Skip to content
CitationEasy
Field-Specific

How to Cite Medical Sources in Vancouver and AMA

Medical papers use numbered citations so a methods paragraph can carry ten trial numbers without a thicket of parentheses. The current authorities are the *AMA Manual of Style*, 11th edition (2020) for JAMA Network journals and US medical schools, and the free ICMJE Recommendations plus NLM *Citing Medicine* for Vancouver. Cite the object you retrieved—a journal article, a guideline chapter, a Cochrane review, a preprint, a ClinicalTrials.gov record, or a package insert—not PubMed, the PDF filename, or a lecture slide about that object.

By 24 min read

The ornate two-storey reading room of the library at the Château de Chantilly
Photo: Gabriel Ghnassia, Wikimedia Commons (CC0)
On this page

Quick Answer: How to Cite Medical Sources

Open the syllabus or the journal's Instructions for Authors first. Use AMA when that document names the AMA Manual of Style or a JAMA Network title. Use Vancouver when it names ICMJE, Uniform Requirements, NLM, or Vancouver. Number sources in the order they first appear and reuse that number later. Look up the journal abbreviation in the NLM Catalog. Prefer a DOI over a URL. Do not invent an NEJM abbreviation or a fake trial identifier.

Medical citation templates

AMA 11 journal article:

Author AA, Author BB. Title of article. Abbreviated Journal. Year;volume(issue):pages. doi:10.xxxx/xxxxx

Vancouver journal article (NLM sample):

Author AA, Author BB. Title of article. Abbreviated Journal. Year Mon Day;volume(issue):pages.

AMA in the sentence:

Solid-organ transplantation in selected patients with HIV infection has been reported.1

ICMJE Vancouver in the sentence:

Solid-organ transplantation in selected patients with HIV infection has been reported (1).

Authors before et al.:

AMA: list 1–6; if more than 6, first 3 + et al. Vancouver samples: first 6 + et al.

Those in-text marks are not interchangeable house styles. ICMJE and NLM state the Vancouver policy in sentences you can take to a professor who says "just number them":

Official Vancouver wording

ICMJE Recommendations

References should be numbered consecutively in the order in which they are first mentioned in the text. Identify references in text, tables, and legends by Arabic numerals in parentheses.

ICMJE, on journal titles

The titles of journals should be abbreviated according to the style used for MEDLINE.

ICMJE, on NLM format

References should follow the standards summarized in the NLM's Sample References webpage and detailed in the NLM's Citing Medicine, 2nd edition.

National Library of Medicine

The recommended style for references is based on the National Information Standards Organization NISO Z39.29-2005(R2010) Bibliographic References as adapted by the National Library of Medicine for its databases.

ICMJE, on abstracts versus articles

References should be made to published articles rather than to abstracts whenever possible.

AMA is stricter about the mark, the author cutoff, and the issue number. The JAMA Network Open Instructions for Authors, which implement AMA 11 for that journal, put the house rule in one paragraph:

JAMA Network Open Instructions for Authors

Number references in the order they appear in the text; do not alphabetize. In text, tables, and legends, identify references with superscript arabic numerals. When listing references, follow AMA style and abbreviate names of journals according to the journals list in PubMed. List all authors and/or editors up to 6; if more than 6, list the first 3 followed by "et al." Note: Journal references should include the issue number in parentheses after the volume number.

What This Guide Covers

Use this page when the hard part is the medical object: an RCT versus the CONSORT statement that tells you how to report it, a Cochrane review versus the trials it pooled, an ADA Standards chapter versus a lecture slide, a medRxiv preprint versus the later journal article, a ClinicalTrials.gov record versus the published results paper, or a package insert versus a review article about the drug. For the AMA versus Vancouver punctuation table, start at AMA vs Vancouver. If the real choice is numbered medicine versus author-date social science, see Vancouver vs APA. Nursing programs that require APA 7 should use the nursing citation guide instead of importing superscripts. The field hub is citation styles for medicine.

Cite the Object You Retrieved

Finding a PDF in PubMed does not make the source a webpage or a database citation. Name the work a later reader can open. In medicine that choice also has a patient-safety edge: a methods paragraph that cites a review as if it were the trial, or a retracted paper as if it were current evidence, can send a reader to the wrong protocol.

