BMJ Formatting Requirements: Complete Author Guide

BMJ Research articles require a structured abstract of 250-300 words (up to 400 words for a CONSORT or PRISMA style abstract). The headings are:

• Objectives (what the study aimed to do)

• Design (study type: RCT, cohort, cross-sectional, etc.)

• Setting (where the study was conducted)

• Participants (who was studied, with numbers)

• Interventions (if applicable)

• Main outcome measures (primary and secondary)

• Results (key findings with effect sizes and confidence intervals)

• Conclusions (what the results mean for practice)

• Trial registration (registry and number, for clinical trials and, if available, for observational studies and systematic reviews)

The BMJ abstract structure is more granular than what you'll find at JAMA or The Lancet. Each section should be concise but specific. The Design section, for example, shouldn't just say "cohort study." It should specify "prospective population-based cohort study" or "retrospective matched cohort study."

In the Results section, include effect estimates (odds ratios, hazard ratios, risk differences) with 95% confidence intervals and, where appropriate, the exact level of statistical significance. The BMJ abstract is expected to contain enough quantitative detail that readers can assess the findings without reading the full paper.

The BMJ asks for a title page that gives:

• Manuscript title (informative; for research papers, add a subtitle with the study design)

• All author names, each with affiliation (job title) at the time the paper was written, and email

• Corresponding author details (name, full address, email); there can be only one corresponding author

• ORCiD iD for the corresponding author on research articles (required if the article is accepted)

You enter the main-text word count when you submit the manuscript. The BMJ also asks for a short summary box, described below.

The summary box is distinctive to the BMJ. It has two sections:

What is already known on this topic:

• 2-3 single-sentence bullets summarizing the state of knowledge before the study, and why the study was needed

What this study adds:

• 1-2 single-sentence bullets giving a simple answer to "What do we now know as a result of this study that we did not know before?", ideally with the one most useful summary statistic, such as NNT

Write it carefully because readers see it early. Each bullet should be one sentence, directly stated, with no hedging.

The BMJ states no maximum number of figures and tables for research articles, but it asks authors to use illustrations only when data cannot be expressed clearly in any other way.

Figure requirements:

• Resolution: The BMJ's image page says ideally at least 300 dots per inch, and images below 200 dots per inch may produce fuzzy images in print; the BMJ Author Hub for BMJ journals says a minimum of 300 dpi for non-vector files and 1200 dpi for line art

• Accepted formats: BMJ says images should be submitted as GIF, TIFF, EPS, or JPEG files, and prefers PDF or JPEG for figures uploaded in ScholarOne

• Supply image files at at least 100% of the intended printed size; BMJ cannot enlarge images by more than 5%

• Figures should be uploaded as separate files during submission, not embedded in the text

• Patient photographs need the patient's written consent if there is any chance the patient could be identified

• When you submit graphs, scattergrams, or histograms, supply the numerical data they are based on

Table requirements:

• The BMJ's submission page asks for tables as separate Word files and says not to use Excel; PowerPoint and Excel files cannot be accepted for tables or figures that accompany the main manuscript. The BMJ Author Hub formatting checklist for BMJ journals instead asks whether tables are embedded in the main Word document, so follow the instructions shown in the ScholarOne upload screens

• Tables should be simple and should fit across a maximum of two landscape pages

• Tables should not duplicate information in the text of the paper

• Keep caption text concise and relevant to the table

Supplementary files:

BMJ accepts supplementary files (for example videos, datasets, research protocols, and checklists or statements). They are not converted to PDF and are available to reviewers or editors exactly as you upload them; they may accompany the final version of the paper online.

The BMJ uses Vancouver reference style, which is the most common format in biomedical publishing. If you've published in any medical journal, you likely know this style already.

Key formatting rules:

• References are numbered in the order in which they appear in the text; reference numbers go after commas and full stops, and before semicolons and colons. The BMJ Author Hub says citations in the text can be in square brackets [6] or superscripts

• References are numbered consecutively in order of first appearance

• For papers with 1-6 authors, list all authors

• For papers with 7 or more authors, list the first 3 followed by "et al"

• Journal titles are abbreviated according to the style of Index Medicus, and online material is cited with its URL

• The BMJ Author Hub says each reference should include a DOI or PubMed ID

Note that BMJ's documents differ on author lists: its editorial instructions say to give the first six authors followed by et al when there are more than six, while its house-style page says to give the first three. Check a recent BMJ article before you finalize.

