax (0117 414 7887)
Mark up the corrections on screen. In this case, please use sticky notes not text edits, and do not use highlighting. The sticky notes should be placed at the text to be corrected rather than in the margin. You can then return the pdf via email, adding the suffix ‘proofread your initials’ to the file name.

Layout and style

Title page


  • Check for sense and succinctness. Abbreviations are acceptable in the title if they make it more readable. Abbreviations should not be defined in the title, but should still be included in the abbreviations list and defined where they first occur in the Abstract and main text..
  • Make sure that any animal species are named (note that animals are used as models of type 1 and 2 diabetes; they cannot have type 1 or type 2 diabetes). However, if the study covers both human work and non-human models, or several different animal models, this may not be necessary, in the interests of succinctness.
  • Use a colon rather than dash to separate title and subtitle. Use lower case after the colon.
  • We do not include a short title, so this can be deleted.

Authors’ details

  • The author list and affiliations will be formatted by the typesetters, so no need to worry about commas/dots or ‘and’ between authors’ names, or to convert the affiliation list numbers to superscript (although superscript numbers are preferable in the author list, as easier to follow)
  • Family and given names should be written out, with other initials; e.g. Matthew N. Burton; J. Maria Johnson; Jean-Marc Dupont. There is no need to format the author list, e.g. names can be separated by centred dot, comma, space, etc. Delete academic titles.
  • If any of the authors’ names include particles (e.g. de, de al, von, del, etc.) or unusual abbreviations (Ch., Ph. etc.), add an author query to find out how the author would like their name to appear on PubMed.
  • Affiliation details should be given in the order: department/division, institution/organisation, town/city, state (for USA, Canada and Australia; also for some other countries, if provided), country. Street addresses and postal codes are not included.
  • There is some flexibility in affiliation order, where organisations have specific requirements for how the affiliation should be set out.
  • Abbreviations can be used in the affiliations without definition, as this wording might be a specific requirement of the organisation.
  • For UK addresses, the affiliation addresses should not include England, Scotland, Wales, Norther Ireland.
  • Present addresses appear as an additional affiliation, e.g.
    • Present address: Department of Endocrinology, University of Bristol, Bristol, UK
    • A retired author giving a private address does not need to provide full details, and can just give city and country if preferred.
  • For US, Canadian and Australian states, use the two- or three-letter Post Office abbreviation e.g. ON not Ontario. The name of the country is not abbreviated, except for the USA and UK.
  • Names of institutes/universities and towns/cities/countries can be left as supplied by the author (i.e. they do not have to be the English spelling).
  • INSERM, which is a French group of research units, should be Inserm (lower case with initial capital letter).
  • The Spanish Centro de Investigacion Biomedica en Red (CIBER) organisations, e.g. CIBERDEM and CIBERESP, are umbrella organisations that are associated with Instituto de Salud Carlos III (ISCII), Madrid, Spain.

Authors can EITHER include full details including the city, e.g.:
1. Centro de Investigación Biomédica en Red de Diabetes y Enfermedades Metabólicas Asociadas (CIBERDEM), Instituto de Salud Carlos III (ISCIII), Madrid, Spain
OR leave out ISCIII and city details and include the URL instead, e.g.:
1. Centro de Investigación Biomédica en Red de Diabetes y Enfermedades Metabólicas Asociadas (CIBERDEM), Spain

  • Affiliations for the German organisation DZD should be set out as follows (changing the city details as appropriate):

German Center for Diabetes Research (DZD), München Neuherberg, Germany

  • If the affiliation includes Singapore, add ‘Republic of Singapore’. Inform the author of the change
    … , Singapore, Republic of Singapore
  • In ‘Queen Mary College University of London’, there is no comma after ‘College’
  • If there are two corresponding authors, ask which of the authors should receive the proofs (see standard queries)
  • Corresponding author details should be below the affiliations, and just include the corresponding author(s)’ name and email address.
  • ORCID iDs can appear alongside each author name, with the corresponding author details, or in a list on the title page.


  • Move statements such as ‘Part of this work was presented/was published as an abstract’ to the acknowledgements section
  • For two or more authors who contributed equally to a study, add a note, for example:
    • Peter A. Bloggs and Rowena J. Smith contributed equally to this study; or
    • Peter A. Bloggs and Rowena J. Smith are joint first authors. Greg P. Stanley and Susan R. Murray are joint senior authors.
  • For authors who have died, a footnote should be included based on the following style:
    • Professor G. Stanley, who supervised this research, died on 17 June 2010 before publication of this work.

Received / Accepted dates

  • These dates should already be on the paper (date format: 4 July 2002). Springer will add an online publication date at a later stage


  • The abstract should have a four-part structure (Headings: ‘Aims/hypothesis’, ‘Methods’, ‘Results’, ‘Conclusions/interpretation’, as italic shoulder headings). Please check that the author has included data to support the main findings of the study and raise an author query or ask the Editorial Office if you are unsure. Please check any data in the abstract against the main text. Please check that the abstract is clear and engaging – remember that it is one of the entry points to the article. If abstracts are written in ‘telegram’ style, i.e. incomplete sentences, amend to full, written-out sentences. There should not be any citations. Include as few abbreviations as possible (these should be written out in full on first mention in the abstract if not in the accepted abbreviations list and again on first mention in text).

Data availability

  • If the data availability statement includes a URL linking to a data repository (e.g. Gene Expression Omnibus [GEO]) or to code used for analysis (e.g. on GitHub), that section of the statement should be duplicated below the abstract, with the side heading ‘Data availability’, as well as appearing in its usual position in the end matter (just after the acknowledgements).

Randomised controlled trials

  • If the study is a randomised controlled trial (RCT), the journal follows CONSORT guidelines for the abstract structure and content. Authors should also complete a CONSORT checklist for abstracts; this should be sent to you with the paper for copy-editing. The abstract must conclude with a clinical trials registration number and details of funding. For example:
    • Trial registration NCT00056378
    • Funding The study was funded by Galactozone.
  • If the trial number is stated in the text, then the style is:
    … ( registration no. NCT00056378)…
  • The paper should also include a flow chart. The Editorial Office has a checklist for RCTs.

Clinical trials registration numbers

  • In line with the requirements of the International Committee of Medical Journal Editors (ICMJE), and CONSORT guidelines, Diabetologia requires authors to have registered their trial in a registry at the time of conception.
  • Acceptable registries include any registry that is a primary register of the WHO International Clinical Trials Registry Platform (see or in (
  • The abstract must conclude with a clinical trials registration number and details of funding. For example:
    • Trial registration: NCT00094757
    • Funding: The study was funded by Galactozone.
  • If there is no abstract, for example in a Research Letter, then the registration number and funding line go before the Acknowledgements.
  • If the registration number is not present, please contact us.

Text boxes and Research in context

  • Text boxes and Research in context boxes will be formatted in house, by the Editorial Office, and are sent for typesetting as separate files, although they will be included in the main Word file for copy-editing.
  • Research in context boxes are always positioned at the top of the second page in the published PDF; for other text boxes, please ask authors where they should be placed, if this is not already clear.
  • If authors have put text in a box or table, there is no need to remove it from the box when copy-editing.
  • Spelling: Text box; Research in context.
  • No full stop at end of each section.
  • No internal gridlines.
  • Remove any font colouring or highlighting.
  • Edit the text in either type of text box for clarity and conciseness; check against the main text and abstract to make sure it is consistent.
  • See examples of Text boxes and Research in context boxes.

Research in context

Authors include a ‘Research in context’ summary (<200 words) below the abstract, which includes the following questions with responses:

  • What is already known about this subject? (maximum of 3 bullet points)
  • What is the key question? (one bullet point only; formatted as a question)
  • What are the new findings? (maximum of 3 bullet points)
  • How might this impact on clinical practice in the foreseeable future? (one bullet point only)

This should be edited to make sure it is clear and to-the-point, and is in line with the findings as reported in the abstract and main text. Abbreviations should generally be defined at first mention, in the same way as in the abstract, although we have a bit more flexibility and may allow abbreviations to be used without definition where this reads better.


Authors are asked to provide a tweet/post that we will use to highlight the article using the @DiabetologiaJnl handle on  If the author has not included a tweet, we may ask copy-editors to suggest one. This can be based on the title or the main findings of the paper and should be less than 250 characters. Abbreviations can be used without definition for brevity, including abbreviations we do not generally use, such as T2D, T1D. Hashtags can be added for key terms.


  • Up to ten keywords can be included (we are flexible on this and a slightly higher number can be included, if appropriate). Check for alphabetical order. Keywords should have an initial capital letter only.
  • For a review, meta-analysis, systematic review or umbrella review, add ‘Review’, ‘Meta-analysis’, ‘Systematic review’ or ‘Umbrella review’, respectively, to the keywords.
  • If the first character or characters in a chemical compound is a prefix or numeral (such as α-, O-, cis-) then the prefix or numeral is ignored for the purposes of alphabetisation. Other terms that begin with symbols or numerals (such as 30 min glucose tolerance test) are listed at the start of the alphabetical list.

Abbreviations list

  • The list should include only abbreviations and acronyms not listed in Instructions to Authors as accepted abbreviations and used more than once. Abbreviations not on the approved list should be defined in parentheses on first use in the abstract and on first use in the main text. If the authors have defined abbreviations that are on the journal’s accepted list, delete the definition here and in the main text.
  • Abbreviations should be in alphabetical order (see Keywords for information on ordering terms with prefixes and numerals).
  • The written-out form of the abbreviation in the list is all lower case but with an initial capital letter (except for proper names, name of trials, etc.).
  • Authors do not have to use the accepted abbreviation in place of the full version, although there are some terms that we would expect to see abbreviated, such as BMI.  Authors should be consistent, and once an abbreviation has been introduced it should be used throughout. There are a few exceptions to this rule, for example ‘s.c.’ might be used when giving experimental details, and ‘subcutaneous’ when mentioned more generally.
  • Abbreviations used only once in the Methods can be handled less strictly. Versus is always abbreviated to vs (no final point)
  • Do not include any gene symbols, as these are not strictly abbreviations.
  • Short abbreviations for diseases are not allowed, such as T2DM, T1DM, DN (diabetic nephropathy), DR (diabetic retinopathy), AZ (Alzheimer’s disease), MetS (the metabolic syndrome). They may be used in figures and tables for space reasons, and defined in the legend/footnote. Please note that DKD (diabetic kidney disease), CKD (chronic kidney disease) and GDM (gestational diabetes mellitus) are allowed to be abbreviated after they are first defined (please ask the Editorial Office before making global changes if you are unsure).
  • We do not have a general rule for avoiding short abbreviations, although abbreviations that could be confusing or unnecessary should be removed, for example FM, flash monitoring; GK, glucokinase. Ask the Editorial Office if you are unsure.
  • Trial acronyms do not need to be written in full, although they can be if authors choose to do so. This doesn’t need to be consistent within the paper (e.g. authors may choose to write out a less well-known acronym). A reference should still be included for trials mentioned in the text.
  • Abbreviations do not need to be defined or written out in the affiliations or the end matter, and do not need to be used consistently in these sections (e.g. a trial acronym may be used throughout the text, but the authors may opt to use the full trial name in the end matter – this is OK).
  • There is no written-out version of BIGTT (a test to measure beta cell function, insulin sensitivity and glucose tolerance simultaneously).
  • Three letter and one letter amino acid abbreviations (e.g. Gly, G) do not need to be defined.

