Cited by 63Academic Writing and PublishingBiomedical Text Mining and OntologiesMeta-analysis and systematic reviews

Accuracy of References in Five Biomedical Informatics Journals

Open access
Aronsky, D.·2004·Journal of the American Medical Informatics Association

Abstract

OBJECTIVE: To determine the rate and type of errors in biomedical informatics journal article references. METHODS: References in articles from the first 2004 issues of five biomedical informatics journals, Journal of the American Medical Informatics Association, Journal of Biomedical Informatics, International Journal of Medical Informatics, Methods of Information in Medicine, and Artificial Intelligence in Medicine were compared with MEDLINE for journal, authors, title, year, volume, and page number accuracy. If discrepancies were identified, the reference was compared with the original publication. Two reviewers independently evaluated each reference. RESULTS: The five journal issues contained 37 articles. Among the 656 eligible references, 225 (34.3%) included at least one error. Among the 225 references, 311 errors were identified. One or more errors were found in the bibliography of 31 (84%) of the 37 articles. The reference error rates by journal ranged from 22.1% to 40.7%. Most errors (39.0%) occurred in the author element, followed by the journal (31.2%), title (17.7%), page (7.4%), year (3.5%), and volume (1.3%) information. CONCLUSION: The study identified a considerable error rate in the references of five biomedical informatics journals. Authors are responsible for the accuracy of references and should more carefully check them, possibly using informatics-based assistance.

Integrity signals

Registry

Crossref · 10.1197/jamia.m1683

Cites

32 of 32 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 10 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    An error in the first author's name was found 39 times (eight errors in the last name and 31 in the first or middle initial).

    “An error in the first author's name was found 39 times (eight errors in the last name and 31 in the first or middle initial).”
    Quote in the paperNot a substantial claim
  • Findingown data

    Inaccurate journal titles accounted for 97 (31.2%) errors.

    “Inaccurate journal titles accounted for 97 (31.2%) errors.”
    Quote in the paperNot a substantial claim
  • Findingown data

    In 18 references, an inaccurate page listing accounted for 23 (7.4%) errors (four in the first page, nine in the last page, and five in both pages).

    “In 18 references, an inaccurate page listing accounted for 23 (7.4%) errors (four in the first page, nine in the last page, and five in both pages).”
    Quote in the paperNot a substantial claim
  • Findingown data

    At the time of this study in summer 2004, a MEDLINE query for ''Int J Med Inform [jour]'' returned no results.

    “For example, at the time of this study in summer 2004, a MEDLINE query for ''Int J Med Inform [jour]'' (International Journal of Medical Informatics) returned no results.”
    Quote in the paperNot a substantial claim
  • Findingown data

    Such an error is present in reference 20 of the present article, in which MEDLINE lists the third author's name as ''Kee WD,'' whereas the original article lists the correct name as ''Ngan Kee WD''.

    “Such an error is present in reference 20 of the present article, in which MEDLINE lists the third author's name as ''Kee WD,'' whereas the original article lists the correct name of the author as ''Ngan Kee WD'' (same author as reference 26).”
    Quote in the paperNot a substantial claim
  • May combine two claims. 12
  • Classified as probably not a substantial claim. 18
  • May combine two claims.

    Findingown data

    The author element was the most frequent error category and contained 121 (39.0%) errors in 99 references.

    “The author element was the most frequent error category and contained 121 (39.0%) errors in 99 references.”
    Quote in the paperSubstantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    At that time, a look up in MEDLINE's journal database returned the applicable MEDLINE abbreviation Int J Med Inf.

    “At that time, a look up in MEDLINE's journal database returned the applicable MEDLINE abbreviation Int J Med Inf.”
    Quote in the paperNot a substantial claim
  • May combine two claims.

    Findingown data

    A single error occurred in 170 (75.6%) and multiple errors in 55 (24.4%) references (two errors in 51 and three errors in four references).

    “A single error occurred in 170 (75.6%) and multiple errors in 55 (24.4%) references (two errors in 51 and three errors in four references).”
    Quote in the paperNot a substantial claim
  • May combine two claims.

    Findingown data

    The title element included 55 (17.7%) errors, the year element 11 (3.5%), and the volume element four (1.3%).