Choose the template from the object you used, not from the search box.
What you usedCite this objectDo not cite
Peer-reviewed articleThe article (authors, abbreviated journal, year, volume, issue, pages or e-locator, DOI)PubMed, MEDLINE, or the PDF filename
Randomized trialThe trial report you opened, plus the registry number if the journal asks for itA news write-up or a textbook paragraph about the trial
Cochrane or other systematic reviewThat review (authors, Cochrane Database or journal, DOI)Every trial the review pooled, unless you also retrieved those trials
Clinical practice guidelineThe issuing agency's dated chapter or full guideline at the URL you usedA guideline aggregator or the retired National Guideline Clearinghouse
PreprintThe preprint server record, marked as a preprint, with the version you readThe later journal article you have not opened
Trial registrationThe ClinicalTrials.gov or other registry record, including the NCT numberAn invented NCT string or the results paper you did not retrieve
Package insert or prescribing informationThe drug name, insert type, manufacturer, year, and URL you openedA lecture slide or a review article that paraphrases the label

Which Style: AMA, Vancouver, or APA?

Medicine is not one citation style. US MD and DO programs, and JAMA Network journals, usually want AMA 11. Most international biomedical journals want a Vancouver-family numbered list built on ICMJE and NLM. BSN, MSN, and DNP programs, and most nursing journals, want APA 7. Do not mix superscripts from a JAMA article you liked into an APA nursing paper, and do not copy an APA parenthetical year into a Vancouver reference list.

AMA 11th edition (2020)

  • One paid manual: AMA Manual of Style
  • In-text: superscript after punctuation1
  • Journal titles: MEDLINE abbreviation, italic
  • More than 6 authors: first 3, then et al.
  • Books: publisher; year (no city)
  • DOI: doi:10.1001/jama.2016.9797 (no space, no period)

Vancouver (ICMJE / NLM)

  • A free policy layer, not one paid book
  • ICMJE in-text: Arabic numerals in parentheses (1)
  • Journal titles: MEDLINE abbreviation, usually roman
  • More than 6 authors: first 6, then et al.
  • Books: Place: Publisher; year
  • DOI: doi: 10.1136/bmj.a2752 (space after the colon in NLM samples)

If the prompt just says "Vancouver," open the target journal before you format the first number. ICMJE names parentheses. Many medical schools teach brackets [1] or superscripts and still call the result Vancouver. The reference list should follow NLM Samples of Formatted References (last reviewed 2 June 2026) unless the journal prints its own samples.

In-Text Citations

Give each source one number the first time it appears. Reuse that number for every later mention, including tables and figure legends. A source cited only in a table is numbered when that table is first mentioned, not dumped at the end. Neither list is alphabetical. AMA places the superscript after commas and periods, with no space before the mark. ICMJE adds the table-and-figure sequence in the same Recommendations page:

ICMJE, sources that appear only in a table or figure

References cited only in tables or figure legends should be numbered in accordance with the sequence established by the first identification in the text of the particular table or figure.

The same Halpern paper, two in-text marks

AMA:

Solid-organ transplantation in selected patients with HIV infection has been reported.1 Later reviews reused the same number.1

ICMJE Vancouver:

Solid-organ transplantation in selected patients with HIV infection has been reported (1). Later reviews reused the same number (1).

Common journal Vancouver variants:

Reported [1]. / Reported.1

Several sources:

AMA: 1,3 or 1-3. ICMJE: (1,3) or (1-3).

Journal Articles: The Default Medical Source

Most evidence-based assignments want a primary empirical report or a systematic review you actually opened. JAMA Network Open restates the AMA 11 author cutoff that students miss on multi-center trials: list every name through six authors; if there are more than six, print the first three plus et al. NLM's sample-references page (last reviewed 2 June 2026) still uses the three-author New England Journal of Medicine paper as its standard journal example. Both styles list every name when there are six or fewer. Italics mark the AMA journal title. Vancouver follows NLM punctuation, including the optional short form that drops month and issue when pagination is continuous:

Official NLM / AMA journal example (Halpern et al., 2002)

AMA 11:

1. Halpern SD, Ubel PA, Caplan AL. Solid-organ transplantation in HIV-infected patients. N Engl J Med. 2002;347(4):284-287.

Vancouver (NLM sample, with issue):

1. Halpern SD, Ubel PA, Caplan AL. Solid-organ transplantation in HIV-infected patients. N Engl J Med. 2002 Jul 25;347(4):284-7.