Body of text

Levels of headings must be clearly differentiated by typographical means by the copy-editor:

  • Style 1: If there are two heading levels:
    • H1 Heading on line by itself in bold (14 pt in Word)
    • H2 Shoulder heading in bold (12 pt in Word, text runs on)
  • Style 2: If there are three heading levels:
    • H1 Heading on line by itself in bold (14 pt in Word)
    • H2 Heading on line by itself in bold (12 pt in Word)
    • H3 Shoulder heading in bold (12 pt in Word, text runs on)
  • Either Style 1 or 2 should be applied throughout the paper (not, for example, Style 1 in the Introduction and Style 2 in the Results).
  • Please check that gene symbols are correctly formatted with italics in bold subheadings.
  • Please see here for examples of articles styled with Style 1 and Style 2.
  • Displayed quotes are styled as follows: the quotation in is italics, left justified; the author/source is upright, left justified on the line below the quote. For example:
    • Life is like a jar of honey!
      John Golding


  • Note that sections are headed Introduction, Methods, Results, Discussion.
  • Check heading hierarchy and style appropriately.
  • If an additional section ‘Conclusions’ has been added by the author, please include this as a subsection of the Discussion.
  • Delete trademark devices or type of company (e.g. Ltd, Inc, GmbH, Co, Corp, LLC).
  • Check that the suppliers’ or manufacturers’ details are given at first mention in the text for animals or other biological materials, or for non-standard materials (supplier name, country). After the first mention, only the manufacturer’s name need be given and not the location.
  • Check that there is an explanation of ± (standard deviation or standard error) in the text or in the Statistical analysis section.

Author names and other names throughout

  • The names of individuals mentioned in the main text and Acknowledgements sections should include initials and last names, but not titles (e.g. J. P. Smith). As an exception, the names of authors mentioned in the text can be written either as initials and last names (as above) or as initials with no full stops; e.g. ‘Two authors (AB and CD) assessed the images in a blinded fashion’. Make consistent as appropriate.
  • In the Funding, Authors’ relationships and activities and Contribution statements, author names are given as initials with no full stops.

Ethical issues

  • Check that the paper includes an ethics statement.
  • Studies on people: there should be a sentence stating that the study has the approval of an ethics committee and/or that studies have been carried out in accordance with the Declaration of Helsinki. It is not mandatory to have both. There should also be a statement to the effect that informed consent was obtained from each participant.
  • Studies on children: for studies in children/minors, informed consent is obtained from parents/guardians on behalf of the child. If the authors have not mentioned ‘assent’ (a less formal process of obtaining agreement to any procedures directly from the child, in addition to formal consent from the parent/guardian), please raise a query asking whether assent was sought/obtained.
  • Studies on animals: there should be a sentence stating either that the study was approved by the local ethics committee or that the study was conducted in accordance with the Principles of Laboratory Care. Details of the breeds and suppliers should also be included.
  • If the relevant statement is missing, raise as an author query.
  • Database studies: for studies involving patient medical data, we would expect to see a statement that the study has the approval of an ethics committee and/or those responsible for data privacy in the health authorities involved (the Caldicott Guardians for UK NHS data). The use of data from certain publicly available databases, in which patient data are not identifiable (such as the DIAGRAM, MAGIC, GIANT, GLGC, ICBP and ADIPOgen databases), does not need ethics approval, although there are certain guidelines that need to be followed for citing the source, and which are given in the respective websites.


  • When mention of race or ethnicity is relevant to an understanding of scientific information, be sensitive to the designations that individuals or groups prefer. Follow author usage where appropriate  (AMA Manual of Style [2007] 10th edition New York Oxford University Press p 415).
  • Avoid the use of Caucasian to indicate white, as it is technically specific to people from the Caucasus region (AMA Manual of Style (2007) 10th edition New York Oxford University Press p 415. Preferred terms are of European descent, Europid or white (raise as an author query) See also Bhopal & Donaldson, 1998.
  • American Indian is generally preferred to the broader term Native American, which is acceptable but also includes Hawaiian, Samoan and Alaskan natives. Wherever possible, specify the nation of peoples (e.g. Inuit, Navajo) rather than using the more general term.
  • Similarly, Asian persons may wish to be described according to their country or geographic area of origin, e.g. Chinese, Indian, Japanese, Sri Lankan, not Oriental or Orientals.
  • For Indigenous Australian, include I with capital letter.

End matter

End matter should be in the following order, using these headings:

  • Acknowledgements
  • Data availability
  • Funding
  • Authors’ relationships and activities
  • Contribution statement

Editing end matter

  • Please see notes below for specific sections, including style for names and initials.
  • Other than this, please minimise changes to the Funding statement, Acknowledgements and other sections referring to funding agencies or study sponsors. If you think changes may be needed, please query this with the author first.
  • There is no need to spell out abbreviations or acronyms in the end matter.
  • If an abbreviation or acronym for an organisation or study group has been defined and used elsewhere in the paper, it doesn’t need to be used consistently in the end matter if the authors have opted to use the full version here.
  • Some funding agencies require that a disclaimer is included. This is particularly important and could necessitate an erratum if it has been left out, so please take particular care when editing these.
  • Please include the standard ‘Funding’ AQ (in the table of general queries) in all instances except where there is a negative funding statement (e.g. ‘This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.’).
  • Trademarks and other symbols are allowed for company names in the end matter (although we delete them in the main text). Company names in the ‘Authors’ relationships and activities’ section only need a light edit, although if there are incosistencies (e.g. ‘Sanofi’ in one place, ‘Sanofi Aventis’ in another), ask if they should be made consistent.


  • Heading: bold, run-on
  • Use an English translation for foreign organisations where appropriate.
  • Initials and last names, but not titles, should be given with brief affiliations (e.g. J. P. Smith, Kings College London, London, UK).
  • If it looks as if there has been assistance in writing or editing the paper, please ask the author to give details of the person/company; raise a query regarding sponsor involvement if it appears that a commercial sponsor may have been involved in editing or writing the paper (see standard queries).

Data availability

  • Statements of data availability should be included under this heading, and should include links to data repositories or to code used for analysis, as well as details of how to access any additional data. This will include a repetition of statements that contain a URL for data access, which are also included below the Abstract. If authors have included links to repositories or code only within the main text or ESM, please copy them to the data availability statement as well as the abstract – details can appear in all three places. Please click here for examples of appropriate wording. This can be tweaked at copy-editing as appropriate.


  • Heading: bold, run-on
  • Author names are given as initials with no full stops.
  • Authors should be asked to provide a statement even where no specific funding is received: This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

Authors’ relationships and activities

  • Heading: bold, run-on
  • Check that a statement is included, even if it is negative.
  • Format for negative statement: The authors declare that there are no relationships or activities that might bias, or be perceived to bias, their work.
  • For a positive disclosure of interest, the author’s initials (no full stops) should be used, e.g.: PTJ has an 80% holding in Random Drug Ltd. All other authors declare that there are no relationships or activities that might bias, or be perceived to bias, their contribution to this manuscript.
  • Check the affiliations and raise a query if any of the authors is listed as employed by a company or commercial organisation, and this is not stated in the statement of the authors’ relationships and activities.

Contribution statement

  • Heading: bold, run-on
  • Author names are given as initials with no full stops (unlike elsewhere in the paper where the format e.g. P. Bateman is used).
  • The authors’ initials should be cross-checked against the author list to ensure that they are correct and complete).
  • Text should be given as prose, rather than a bullet list.

Note added in proof

  • A note added in proof can be included as a footnote on the title page or placed before the Acknowledgements in the same style


  • Any appendices (e.g. lists of study group members) are placed before the references.

Electronic supplementary material

  • Electronic supplementary material (ESM) is provided by the authors as a PDF ready to be posted online. The PDF should contain all the ESM (with the exception of video files or ESM tables supplied as Excel files), in the order: ESM text (methods and results), tables, figures, and finally lists of study group members, if applicable.
  • Check that there are call outs to the ESM at appropriate points in the text and that the citations are in the correct order.
  • Call outs should be in the form, for example: ‘ … (see electronic supplementary material [ESM] Table 1)’. For subsequent call outs ‘ …(ESM Table 1)’. There is no need to include ESM in the list of abbreviations.
  • Video/movie files should be named ESM Video 1, ESM Video 2 etc. with corresponding call outs. If a paper contains video files, raise an author query to ask authors if they would like them to be streamed in the HTML version of the online publication (multimedia manuscripts).
  • Add the standard author query to give the author a chance to amend the ESM in the light of the copy-editor’s queries on the paper.

Tables and figures

  • Check that tables and figures are cited in the main text in consecutive order. Individual figure parts, however, do not need to be cited in the main text, or in consecutive order. Use ‘Fig.’ throughout text unless the word occurs at the start of a sentence, in which case use ‘Figure’. For multipart figures use ‘Fig. 1a,b’ and ‘Figs 2, 3’ when referring to separate figures; spacing will be standardised by typesetters.
  • There is no need to redefine abbreviations used in figure legends and table footnotes if they have already been defined in text and are included in abbreviations list.
  • In general, use superscript letters rather than symbols for footnotes, except for significance values (use *p<0.05, **p<0.01, ***p<0.001 as far as possible; the sequence for symbols for other levels of significance is: †, ‡, §, ¶; please don’t use # or $ signs).
  • Proofreaders: note that if a figure spans two columns and there are more than two lines in the legend, then the legend is spilt into two columns. If the legend is only one or two lines, then the legend runs across the two columns.
  • Reference citations in a table or figure legend that do not also occur in the text should EITHER be listed at the end of the reference list OR listed in numerical order at the position of the first call out to that figure or table in the main text. Either system is acceptable, but authors should be consistent.
  • If comparing different studies, citations in a table should be given in the format Name of study or first author’s name followed by et al (give names of both authors if there are only two authors), year of publication (optional), the reference citation number in square brackets, e.g. Smith et al (2008) [14]. PMIDs should be replaced with these details. See also Large tables below.
  • If figures or tables are from another paper (or modified or adapted from another paper) then the corresponding author will need to provide written proof that he/she has permission to reproduce the material, unless the material has been published as open access under a CC BY licence ( This will apply even if the author is the same, or if the material is from a previous Diabetologia (copyright Springer) paper, although in these cases there will generally be no charge.  Let the Editorial office know if you think permission may be required.


