Chain retraction: how to stop bad science propagating through the literature

• August 29, 2024

The retraction of flawed research papers, while essential for scientific integrity, is not enough to prevent the spread of bad science. Retractions have skyrocketed, reaching nearly 14,000 in 2022, yet faulty studies still influence the literature. A key issue is that papers citing retracted work often remain unaddressed, allowing erroneous data to propagate. Despite the growing problem, there is no systematic method for alerting authors when studies they cited are retracted.

To combat this, we need large-scale mechanisms to halt the spread of errors. The academic system, driven by publications and citations, often overlooks dubious work, and unscrupulous practices like paper mills exploit this system. Even reputable publishers may inadvertently contribute by prioritizing volume over quality.

Scientists, publishers, and institutions must work together to maintain the integrity of research. Tools like PubPeer and the Problematic Paper Screener (PPS) can flag suspicious studies, while RetractoBot and the Feet of Clay Detector can identify retracted references. However, these efforts need widespread adoption and integration into editorial processes.

Publishers should take a proactive role by routinely screening manuscripts for plagiarism, doctored data, and unreliable references. They must also improve retraction practices, ensuring retracted articles are clearly marked and regularly checking for overlooked retractions. Additionally, the scientific community should value contributions to correcting the literature, and funders and universities should prioritize solid science over sheer publication numbers.

Ultimately, stopping bad science from spreading requires accountability, preventive measures, and collaboration from all involved to uphold the credibility of scientific research.

Nature 632, 977-979 (2024)

doi: https://doi.org/10.1038/d41586-024-02747-1

Original article via Nature By Guillaume Cabanac

Illustration: Phil Wheeler