Vancouver (NLM optional short form):

1. Halpern SD, Ubel PA, Caplan AL. Solid-organ transplantation in HIV-infected patients. N Engl J Med. 2002;347:284-7.

When the byline has seven or more names, the cutoff splits. NLM's sample for that case is Rose et al. in Brain Research. JAMA Network Open's own worked list uses Youngster et al. in JAMA (six authors, so every name stays). The Obama JAMA special communication and the Wang JAMA Pediatrics e-locator remain the AMA 11 DOI and e-locator patterns used on this page:

Official multi-author and DOI examples

Vancouver (NLM sample, first 6 + et al.):

2. Rose ME, Huerbin MB, Melick J, Marion DW, Palmer AM, Schiding JK, et al. Regulation of interstitial excitatory amino acid concentrations after cortical contusion injury. Brain Res. 2002;935(1-2):40-6.

AMA 11 (same paper, first 3 + et al.):

2. Rose ME, Huerbin MB, Melick J, et al. Regulation of interstitial excitatory amino acid concentrations after cortical contusion injury. Brain Res. 2002;935(1-2):40-46.

JAMA Network Open worked example:

3. Youngster I, Russell GH, Pindar C, Ziv-Baran T, Sauk J, Hohmann EL. Oral, capsulized, frozen fecal microbiota transplantation for relapsing Clostridium difficile infection. JAMA. 2014;312(17):1772-1778. doi:10.1001/jama.2014.13875

AMA 11 (Obama, JAMA 2016; PMID 27400401):

4. Obama B. United States health care reform: progress to date and next steps. JAMA. 2016;316(5):525-532. doi:10.1001/jama.2016.9797

AMA 11 (Wang et al., e-locator):

5. Wang G, Hu FB, Mistry KB, et al. Association between maternal prepregnancy body mass index and plasma folate concentrations with child metabolic health. JAMA Pediatr. 2016;170(8):e160845. doi:10.1001/jamapediatrics.2016.0845

Group authors stay as published. Do not write "Anonymous" unless that word is in the byline. AMA 11 uses the World Medical Association's Declaration of Helsinki and a named writing committee as the pattern. NLM's organization-as-author sample is the Diabetes Prevention Program Research Group; its personal-plus-organization sample is Vallancien et al. with the Alf-One Study Group:

Official group-author examples

AMA 11 (WMA, JAMA 2013):

6. World Medical Association. World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects. JAMA. 2013;310(20):2191-2194. doi:10.1001/jama.2013.281053

AMA 11 (writing committee):

7. Writing Committee for the Diabetic Retinopathy Clinical Research Network. Panretinal photocoagulation vs intravitreous ranibizumab for proliferative diabetic retinopathy. JAMA. 2015;314(20):2137-2146. doi:10.1001/jama.2015.15217

Vancouver (NLM organization-as-author sample):

6. Diabetes Prevention Program Research Group. Hypertension, insulin, and proinsulin in participants with impaired glucose tolerance. Hypertension. 2002;40(5):679-86.

Vancouver (NLM personal + organization sample):

7. Vallancien G, Emberton M, Harving N, van Moorselaar RJ; Alf-One Study Group. Sexual dysfunction in 1,274 European men suffering from lower urinary tract symptoms. J Urol. 2003;169(6):2257-61.

Journal Abbreviations

A widely copied chart says AMA prints full journal names and Vancouver abbreviates them. That chart is wrong. Both styles use MEDLINE abbreviations from the same NLM list. JAMA Network Open is explicit: abbreviate names of journals according to the journals list in PubMed. AMA then italicizes that abbreviation.

AMA journal-title practice (JAMA Network implementation)

When listing references, follow AMA style and abbreviate names of journals according to the journals list in PubMed.

Look the abbreviation up. Do not guess J. Am. Med. Assoc. or NEJM. Search the NLM Catalog and copy the MEDLINE / PubMed abbreviation. One-word titles such as Lancet or JAMA stay unabbreviated because there is nothing to shorten. If a journal later changed its name, cite the title that was in use when the article was published: Arch Intern Med for a 2012 paper, JAMA Intern Med for a 2015 paper.