  • Figures can be marked up as PDFs using electronic comments and mark-up (for example in Adobe Reader, Adobe Acrobat or other PDF readers) or by hand and scanned. Our preference is for electronic mark-up, as it allows us to edit and add to your changes. Please return them with the copy-edited manuscript and author query form.
  • See examples of marked up figures and refer to the Artwork Guidelines in the Instructions to authors.
  • The Editorial Office might ask the authors to make further changes to the figures, or we may ask the typesetters to make all changes. This can depend on the complexity of the changes needed and the file format (the typesetters cannot easily amend non-vector/non-editable files [e.g. Tiffs], and have to re-key all the text throughout the figure if any changes are made). We can’t always decide on the best approach until we have checked the edited manuscript, so please use clear mark-up that can be understood by either the author or the typesetter (proof symbols should be avoided, as the authors may not understand these, and the typesetters are now used to working with comment boxes, provided the instructions are clear).
  • Where it is clear that the authors will need to supply new figures (e.g. from notes to the authors at the top of the manuscript), please add comments/queries linked to the relevant figure legend, listing additional changes required, as well as marking them up on the figure, so it is clear what the authors need to do (they may not be able to view the changes in the marked up PDF).
  • On the first page of the marked up figure file, include a box with the first author’s name and the paper number, and include a box with the figure number for each figure.
  • Add global instructions as necessary, e.g.:
    • Figure part labels in bold
    • Axis labels not in bold
    • Make font size consistent (raise for individual figures only if needed, to avoid unnecessary work)
    • Reduce weight of axis lines to 0.75 pt (raise only where lines are particularly thick)
  • Each separate figure part should generally have a label, although we make exceptions, including: gel plus associated bar chart; micrograph plus associated quantification; related micrographs; associated AUC graphs; linked grids of graphs where relabelling would create extensive extra work. Generally, if the discussion of figures is not understandable without use terms such as ‘upper’, ‘lower’, ‘left’ and ‘right’, then figures should probably be given individual part labels. If you think a figure may need to be relabelled, please check with the Editorial Office.  You would then need to amend the figure legend and main text figure citations, and ask the authors to check carefully.
  • Change figure part identifiers to lower-case bold (abc, etc.) and align them with y-axis label if they are particularly misaligned (no need to change them if they are only slightly out of place and look tidy); add a global instruction for this if appropriate.
  • Watch-out for non-SI units. Plasma glucose or plasma insulin and units derived from them are the usual suspects. If non-SI units are used, please ask for a replacement figure. Details of the correct units as well as conversion factors can be found on
  • Please ensure it is clear in the figure or main figure legend how the data are presented. In addition, where box and whisker plots are presented, please raise standard AQ to ask authors to describe what is shown if not already included.
  • Figures should be correctly worded and labelled, with capital letters for first letter of the first word of axis legends/labels only (or for abbreviations or acronyms, as needed).
  • Check that curves and experimental points correspond with axes, and that the upper and/or lower axis value shown exceed the highest/lowest data points.
  • Frames or headings may need to be deleted from graphs, but they can be retained if useful. Check with the Editorial Office if you are not sure.
  • If a graph has two y-axes, the labels for the axes should both face inwards towards the body of the graph.
  • Indicate a line break in an axis label if the label is much longer than the axis (but first consider whether there is sufficient space across the page to allow for this – we can be flexible if space is tight).
  • Indicate that x-axis labels set on the vertical should be rotated if there is space so that the text is horizontal or diagonal. You may then need to indicate line breaks if the labels are too wide.
  • Visual keys are preferable unless they make the figure too large to be set at a suitable size, in which case they should be deleted and information added to the legend (asking the authors to check). Ask the Editorial Office if unsure.
  • The typesetters can’t accept pictorial symbols in the figure legends so, in the absence of a visual key, they should be written out.
  • If a visual key applies to more than one figure part, check whether this is obvious. If necessary, add clarification to the legend (e.g. ‘The key in a also applies to c, e and g’), or check with the author.
  • If the authors have duplicated the key by including both a visual key and a description in the legend there is no need to delete the key or the legend text (although check for consistency across figures – if they have used this approach for all figures, it’s OK to keep it, but consider deleting the redundant text if inconsistent).
  • Authors are generally requested to provide column scatter graphs or bar charts with the data points superimposed on the bar. Please make sure that means and error bars can be seen clearly through the data points (e.g. by using grey fill for symbols and black for the means and error bars).
  • If the authors have included the ‘n=’ number in the figure legend, make sure it matches the number of data points on the graph. If authors have not provided n values or have provided n value ranges (e.g. n=3–6) for graphs without individual data points, then please request specific n
  • Add commas for numbers ≥10,000.
  • Indicate prime signs and multiplication signs if they do not appear to be correct.
  • No need to mark up hyphens to be changed to minus signs, as this can create unnecessary extra work.

Statistical significance

  • Significance is usually shown by superscript symbols. Amend the symbols on the figures and in the legends as necessary, or raise a query for the authors to make changes.
  • Where statistical symbols are included on graphs, please make sure it clear what the comparison is (e.g. in time curves, are data being compared between groups or vs baseline).
  • Symbols are used in the order: *†‡ § ¶, with different symbols used for different comparisons. We can be flexible in the order the symbols are used, so no need to make extra work if symbols are used sensibly by the authors, but in a different order.
  • Reserve *, ** and *** for p<0.05, p<0.01 and p<0.001, respectively.
  • The hash symbol (#) and dollar sign ($) are not accepted.
  • Asterisks do not need to be in superscript in the legend, although other symbols should be.
  • Use double or triple symbols to show a second significance level for comparison of the same variables (e.g. *p<0.05 for plasma glucose vs time; p<0.05, †††p<0.001 for plasma insulin vs time).
  • We generally avoid showing p values on the figure—statistical symbols should be used instead, although there are exceptions, for example: occasional non-significant p values (but query with the authors whether they can be removed, as we do not generally include non-significant p values); p and r2 values can be retained in a correlation plot provided there is space for them within the axis; a p value shown for group comparisons on a line graph.

Forest plots

  • In the first column the author should include either:
    • the name of the first author of the study plus et al (without a full stop) and the year of publication, for example: Green et al, 2009 [21] (if there are only two authors on the paper the name of the second author is included. For example: Green and Brown, 2009 [22]); or
    • the name or acronym of the study, for example ADVANCE [23] (the year(s) for the study can be included, but this is optional); or
    • a grouping identifier, depending on how cohorts/participants/study items were grouped for the study; e.g. Type 2 diabetes, Type 1 diabetes, Gestational diabetes; or Carbohydrate, Protein, Fat, Fibre
  • You may need to add an author query to ask the author to amend the figure if the reference numbers are missing.
  • IV (inverse variance) and M–H (Mantel–Haenszel) should be explained in the figure legend.
  • I2 with upper case italic I. Use χ symbol for Chi2

Flow charts

  • If we have an editable version of a flow chart in PowerPoint and Word, this will be sent on to you to edit. See Tips for editing flow charts for suggestions.



Check that tables are cited in consecutive order in the main text. Blank cells in tables are permitted but blank rows will be removed by the typesetter.

  • The typesetters will not print any internal gridlines or rows of space. You may need to edit the table and/or raise author queries with this in mind.
  • In order to clarify the width of spanner rules in column headings where necessary, add a narrow blank column between the appropriate columns. Merge the cells within the column in the header rows (see examples). It may not be possible to add a column without distorting the layout of the table: in this case see instructions in examples for how to proceed.
  • Add a heading to the left-hand column header cell if it is blank (note that this should be in the singular, for example, ‘Variable’ or ‘Characteristic’).
  • Superscript letters used to link to footnotes should appear in the following order in the table: left to right along column headers, left to right along each subsequent row (including the row header).
  • Entries in the left-hand column subordinate to those above them (in contrast to independent entries) should be indented by two em spaces (select an em space from the special character tab in the symbol menu or set up a keyboard shortcut). If you have more than two levels, indent level 2 by two em spaces, then level 3 by four em spaces.
  • Delete % symbols from within the body of the table, and instead ensure it is clear in a footnote or in the left column or heading how data are presented (e.g. ‘data are n (%)’); there may be exceptions to this rule, however, for large tables with data presented in various ways.
  • The typesetters will format the tables, so there is no need to align values.
  • The typesetters will not print any bold within the tables. If the author has used bold type to show significance, for example, you will need to use superscript symbols within the table with a linked footnote to explain the meaning of the (previously) bold type.
  • If you’re having problems with tables where you can’t get the heading to stay at the top, or the footnotes disappear behind or wrap around the table, then it’s usually because the text wrapping feature is on. To fix this put the cursor in the table then Layout tab/Cell size/Table/Text wrapping/None. If that still doesn’t fix it, try Layout tab/Cell size/Cell/Options/Untick wrap text as well.

Large tables

  • Look out for large tables summarising data from other sources, particularly in meta-analyses and systematic reviews, as these can present a problem at typesetting in terms of layout.
  • Concentrate on content not format, as much of the formatting, such as indents within cells, is stripped out at typesetting. However, it is useful to adjust column width to make the text-heavy columns as wide as possible.
  • Bullets can be used within cells if this is the clearest way of showing the data.
  • Is the style of information consistent across cells e.g. bullet points vs prose? Try to be as consistent as possible between different sources-while different research papers will report things in different ways, try to reword for brevity and uniformity where possible.
  • Try to make text as concise as possible.
  • Consider whether text can be abbreviated e.g. the abbreviation T2D would be fine in a table, although not allowed in the main text. Can you suggest other non-standard abbreviations to reduce the content of cells?
  • Please define abbreviations in footnotes rather than within the table.
  • Can you suggest footnotes that could be used to reduce the content of cells, for example where the same information is repeated in numerous cells?
  • Could the content of any columns be merged to save space? This should generally be raised as a query initially, rather than spending time on it, as the authors may object or have alternative ideas.
  • Consider the order of publications within the table, and raise a query for the author if it does not seem to follow a pattern, e.g. date order, alphabetical order, date order within different categories. We have had authors wanting to change the order of tables completely at the proof stage, which was quite a headache, so much better to consider it at copy-editing!

Figure legends, table headers and table footnotes

  • The typesetters carry out minimal formatting to the figure legends, table headers and table footnotes, so they should be edited as they are to be printed
  • In general:
    • pnrt, etc. in italics
    • En dashes for number ranges, etc.
    • Add commas in values >9999 (e.g. 1000, 10,498)
    • True multiplication signs, degree symbols, etc.
    • Abbreviations not in italics. If they are in a list, arrange alphabetically and separate with commas and semicolons, e.g.
      FFM, fat-free mass; PAD, peripheral arterial disease; RER, respiratory exchange ratio
    • Abbreviations such as NA (not available or not applicable), ND (no data) and NT (not tested) may also be used where appropriate and should be explained in the table footnotes
  • Table headers
    • Table 1(emspace)One sentence of text (no full stop)
  • Table footnotes
    • Arrange each distinct footnote on a separate line (no full stop at the end of each footnote) and in an appropriate order, e.g.
      Data are means ± SEM
      Explanation of superscript letters (preferably one per line)
      Statistical information (p values, etc.)
    • Superscript letter(no space)text
  • Figure legends
    • Fig. 1 (emspace)Text(no full stop)
    • Letters to indicate panels are in bold and are enclosed in brackets (non-bold). e.g.
      Fig.3 (emspace)(a) Respiratory exchange ratio and (b) electron transport
    • Figure symbol descriptors not in italics (e.g. black squares, high-fat diet) (the typesetters cannot accept pictorial symbols – they must be written out)


Reference citations

  • References should be cited in numerical order in the text; this will have already been checked prior to copy-editing, and a query may be raised at the top of the paper for you to pass on to the authors if there is a problem. Please double-check reference order again, particularly if new references are added, or if general-information websites are removed from the reference list.
  • Citations in a table or figure legend that do not also occur in the text should EITHER be listed at the end of the reference list OR listed in numerical order at the position of the first call out to that figure or table in the main text. Either system is acceptable, but authors should be consistent.
  • References in the text should be given as a range when applicable, i.e. [1−5, 7].
  • No full stop in ‘et al’ (to match the style of the reference list).
  • Where there is reference to a paper by two authors with their names written out in the text, use ‘and’ rather than ‘&’ to link them. For example: ‘ … shown by Sawyer and Gale [39] … ‘.
  • Authors sometimes refer to a reference in the text using the last (senior) author’s name. This is OK, but should be rephrased as ‘Smith and colleagues’ or ‘Smith and co-authors’, rather than ‘Smith et al’ if Smith is not the first author.
  • There is no need to convert round brackets to square ones, nor to convert hyphens to en rules (the typesetters will do this automatically).

