We are all sadly familiar with the concept of fake news, but what about fake research? An increasing amount of evidence suggests that we can not assume that research integrity is all that it should be.
An analysis of trials submitted to Anaesthesia between 2017 -20 found 14% had false data and 7% had enough false data to invalidate the whole trial (labelling them 'zombie trials'). The trials with false data were not evenly distributed with a few countries being over-represented, raising the possibility that poor research infrastructure and governance is a contributing factor.
Another concern is paper mills - sites that generate papers on demand for a fee, with no research being done at all. The editor of RSC Advances retracted 68 articles from the journal when investigation found them to be fake manuscripts from paper mills. Editors of journals such as The International Journal of Cancer have raised similar concerns and adjusted submission policies in response to the problem.
So what does this mean for the researcher? The risks of basing treatment or further research on false data are obvious. A recent blog in BMJ suggested there are so many trials include faked data that anyone doing a systematic review should assume a study is flawed until proven otherwise. But faked papers and data can be hard for the uninitiated to spot. Many of the papers in the Anaesthesia example were only identified on examining the original data. Telling clinicians to check for dodgy research is easy to say but hard to do - here are a few tips to help:
We need to talk about systematic fraud. Nature 566, 9 (2019)
The raw truth about paper mills. FEBS Lett, 595: 1751-1757.
The science institutions hiring integrity inspectors to vet their papers. Nature 575, 430-433 (2019)
Journal retractions in oncology: a bibliometric study. Future Oncology 2019 15:31, 3597-3608
Continued Citation of Retracted Radiation Oncology Literature—Do We Have a Problem? International Journal of Radiation Oncology Biology Physics. 2019