NLM / MEDLINE abbreviations used by both styles
Full title (for recognition)Abbreviation in the reference
The New England Journal of MedicineN Engl J Med
JAMAJAMA
The LancetLancet
BMJ (formerly British Medical Journal)BMJ
Annals of Internal MedicineAnn Intern Med
Diabetes CareDiabetes Care
Cochrane Database of Systematic ReviewsCochrane Database Syst Rev

DOI, URL, and PMID

A DOI is a permanent identifier for the article. A PMID is a PubMed locator. A URL is a web address that can move. AMA 11 treats the DOI as the last element. University restatements of AMA 11 (USC, UNE, James Cook) warn against a trailing period that would break the copy-paste: write doi:10.1001/jama.2016.9797 with no space after the colon and no period after the identifier. Include a DOI when one exists. No accessed date is required for a DOI because it is a permanent identifier.

NLM's optional DOI sample is Zhang et al. in BMJ. Note the space after doi: in the NLM line. Some journals also want a PMID. Include it when the instructions ask; do not substitute a PMID for a DOI. If you still need to recover a DOI, see How to Find a DOI or How to Use DOIs.

Official NLM DOI sample

Vancouver (NLM sample):

Zhang M, Holman CD, Price SD, Sanfilippo FM, Preen DB, Bulsara MK. Comorbidity and repeat admission to hospital for adverse drug reactions in older adults: retrospective cohort study. BMJ. 2009 Jan 7;338:a2752. doi: 10.1136/bmj.a2752. PubMed PMID: 19129307; PubMed Central PMCID: PMC2615549.

Ahead of Print, Supplements, and Errata

Students often treat an online-first paper, a journal supplement, and a corrected republication as if they were ordinary paginated articles. NLM's sample-references page gives a separate line for each case. Use the Epub date when the article appeared electronically before print. Use Suppl when the pagination sits in a supplement—exactly the situation for the ADA Standards of Care chapter later on this page. If you cite a paper that later published an erratum, keep the original citation and add the erratum note:

Official NLM ahead-of-print, supplement, and erratum samples

NLM #18, published electronically ahead of print:

Yu WM, Hawley TS, Hawley RG, Qu CK. Immortalization of yolk sac-derived precursor cells. Blood. 2002 Nov 15;100(10):3828-31. Epub 2002 Jul 5.

NLM #6, volume with supplement:

Geraud G, Spierings EL, Keywood C. Tolerability and safety of frovatriptan with short- and long-term use for treatment of migraine and in comparison with sumatriptan. Headache. 2002;42 Suppl 2:S93-9.

NLM #17, article with published erratum:

Malinowski JM, Bolesta S. Rosiglitazone in the treatment of type 2 diabetes mellitus: a critical review. Clin Ther. 2000;22(10):1151-68; discussion 1149-50. Erratum in: Clin Ther. 2001;23(2):309.

Clinical Practice Guidelines

Guidelines are how a medical paper moves from a single trial to a recommended action. Cite the dated chapter or full document you opened, not a blog summary. The American Diabetes Association publishes Standards of Care in Diabetes each year as a Diabetes Care supplement. The 2026 glycemic-goals chapter (PMC12690178; doi:10.2337/dc26-S006) prints this suggested citation:

ADA suggested citation, Standards of Care in Diabetes—2026

ADA's own suggested form:

American Diabetes Association Professional Practice Committee for Diabetes. 6. Glycemic goals, hypoglycemia, and hyperglycemic crises: Standards of Care in Diabetes—2026. Diabetes Care 2026;49(Suppl. 1):S132–S149

AMA 11 (same chapter, with DOI):

8. American Diabetes Association Professional Practice Committee for Diabetes. 6. Glycemic goals, hypoglycemia, and hyperglycemic crises: Standards of Care in Diabetes—2026. Diabetes Care. 2026;49(suppl 1):S132-S149. doi:10.2337/dc26-S006

Do not cite the retired National Guideline Clearinghouse. If you used a NICE, WHO, AHA, or IDSA guideline, name that agency and the year on the document, then add the URL you actually opened. A grey-literature report that is not a journal article still needs a recoverable title and date.

Package Inserts and Prescribing Information

A drug label is a source when you are citing the approved indication, a boxed warning, or a dosing table from the insert you opened. AMA 11 §3.13.6 treats the drug name as the title. USC's AMA 11 restatement of that section uses a print insert and an online prescribing-information PDF. Cite the manufacturer document, not a lecture slide about the drug:

AMA 11 package-insert pattern (USC restatement of §3.13.6)

Print package insert:

Lamasil. Package insert. Sandoz Pharmaceutics Corporation; 1993.