Example journal article (6 or fewer authors):

1. Davies NM, Holmes MV, Davey Smith G. Reading Mendelian randomisation studies: a guide, glossary, and checklist for clinicians. BMJ 2018;362:k601. doi:10.1136/bmj.k601.

Example journal article (7+ authors):

1. Patel R, Singh A, Kumar V, et al. Long-term cardiovascular outcomes after bariatric surgery. BMJ 2026;388:e074521. doi:10.1136/bmj-2025-074521.

Example book:

1. Rothman KJ, Greenland S, Lash TL. Modern Epidemiology. 4th ed. Wolters Kluwer; 2021.

Note that the BMJ uses a semicolon between volume and page number (362:k601) rather than a comma. Also, the BMJ has increasingly used article IDs rather than page numbers, so check the current format for recent BMJ articles.

This is one of the BMJ's most distinctive requirements. BMJ asks authors of research papers to provide a short paragraph, as a subsection within the Methods, entitled "Patient and Public Involvement" describing how patients or members of the public were involved in the study. BMJ calls this mandatory patient and public involvement reporting.

The statement should address:

• Were patients involved in setting the research question or outcome measures?

• Were patients involved in the design of the study?

• Were patients involved in the recruitment or conduct of the study?

• Will results be disseminated to study participants?

If patients were not involved, you must explicitly state that. "No patients were involved in setting the research question or the outcome measures, nor were they involved in developing plans for the design or implementation of the study" is acceptable, but the BMJ editors prefer actual involvement.

This requirement reflects the BMJ's editorial commitment to patient-centered research. Papers that demonstrate genuine patient involvement in study design tend to be viewed more favorably during review.

The BMJ asks that all manuscripts be written in accordance with the appropriate reporting guideline, and asks you to submit the completed checklist as supplemental material. Treat a missing checklist as a likely cause of a return.

Checklists by study type (The BMJ's article-types page names CONSORT, PRISMA, STARD, STROBE, GRIPS, CHEERS, TRIPOD and CODE-EHR; the BMJ Author Hub list also names SQUIRE and ARRIVE; both say others may apply, and COREQ is on neither list):

Download the appropriate checklist from the EQUATOR Network and complete it before submission. In the checklist, specify the page number or line number where each item is addressed in your manuscript.

If you check "yes" for an item but the information isn't actually in the paper, reviewers and editors can see the gap. Fill out the checklist honestly and use it as an editing tool to ensure completeness.

The BMJ practices open (signed) peer review. This doesn't directly change how you format your manuscript, but it affects the review process in ways authors should understand.

What open peer review means at the BMJ:

• Reviewers sign their reports, and accepted research and analysis papers usually have their prepublication history posted alongside them on bmj.com

• That history generally includes previous versions, the committee report, the reviewers' comments and the authors' responses

• Reviewers agree to have their signed comments posted if a paper is published, but not otherwise

• Authors should write responses to reviewers knowing they'll be published

This transparency tends to produce more constructive and detailed reviews. When you write your response to reviewers, be thorough and professional. Your response letter becomes a permanent part of the published record.

1. The "What this study adds" box. This is published prominently with every Research article. It's not optional and isn't the same as the abstract. Think of it as the elevator pitch for your paper.

2. Patient involvement statement placement. This goes in the Methods section, as a subsection entitled Patient and Public Involvement, not at the end of the paper.

3. Data sharing statement. The BMJ requires a data sharing statement for all research papers. For reports of clinical trials submitted on or after 1 May 2024, the underlying data must be available at the time of publication; for other research, BMJ does not require data sharing but encourages authors to state that they will share data on reasonable request.

4. Transparency declaration. The lead author must include a transparency declaration: "The lead author affirms that the manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as originally planned (and, if relevant, registered) have been explained."

5. Copyright and open access. BMJ Research articles are published open access. The default licence is Creative Commons Attribution Non Commercial (CC BY-NC 4.0); where a funder requires it, the author can select CC BY 4.0 during submission. BMJ asks authors for an open access fee on acceptance (see the BMJ APC guide).

6. Competing interests format. The BMJ uses a standardized competing interests statement. Each author must complete the ICMJE disclosure form, and the statement published in the paper must match.

