Last June, C&EN covered a paper in Science showing that polystyrene microplastics from personal care products ingested by perch could “inhibit hatching, stunt growth, and boost predation” of the fish (2016, DOI: 10.1126/science.aad8828). The paper has now been retracted.
After Oona Lönnstedt and Peter Eklöv of Uppsala University published the study, seven researchers filed a complaint with Uppsala, claiming missing data, statistical design and analysis problems, and discrepancies between how the researchers said they carried out the study and eyewitness accounts of the experiments.
A university-convened panel of researchers concluded on Aug. 31 that no evidence of research misconduct existed. At about the same time, Lönnstedt and Eklöv reported that a computer containing data from the study had been stolen and there was no backup.
The university also asked Sweden’s Central Ethical Review Board to consider the complaint. On April 21, the board decided that the authors’ responses to questions were deficient and contradictory, that they had failed to get required animal ethics approval, that the data loss suggested the research had not been conducted to the reported extent, that Science should not have accepted the paper, that the Uppsala panel appeared to have missed evidence of research misconduct, and that the authors were “guilty of scientific dishonesty.” Science retracted the paper on May 3.
Lönnstedt and Eklöv tell C&EN that they did get the required animal ethics permit and have documentation, that they responded straightforwardly to board inquiries, that the loss of data “was an unfortunate mistake and not dishonesty,” and that board deliberations were compromised by publicity about the case, including on social media, by researchers who had filed the complaint.
Chelsea M. Rochman of the University of California, Davis, an expert on fish exposure to microplastics, comments that she doesn’t believe the retraction will hurt the field. But in general, she says, “the whole situation is a shame. Unethical research activities do not help the credibility of science.”