Online prescribing information (URL verified):

Lovenox. Prescribing information. sanofi-aventis U.S. LLC. Accessed April 15, 2020. https://products.sanofi.us/Lovenox/Lovenox.pdf

Systematic Reviews, Cochrane, and PRISMA

A systematic review is a journal article or a Cochrane review, not a new source type. Cite the review you opened. Do not list every included trial unless you also retrieved those trials. Cochrane's own suggested citation for the 2020 aspirin-for-perineal-pain review (PMID 32702783) still uses "Art. No." and the spelled-out journal title. AMA and NLM drop "Art. No." and use the MEDLINE abbreviation Cochrane Database Syst Rev, with the CD number in the page slot and no volume:

Verified Cochrane review (Shepherd and Grivell, 2020)

Cochrane's own suggested form:

Shepherd E, Grivell RM. Aspirin (single dose) for perineal pain in the early postpartum period. Cochrane Database of Systematic Reviews 2020, Issue 7. Art. No.: CD012129. DOI: 10.1002/14651858.CD012129.pub3.

AMA 11:

9. Shepherd E, Grivell RM. Aspirin (single dose) for perineal pain in the early postpartum period. Cochrane Database Syst Rev. 2020;(7):CD012129. doi:10.1002/14651858.CD012129.pub3

Vancouver (PubMed / NLM punctuation):

9. Shepherd E, Grivell RM. Aspirin (single dose) for perineal pain in the early postpartum period. Cochrane Database Syst Rev. 2020 Jul 24;7(7):CD012129. doi: 10.1002/14651858.CD012129.pub3.

If you are writing a systematic review, follow PRISMA 2020. The statement paper is itself a citeable BMJ article:

PRISMA 2020 statement (verified BMJ article)

Vancouver:

10. Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021 Mar 29;372:n71. doi: 10.1136/bmj.n71.

AMA 11:

10. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. doi:10.1136/bmj.n71

Randomized Trials and CONSORT

Cite the trial report, not a press release. If you are reporting a randomized trial, the current reporting guideline is CONSORT 2025, published simultaneously in several journals. The BMJ version (published 14 April 2025) is BMJ 2025;389:e081123, not volume 388. A landmark trial you actually opened—here, the Framingham six-year follow-up, confirmed in MEDLINE as PMID 13751193—is still just a journal article:

CONSORT 2025 and a verified landmark trial

CONSORT 2025 (BMJ 2025;389:e081123; PMID 40228833):

11. Hopewell S, Chan AW, Collins GS, Hróbjartsson A, Moher D, Schulz KF, et al. CONSORT 2025 statement: updated guideline for reporting randomised trials. BMJ. 2025;389:e081123. doi: 10.1136/bmj-2024-081123.

AMA 11 (same CONSORT paper):

11. Hopewell S, Chan AW, Collins GS, et al. CONSORT 2025 statement: updated guideline for reporting randomised trials. BMJ. 2025;389:e081123. doi:10.1136/bmj-2024-081123

Framingham 1961 (MEDLINE PMID 13751193):

12. Kannel WB, Dawber TR, Kagan A, Revotskie N, Stokes J. Factors of risk in the development of coronary heart disease--six year follow-up experience. The Framingham Study. Ann Intern Med. 1961;55:33-50.

Books, Chapters, and Medical Textbooks

Textbooks are acceptable for physiology, physical examination, and the textbook author's own framework. They are secondary sources for the trials they narrate. AMA 11 dropped the city of publication. Vancouver still prints Place: Publisher. NLM's personal-author book sample is Murray Medical Microbiology. Its chapter sample is Meltzer, Kallioniemi, and Trent in Vogelstein and Kinzler. AMA 11 §3.9.3 uses an edited Clostridium difficile methods book:

Official book and chapter examples

Vancouver (NLM book sample):

13. Murray PR, Rosenthal KS, Kobayashi GS, Pfaller MA. Medical microbiology. 4th ed. St. Louis: Mosby; 2002.

Vancouver (NLM chapter sample):

13. Meltzer PS, Kallioniemi A, Trent JM. Chromosome alterations in human solid tumors. In: Vogelstein B, Kinzler KW, editors. The genetic basis of human cancer. New York: McGraw-Hill; 2002. p. 93-113.

AMA 11 (official edited book):

13. Mullany P, Roberts AP, eds. Clostridium difficile: Methods and Protocols. Humana Press; 2010.

If you used Harrison's Principles of Internal Medicine, cite the edition on the copyright page you opened. Do not cite a previous-edition chapter you never held. When the byline has more than six editors, AMA prints three names plus et al. and NLM samples print six plus et al.

