A Post-Publication Notice in a High-Profile Screening Trial
Nature Medicine has published an Author Correction to the SCREESCO randomized controlled trial, a study whose title frames a direct comparison between colonoscopy and fecal immunochemical testing against usual care in diagnostic colorectal cancer screening. The notice carries the DOI 10.1038/s41591-026-04702-1 and was published online on 10 September 2026, with the record appearing in the journal's feed the following day.
The correction is attributed to the study's authors rather than to the journal's editorial office, a distinction that matters for how readers should interpret it. It is filed under the heading "Author Correction," the label Nature Medicine applies when the people who wrote a paper are the ones initiating a fix to their own published record.
What exactly was amended — a figure, a table, a statistical value, an author affiliation, or a line of text — is not described in the notice's title or the publication metadata available at the time of writing. The full text of the correction was not retrievable when this report was compiled, so the specific substance of the change remains, for now, unconfirmed from the source itself. What can be stated with confidence is narrower but still meaningful: a correction has been issued, and it is attached to a randomized controlled trial of colorectal cancer screening.
What an Author Correction Actually Means
Scientific publishing operates on the assumption that the record is self-correcting. Corrections are one of the mechanisms that make that assumption more than a slogan. They exist because research is a human enterprise conducted by human beings, and because a paper's published version is meant to be the definitive account of what was done and what was found.
Author Corrections sit on the milder end of the spectrum of post-publication notices. They generally signal that an error was identified, acknowledged and fixed, without any suggestion that the underlying findings should be withdrawn or that misconduct occurred. They are distinct from retractions, which remove a paper's claims from the literature, and from editorial expressions of concern, which flag unresolved questions while an investigation proceeds.
That distinction is not semantic. For clinicians, systematic reviewers and guideline panels who cite trial evidence, knowing which category a notice falls into determines whether the study can still be relied upon, and in what form. A corrected table is a paper that has been improved. A retracted trial is a paper that must be set aside.
The Study Beneath the Correction
SCREESCO is a randomized controlled trial, and the comparison its title describes sits at the center of a long-running debate in cancer screening: how should health systems look for colorectal cancer in people without symptoms?
Colonoscopy is the more definitive of the two approaches named in the study. It allows a physician to visualize the entire colon and to remove precancerous growths during the same procedure, which is why it functions as both a diagnostic and a preventive intervention. The trade-offs are well known in clinical practice: it is invasive, it requires bowel preparation, it carries small but real procedural risks, and it depends on a supply of trained endoscopists and equipped facilities.
Fecal immunochemical testing takes the opposite approach. It is non-invasive, can be completed at home, and requires no sedation or specialist time to administer. Its limitation is that it detects the presence of blood in a stool sample rather than the lesion itself, which means abnormal results must be followed by colonoscopy, and its sensitivity for detecting advanced disease is not equivalent to direct visualization.
Usual care, the third arm implied by the trial's title, provides the baseline against which organized screening programs are measured. Randomized trials of this design exist because observational data cannot answer the central question: does offering a screening test to a population actually reduce deaths from the disease, compared with what those populations already receive?
Why Corrections Matter More in Randomized Trials
Randomized controlled trials occupy a privileged position in medical evidence. They are expensive, slow and difficult to run, and they are the design most capable of supporting causal claims about an intervention. Regulatory decisions, screening recommendations and clinical guidelines lean heavily on them.
That influence is precisely why transparency around their published record carries weight. A correction to a large trial is not a minor administrative event, even though the correction itself may be small. Trials are cited for years, sometimes decades, and downstream documents — meta-analyses, cost-effectiveness models, national screening policies — frequently quote numbers that originated in a single published table.
The research community has spent the past decade paying closer attention to how often corrections, retractions and data discrepancies appear in the biomedical literature, and how slowly they propagate. The consistent finding is that notices are issued far more often than the citations that would alert readers to them. A correction published in a journal does not automatically update the thousands of PDFs and reference managers holding the earlier version.
What Readers Should Take From This Notice
For anyone following colorectal cancer screening research, the practical takeaways are limited but clear:
- The trial's existence is unaffected. A correction notice does not erase a study. The randomized comparison between colonoscopy, fecal immunochemical testing and usual care remains part of the literature.
- The corrected version supersedes the original. Readers citing the SCREESCO trial should consult the notice alongside the paper and cite the version of record as amended.
- The scope of the change is unknown from the metadata alone. Until the notice's full text is accessible, readers cannot assess whether the correction touches a minor detail or a substantive element of the analysis.
- Corrections are a sign of a functioning record. The alternative is not fewer errors; it is errors that go uncorrected because no mechanism exists to surface them.
What to Watch Next
The most useful next step is straightforward: read the correction in full. Author Corrections typically state the original passage, the corrected passage and, in many cases, an explanation of how the error arose. That level of detail determines whether the change is cosmetic or consequential.
It is also worth watching whether the correction is reflected in the trial's registration record, in any linked supplementary material, and in the reference lists of reviews that have already cited the study. Journals increasingly push corrections into indexing databases and reference managers, but the lag between publication and propagation remains a persistent weakness in how the scientific record stays accurate.
For now, the SCREESCO correction stands as a reminder that even rigorous, randomized, peer-reviewed evidence is a living document — one that gets edited in public, and that asks its readers to keep track of the edits.
This article is based on reporting by Nature Medicine. Read the original article.
Originally published on nature.com