Reference list

  • Please avoid unlinking fields if the reference list has been created in EndNote, as this would make it difficult for the author to make further changes to the references. Please raise an author query asking authors to unlink after checking copy-editing changes.
  • The reference list is reformatted by the typesetters provided all the correct information is there.
  • Please check the following during copy-editing:
    • scan references for potential problems and, where possible, correct any mistakes by checking on PubMed or raise as an author query
    • refer to the instructions below for the style for specific types of article
    • if there are more than six authors, change reference to name of first three authors followed by et al
    • if authors have reported one author plus et al, check PubMed for additional authors; if authors have done this throughout the list, raise a query asking the authors to amend their reference list
    • article and book titles should be all lower case except for an initial capital on the first word and proper nouns, protein and chemical symbols, trial acronyms etc. (sentence case)
    • check the online PDF version for Greek letters or other symbols in titles and reinsert these—they are often stripped out in the online PubMed title and when imported into the reference manager
    • page numbers are always given in full, i.e. 1074−1083 not 1074−83; although this may be corrected by the typesetters, it might be missed if inconsistent in the original
    • there is no need to add issue number or DOI at copy-editing (please do not delete if these are already present at copy-editing)
  • The typesetters should make the following changes during typesetting, as required:
    • change the position of the year
    • abbreviate written-out journal names: this is done automatically by the typesetters; however, if abbreviations have been given but look incorrect refer to the National Center for Biotechnology Information (NCBI) for correct journal abbreviations
    • insert en-rules into number ranges
    • delete full stop after ‘et al’
    • reformat access date
Styles for particular types of reference
  • The style for a journal article is: Authors (year) Article title. Journal Volume number(Issue number): page range. DOI. For example, Tanaka Y, Tran PO, Harmon J, Robertson RP (2002) A role for glutathione peroxidase in protecting pancreatic ß cells against oxidative stress in a model of glucose toxicity. Proc Natl Acad Sci U S A 99(19):12363–12368.
  • The style for a reference available online but not yet in print is to include a DOI number, for example:
    … Diabetologia doi: 10.1007/s00125-009-1361-4
  • The style for a whole book is: Authors (year) Title, edition. Name and location of publishers. For example:Ekoe JM, Zimmet P, Williams R (2001) The epidemiology of diabetes mellitus: an international perspective. Wiley, Chichester
  • The style for a chapter from a book is: Authors (year) Chapter title. In: editors’ names and initials (eds), Title of book, vol. no. Name and location of publishers, page range [if available]. For example:Hopper JL (2000) Why ‘uncommon environmental effects’ are so uncommon in the literature. In: Spector TD, Sneider H, MacGregor AJ (eds) Advances in twin and sib-pair analysis, 1st edn. Oxford University Press, London, pp 151-165
  • The style for supplements is: Int J Obes Relat Metab Disord 25(Suppl 5): S56−S62, i.e. round brackets, no space.
  • The style for website URLs is: Authors (year) Title. URL, accessed date. ‘http://’ may be deleted if the URL includes ‘www’. The access date should be included for all websites. For example:European Medicines Agency, Committee for Proprietary Medicinal Products (2002) Note for guidance on clinical investigation of medicinal products in the treatment of diabetes mellitus. Available from Accessed 17 Apr 2008
  • The style for articles published only electronically may include a pii number or another form of electronic page number. For example Bertinato J, Lavergne C, Rahimi S et al (2016) Moderately low magnesium intake impairs growth of lean body mass in obese-prone and obese-resistant rats fed a high-energy diet. Nutrients 8(5): pii: E253.
  • Websites from which specific published data, e.g. incidence of diabetes, are reported should be included in the reference list; websites for online calculation tools, databases or general information databases (i.e. the author is not citing a specific piece of information) should be in the main text, not the reference list.
  • The style for archived website URLs is to include ‘(archived)’ in parentheses after the access date. For example: ISD Scotland (2009) General practice – Quality & Outcomes Framework (2009/2010). Available from . Accessed 15 Mar 2011 (archived)
  • The language that a paper is printed in (if not English) should be specified after the page number, in square brackets, e.g. … in type 1 diabetic patients. Diabetologia 45(2):212–215. [article in Japanese]. The title may be given in English or in the original language.
  • The style for a reference including an erratum is, for example:
    Turnbull FM, Abraira C, Anderson RJ et al (2009) Intensive glucose control and macrovascular outcomes in type 2 diabetes. Diabetologia 52(11): 2288–2298. DOI 10.1007/s00125-009-1470-0. Erratum 52(11):2470
  • The style for Cochrane reviews (see is, for example: Siebenhofer A, Plank J, Berghold A et al (2006) Short acting insulin analogues versus regular human insulin in patients with diabetes mellitus. Cochrane Database Syst Rev, Issue 2, Art. no.: CD003287. DOI: 10.1002/14651858.CD003287.pub4
  • The Cochrane review handbook is referenced as follows: Higgins JPT, Green S (eds) (2011) Cochrane handbook for systematic reviews of interventions 5.1.0 [updated March 2011]. The Cochrane Collaboration. Available from (or updated version when available: see
  • The style for package inserts is to include ‘[package insert]’ as well as the date, manufacturer details and location; the URL for the package insert may also be included, for example: Cialis [package insert] (2003). Indianapolis, In: Eli Lilly & Co; Victoza [package insert] (2010). Novo Nordisk A/S. Available from Accessed 21 Sep 2015.
  • The style for patents is, for example: Rabiner RA, Hare BA, inventors; OmniSonics Medical Technologies Inc, assignee. Apparatus for removing plaque from blood vessels using ultrasonic energy. US patent 6,886,670. 15 Mar 2005. Please see the AMA manual of style, the European Patent Office or the US Patent and Trademark Office for further details.
  • The style for additional data and data sets included in a public repository should include the DOI and/or the URL specific to the DOI, for example: Heinonen S, Muniandy M, Buzkova J, et al (2016) Additional data for ‘Mitochondria-related transcriptional signature is downregulated in adipocytes in obesity: a study of young healthy MZ twins’. figshare:
  • If data in a repository has not been assigned a DOI (this may depend on the repository used) it should not be cited as a reference, and instead the URL should be given in the main text, together with the access date.
  • Authors can cite records using the recommendations in the following link:; however, most authors just provide the trial registration number in the text or abstract, and this is also acceptable.
  • Use the ‘core name’ of a publisher, e.g. Wiley instead of John Wiley. Give only one location for a publisher, using either the first supplied in the manuscript (if more than one is given) or the publisher’s main location
  • Use ‘in press’ only in place of volume and page numbers that are not yet available [e.g. … JAMA (in press)], not as a substitute for the (probable) year of publication or for a journal name, both of which are required. Add an author query asking for publication details or DOI. If the paper is ‘in press’ and no DOI is available, then we require written confirmation of acceptance.
  • Abstracts and articles on preprint servers are allowed only for the current year and the preceding year. If the author uses older abstracts/preprint articles, use an author query to ask for a full reference. If this is not available, the data will be cited as ‘unpublished results’ or ‘unpublished data’ (whichever the author prefers) (see section below on obtaining written consent.) We allow for more flexibility in meta-analysis, systematic review and umbrella review papers: if you find an older abstract in one of these article types please let us know.
  • The format for references that are abstracts is the same as for journal articles but with (Abstract) after the page numbers.
  • References to research on preprint servers such as BioRxiv is as follows, and the word ‘preprint’ and the version date are required: e.g. Bar DZ, Atkatsh K, Tavarez U, Erdos MR, Gruenbaum Y, Collins FS (2016). Biotinylation by antibody recognition–a novel method for proximity labeling. BioRxiv 069187 (Preprint). 11 Aug 2016.  Available from:
  • We allow reference to unpublished and ongoing work in the main text but not in the reference list. If references are made to unpublished results by one or a number of the authors of the manuscript, these should include the author’s name and should be written in parentheses, e.g. (J. Smith, unpublished data) or (J. Smith, unpublished results). The terms ‘unpublished data’ or ‘unpublished results’ are preferable to alternatives such as ‘manuscript submitted for publication’. Personal communications are also inserted in the main text, not in the reference list (e.g. J. Smith, Department of Medicine, University of Bristol, UK, personal communication).
  • Written consent is needed if the person referred to in regard to unpublished results or in a personal communication is not one of authors. Raise an author query asking the author to provide written evidence of consent and details of affiliation (or to remove reference to personal communication).

Formats for other types of articles

All types of articles include a declaration of the Authors’ relationships and activities, except for In Memoriams and Book Reviews

Short communication

  • Rubric: Short communication
  • Short communications are identical to full papers in every respect, except that they have a reduced word count and restrictions on the numbers of tables and figures


  • Rubric: Editorial
  • There are no received/accepted dates. There is no abstract. A list of abbreviations may be included. Keywords are optional


  • Rubric: Commentary
  • Include received/accepted dates. The inclusion of an unstructured abstract is encouraged. A list of abbreviations may be included. The inclusion of keywords is mandatory

For Debate

  • Rubric: For debate
  • Include received/accepted dates
  • Include a list of abbreviations
  • Inclusion of keywords is optional
  • Abstract is unstructured
  • The use of colour in figures is strongly encouraged


  • Rubric: Review
  • Include received/accepted dates. Include a list of abbreviations
  • The keywords should include Review
  • The abstract is unstructured
  • The use of colour in the figures is strongly encouraged


  • Rubric: Article
  • Include received/accepted dates. Include a list of abbreviations
  • The keywords should include Meta-analysis
  • The abstract is structured
  • There is no limit on the number of references

Systematic review

  • Rubric: Article
  • Include received/accepted dates. Include a list of abbreviations
  • The keywords should include Systematic review
  • The abstract is structured
  • There is no limit on the number of references

Umbrella review

  • Rubric: Article
  • Include received/accepted dates. Include a list of abbreviations
  • The keywords should include Umbrella review
  • The abstract is structured
  • There is no limit on the number of references

Book review

  • Rubric: Book Review
    Book review titles are styled, for example:

    • Jones F, Smith AB, Brown W (eds) The diabetic foot (second edn). Humana Press, Totowa, New Jersey, 2006
      567 pp (ISBN 2 98567 564 4). Hardcover £15
    • P. White
  • There are no received/accepted dates. The printed version will include an online publication date.
  • A colour illustration of the cover of the book is always included


  • Rubric: Letter
  • The titles of comments should be styled with a unique title, for example:
    • Further investigation of the metabolic consequences of fetal and early postnatal growth restriction.
    • You may need to ask the author to provide this.
  • A title of a response to a comment would be styled as the unique title provided by the initial respondent (with no more than three authors listed plus et al if necessary). For example:
    • Further investigation of the metabolic consequences of fetal and early postnatal growth restriction. Reply to Peters JB, Smith M, Harris P, et al [letter]
  • If the letter is a reply to two comments on an original article, the format for the title is, for example:
    • Congenital rubella: citation virus or viral cause of type 1 diabetes? Reply to Honeyman MC, Harrison LC [letter] and Burgess MA, Forrest JM [letter]
  • Letters begin ‘To the Editor:’ with the text run on
  • There are received/accepted dates but no abstract. The inclusion of keywords is encouraged but not mandatory. Try to keep the number of abbreviations to a minimum

Research letter

  • Rubric: Research letter
  • Research letters begin ‘To the Editor:’ with the text run on
  • There are received/accepted dates but no abstract. The inclusion of keywords is encouraged but not mandatory. Try to keep the number of abbreviations to a minimum
  • There are no internal headings

Consensus statements (simultaneous publications with the ADA)

  • Consensus statements must include a footnote as follows (update year if necessary):
    • Simultaneous publication: This article is being simultaneously published in 2009 in Diabetes Care and Diabetologia by the American Diabetes Association and the European Association for the Study of Diabetes. Copyright 2009 by the American Diabetes Association and Springer. Copying with attribution allowed for any non-commercial use of the work.
  • The ADA will use American spelling and terminology (for example HbA1c, glycemia), but normal house style for Diabetologia still applies
  • The online and print publication dates must be coordinated between Diabetologia and the ADA