Clinical Trial Registrations

A registry record is a source when you are citing the protocol, the status, or a result that has not yet appeared as a journal article. NLM's sample adds a ClinicalTrials.gov number to a journal article that already has a PMID. Use the real NCT from ClinicalTrials.gov; do not invent one.

Official NLM trial-registration sample

14. Trachtenberg F, Maserejian NN, Soncini JA, Hayes C, Tavares M. Does fluoride in compomers prevent future caries in children? J Dent Res. 2009 Mar;88(3):276-9. PubMed PMID: 19329464. ClinicalTrials.gov registration number: NCT00065988.

Preprints

medRxiv, bioRxiv, and similar servers publish manuscripts before peer review. Mark the preprint status so a reader does not treat the record as a journal article. ICMJE's overlapping-publications page (and the January 2026 Recommendations PDF) now states both the label rule and the later-article rule:

Official ICMJE preprint wording

ICMJE, referencing preprints

When preprints are cited in submitted manuscripts or published articles, the citation should clearly indicate that the reference is a preprint.

ICMJE, when a later journal article exists

When a preprint article has been subsequently published in a peer-reviewed journal, authors should cite the subsequent published article rather than the preprint article whenever appropriate.

AMA 11 §3.11.4.1 says the version cited should be the version used, and gives this bioRxiv example. NLM prefers "Forthcoming" rather than "In press" and marks preprints in square brackets:

Official AMA and NLM preprint examples

AMA 11 (Bloss et al., bioRxiv):

15. Bloss CS, Wineinger NE, Peters M, et al. A prospective randomized trial examining health care utilization in individuals using multiple smartphone-enabled biosensors. Preprint. Posted online October 28, 2015. bioRxiv 029983. doi:10.1101/029983

Vancouver (NLM sample):

15. Bar DZ, Atkatsh K, Tavarez U, Erdos MR, Gruenbaum Y, Collins FS. Biotinylation by antibody recognition- A novel method for proximity labeling. BioRxiv 069187 [Preprint]. 2016 [cited 2017 Jan 12]. Available from: https://www.biorxiv.org/content/10.1101/069187v1

If the preprint later appears in a peer-reviewed journal, cite the version you actually used for a student paper that is already written from that preprint. For a manuscript you are still submitting, ICMJE wants the later journal article whenever it is appropriate. Do not silently upgrade a preprint citation to the journal article if you never opened the journal article.

Retracted Articles

If you must cite a retracted paper—for example, to discuss why the evidence changed—NLM requires the retraction note on the original citation. Do not present a retracted trial as current clinical evidence. ICMJE now names PubMed as the MEDLINE authority for retraction checks:

Official ICMJE retraction wording

Authors are responsible for checking that none of the references cite retracted articles except in the context of referring to the retraction. For articles published in journals indexed in MEDLINE, the ICMJE considers PubMed the authoritative source for information about retractions.

The official NLM pair is Feifel, Moutier, and Perry in the Journal of Clinical Psychiatry. Item 15 is the retracted article; item 14 on the NLM page is the retraction notice:

Official NLM retracted-article sample

NLM #15, article retracted:

16. Feifel D, Moutier CY, Perry W. Safety and tolerability of a rapidly escalating dose-loading regimen for risperidone. J Clin Psychiatry. 2000;61(12):909-11. Retraction in: Feifel D, Moutier CY, Perry W. J Clin Psychiatry. 2002;63(2):169.

NLM #14, article containing retraction:

16. Feifel D, Moutier CY, Perry W. Safety and tolerability of a rapidly escalating dose-loading regimen for risperidone. J Clin Psychiatry. 2002;63(2):169. Retraction of: Feifel D, Moutier CY, Perry W. J Clin Psychiatry. 2000;61(12):909-11.

