
Science publishers take too long to retract or correct research papers, a report by the US National Academies of Sciences, Engineering and Medicine (NASEM) finds. Standards for fixing errors are also inconsistently applied across research journals, it says.

Why retractions data could be a powerful tool for cleaning up science
There are no data on how long retraction or correction currently takes, as journals do not generally disclose when they were alerted to a problem in a paper; the NASEM report says they should. On average, according to the Retraction Watch database, retractions occur around 2.4 years after a paper is published. The report recommends that editorial review processes be completed within 12 weeks for most cases and should rarely stretch beyond 6 months — at least when editors are looking at concerns over a single paper.
“The big picture is, the process is too slow,” says Randy Schekman, a cell biologist at the University of California, Berkeley, who chaired the 14-person committee that produced the NASEM report. A prompter and more transparent process would strengthen the reliability of the scientific record, he says.
The report was due to be published online on 19 November, but publication was delayed after journalists at Science noted that it contained some references with fabricated first names for authors, a possible sign that large language models had been used to write or edit the material and had been insufficiently checked.
“It's unfortunate that there were these flawed references in a report about fixing the literature,” says Jennifer Byrne, a cancer researcher and research integrity sleuth at the University of Sydney who first flagged the concern to Science when reviewing the report, which was shared ahead of publication with media outlets.
Fixing the literature
The report had two main aims: to speed up the process of retraction or correction and to lower the “tremendous emotional barriers” for authors to correct honest mistakes in the record, Schekman says. “Part of the problem is that it’s too rare for a paper to be retracted,” he adds.

Retractions of old papers are on the rise: when should bad science be culled?
The report also suggests that journals shouldn’t just wait for complaints to reach them, but instead “take proactive steps” to identify errors and integrity issues. That might include journal-wide audits, or regularly monitoring social media and online platforms where post-publication issues are raised, it says.
Existing guidelines for publishers, such as those from the Committee on Publication Ethics (COPE) in London, only urge a “timely” response to concerns without setting out expected schedules, and don’t suggest journals scan online platforms for complaints.
“A major theme running through the report is that we have to distinguish between retraction or correction, and research misconduct,” notes Dorothy Bishop, a research-integrity sleuth in Oxford, UK, who was on the NASEM committee. Authors shouldn’t associate retracting a paper with the stigma of misconduct, she says. And although journal editors sometimes wait for a university to investigate whether there is research misconduct before correcting a paper, they shouldn’t, she adds.

More than 10,000 research papers were retracted in 2023 — a new record
Calls for changing these attitudes aren’t new: journals have long emphasized that retractions can cover honest mistakes. When it comes to journal processes, Holden Thorpe, the editor in chief of Science, urged that issuing corrections or retractions and making a determination of misconduct be separated in an editorial in 2022, and COPE guidelines already emphasize the point. But “there are still plenty of people who see a correction or retraction as an allegation of fraud,” says Thorpe, who was also on the NASEM committee. “Hopefully, this report will accelerate the recognition of the distinction,” he says.
Enjoying our latest content?
Log in or create an account to continue
- Access the most recent journalism from Nature's award-winning team
- Explore the latest features & opinion covering groundbreaking research