Annual lectures

Named lectures, such as the Minkowski or Claude Bernard lectures, should not appear as such in the title, but should include a footnote so that it is clear that this it is an invited review based on the lecture. For example:
This review was invited based on the [Claude Bernard/Minkowski] lecture at the EASD, September 2011

In memoriam

  • Rubric: In memoriam
  • The title is styled, for example: [Name of deceased]: [Dates or short text]
  • A photograph is included in the left-hand column before the text

Top of section

Grammatical conventions

Terms to avoid

  • Type 1 or type 2 diabetes associated with an animal. Re-phrase to indicate that the animal is being used as a model for type 1 or 2 diabetes in the title and at first mention in the abstract and in the main text.
  • Prediabetic, unless used in a strictly temporal sense; where the term is used throughout the paper, it is OK  to retain it, provided it is clearly defined at an appropriate point in the Methods.
  • Antidiabetic: replace with glucose-lowering drugs/medication.
  • Hypoglycaemic drug/medication: replace with glucose-lowering drug/medication. Hypoglycaemic can still be used when referring to hypoglycaemic episodes etc.
  • Glycaemia: ask authors to consider changing to blood or plasma glucose or HbA1c (see standard AQ). Note that terms such as ‘hypoglycaemia’, ‘hyperglycaemia’ etc. are fine to use.
  • Increased/reduced insulin resistance: state ‘more severe insulin resistance’ or just ‘insulin resistance’ (depending on the context) or ‘improved insulin sensitivity’.
  • ‘Regular insulin’ (American term) should be replaced by ‘short-acting insulin’.
  • Diabetics: use ‘individuals with diabetes’, ‘people with diabetes’ (using person-first terminology, where practical) or ‘diabetic individuals’ (if using person-first makes the phrasing too wordy).
  • Patients: try to avoid overuse of the term ‘patient’ where practical to do so; for example patient is fine where a person is being referred to in terms of their treatment, but it may be more appropriate to use ‘individual’ or ‘person’ in more general situations.
  • Controls: avoid the use of ‘controls’ for humans; instead use ‘control group’, ‘control participant’.
  • Persons (plural): change to people
  • Subjects: as far as possible, use ‘participants’ or ‘individuals’ to convey the notion of active participation; there are two exceptions:
    • where the study uses a database (and so the individuals haven’t actively participated). Please retain ‘subjects’
    • ‘between-subject factor’ and ‘within-subject factor’ are statistical terms and should not be replaced by between-participant and within-participant.
  • Caucasians – replace with Europids, or people of European extraction (see Ethnicity).
  • Different to – change to different from.
  • Compared to – change to ‘compared with’ in most instances. ‘Compared to’ is often used in a figurative sense, mainly in poetry, although ‘comparable to’ is generally correct, as it is used to demonstrate similarities between two conditions. See Troublesome words for more details.
  • # symbols, for example in manufacturers’ catalogue numbers or grant numbers. Please replace these with ‘no.’.


  • Use Concise Oxford Dictionary, -ise endings.
    • Use ph spelling in:
    • sulphur, sulphate, sulphated
    • but f in sulf(o) prefix: sulfatide, sulfotransferase, sulfonylurea
  • Note dys- (prefix meaning bad) vs dis e.g. dysregulation, dyslipidaemia
  • Note -penia (suffix meaning lack of) not -poenia, or -paenia
  • Note the following, which are spelled ‘e’ not ‘ae’:
    • postrema
    • gastrocnemius muscle
    • Hematocrit
    • hemangioma
    • erythema
  • Analyzer (not Analyser) should be written when it is part of a trade name, e.g. ‘Technicon AutoAnalyzer’ but not when it is not written as part of the trade name, e.g. ‘autoanalyser (Technicon)’ or ‘a Technicon autoanalyser’
  • ‘Hyperlacticacidaemia’ is commonly misspelled as ‘hyperlacticaemia’
  • Haematopoeisis vs haemopoeisis – use whichever form the author uses as long as it is consistent and has the UK ‘haem’ spelling

Preferred spellings and usage

UK vs US spelling


  • With units of measurement always use the singular verb form:
    • Five millilitres was injected
    • In our opinion, 100 km is far enough
  • A value expressed as a percentage uses the singular or plural verb form depending on the noun:
    • Within the group, 15% of the women were glucose intolerant
    • Within the group, 15% of the population was glucose intolerant
  • The verb in either/or and neither/nor constructions should always agree with the noun nearest to it:
    • Either nausea or headache is …
    • Neither glucose nor HDL-cholesterol was increased by exercise
    • Neither the mouth nor the feet were affected
  • This also applies to ‘and/or’
  • A ‘with’ phrase that is part of a nominative phrase does not determine verb number, which must agree with the grammatical subject:
    • Isoprenaline given together with propranolol is effective in …
  • The number is singular and a number of is plural:
    • The number that responded was surprising.
    • A number of respondents were verbose in their answers


  • Ensure that adjectives qualify an appropriate noun, e.g.
    • ‘Cardiac diet’ should be rewritten: ‘a diet for patients with cardiac disease’ (a diet cannot be ‘of the heart’)
    • ‘Cardiac patient’ should be rewritten – a patient may be weak or depressed, but not cardiac

Troublesome words

Here is a full list of troublesome words.


Full stops

  • Abbreviations require full stops, but contracted words do not. To make the distinction, contractions finish with the same letter as the full word; so, versus contracts to vs (not vs.), Doctor to Dr, Saint to St, Junior to Jr
  • The word ‘number’ may be abbreviated using a full stop to ‘no.’ in the body of tables and figures, or in parentheses in the main text. Do not use the number sign (#)
  • Use the full stop as a decimal indicator (r = 0.75) – not a decimal comma


  • Use a comma before ‘respectively’, for example:
    • … the apples and bananas were green and yellow, respectively
    • no Oxford (serial) comma, except for clarity. For example: Diabetes is common in Somerset, Cumbria and Lancashire. The incidence was measured in Somerset, Cumbria, Hereford and Worcester, and Lancashire

Quotation marks

  • Single quotes should be used in preference to double quotes; double quotes may be used for a quotation within a quotation, but this would be unusual.


  • Use unspaced em dash rather than spaced en dash; however, consider whether a dash is the most appropriate punctuation in the context, as commas or parentheses may work better in formal writing.


  • Our policy is for minimal hyphenation. There is no hyphen in beta cell, even if used adjectivally. There are no hyphens between values and units, e.g. 3 ml tube not 3-ml tube; 6 week checkup, not 6-week checkup; but hyphenate 6-day-old mice (to avoid confusion).
  • Hyphenate a compound containing:
    • a noun/adverb and a participle that together make up an adjective if they precede the noun, but not if they follow the noun, e.g.
      decision-making methods
      methods of decision making
    • An adjectival phrase when it precedes the noun, but not if it follows the noun:
      end-to-end anastomosis
      the anastomosis was end to end
    • An adjective and noun when they precede and modify another noun, but not when they follow a noun:
      upper-class values        the values were upper class
    • low-density resolution    the resolution was low density
    • Prefixes that precede a proper noun
      anti-American demonstration
    • Numbers from 21 to 99 (i.e. compound numbers) when written out (e.g. at the start of a sentence):
      Thirty-six patients were examined
    • Fractions:
      A two-thirds majority was needed
    • When two or more compounds have a common base, omit the base in all but the last. In unhyphenated compounds written as one word, there is no need to repeat the base:
      10- and 15-year-old boys
      HDL- and LDL-cholesterol were measured
      Macro- and micronutrients were analysed
    • Use hyphens to avoid ambiguity:
      a small-bowel constriction (constriction of the small bowel)
      a small bowel constriction (a small constriction of the bowel)
    • Furthermore, the unhyphenated word may have a different meaning
      re-treat       retreat
    • re-creation   recreation


  • In the text, the ordering of parentheses is ([ ]), for example:… the patient’s energy intake was measured (40 kJ/kg [10 kcal/kg]) …
  • In mathematical terms and equations, the ordering is [()], for example:… z = k[(a + b) – y(c + d)]; when deciding on brackets in equations, usually work from inside () to outside.
  • The ordering will only change if the style of bracket is essential to the meaning or format, for example [14C]glucose.


  • Blots: Southern blot has a capital ‘S’ because it was named after Professor Southern. Other types of blots (western blotting, northern blotting) have a lower case initial letter.
  • Use a capital letter after a numerical or Latin prefix at the start of a sentence, for example:
    2,3-Dihydroglycerol was added …
    trans-Fatty acids are found …

Small capitals

  • Used for prefixes in chemical formulas, such as L for laevo and D for dextro, e.g. D-fructose. You need to have the letter styled as lower case before or after adding the ‘small caps’ format, otherwise the small capitals will appear as normal capitals on-screen


  • Abbreviations and acronyms should be defined at first mention in the Abstract and main text, in the same way, and added to the abbreviations list if they occur more than once.
  • An abbreviation is a shortened or contracted representation of a word or phrase. For a list of abbreviations accepted without explanation, see accepted abbreviations.
  • An acronym is a word formed from the initial letters or groups of letters of words in a set phrase or series of words. Acronyms are pronounced as words (e.g. ELISA: enzyme-linked immunosorbent assay; avoid ELISA assay).
  • Read abbreviations/acronyms as separate letters, e.g. we write ‘an RR’ not ‘a RR’ (because we pronounce ‘R’ as ‘are’, as if it has a vowel in front). However, some abbreviations are pronounced like words, e.g. SNP (pronounced ‘snip’) and ROC (pronounced ‘rock’), so we would write ‘a SNP’, ‘a ROC’.
  • The use of single letter abbreviations for amino acids, e.g. DEAD box protein, where D, E, A, D stand for individual amino acids (aspartic acid, glutamic acid, alanine and aspartic acid, respectively), is acceptable.
  • Abbreviations should not be defined in the title of the paper or in headings or subheadings. In this case, retain the full version in the heading and define at first mention in the text. The abbreviation should be used in the headings from then on. For example:
    The role of fasting blood glucose level The role of fasting blood glucose (FBG) level in the weight gain of the mice …
    However, if there are complex or numerous abbreviations (particularly protein abbreviations) in the title or headings, it is sometimes clearer to retain the abbreviations in the title/heading, provided they are clearly defined in the text and abbreviations list: e.g.
    RANKL-OPG and RAGE modulation in vascular calcification and diabetes: novel targets for therapy
    reads better than:
    Receptor activator for nuclear factor κB ligand-osteoprotegerin and receptor for AGE modulation in vascular calcification and diabetes: novel targets for therapy
    Raise it as a query for the Editorial Office if you are unsure.
  • If abbreviations are used for apolipoproteins, use the format: ApoA-II, ApoB-100, etc.
  • For trial names in English write the trial name out in full and then add the abbreviation in parentheses at first mention in the text. For trial names not in English the abbreviation should be used first, then the written-out form given in parentheses [e.g. PREDIMED (Prevencion con Dieta Mediterranea)].


  • Use numbers in lists in the following form, with a colon and separated by a semicolon: (1); (2); (3) not (i), (ii), (iii). Semicolons should still be used to separate list items even if the phrase contains ‘and’ or ‘or’, e.g. (1) mice; (2) rats; and (3) zebrafish.