Failure Modes Medical Students Actually Hit

  • Invented journal abbreviations. NEJM, J. Am. Med. Assoc., and Brit Med J are not MEDLINE abbreviations. Copy N Engl J Med, JAMA, and BMJ from the NLM Catalog.
  • The wrong et al. cutoff. AMA lists three names when there are more than six authors. Vancouver samples list six. Copying a PubMed display that now lists every author is not the manuscript rule.
  • DOI written as a URL, or a period after the DOI. AMA wants doi:10.1001/jama.2016.9797 with no trailing period. NLM samples use doi: 10.1136/bmj.a2752 with a space after the colon.
  • Citing PubMed as the source. PubMed is how you found the article. The source is the journal article.
  • A fake NCT or DOI. If you cannot find the registry number or DOI, omit it. Do not pad the reference with a plausible-looking identifier.
  • A review cited as if it were the trial. Open the primary report when you discuss that trial's methods or results.
  • A preprint cited as if it were peer-reviewed. Keep the preprint label. If a later journal article exists and you are still writing, ICMJE wants that article whenever it is appropriate.
  • A retired guideline clearinghouse. Cite the agency document you opened.
  • An online-first paper cited as if it already had final pagination. Use NLM's Epub line until print pagination exists.
  • Alphabetical numbering. Both styles number in order of first appearance.
  • Mixing APA years into a numbered list. Nursing papers that need APA belong on the nursing citation guide.

Official Pages and Recovery Tools

Use PubMed to recover authors, the official abbreviation, the PMID, and a DOI when one is indexed. Use the NLM Catalog when PubMed's abbreviation line is missing. EndNote, Zotero, and Mendeley can store those fields; they cannot invent an issue number you never captured. For one source type at a time, see how to cite a journal article in Vancouver or how to cite a journal article in AMA.

Medical citation questions

Is AMA the same as Vancouver?
No. Both number sources in order of appearance and use MEDLINE journal abbreviations, but they are not one style. AMA 11 uses superscripts and a three-author et al. cutoff. ICMJE Vancouver uses Arabic numerals in parentheses and a six-author cutoff in the NLM samples. See AMA vs Vancouver for the punctuation table.
How many authors do I list before et al. in a medical paper?
AMA 11 lists every name through six authors, then the first three plus et al. NLM's Vancouver samples list the first six plus et al. MEDLINE/PubMed now displays every author; that database display is not the manuscript rule unless the journal says it is.
Do I italicize journal titles in Vancouver?
Usually no. NLM samples set journal abbreviations in roman type. AMA 11 italicizes the abbreviated title. Follow the journal you are submitting to if it prints its own samples.
Should I include a DOI in AMA and Vancouver?
Yes when one exists. AMA 11 wants the DOI as the last element, written doi:10.… with no space and no trailing period. NLM samples write doi: 10.… with a space after the colon and may also add a PMID. Do not add an accessed date when a DOI is present.
How do I cite a clinical practice guideline?
Cite the issuing agency and the dated document you opened. The ADA 2026 Standards chapter used on this page is a Diabetes Care supplement with a DOI. Do not cite the retired National Guideline Clearinghouse or a slide deck that paraphrases the recommendation.
How do I cite a Cochrane review or a systematic review?
Cite the review as a journal article or as an article in the Cochrane Database of Systematic Reviews. Include the DOI and the CD number. Cochrane's own line for Shepherd and Grivell 2020 still says Art. No.: CD012129; AMA drops that phrase and italicizes Cochrane Database Syst Rev. Do not list every included trial unless you also retrieved those trials. If you are writing a review, follow PRISMA 2020.
How do I cite a preprint or a ClinicalTrials.gov record?
Mark a preprint as a preprint. ICMJE says that when a later journal article exists, cite that article whenever it is appropriate. For a registry record, use the real NCT from ClinicalTrials.gov. NLM's sample attaches NCT00065988 to Trachtenberg et al. in the Journal of Dental Research.
How do I cite a package insert or prescribing information?
AMA 11 §3.13.6 starts with the drug name, then Package insert or Prescribing information, the manufacturer, and the year. For an online label, add the accessed date and the URL you opened. The Lovenox PDF on this page is the USC-restated AMA example; the file was still at that Sanofi URL when this page was checked.
How do I cite an article that was retracted or later corrected?
NLM requires a Retraction in or Erratum in note on the original citation. ICMJE tells authors to check PubMed so that none of the references cite retracted articles except when discussing the retraction. Do not present Feifel 2000, or any later retracted trial, as current clinical evidence.
Should medical students use APA instead?
Only if the course or journal asks for APA. That is the default in nursing, not in most MD, DO, or biomedical-journal workflows. If your syllabus says APA, use the nursing citation guide or Vancouver vs APA rather than forcing superscript numbers into an author-date paper.

Format the record you actually captured

Lock AMA or Vancouver first. Then format the journal article, guideline, or book from a complete record. A generator cannot invent an issue number, a MEDLINE abbreviation, or an NCT you never stored.

Vancouver citation generator →AMA citation generator →