    “The title element included 55 (17.7%) errors, the year element 11 (3.5%), and the volume element four (1.3%).”
    Quote in the paperNot a substantial claim
  • May combine two claims.

    Findingown data

    The five journal issues included 37 publications with 1,356 references, of which 656 (48.4%) met the inclusion criteria.

    “The five journal issues included 37 publications with 1,356 references, of which 656 (48.4%) met the inclusion criteria.”
    Quote in the paperSubstantial claim
  • May combine two claims.

    Findingown data

    50 punctuation errors in any part of the 656 study references and one duplicate reference were found; neither error type was included in the overall error rate.

    “We found 50 punctuation errors in any part of the 656 study references and one duplicate reference; neither error type was included in the overall error rate.”
    Quote in the paperNot a substantial claim
  • May combine two claims.

    Findingown data

    An incorrect journal abbreviation was listed 93 times, and an incorrect journal name appeared four times.

    “An incorrect journal abbreviation was listed 93 times, and an incorrect journal name appeared four times.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Most errors affect the names of authors, followed by errors in the title element of a reference.

    “Most errors affect the names of authors, followed by errors in the title element of a reference.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites 5,6,[20][21][22][23]25

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Premisenone given

    Previous investigators have reported on reference accuracy in various general and specialist biomedical journals and found error rates ranging from 3% to 60%.

    “Previous investigators have reported on reference accuracy in various general and specialist biomedical journals and found error rates ranging from 3% to 60%.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

    Interpretationreasoning

    For readers who are unfamiliar with the biomedical informatics field, it is more difficult to identify the correct journal.

    “For readers who are unfamiliar with the biomedical informatics field, it is more difficult to identify the correct journal.”
    Quote in the paperNot a substantial claim
  • Classified as probably not a substantial claim.

    Premisecitation

    The Uniform Requirements for Manuscripts Submitted to Biomedical Journals recommends an abbreviated list of authors using et al. after the sixth author.

    “The ''Uniform Requirements for Manuscripts Submitted to Biomedical Journals'' recommends an abbreviated list of authors using et al. after the sixth author.”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [1] 1. Arrowsmith JE. The accuracy of references in journal articles. Anaesthesia. 1996;51:517-8.

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [29]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).

    Interpretationnone given

    Errors in the author's name may only demonstrate a lack of attention to detail.

    “Errors in the author's name may only demonstrate a lack of attention to detail.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Interpretationreasoning

    Unsuccessful linking increases the visibility of errors to the readers.

    “Furthermore, unsuccessful linking increases the visibility of errors to the readers.”
    Quote in the paperNot a substantial claim
  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation

    The names of authors are critical for retrieving articles and are relevant when author citations are used to measure research productivity.

    “However, the names of authors are not only critical for retrieving articles but are also relevant when author citations are used to measure research productivity.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [30]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Classified as probably not a substantial claim. May combine two claims.

    Interpretationreasoning

    Misspellings and variations in an author's name may prevent the program from retrieving all relevant citations and influence bibliometric measures and information retrieval approaches.

    “Misspellings and variations in an author's name may prevent the program from retrieving all relevant citations and influence bibliometric measures and information retrieval approaches.”
    Quote in the paperSubstantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Interpretationnone given

    These error rates are within the range of reference accuracy rates found in other biomedical journals.

    “These error rates are within the range of reference accuracy rates found in other biomedical journals.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductnone given

    Our study examined only one issue of each journal, and it is possible that other issues contain fewer errors in references.

    “Our study examined only one issue of each journal, and it is possible that other issues contain fewer errors in references.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    Disagreements among the two reviewers were resolved by consensus involving the three participating reviewers.

    “Disagreements among the two reviewers were resolved by consensus involving the three participating reviewers.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The Science Citation Index from Thomson ISI allows users to examine the citation frequency of an author's publications.

    “For example, the Science Citation IndexÒ from Thomson ISI allows users to examine the citation frequency of an author's publications.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [31]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Classified as probably not a substantial claim.

    Study design and conductnone given

    Each included reference was independently checked by two reviewers.

    “Each included reference was independently checked by two reviewers.”
    Quote in the paperSubstantial claim
  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductreasoning

    This study included only references to journal articles that were indexed in MEDLINE; error rates for excluded references may differ from our results.