Top of section

Scientific conventions

Units and spacing

  • The typesetters will insert spacing between units and symbols automatically so there is no need to format them. There will be a space between the number and unit symbol except in the case of degrees, % and ‰, e.g.
    3 mg
    but 37ºC, 9%
  • Note that the units for blood pressure are closed up: mmHg
  • House style is for no space around mathematical symbols (±, ×, =, >, :, etc). Again, this will be done automatically by the typesetters. For example:
    8×9 [use multiplication symbol × not letter x]
    1:4 ratio
    >10 mm
  • Multiple units are formatted as number/unit or number unit−1 unit−1 (the solidus symbol [/] is used only once; house style is to separate units from each other by spaces, not centred dots), e.g. 15 m/s, 25 mg kg−1 day−1
  • The percentage sign should be repeated in pairs of values (e.g. 3.5% vs 6%), but need not be repeated for ranges (e.g. 3.5−6%) or with 95% CIs (e.g. −0.03% [−0.43, 0.36]), or for other units (e.g. 6.7 vs 8.9 mmol/l).
  • Authors can opt to repeat units with 95% CI or not – stick to the authors’ style, and make consistent if needed.
  • There are no units for ratios if the numerator and denominator have the same units. If the if the numerator and denominator do not have the same units, then the ratio will have units. For example:
    • The ratio of energy intake to energy expenditure: both terms are in kJ/day, and so the ratio has units [kJ/day]/[kJ/day], which simplifies to no units (cancel out).
    • The albumin/creatinine ratio (ACR): the numerator is mg/l, the denominator is mmol/l, so the units are [mg/l]/[mmol/l], which can be simplified to mg/mmol.
  • Units of measure do not need to be abbreviated when they are not preceded by a numeral: either ‘… is measured in milligrams per litre’ or ‘…is measured in mg/l’ is acceptable; for simple units, the unabbreviated form is usually better, e.g. ‘…is measured in grams’.
  • House style is to use SI units throughout.
  • We only insist on SI units for circulating levels of glucose, insulin, etc. (e.g. blood, urine, plasma), but not for tissue levels (e.g. liver, pancreas, etc.).
  • Check especially for the following in the text, tables and figures:
    • serum insulin (should be pmol/l, but doses of insulin may be in U)
    • serum glucose (should be mmol/l)
    • radioactivity (should be Bq not Ci)
    • energy (should be kJ not kcal). For dietary energy, the non-SI units (kcal) may be placed in parentheses after the SI units (note also: low-energy diet, not low-calorie diet; however VLCD [very low calorie diet] is acceptable as the term is widely used – it should be defined as ‘a diet very low in energy [very low calorie diet, VLCD] at first mention)
    • centrifuge speed (should be g not rev/min or rpm; please note ‘g’ in italics to distinguish it from grams; shaker speed can be in rev/min)
    • we will accept alternative molar prefixes, e.g. nmol/l rather than pmol/l, as this could be more appropriate for some quantities
  • HbA1c should be reported in dual units (both mmol/mol and %), for example:
    • HbA1c level was 31 mmol/mol (5%)
    • If the author reports HbA1c results in % or mmol/mol only, use the standard author query to ask to author to dual report HbA1c results. The values for the different units should be presented on separate rows in a table. HbA1c values should be presented in mmol/mol in figures. If referring to previously reported results (e.g. in the Discussion section of the paper), mmol/mol should be used.
  • The use of ‘parts per million’ (ppm) is acceptable for mass spectrometry
  • Insulin doses are given in U not IU
  • Use the conversion 1 kcal = 4.184 kJ for dietary energy (see Food and Agriculture Organization of the United Nations (2003) Food energy – methods of analysis and conversion factors. FAO food and nutrition paper 77)
  • Values in ppm or ppb should also be given in SI units
  • Values for non-esterified fatty acids (NEFA) can be given in mmol/l or mEq/l (analysed fatty acids are a mixture of undetermined molecular masses so mmol/l is not strictly correct)
  • Values in deciliters (dl) are generally not acceptable:
    amino acids (should be µmol/l not mg/dl)
    glucose, lipoproteins, LDL-/HDL-cholesterol (should be mmol/l not mg/dl)
    C-peptide (should be nmol/l not ng/dl)
    cortisol (should be nmol/l not µg/dl)
    glucagon (should be ng/l not pg/ml)
    with some exceptions:
    lipoprotein a [Lp(a)] is generally given in mg/dl
    blood flow across the arm is generally given as dl or ml per unit of time
  • The JAMA conversion calculator  includes SI units for standard biochemical components (raise the conversion as an author query).
  • When the authors denote mol/l as ‘M’ (or ‘N’) in concentrations it should be changed as follows (N values do not always stay the same – check with the author):
    M → mol/l
    mM → mmol/l
    µM → µmol/l
    nM → nmol/l
  • Molecular mass is given in kDa. Relative molecular mass (Mr) has no units. Molecular weight is often used incorrectly to mean molecular mass. Raise an author query
  • Relative units are defined against a specific variable (e.g. relative to an internal standard or control values). Arbitrary units are typically not defined and will apply when certain types of measurement are used, for example fluorescence intensity.
  • The format for weight/volume is wt/vol. (upright, not italic). (Also vol./vol., wt/wt)
  • The units for the area under the curve are the y-axis units multiplied by the x-axis units, e.g. the units glucose AUC are mmol/l × min
  • The HOMA index has no units. We cannot accept the equation written out with non-SI units. If the author provides the equation in non-SI units, raise an author query asking the author to provide the equation with SI units, bearing in mind that the constant will also be altered (a fixed constant for all papers cannot be provided as insulin assays vary in their standardisation and in their cross-reactivity with proinsulin species). Note that the value of the index does not change and there is no re-calculation required. Alternatively, the author can omit the equation and simply provide a reference. If they used the computer-generated index, the website can be referred to directly ( (see Levy et al 1998).
  • The units for insulin sensitivity index (glucose infusion rate/plasma insulin per unit body surface area per time) are µmol m−2 min−1 (pmol/l)−1. It may also be expressed on a body weight basis: µmol kg−1 min−1 (pmol/l)−1. This can also be written as µmol kg−1 min−1 pmol−1 × l.
  • The MINMOD program generates the insulin sensitivity index in non-SI units (× 10−4 min−1 [µU/ml]−1 or [µU/ml]−1 × min−1). To enable comparison with other papers, we can allow these units, but add a conversion factor to the text or a table footnote (ask the author to check, as conversion factors may change with different versions of MINMOD):
    • To convert values to SI units multiply by 0.167
    • Under the same circumstances, the disposition index (DI) can be given without units, but add a statement that DI is calculated as the insulin sensitivity index generated by the MINMOD program ([µU/ml]−1 × min−1) multiplied by AIRg (pmol/l × min); alternatively, if AIRg is also generated by MINMOD, units can also be given in MINMOD format (mU/l x min) and a statement can be added:
    • To convert AIRg and DI to SI units, multiply by 6
  • M value (marker of insulin sensitivity): M in italics, no hyphen. Units are µmol kg−1 min−1 or µmol m−2 min−1 or µmol/min)
  • Insulin secretion rate (ISR) units are l/min
  • Units for eGFR are styled as ml/min per 1.73 m2


  • In running text with non-unit quantities, use words and numerals for numbers as described here: one to ten, 11, 12, 13, … , n
    first to tenth, then 11th, 12th, 13th, … , nth (‘th’ is not superscript). In a numerical or scientific context, numerals can be used below 11, e.g. ‘1st quartile’
    twofold to tenfold, then 11-fold, 12-fold, 13-fold, … , n-fold (4.5-fold)
    one million, 1.3 million
  • Use a numeral if the context of the number is mathematical or followed by a unit: a factor of 3, a multiple of 5, a ratio of 1:4 (but four times as much), 5 mm, 5 days
  • House style is to use a comma to separate groups of three digits to the left of the decimal point (if any) for numbers greater than 9999 (e.g. 64,000 and 12,299). This will be formatted automatically by the typesetters so there is no need to add commas to numbers when copy-editing, except on figures, in figure legends and in table headers and footnotes, as these are not formatted by the typesetters. Commas are not generally included for genetic base pair positions. Proofreaders should not ask for commas to be replaced by thin spaces.
  • Make sure there is a value before a decimal point (i.e. 0.37, not .37).
  • Numbers should always be expressed as words at the beginning of a sentence. Rewording may be easier (e.g. ‘Thirty of 68 patients…’ or ‘Of 68 patients, 30 …’). Chemical names are an exception to this rule (‘5-Methylsulfadiazine was added …’).
  • Repeat all digits in a series or range (e.g. 1960–1975 and 29,143–145,147).
  • Spans of large (or very small) numbers (e.g. 1–2 × 107) should be expanded (e.g. 1 × 107 – 2 × 107) for clarity if that is the intended meaning.
  • Gestational age may be given in the format xxy, corresponding to xx gestational weeks and y days e.g. 291±25 weeks. Terminology must be explained on first mention (e.g. gestational age is given as weeksdays).


  • Commonly used currency symbols that are unique may be used without definition (e.g. £, €, $, ¥). Currency symbols that are less well known (e.g ₹), should be defined on first mention.
  • Symbols should precede the amounts e.g. €500.
  • For currency symbols that are used by more than one country and hold a different value within each country, the country should be specified before use of the symbols e.g. US$5 million. Abbreviations may be used for the country e.g. (AUS, CAD), but these must be defined on first mention e.g. Australian (AUS)$; Canadian (CAD)$ (this does not apply to ‘US’)


  • If a unit is preceded by a numeral make the following changes:
    • hours/hrs change to h
    • minutes to min
    • seconds to s
  • Days, months, weeks, years are always written in full.
  • 24 hour clock: use format 15:00 hours. (with colon)


  • Express dates as: day month year (e.g. 9 September 1979)
  • Year ranges should be written out in full (e.g. 1995–1997)

Statistics and confidence intervals

df   degrees of freedom
F    F test (a statistical test in which the test statistic has an F distribution)
N   sample size of population
n    sample size of subgroup
p    statistical probability
r    bivariate correlation coefficient
R   multivariate correlation coefficient
r2  bivariate coefficient of determination
R2  multivariate coefficient of determination
t    Student’s t statistic (determined using Student’s t test)
U   Mann–Whitney U (Wilcoxon) statistic (determined using Mann–Whitney U (Wilcoxon) test

  • Ratios: house style is for a closed-up colon or solidus for ratios, e.g. total cholesterol:HDL-cholesterol or albumin/creatinine ratio. Spaces around the colon or solidus will be removed automatically by the typesetters
  • Chi-squared should always be referred to as χ2 written with a Greek lower case letter χ and superscript 2.
  • Student’s test – published under the pseudonym Student so always capital S, apostrophe between t and s and t in italics with no hyphen. Should always state whether paired or unpaired.
  • Do not use an equals sign for OR, HR, RR or CI (for example, OR 4.3 [95% CI 4.0, 4.6]).
  • The format for confidence intervals is: 95% CI 96, 103; do not include a colon in 95% CIs (e.g. 95% CI 4, 12, not 95% CI: 4, 12).
  • The format for a range or interquartile range (IQR) is with a closed dash (5–8), but authors can also opt to use commas for (5, 8), particularly where some of the values are negative. If authors give the IQR as a single number (which is not, technically, incorrect) they should be asked to provide the limits to the interquartile interval (often referred to as [p25–p75] or [Q1–Q3]). Make sure it is always clear in the text whether a 95% CI, range or IQR  is being reported.
  • Use a multiplication sign for group interactions (for example, diet × time rather than diet-by-time).
  • RR is the standard abbreviation for relative risk, also sometimes referred to as risk ratio. Authors occasionally use RR as an abbreviation for ‘rate ratio’, which is a slightly different measure. Rate ratio should always be written in full throughout, and not abbreviated.


