Reference Style (nejm-citation)

When to trigger

• References are in author–date (APA/Harvard) or another numbered convention and must become Vancouver/ICMJE.
• The reference list is alphabetical instead of ordered by citation appearance.
• Author lists are not handled per the "list N then et al." convention.
• The reference count exceeds the article-type cap.

Vancouver / ICMJE mechanics (NEJM)

• Numbered, in order of appearance in the text; in-text citations are superscript numbers (e.g., text.¹ or text.²,³).
• A single reference list numbered in citation order — not alphabetical.
• Author lists: the Vancouver/ICMJE convention lists the first six authors then "et al." when there are more than six. NEJM uses a limited author list of this kind — confirm the exact cutoff in the current author guidelines.
• Journal titles abbreviated per NLM / Index Medicus abbreviations.
• Include volume, issue/pages (or article identifier), and year; include the DOI where available.
• Keep the reference list within the article-type cap (an Original Article runs a limited list, on the order of ~40).

Reference formats (shape)

• Journal article: Author AB, Author CD, Author EF, et al. Title of the article. Journal Abbrev Year;Volume:first-last.
• Book: Author AB. Book title. Edition. City: Publisher; Year.
• Chapter: Author AB. Chapter title. In: Editor CD, ed. Book title. City: Publisher; Year:pp-pp.
• Online / dataset: Author/Org. Title. URL (accessed date). — include DOI when present.

Surname precedes initials (no periods between initials in Vancouver); article title in sentence case; journal abbreviated and not italicized in the classic style. Use a reference manager's NEJM/Vancouver style and verify mechanically.

Conversion from author–date

Tooling

• Use Zotero/EndNote with an NEJM / Vancouver (ICMJE) style; do a final manual pass on the author cutoff and NLM abbreviations.
• Verify every superscript number resolves to a list entry, with no gaps or duplicates, and that the count is within the cap.

Citation pass for New England Journal of Medicine

Use this as a second-pass capability check. First lock the clinical question, population, endpoint, effect size, safety signal, and practice implication; then test whether the manuscript addresses clinical-medicine reviewers who expect practice-changing evidence, patient relevance, safety, and exact reporting discipline.

• Primary move: Audit references and notes as evidence discipline: each source should position, document, or delimit a claim.
• Decision ledger: return claim / evidence / blocker / next edit rows so the next pass can patch the manuscript directly.
• Neighbor test: compare against JAMA for broad clinical medicine, Lancet for global-health/public-health reach, specialty journals for narrower disease domains; if the neighboring outlet has the stronger audience claim, recommend re-routing before polishing.
• Verification floor: before submission-ready advice, re-open resources/official-source-map.md for volatile rules and name the one unresolved fact that could change the recommendation.

Output format

【Style detected】 author-date / other-numbered → Vancouver/ICMJE
【Numbering】 by appearance? single list, not alphabetical? yes/no
【Author lists】 first six then et al. (confirm cutoff)? yes/no
【Journal abbreviations】 NLM/Index Medicus applied? yes/no
【Reference count】 N vs ~40 cap (article-type) → ok / over
【Unresolved/duplicate refs】 [...]
【Next】 nejm-submission

Anti-patterns

• Do not leave author–date citations anywhere in an NEJM manuscript.
• Do not alphabetize the reference list — order is by appearance.
• Do not spell out full journal names — use NLM abbreviations.
• Do not list all authors when there are more than six — use "et al." per the convention.