456. This is the impressive number of scientific publications from the Marseille IHU (IHUm) containing possible ethical or legal breaches, according to a study published by a team of researchers in Research Integrity and Peer Review. This meticulous and long-term work began more than two years ago. Its beginnings had moreover been reported by L’Express, in a survey which already mentioned nearly 247 clinical trials with the same authorization number (IFR48 09-022), whereas such a number cannot – in almost -all cases-, to be used only once. In addition to these serious facts, other problems were raised, including the production of dozens of two or three-page studies of little or no scientific interest, sharing up to 75% similarities between them and published one or two days only after their reception in scientific journals… Directed by members of the IHUm.
These works, to which were added other surveys on the IHUcontributed to triggering the investigations of the National Agency for Health and Medicine and that, jointly, of the General Inspectorate of Social Affairs (IGAS) and the General Inspectorate of Education, Sport and Research (IGESR), which led to the departure of Didier Raoult and a (partial) reorganization of the IHU.
In the meantime, the researchers continued their investigations, which they compiled in this new study which pinpoints nearly twice as many problematic trials. “We explain in this article that we have analyzed 456 IHUm trials and we raise questions about the ethics of this work”, summarizes Lonni Besançon, co-author of the study, researcher in “human-computer interaction” who has , moreover, obtained the indictment of Didier Raoult for public defamation for having affirmed, in a video broadcast on the YouTube channel of the IHU Méditerranée and viewed 1.2 million times, that Lonni Besançon threatened “to send a suicide car on the IHU”.
Unauthorized studies published in friendly journals
Their work confirms that of the 456 studies they examined, 248 were conducted with the same ethics approval number (IFR48 09-022), even though the people involved in the trials, the samples and the countries where the samples were taken were conducted are different. “For example, there is a sample of breast milk from a woman in Mali, and samples taken from HIV-positive people in Marseille, so two different countries, two types of individuals and different samples, but to be sure that these samples are taken within the framework of the law and to protect the participants, one must normally ask, each time, an agreement to the ethical committees”, explained Lonni Besançon during an interview granted to France Televisions. “And since the publication of our work, a new IHU trial using the number IFR48 09-022 has been published”, adds Fabrice Franck, independent researcher and main author of the study.
This work also made it possible to show that 39 studies from the IHUm did not even contain a reference to any agreement number of an ethics committee, even though they clearly presented procedures relating to Research Involving the human person. (RIPH), which require such agreements. A closer examination of the work of the IHUm reveals that 456 studies are likely to pose problems of this type.
Finally, the authors indicate that they contacted the editors-in-chief of all the journals that published the IHUm studies with obvious shortcomings. “The number of editors contacted may seem very small compared to the number of publications we report, but many of them were published in the same journal, in particular New Microbes and New Infections, who published 135″, writes Lonni Besançon on Twitter. A situation that is all the more worrying as some of their authors are members of the journals that published them. Pierre-Edouard Fournier, the new director of the IHUm who took over from Didier Raoult, was for example a member of the editorial board of New Microbes and New Infection for years. A third of the work published by this journal is also co-signed by Prof. Didier Raoult, which is highly unusual in the world of scientific publication. Of the 85 journals contacted by the authors of the study published in Research Integrity and Peer Reviewonly 19 editorial teams responded.
The researchers therefore suggest that the rules of scientific publication evolve, in particular that the editors of scientific journals systematically require the deposit of the ethics approval number and that these are provided as metadata, so that the post-hoc analysis can be done more systematically. Finally, they propose that the submission processes be modified and that the potentially confidential deposit of ethical documents related to clinical research be required, but also that the editorial procedures take into account the international ethical framework for research. “It is urgent that publishers require approval of clinical research. To do this, they could request validation from the sponsoring organization or the authority that issued this authorization number”, adds Lonni Besançon, who calls the French authorities to take seriously its concerns, shared by a large part of the scientific community.