BMJ (British Medical Journal) Submission Guide: What Editors Screen Before Review

That is the real BMJ screen. The journal is not mainly checking whether the paper is statistically respectable. It is deciding whether the work belongs in a broad medical conversation and whether the package is credible enough for BMJ's transparency model.

The official BMJ resources are unusually clear that research papers need a Patient and Public Involvement statement, a Data Availability Statement, and the relevant reporting checklist. That means the submission package is doing editorial work before peer review even begins.

The BMJ editor-in-chief is Kamran Abbasi. That matters less as a name to cite than as a reminder that The BMJ is edited for a broad medical readership, not only for the subspecialty that generated the data.

The specialty paper pushed upward for brand reasons. This is still the most common mismatch. The study may be excellent in oncology, GI, cardiology, infectious disease, or another field, but the broader general-medical case is not strong enough.

The practical consequence is still rhetorical. The manuscript says the work matters for care or policy, but the data do not yet force a practical change in thinking.

The transparency layer looks late-built. Thin PPI language, vague data-sharing wording, or half-finished reporting checklists make the paper feel less mature than the results deserve.

The paper is too locally framed. Local data can still work, but the transfer case has to be visible. When it is not, the journal-readership fit weakens immediately.

The first screen is too slow. If the title, abstract, and first display cannot explain the consequence quickly, BMJ editors often assume the general-medical case is weaker than the authors think.

Submission caps: Research articles cap at 4000 words of body text with a 250-word structured abstract and no fixed limit on figures and tables (editorial restraint preferred, with most accepted papers using 4 to 6 figures or tables combined). Vancouver-style references with numbered superscript citations. A mandatory "Patient and Public Involvement" section is required. Supplementary files commonly accept up to 50 MB per upload.

• Day 0: BMJ submission upload. The bmj.com/submission portal and BMJ ScholarOne site accept the package (manuscript, structured abstract, ORCID identifiers, cover letter explaining broad-audience consequence, ICMJE conflicts of interest disclosure, funding statement, author contributions, data availability statement, Patient and Public Involvement section, reporting checklist matched to study type, suggested reviewers), run BMJ Publishing Group integrity checks, and route to a BMJ handling editor.
• Days 1 to 21: First editor read. The editor evaluates whether the paper changes practice, policy, or clinical thinking outside one specialist lane, reporting-checklist completeness, and methodological transparency. About 70 to 80% of submissions are desk-rejected here.
• Days 21 to 60: Peer review. Two to four reviewers spanning the clinical specialty, biostatistics, and clinical-trials methodology. Reviewer reports return on a 4 to 8 week cadence.
• Days 60 to 90: First editorial decision. Major revision is the most common outcome for papers that pass desk review.
• Days 90 to 180: Revision rounds and publication. BMJ production typically pushes accepted Research articles online within 2 to 4 weeks of acceptance.

This guide tells you what BMJ editors look for on the public page; the review tells you whether your paper passes that screen before you spend the submission. Manusights checks are covered by a 60-day money-back guarantee, and we do not train models on submitted manuscripts.

In our analysis of BMJ submission pages, the specific failure pattern is almost always visible in the manuscript components before the portal is opened: the abstract, cover letter, first figure, methods transparency layer, or references are carrying a narrower story than the title promises.

The specialty result without a general-medical decision

In our pre-submission review work with clinical manuscripts targeting BMJ, the most common fit problem is not weak methods. It is a title, abstract, and first figure that still read like a specialty paper that has been pushed upward for brand reasons. The data may be real, the statistical analysis may be respectable, and the discussion may be clinically interesting, but the manuscript does not yet state what a general clinician, public-health reader, guideline committee, or health-system decision maker should do differently after reading it.

For BMJ, that weakness has to be fixed before upload. The cover letter should name the current clinical decision or policy uncertainty, the abstract should quantify the practical consequence, and the first table or figure should make the patient-facing or system-facing result legible without specialty shorthand. If the result belongs more naturally in JAMA, The Lancet, NEJM, Annals of Internal Medicine, or a specialist BMJ title such as Gut or Heart, the manuscript should say why BMJ is still the right broad-medical venue rather than relying on prestige fit.

Check whether your BMJ manuscript passes the broad-audience decision screen →

The transparency package that looks added after the science

In our pre-submission review work with BMJ-targeted manuscripts, the second recurring pattern is a research package where the Patient and Public Involvement statement, Data Availability Statement, reporting checklist, trial registration, conflicts disclosure, and supplementary methods look like compliance material assembled after the paper was finished. That is dangerous for BMJ because the journal's author guidance treats patient partnership, reporting discipline, and data sharing as part of research trust, not as paperwork.

A stronger BMJ submission makes those components cohere with the manuscript. The methods section should explain how patient or public input affected the research question, outcome selection, recruitment materials, interpretation, or dissemination plan. The data availability statement should be specific about repository access, request conditions, de-identification limits, code, protocols, and statistical analysis files. The reporting checklist should map to the paper rather than merely exist as an attachment. If the paper would be more credible at BMJ Open, PLOS Medicine, BMC Medicine, JAMA Network Open, or a specialty society journal after a transparency rebuild, that is a fit signal, not just a formatting issue.

Check whether your BMJ submission package has the right transparency signals →

The cover letter that describes importance instead of changing the editor's triage

In our pre-submission review work with manuscripts targeting BMJ, weak cover letters usually restate the abstract: background, methods, findings, conclusion. Stronger BMJ cover letters do something more useful for triage. They tell the editor what decision problem the paper resolves, why that problem matters across settings, why the study design can support the claim, and why a specialty journal would under-serve the audience that needs the finding.

The manuscript components have to support the same argument. The abstract needs a clear effect size, confidence interval, and patient or policy implication. The discussion should not bury limitations in generic language; it should explain why the result travels beyond the sampled hospital, country, population, or intervention. The references should show awareness of recent BMJ, JAMA, Lancet, NEJM, and Annals conversations rather than only field-specific precedent. A BMJ editor should be able to use the cover letter, abstract, and first display to defend the paper's broad-medical relevance in one minute.

Check whether your BMJ cover letter and abstract make the editorial case →

Ask these three questions before you submit:

1. Would a general clinician or health-policy reader still care if the specialty labels were removed?
2. Does the title, abstract, and first display make the practical consequence visible immediately?
3. Does the package already look publishable under a high-transparency model?

If the answer to any of those is weak, the fit problem is probably larger than any portal or formatting issue.