Read our COVID-19 research and news.

MirageC/Getty Images, edited by E. Petersen/Science

What is research misconduct? European countries can’t agree

In Sweden, a national code takes 44,000 words to define research misconduct and discuss scientific values. Next door, Norway’s equivalent is a brisk 900 words, little more than in this news article. And it’s not just the size of the codes that differs across Europe: A new analysis of scientific integrity policies in 32 nations has found widely varying standards and definitions for research misconduct itself, despite a 2017 Europe-wide code of conduct intended to align them.

Research ethicists say the differences threaten to create confusion and disputes for international scientific collaborations. Teams often include members working in different countries; if a team member is accused of research misconduct, which country’s rules should apply? The decision affects who can be held responsible, and which behaviors are considered unethical. “It really is a difficult issue,” says Nicole Föger, managing director of the Austrian Agency for Research Integrity.

The mismatched standards have already led to practical problems, Föger says. She cites a case of an Austrian postdoctoral researcher who applied Austrian ethical standards while working at a university in another European country. The Austrian standards—mandated by the postdoc’s Austrian research funding contract—forbid “honorary authorship” for researchers who did not contribute substantially to a paper. But after leaving a senior researcher at this university off a paper because of a lack of contribution, the postdoc faced a university investigation and was found to have been in the wrong.

The 2017 European Code of Conduct for Research Integrity, developed by the European Federation of Academies of Sciences and Humanities, was designed to be easy for European countries to adopt, offering a nonbinding framework that they could add to as needed to fit their circumstances. It updated an unwieldy 2011 code and was more concise, Föger says. The 2017 version encourages core principles, including honesty, respect, and accountability, describes good research practices, and gives specific examples of misconduct.

But pickup of the European framework has been spotty, according to a study by Hugh Desmond, a philosopher of science at the University of Antwerp, and KU Leuven bioethicist Kris Dierickx. Of 32 countries in the analysis, only two—Bulgaria and Luxembourg—have adopted the European Code wholesale, the authors reported last month in Bioethics. There’s just one policy that all countries have agreed on: that fabrication, falsification, and plagiarism of data and findings constitute research misconduct.

Beyond that, the national policies stray significantly from the European model. “If they seek to rephrase things, that’s already significant in itself,” Desmond says—a signal that the authors of the document intended something different from the Europe-wide code. Many do not address behaviors, besides fabrication, falsification, and plagiarism, that the European code defines as misconduct, such as financial conflicts of interest, manipulating authorship, and self-plagiarism. Some countries say misconduct requires an intention to deceive; others define it as any violation of the code, even negligent ones. Some nations hold all co-authors jointly responsible for fraudulent work, whereas others don’t specify who is responsible.

But the study’s method overestimates differences between countries, says Daniele Fanelli, a scientific misconduct researcher at the London School of Economics. Just because wordings differ slightly from the European code doesn’t mean they don’t endorse the same underlying principles, Fanelli says. Another limitation of the study: Many countries—including Austria—have not yet updated their policies in response to the 2017 code.

The findings echo an ongoing debate in the United States about how research misconduct should be defined, says Lisa Rasmussen, a bioethicist at the University of North Carolina, Charlotte. In 2000, the U.S government defined misconduct as fabrication, falsification, and plagiarism, but some researchers have argued since that other behaviors—such as sexual harassment—should be included.

The practical problems raised by a lack of consensus aren’t limited to Europe, says David Resnik, a bioethicist at the U.S. National Institutes of Health. The potential for serious complications with international collaborations is a worry lurking in the background that “really hasn’t gotten the attention that it deserves,” he says.

Global alignment of standards and policies would likely be even more difficult than it has proved to be in Europe, Desmond says. And with calls for harsher penalties and even criminalization for misconduct, he worries the complications caused by policy mismatch may become “a much more pressing problem.”