2 in 5 citations still cite a retracted paper as support, even when the citing paper's own reference list marks it “retracted”.
That is 40.4% of 3,628 such citations, about the same as for all citations (40.7%). Only 12.9% flag it as retracted or unreliable in the text.
Cited it as support
Only 12.9% of the marked citations flagged it as retracted or unreliable in the text.
Source: SaneCite Phase 2 census: 51,626 citations from open-access papers citing retracted biomedical papers (PubMed-indexed, retracted 1990–2023). Dot = estimate; line = 95% CI. Provisional: read by three AI models; a human check comes next. Data as of 14 Oct 2026.
Source: SaneCite Phase 2 census, 51,626 citations, data as of 14 Oct 2026 · sanecite.com/charts
Show the numbers
| Group | Citations | Cited it as support (95% CI, %) | Flagged it (95% CI, %) |
|---|---|---|---|
| Reference list says “retracted” | 3,628 | 40.4% (37.4–43.5) | 12.9% (7.6–18.0) |
| All citations | 51,626 | 40.7% (39.9–41.5) | 3.3% (2.6–4.3) |
“Reference list says retracted” means the word “retract” appears in the citing paper's own reference entry for the work. Most entries carry a clear mark: a title prefix such as “RETRACTED:” (2,664) or a retraction note (523). The other 441 mention it in other formats. On the 3,187 clear marks alone, 41.5% (1,322) cite the work as support. The mark may come from the authors' reference software or from the journal. 95% confidence intervals (the likely range) come from re-running the numbers 400 times, drawing whole retracted papers at random (a cluster bootstrap).
Free to share, with credit.
Download the image, paste the caption, or link to this page. Please credit SaneCite (sanecite.com). The numbers come from the open census data; nothing here is picked by hand.
- Provisional.Three AI models read every citing passage. On a blind sample of 150 citations they agreed with a separate AI reader on whether each cited the paper as support in 142 (94.7%); a check by two human labellers comes next, and the figures may move. The rates are not corrected for reader error: corrected using that one AI reader's checks, the 40.7% headline estimate would be about 38.1% (95% CI 33.5–41.8%).
- What a citation is here.One later paper citing one retracted paper, in a year after the retraction. Citations to paperwork retractions, which find no fault in the science, are reported apart.
- What it covers.Open-access papers in Europe PMC citing PubMed-indexed papers retracted from 1990 to 2023. About 43% of all later citations could be read.
- Check it yourself.This chart's data · every citation, searchable · method and limits · code.