  • Matching has a very particular meaning in relation to clinical studies, whereby an individual in one arm of the study is matched with one or more individuals with the same characteristics, e.g. age, sex, BMI, who would be in the control arm
  • Ensure that matching criteria are given for continuous variables e.g. age ± 2 years, BMI ± 1 kg/m2. If these variables were not actually matched and were simply similar between the three groups, then please make this clear throughout the manuscript
  • The term ‘matching’ should not be used when referring to two selected cohorts with similar characteristics, because the individuals have not been directly matched
  • There is no need to request matching criteria for studies with animal models because litters of the same age can easily be bred, and because matching is not generally carried out for individual animals


  • Ask the author which base logs are to if not specified. We prefer to use loge rather than ln


  • Latin and foreign terms that appear in the Oxford English Dictionary (e.g. in vitro) should be written in roman script rather than italics, and used sparingly. Species names should still be italic.

Mathematical symbols and equations

We can accept equations written in Word or in Equation Editor (this is embedded in later versions of Word), but you may need to adjust the formatting. If you have Equation Editor, you can access it by selecting Equation in the Insert tab. If you are uncertain how to use Equation Editor, ask us to format the equation as necessary. Variables should be in italics (check that the values in the equations match those in the text) and constants in roman. Lower case and upper case Greek symbols are upright, non-italic. There is no punctuation around the equation:

Example of equation

  • Mathematical symbols should generally be retained in prose and e.g. ‘ ×3’, not ‘multiplied by 3’, ‘Δ’ not ‘delta’
  • Italicise the abbreviation for acceleration due to gravity, g, to distinguish it from g for gram. It should not be preceded by ×
  • Various pre-formatted terms are included in the Equation Editor document, including VO2max with overdot on the V and 2ΔCt  (copy and paste the terms from Equation Editor).
  • Exponential numbers should be styled as 1.0 × 10–10 rather than 1.0E–10.

Chemical nomenclature

Element symbols and names

  • Either names or symbols (e.g. iron, Fe) may be used in running text; symbols do not need defining unless you think it may be unclear in the context. Use of symbols and names does not need to be consistent – it can depend on context.
  • When many elements or compounds are mentioned or discussed together, it is often clearer and saves space to use the symbols.
  • When ions are being discussed, the symbol is sometimes more precise and usually shorter (e.g. Fe2+, Fe3+, not iron ions).
  • When elements or compounds are mentioned relatively infrequently, use of symbols may interfere with the flow of the writing and the spelled out name is preferable
  • If chemical bonds are used in running text, use en dashes. e.g. CH3–COOH
  • Note: In protein structures featuring the three-letter amino acid abbreviations, hyphens are used, not en rules


  • Symbols for elements that occur as locants (i.e. the portion of a chemical name that designates the position of an atom or group in a molecule) in chemical names are set in italics. This rule covers any single capital letter except the small caps d or l (these refer to configurations). For example:
    • O-methyltyrosine
    • S-benzyl-N-phthalocysteine
    • hexahydro-2H-azophin-2-one
  • However, when a chemical name is not used, the element symbol remains roman. For example:
    • N-linked glycan
    • N-terminal, N-terminus or N terminus


  • Symbols for configuration or rotation precede the name of the compound and are joined to it by a hyphen. The symbols are dl, and dl (italics) or (+), (–), and (±) for optical rotation; d, l, or dl (small capitals) for configuration in carbohydrates and amino acids; and R and S (italics) for absolute configuration. For example:
    • d-6-hydroxytryptophan
    • emodin-l-rhamnoside
    • (+)-6-hydroxytryptophan
    • d-(+)-alanine
    • (1R,3R,5S)-[(1S-sec-butoxyl)]-3-chloro-5-nitrocyclohexane


  • Symbols (numbers, Greek letters, etc.) are separated with hyphens, even in fairly simple structures such as α-amino acids
    Other prefixes indicating configuration that should be marked for italics include:

    • allo
    • cis
    • trans
    • ortho
    • para
    • meta
  • These prefixes may also be used alone in the text without being attached to a specific chemical name and should also be marked for italics. For example:
    • … encoding a trans-acting product
    • … ortho coupling
    • … via the meta pathway
  • Latin and Greek prefixes are set in Roman type and closed up to the word
    Latin: di, tri, tetra, penta, hexa, hepta, octa, nona, deca
    Greek: bis, tris, tetrakis, pentakis, hexakis, heptakis, octakis, nonakis, decakis
  • The first letter of the chemical name should be capitalised at the beginning of a sentence. When a prefix is at the beginning of a name that starts a sentence, capitalise the first letter of the chemical name itself. For example:
    S-Benzyl-N-phthalocysteine was added.
  • Use n rather than the Greek letter omega (ω) in fatty acid nomenclature (e.g. n-3 fatty acids, 18:2 n-6).

Ions, nuclides and other species

  • Ionic charges are indicated by superscript numbers and plus and minus signs. For example:
    • Na+
    • Cl
    • Zn2+ (not Zn++ or Zn+2)
    • PO42– (note that the superscript should be placed outside the subscript, not aligned with it)


  • Style vitamins as ‘vitamin B12‘, ‘vitamin D3‘ etc.
  • Chemical names, such as cholecalciferol (vitamin D3) are also acceptable


  • Written-out version of L-NAME is Nω-nitro-L-arginine methyl ester (note that both capital Ls are small capitals)
  • ‘-tropin’ is preferred to ‘-trophin’ for peptide hormones (e.g. neurotropin) (see recommendation of IUPAC-IUB CBN)
  • The addition of ‘+’ to indicate ‘positive’ is acceptable and does not need to be defined, e.g. insulin+, CD4+


  • For simple molecules the labelling is indicated by writing the chemical formula with the prefixed superscripts attached to the atomic symbol (i.e. 14CO2).
  • For more complex formulas the symbol for the isotope is placed in square brackets in front of the name (closed up), e.g. [14C]urea.
  • In a complex formula the isotope should relate to the appropriate part of the molecule, e.g. deoxycytidine 5′-[32P]triphosphate not [32P]deoxycytidine 5′-triphosphate; 2-deoxy[3H]glucose not 2-[3H]deoxyglucose.
  • If the isotope appears in more than one position, the number of labelled positions is added as a right-hand subscript, e.g. [14C2]glycolic acid. If the isotope is uniformly distributed among the positions the symbol ‘U’ is used, e.g. [U-14C]glucose. If it is generally distributed among the positions, but not necessarily uniformly, then [14C]glucose is sufficient.
  • Where the native compound does not contain the atom labelled, the square brackets are not used. Instead it is written as, for example, 131I-labelled insulin, not [131I]insulin.
  • Similarly, 14C-labelled amino acids should not be written as [14C]amino acids as the amino group does not contain C
  • Deuterated, tritiated should be written 2H-labelled, 3H-labelled, respectively, or similar (e.g. 2H2O).
  • NOTE: it follows that formats such as 18F-fluorodeoxyglucose or 14C-glucose are incorrect – they should be [18F]fluorodeoxyglucose and [14C]glucose respectively.


  • Drugs can be referred to by several names, including the non-proprietary name (generic), the proprietary name (brand name, trademark – selected by the manufacturer of the drug), the chemical name, the unofficial name or the code designation.
  • Except at the beginning of a sentence, non-proprietary (generic) names are in lower case. Proprietary names have an initial capital. Consult the British National Formulary (BNF).
  • Recommended International Non-proprietary Names (rINNs) for drugs should be used instead of British Approved Names (BANs). A list of former BANs and their corresponding rINNs is available.
  • An exception to this rule: adrenaline and noradrenaline are the official names used by EU members, so adrenaline and noradrenaline should be used throughout, but epinephrine or norepinephrine should be added in parentheses at first mention in the abstract and in the text.
  • Anti-diabetic or hypoglycaemic drugs should be edited to glucose-lowering drugs. Add a note to the query sheet to let the author know.
  • Replace ‘glyburide’ by ‘glibenclamide (known as glyburide in the USA and Canada)’.
  • Insulin and insulin analogues: please ensure authors have included supplier details and trade names for any forms of insulin or insulin analogues used in the study.

Disease and disease classification

Taxonomy and nomenclature

  • Diabetologia uses standard taxonomic styling (see CSE Manual of Scientific Style and Format). To summarise:
  • For animals, plants and bacteria:
    • Names of taxa at genus level and below are in italics.
    • Names of taxa at family level and above are in Roman type.
    • Species should generally be referred to using the binomial at first mention: genus and species name, in italics, with initial capital for the genus name but not the species name, e.g. Staphylococcus aureus.
    • Subsequent mentions can be abbreviated, e.g. S. aureus.
    • Latin vernacular names are often used for certain species, and are neither italicised nor capitalised. e.g. Drosophila melanogaster is the species binomial, Drosophila refers to the genus, and drosophila is the vernacular name.
  • When a species is not identified, ‘sp.’ or ‘spp.’ may be use. Sp. refers to one species alone, whereas ‘spp.’ refers to 2 or more species e.g. Angelica sp. or Angelica spp. Please note that sp. and spp. are set in Roman, whereas the genus is in italics. When referring to a genus as a whole, the genus name alone should be used e.g. ‘Angelica occurs in…’.
  • For viruses:
    • Virus species do not normally have Latin names, but the formal virus name (as approved by the International Committee on Taxonomy of Viruses [ICTV]) is written in italics and the first letter of the first word is in capitals, e.g. Maize dwarf mosaic virus.
    • Genus names are not generally included in virus species names.
    • This system is not universal, and many researchers of medically important viruses do not use the italic style. If the author has been consistent, there is no need to make changes.
    • This system of italics and capitalisation is only used in a taxonomic sense, and virus names are often used in the vernacular, and have neither italics nor initial capitals, e.g. poliovirus.
    • Genus names used in the vernacular sense are also non-italic and not capitalised, e.g. enterovirus.
  • Raise a query for the Editorial Office or the authors if you are unsure.


  • Use alpha and beta cell, but use Greek symbols for chemicals and statistical terms.
  • The names of dyes have all initial capitals, e.g. Trypan Blue, Congo Red
  • For magnification, the multiplication sign should come before the value, for example: magnification ×2.5
  • For saline solution, add ‘154 mmol/l NaCl’ in parentheses at first mention in abstract and in text, and raise a query with the author to confirm this is correct.

Top of section

Genes and proteins

General summary

Genes and gene symbols

  • Gene symbols are styled as follows:
    • italic
    • initial capital for mouse or rat (Abc1)
    • upper case for human and some other animals (ABC1)
  • Gene symbols do not need to be included in the abbreviations list or written in full at first mention, although for clarity, particularly with a less well known gene, the authors may give full details of the gene by writing, for example, ‘Hoxa1, which encodes Homeobox A1′.
  • Gene symbols should be checked against NCBI Gene to ensure the official gene symbol has been used, and the official symbol should be given at first mention if an alternative symbol or alias has been used, e.g. Glut4 (also known as Slc2a4). Raise a standard query if this has been done.
  • mRNA and cDNA should also be referred to in italic text using the correct gene symbol, e.g. Hoxa1 mRNA.
  • Written out gene names, for example β-actin, should be in upright text (or the authors can use the gene symbol, ACTB/Actb).

Proteins and abbreviations

  • Protein abbreviations are styled as follows:
    • non-italic
    • upper case (ABC1), although there are some common exceptions, listed below
  • Protein abbreviations should be included in the abbreviations list if they occur more than once in the text and the written-out version should be added to the text at first mention in the Abstract and main text, and usually in the Research in context box (although this rule can be relaxed for readability — ask the Editorial Office if you are unsure).
  • Protein abbreviations do not need to be checked against the NCBI Gene or Protein databases.
  • Exceptions to the usual upper case styling for proteins include:  Akt, eIF, Apo, MafA, mTOR, mTORC, Myc, Slit, Robo, Rab5, Fas, FasL, Nrf2, Noc2, Smad; if you are unsure, ask the Editorial Office and we will check the literature. Restriction and modifying enzymes are also an exception (see more details below).