A BMJ Research article follows the IMRaD style (introduction, methods, results, and discussion). A typical order of elements is:

1. Title page (title with study design, authors, affiliations, corresponding author)

1. Structured abstract (250-300 words, with Objectives/Design/Setting/Participants/Interventions/Main outcome measures/Results/Conclusions/Trial registration) and the summary box ("What is already known on this topic" and "What this study adds")

1. Introduction (succinct, no more than three paragraphs on the background to the research question)

1. Methods (study design, participants, data sources, outcomes, statistical analysis, patient involvement)

1. Results (organized to match the methods, with key figures and tables)

1. Discussion (principal findings, comparison with other studies, strengths and limitations, policy implications)

1. Conclusion (1 paragraph, state the main finding and its implication)

1. Data sharing statement

1. Transparency declaration

1. Author contributions

1. Funding

1. Competing interests

1. References

1. Figure legends

1. Tables

1. Supplementary appendices

The Discussion section at the BMJ has a specific expected structure and should run to no more than a page and a half. It covers: a statement of principal findings; strengths and weaknesses of the study; strengths and weaknesses in relation to other studies, discussing important differences in results; the meaning of the study, with possible explanations and implications for clinicians and policymakers; and unanswered questions and future research. You do not need to signpost these elements with subheadings.

The BMJ has specific expectations for statistical reporting that go beyond most journals:

• Accompany p-values with supporting data, and give denominators for percentages

• Include 95% confidence intervals for all primary comparisons

• Whenever possible, state absolute rather than relative risks (don't report only odds ratios)

• Report numbers needed to treat (NNT) when clinically appropriate

• Pre-specify primary and secondary outcomes; clearly label any post-hoc analyses

• Reference the statistical package used; for a very widely used package such as Stata, no formal reference is needed, but say which version you used

For clinical trials, the BMJ asks for the CONSORT checklist and flowchart showing participant flow.

These errors cause the most delays at the BMJ:

• Missing the EQUATOR reporting checklist entirely

• Omitting the patient involvement statement

• Using "Background" instead of "Objectives" as the first abstract heading

• Forgetting the "What this study adds" box

• Not including a transparency declaration

• Reporting only relative effects without absolute measures

• Submitting without a data sharing statement

• Missing the competing interests disclosures for all authors

• Not following the structured Discussion format

For comparison with another major medical journal, see our JAMA formatting requirements guide. If you're still deciding between these journals, our BMJ vs JAMA comparison covers scope and editorial philosophy.

For the official and complete specifications, visit the BMJ author guidelines page.

Method note: this formatting guide was checked against BMJ author resources, the ScholarOne route at mc.manuscriptcentral.com/bmj, public BMJ research records with prepublication history, SciRev community timing reports, and Manusights reviews of general-medicine submissions. In our review work on BMJ formatting, we see four patterns generate the most consistent technical-return or desk-risk outcomes before editors spend time on the science.

Patient and Public Involvement statement absent or pro forma. The BMJ author guidelines call for mandatory patient and public involvement reporting, as a Patient and Public Involvement subsection within the Methods. If the information is missing from a submitted manuscript, BMJ says it will ask authors to provide it. The guidelines ask authors to describe how patients or the public were involved in setting priorities, defining questions and outcome measures, design, conduct, and dissemination, or to document formally that they were not involved. BMJ also says it continues to consider papers where patients were not involved.

"What is already known / What this study adds" summary box missing. BMJ asks research authors for a two-part summary box: "What is already known on this topic" (2-3 single-sentence bullets) and "What this study adds" (1-2 single-sentence bullets of the study's specific new contribution). This box is separate from the abstract.

Structured abstract headings do not match BMJ format. The BMJ asks for a structured abstract with these headings: Objectives, Design, Setting, Participants, Interventions (where there were interventions), Main outcome measures, Results, Conclusions, and Trial registration for clinical trials. Common substitutions we see include "Aim" instead of "Objectives," "Materials and Methods" instead of "Design, Setting, Participants," or "Background" as an opening section.

Reporting checklist submitted as blank or partially completed. The BMJ asks for the completed reporting guideline checklist as supplemental material. Reviewers check whether each item is addressed and the page number cited. Checklists where items are marked "N/A" without justification, or where the page-number column is left blank, generate revision requests before or during peer review.

The practical formatting test is whether a BMJ editor can verify the manuscript package without searching. The abstract headings, PPI statement, summary box, EQUATOR checklist, references, figures, tables, ethics approval, and data-availability statement should all point to the same study design. If one component contradicts another, the formatting problem becomes a trust problem.

A BMJ submission readiness check evaluates your PPI statement, summary box, abstract headings, and reporting checklist completeness against these patterns.