OBJECTIVE: To investigate whether there is a difference in citation rate between open access and subscription access articles in the field of radiology.
METHODS: This study included consecutive original articles published online in European Radiology. Pearson χ2, Fisher's exact, and Mann-Whitney U tests were used to assess for any differences between open access and subscription access articles. Linear regression analysis was performed to determine the association between open access publishing and citation rate, adjusted for continent of origin, subspeciality, study findings in article title, number of authors, number of references, length of the article, and number of days the article has been online. In a secondary analysis, we determined the association between open access and number of downloads and shares.
RESULTS: A total of 500 original studies, of which 86 (17.2%) were open access and 414 (82.8%) were subscription access articles, were included. Articles from Europe or North America were significantly more frequently published open access (p = 0.024 and p = 0.001), while articles with corresponding authors from Asia were significantly less frequently published open access (p < 0.001). In adjusted linear regression analysis, open access articles were significantly more frequently cited (beta coefficient = 3.588, 95% confidence interval [CI] 0.668 to 6.508, p = 0.016), downloaded (beta coefficient = 759.801, 95% CI 630.917 to 888.685, p < 0.001), and shared (beta coefficient = 0.748, 95% CI 0.124 to 1.372, p = 0.019) than subscription access articles (beta coefficient = 3.94, 95% confidence interval 1.44 to 6.44, p = 0.002).
CONCLUSION: Open access publishing is independently associated with an increased citation, download, and share rate in the field of radiology.
KEY POINTS: • A minority of articles are currently published open access in European Radiology. • European and North American authors tend to publish more open access articles than Asian authors. • Open access publishing seems to offer an independent advantage in terms of citation, download, and share rate.
Open access publishing is independently associated with an increased citation, download, and share rate in the field of radiology.
single-journal observational evidence with incompletely reported download and share analyses
Open access articles were significantly more frequently cited than subscription access articles (beta coefficient = 3.588, 95% confidence interval [CI] 0.668 to 6.508, p = 0.016).
consistent adjusted citation estimate despite residual self-selection in the observational design
Open access articles were significantly more frequently downloaded and shared than subscription access articles.
numerical estimates appear in the Abstract without corresponding Results-body reporting or measurement detail
European and North American authors tend to publish more open access articles than Asian authors.
directly reported regional proportions and significance tests within the sampled journal
Derived from the full evaluation — not a separate score.
Strengths
The paper distinguishes its article-level comparison from earlier radiology work based on journal-level impact measures, providing a narrow but identifiable addition to the literature.
↳ Introduction, final paragraph
The Methods describe consecutive inclusion, an approximately 500-article power target, and a regression model adjusting for seven article characteristics plus time online.
↳ Methods—Data Collection and Statistical Analysis
The Introduction cites a randomized null study, conflicting findings in other specialties, and a radiology journal-level null comparison rather than presenting the field as settled.
↳ Introduction, paragraph 1
Limitations
Download and share coefficients appear in the Abstract and are repeated in the Discussion and Conclusion, but the Results body provides no corresponding estimates, definitions, or analysis detail. This reporting gap limits confidence in two parts of the headline conclusion.
↳ Abstract; Results; Discussion, paragraph 2; Conclusion
The trailing beta coefficient of 3.94 in the Abstract matches neither the primary adjusted estimate of 3.588 nor the Europe/North America estimate of 4.086 reported in the Results.
↳ Abstract, Results; main Results, final paragraph
The Discussion acknowledges that authors may select higher-quality articles for open access, but the regression includes no direct quality, funding, or prestige measure and no internal sensitivity analysis.
↳ Methods—Statistical Analysis; Discussion, limitations paragraph
The article-level radiology comparison is a genuine but narrow contribution to a literature the Introduction accurately presents as mixed. The consecutive sample, power calculation, comparable follow-up, and adjusted model support the reported citation association, although the nonrandomized design leaves quality-related self-selection unresolved. The principal citation coefficient is consistent between the Abstract and Results, but the Abstract also contains an unexplained second coefficient. Download and share findings are numerically reported in the Abstract yet lack definitions and corresponding analysis output in the Results body, materially limiting the broader conclusion.
Nabu’s assessment, alongside the field’s view.
Are you an author of this paper?
Sound3.1
Confidence mediumThe study adds article-level evidence from radiology to a contested open-access literature previously represented in the specialty mainly by journal-level comparisons. Its advance remains incremental because it covers one journal and cannot resolve the broader debate.
“no such research has been performed yet for individual radiological articles”
The consecutive sample, power calculation, matched follow-up, and multivariable model support estimation of an adjusted citation association. Residual quality self-selection remains unresolved, and download/share definitions and full analysis outputs are not presented in the Results body.
“authors may have self-selected only those articles of higher than average quality”
The objective-to-conclusion structure is readily followable, and the principal citation estimate is clearly reported. Reporting is reduced by an unattributed second citation coefficient in the Abstract and by secondary findings that are not developed in the Results body.
“beta coefficient = 3.94, 95% confidence interval 1.44 to 6.44”
The paper presents the prior evidence as mixed, including a randomized null study and conflicting specialty findings. It also discusses self-selection, single-journal scope, follow-up, and article-type restrictions, although residual selection uncertainty is not resolved internally.
“there is currently no unambiguous evidence”
Caveats4 of 4 checks
The primary adjusted citation estimate is internally consistent, but an additional unexplained coefficient appears in the Abstract and the download/share analyses lack corresponding reporting in the Results body.
Ethics approval is expressly declared not applicable for this literature study, and no specific conduct concern is identified in the supplied text.
Flags: 1 declared / 5 total
15 of 15 checkable references verified
16 references in manuscript 1 are books, websites or datasets — counted, but not index-checkable citation details diverge from the cited records for 2 references
References not verified automatically are resolved manually within 24 hours.
No retraction or expression of concern found in Crossref (includes Retraction Watch).
Sources: Crossref (incl. Retraction Watch) ✓
Where this paper’s evidence sits on the path from initial observation to real-world use.
The result is an initial observational association rather than actionable guidance about whether authors should pay open-access fees or institutions should alter policy. Unresolved confounding and incomplete secondary-outcome reporting require further validation.
“a retrospective, nonrandomized observational study”
AI-generated, human-governed. Something look off? Contact us to request a review.
Keep reading
Other papers Nabu has evaluated blind, starting with the same field.
Baptiste Scancar, Jennifer Anne Byrne, David Causeur, et al. · 2026 · Media and communications
These findings suggest that paper mills inflate the citation metrics of supported publications and affected journals at scale.
TentativeChao Ma, Yiwei Li, Feng Guo, et al. · 2019 · Economics
papers published between October and December on average get as much as 18.5% fewer citations than those published in the other months in the year
TentativeMasashi Kawai, Koichi Ojiro, Makoto Asaoka, et al. · 2022 · Other social sciences
日本の機関リポジトリによるオープンアクセス論文の数は限られるものの，その被引用数と様々なタイプのオープンアクセス論文の被引用数の比較からは，日本の機関リポジトリによるオープンアクセスのメリットは確認できなかった．
Tentative