Detailed points

  • If both the gene/mRNA and protein forms of a particular molecule are mentioned in the text, treat them independently in terms of protein abbreviations (defined at first mention of the protein) and gene symbols (official gene symbol given at first mention of the gene, if needed).
  • Both genes and proteins can be referred to as being expressed in the cell in general terms. However, when discussing protein quantification, for example in a western blot or quantified micrograph, the terms ‘levels’ or ‘content’ are preferred.
  • There is no need for the protein abbreviation to ‘match’ the gene symbol; e.g. protein abbreviation can be given as GLUT4, gene as SLC2A4
  • Gene families are not styled in italics – only the actual gene symbol, e.g. ‘Hox genes such as Hoxa1

Transgenic animals

  • Gene symbol styling should be used for most transgenic mice and rats, using mouse/rat gene symbol styling, e.g. Ucp1−/−. Use superscript upright −/−, with correct minus symbols (minus sign Unicode 2212, found under Mathematical Operators in the Word symbol grid), rather than hyphen-minus, as these are not automatically formatted by the typesetters.
  • Use italics for db/db, ob/ob, fa/fa mice etc, in line with the usual convention, although these terms are not always ‘official’ gene symbols. Some other well known transgenic animals do not include gene symbol styling, e.g. NOD mice – this is OK.
  • The full international strain nomenclature ( should generally be included for transgenic mice at first mention, or at the most appropriate point in the Methods. Raise a standard query if it has not been provided. This information is not needed for the more common strains used, e.g. NOD mice, C57/BL mice, ZDF rats. The system for naming rats is less clear cut, so just make sure the strain is clearly defined, with supplier details and references as needed.

Knockout and knockdown

  • Gene symbol styling should be used when discussing knockout of genes, which is usually a permanent modification to the DNA in transgenic mice or rats or genetically modified cell lines.
  • ‘Knockdown’ refers to the process whereby levels of a protein are reduced or temporarily wiped out, usually by using siRNA (small interfering RNA, an oligonucleotide that silences an mRNA sequence and prevents synthesis of the relevant protein).
  • In this way, ‘knockdown’ refers to a protein, which is knocked down, and protein formatting should generally be used, although some authors prefer to use gene symbol formatting (on the basis that the mRNA levels are knocked down), and this is OK  provided it is consistent.
  • Gene symbol styling should be used for the oligonucleotide used to knock down the protein; for example, ‘TCF7L2 knockdown was carried out using Tcf7l2 siRNA’.
  • Plasmids and vectors are generally styled using gene formatting for the gene spliced into the plasmid and Roman elsewhere, e.g. pEGFPC1-FABP5.; pGEM-T-Ins1.

Non-coding RNA

  • A microRNA (miRNA) is a non-coding RNA sequence (i.e. not translated into protein) that is involved in regulating gene expression. Unlike mRNA, miRNA should not be styled in italics using a gene symbol, except where its location on the chromosome is discussed. The usual format is miR-122, miR-184 etc. ‘miR-‘ is pronounced as a word, so write, for example, ‘a miR-122-dependent pathway’ but ‘an miRNA’.
  • 18S RNA is styled upright as it is the written-out gene name. The gene symbol Rn18s (or RN18S for humans) is styled in italics as usual.


  • For western blots, protein abbreviations should be used; in a western blot, a mixture of proteins is run on a gel (to separate them) and transferred onto a membrane, then the protein of interest is probed using a tagged antibody specifically against the protein. The abbreviation ‘p-‘, indicating protein phosphorylation, can be used without definition (e.g. p-AMPK).
  • For northern blots, gene symbols should be used as these are used to measure gene expression;  in a northern blot, a mixture of mRNAs is run on a gel (to separate them), transferred, then probed with a tagged cDNA directed against the mRNA sequence of interest (the cDNA has a sequence complementary to the mRNA and so the two bind together).
  • Gene symbol styling should also be used for Southern blotting (blotting of DNA fragments). Unlike western and northern blotting, use an upper case S in Southern, as it is named after its inventor, Edwin Southern.


  • There is no hyphen in A allele, etc.
  • Alleles are alternative forms of a particular gene. Allele symbols consist of the gene symbol plus an asterisk and the allele designation:
  • Different alleles of the same gene are indicated using a superscript next to the gene symbol. If the allele is dominant, the superscript is upper case with an initial capital letter; if the allele is recessive, the superscript is lower case.
  • Alleles created using recombinant DNA techniques are often named according to how they are created (deletions, insertions, fusions, etc.) rather than their effects. A different naming scheme for proteins is sometimes seen, in the form p29Ras (especially in vertebrates). In these cases, the number indicates the size of the protein in kDa and the superscript (which must be italic) indicates the name of the gene from which the protein is produced.

HLA genes

  • The gene symbols are in italics, e.g. HLA-DQB1, HLA-DQA1, HLA-DRB1HLA is in italics because it is part of the gene symbol.
  • Full allele symbols are also in italics, e.g. DQB1*06:02, DQB1*03:01.
  • A change was made to HLA gene nomenclature in 2010 to take account of the high number of alleles identified. These changes are described clearly in, which gives a useful guide to the naming system and what the different parts of the allele symbol mean.
  • If HLA alleles look like they may be in the old format, please add the standard query. The most obvious way of distinguishing between them is that alleles in the old system may contain longer strings of numbers, which would be separated by colons in the new system, e.g. HLA-DRB1*13010102 under the old system; HLA-DRB1*13:01:01:02 under the new system.
  • Gene families are always in roman, e.g. HLA genes or HLA gene region.
  • Serological specificities usually feature a capital letter(s) followed by a number, e.g. A1, DR4, DQ2. They should be in Roman. (A useful clue is that the methods may discuss serotyping [i.e. testing for antigens in serum] rather than genotyping).

Genetics terminology

  • See the Human Genome Variation Society (HGVS) recommendations ( for general information on sequence variant nomenclature.
  • The following abbreviation is accepted without explanation for the description of variants: ‘c’ for coding DNA sequence (e.g. c.75A>T). Similarly ‘g.’ (genomic), ‘m.’ (mitochondrial), ‘r.’ (RNA sequence) and ‘p.’ (protein sequence) are also accepted without explanation (e.g. g.476A>T, p.Lys76Asn, r.76a>u, etc).
  • Structural mutations are indicated using a fairly standard code, although this code does vary between organisms. The most common and universal symbol is Δ, which indicates that the gene named has been deleted (e.g. HsdΔ)

Gene sets and populations

  • HapMap is no longer widely used (see, but if authors refer to HapMap populations, please include full population names, which are detailed here:; e.g. ‘CEU (Utah residents with northern and western European ancestry)’.
  • If an author uses GSEA software, the website address should be included ( and the reference list must include two specific references:
    • Subramanian A, Tamayo P, Mootha VK et al (2006) Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A 102(43):15545-15550. Available from
    • Mootha VK, Lindgren CM, Eriksson K-F et al (2003) PGC-1-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes. Nat Genet 34(3):267-273. Available from
  • The gene set may include underscored spaces. The authors should include a clear explanation for gene sets or search terms in the figure legend.

Citing genetic analysis software

  • SHEsis software ( The following reference should also be cited:
    • Shi YY, He L (2005) SHEsis, a powerful software platform for analyses of linkage disequilibrium, haplotype construction, and genetic association at polymorphism loci. Cell Res 15(2):97-98.
  • If the haplotype analysis function is cited, then the following reference should be cited in addition to Shi & He:
    • Li Z, Zhang Z, He Z et al (2009) A partition-ligation-combination-subdivision EM algorithm for haplotype inference with multiallelic markers: update of the SHEsis ( Cell Res 19(4):519-523.
  • PLINK software ( Please cite the appropriate reference, depending on the version used; e.g.
    • Chang CC, Chow CC, Tellier LC, Vattikuti S, Purcell SM, Lee JJ (2015) Second-generation PLINK: rising to the challenge of larger and richer datasets. Gigascience 4: 7. doi: 10.1186/s13742-015-0047-8 [PLINK 1.9 and later]
    • Purcell S, Neale B, Todd-Brown K et al (2007) PLINK: a toolset for whole-genome association and population-based linkage analysis. Am J Hum Genet 81(3):559-575. [earlier versions of PLINK]
  • Genetic Power Calculator software ( The following reference should also be cited:Purcell S, Cherny SS, Sham PC (2003) Genetic Power Calculator: design of linkage and association genetic mapping studies of complex traits. Bioinformatics 19(1):149-150.
  • MeV (MultiExperiment Viewer) software ( The following references should be cited in addition to the website URL:
    • Saeed AJ, Bhagabati NK, Braisted JC et al (2006) TM4 microarray software suite. Methods Enzymol 411:134–193.
    • Saeed AJ, Sharov V, White J et al (2003) TM4: a free, open-source system for microarray data management and analysis. Biotechniques 34(2):374–378.

Modifying and restriction enzymes

Previously the first three letters of modifying and restriction enzymes were italicised, but recent guidelines have recommended that italics are no longer used (see: A nomenclature for restriction enzymes, DNA methyltransferases, homing endonucleases and their genes). It is also recommended that all numerals (Roman and Arabic) are closed up. Capital T is preferred for the designation of the main types of REases as Type I, Type II and Type III. See also REBASE: the Restriction Enzyme Database

Modifying enzymes

These are enzymes that synthesise DNA and RNA (polymerases), cleave DNA (nucleases), join nucleic acid fragments (ligases), methylate nucleotides (methylases), and synthesise DNA from RNA (reverse transcriptases). Those in laboratory use come from living systems, often from the same organisms that furnish restriction enzymes. Because the names may be similar, it is essential to specify the type of enzyme, for example:

  • AluI methylase
  • Pfu DNA polymerase
  • TagI methylase
  • Taq DNA ligase

Restriction enzymes

For restriction enzymes, the first three letters indicate the genus and species from which the enzyme is derived. For example:

  • Bme899
  • Sau3A1
  • AccI
  • HpaI
  • DpnI


  • According to WHO recommendations, classes of immunoglobulins should be abbreviated by Ig (not Ig) plus a capital Roman letter indicating the class (e.g. immunoglobulin G is abbreviated IgG).
  • A subclass is indicated by an Arabic numeral after the class letter (closed up, not hyphenated, e.g. IgG1, IgG3). Potentially new subclasses are designated by an abbreviation in parentheses following the class. IgG(Pr), for instance, indicates a potentially new subclass identified in Prague.
  • Any of these abbreviations may begin a sentence or entry in a list or table, with no other capitalisation necessary.
  • In the WHO system, Greek gamma (γ) is used only to designate heavy polypeptide chains of IgG.
  • In the old system gamma- (or γ-)globulin was used as the broad description of immunoglobulins. Although this is not the preferred system, if an author follows it consistently, retain it; if the author is not consistent, use IgG. In the old system, γ-globulin is preferred to gamma-globulin. If the word begins a sentence or an entry in a table or list, capitalise G in globulin.
  • Immunoglobulin chains are known as light (L) and heavy (H). They contain regions designated constant (C), variable (V), diversity (D), and joining (J). For example, the variable region of the heavy chain can be abbreviated as VH. If the gene is referred to, it should be italicised (VH); if the gene product (i.e., the chain itself) is referred to, it should be written in Roman letters. Heavy chains are designated by a lower case Greek letter corresponding to the roman capital letter of the class. Light chains are divided into two types, kappa (κ) and lambda (λ). Use of the Greek letter is preferred (κ-chain).

Top of section