    “This study included only references to journal articles that were indexed in MEDLINE; error rates for excluded references may differ from our results and need to be examined in a separate study.”
    Quote in the paperSubstantial claim
  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductnone given

    Punctuation errors and duplicate references were not included in the calculation of the overall error rates.

    “Punctuation errors and duplicate references were not included in the calculation of the overall error rates.”
    Quote in the paperSubstantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    Our study did not examine MEDLINE's reference error rate.

    “In addition, our study did not examine MEDLINE's reference error rate”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    MEDLINE's reference error rate has been reported to be 1.5% in a study analyzing publications from 1976 to 1998.

    “In addition, our study did not examine MEDLINE's reference error rate, which has been reported to be 1.5% in a study analyzing publications from 1976 to 1998.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [32]

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

24 of 24
  • The paper relies on this without support.

    The reference accuracy in biomedical informatics journals has not been previously examined.

    “The reference accuracy in biomedical informatics journals, the aim of this study, has not been previously examined.”
1 of 1

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • It is well established that reference inaccuracies are highly prevalent in the peer-reviewed literature, ranging from 4% to 48% of citations.

    “It is well established that reference inaccuracies are highly prevalent in the peer-reviewed literature, ranging from 4% to 48% of citations [8-11].”

    From High Rates of Fabricated and Inaccurate References in ChatGPT-Generated Medical Content (2023) · cites this paper as [8]

    The Introduction reports substantial reference-error rates in biomedical journals, and the Results find errors in 34.3% of the references examined. But the paper gives prior rates of 3%–60% or 4.1%–40.3%, not the claimed 4%–48%; its 48.4% figure is the proportion of references meeting the study’s inclusion criteria, not an error rate. Page numbers are unavailable in our copy. (LLM, conf=0.88)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Included references exceed total references in two journals

    Minor

    Table 1 lists 181 included references against 141 total references for the first journal, and 175 against 130 for the second. An included subset cannot exceed its corresponding total.

  • The cited reference 20 has no third author

    Minor

    The Discussion describes reference 20 as having a third author whose name is Ngan Kee WD, but the bibliography lists reference 20 as a single-author paper by Siebers.

  • Reference 26 does not list Ngan Kee WD

    Minor

    The Discussion identifies Ngan Kee WD as an author of reference 26, but that bibliography entry lists Hecht F, Hecht BK, and Sandberg AA.

  • Error multiplicities account for fewer than 311 errors

    Minor

    The stated breakdown gives 170 single errors, 51 references with two errors, and four with three errors: 284 errors altogether, rather than 311.

    Reported 311 · recomputed 284

4 of 4

Cites

Flagged first, then by citations

2 of the 34 references couldn’t be matched to a known paper.

Cited by

Sorted by citations

Authors
Aronsky, D.
Affiliations
Meharry Medical College (US)
Corresponding author
Aronsky, D.
Venue
Journal of the American Medical Informatics Association
Published
2004-11-23
Volume
12
Issue
2
Pages
225-228
ISSN
  • 1067-5027 (ISSN-L)
  • 1527-974X
OpenAlex
W2171590415
Cited by
63
References
34
Citations by year
2026: 2 · 2025: 1 · 2023: 3 · 2022: 1 · 2021: 3 · 2020: 1 · 2019: 1 · 2018: 2 · 2017: 3 · 2016: 6 · 2015: 2 · 2014: 3 · 2013: 5 · 2012: 4
FWCI
9.78
Type
journal article
Language
en
Open access
Locations
  • Journal of the American Medical Informatics Association (publishedVersion)
  • PubMed (publishedVersion)
  • PubMed Central (submittedVersion)
Topics
  • Social Sciences › Arts and Humanities › History and Philosophy of Science › Academic Writing and Publishing
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Biomedical Text Mining and Ontologies
  • Social Sciences › Decision Sciences › Statistics, Probability and Uncertainty › Meta-analysis and systematic reviews
Keywords
citation errors, MEDLINE, bibliographic references, author name errors
MeSH terms
Bibliographies as Topic, Medical Informatics, Periodicals as Topic, Bibliometrics

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-11.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-11.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review