Public attention to the use of animals in research is on the rise, and with good reason. As scientists, we have a responsibility to avoid using animals in our work whenever possible. Not only does this prevent needless suffering and waste of resources, it also leads to better science, because findings from animal experiments often fail to hold up in humans. If studies can be conducted ethically with human subjects, tissues, or organs, they should not use animals.
On paper, some journals appear to clear this bar. In reality, however, they fall short of carrying out their ethical responsibility: We see many examples, especially in journals in the nutrition field, of published research that was conducted in animals but could have been carried out in humans or using human-relevant methods.
For example, a recent study fed monkeys Western- and Mediterranean-style diets to produce information about the diets’ effects on human mood and behavior. Another experiment used pigs to evaluate how diets rich in fruits and vegetables can improve human microbiome health.
This should give pause to the National Library of Medicine (NLM). When deciding if a journal merits inclusion in MEDLINE, the leading bibliographic database for life sciences, NLM may look at whether the journal’s ethical policies align with best practices and how well individual articles adhere to those policies.
Consider the journal Nutrients, which published the studies cited above. In our view, it has a particularly bad record of accepting studies with egregious violations of ethical norms when it comes to experiments on animals. Nutrients’ own guidelines require the “replacement of animals by alternatives wherever possible.” However, the journal, and others like it, continuously publishes research using animals to study human nutrition when alternatives exist – and better ones at that.
Here is one more example: The manufacturer of a saffron-based nutrition supplement explored the spice’s purported mood-boosting properties. But rather than administering the supplement to volunteers and tracking their depressive symptoms with validated instruments, as other researchers have done, the manufacturer force-fed it to mice via gavage and dropped the animals in water tanks to conduct forced swim tests as a measure of their mood.
The study was approved by an institutional ethics committee, but we believe the journal’s ethics guidelines should still have blocked the article from being accepted. Ethics-committee approval is irrelevant to the question of whether such an experiment met the journal’s explicit guidelines for the use of animals in research.
In 2022, Nutrients published more than 4,000 research papers, and just under 800 of those (roughly 20%) used live animals. After spending more than a year investigating the journal, we find that most, if not all, of these studies could have been conducted with humans or human-relevant alternatives. This would not only have spared tens of thousands of animal lives, but would also have provided clinically meaningful data rather than results of uncertain relevance.
Nutrients, which is published by MDPI and indexed for MEDLINE, has a history of editorial problems. In 2018, the journal saw 10 of its senior editors resign, including the editor-in-chief, over alleged pressure to accept “mediocre” papers. The change in leadership and the ethical failings we have identified both warrant a fresh look at the journal’s participation in MEDLINE per the database’s policy for reexamination.
The Physicians Committee for Responsible Medicine, along with the support of more than 700 physicians and health care professionals, brought these issues to the journal, publisher, MEDLINE staff, the National Library of Medicine, and the Literature Selection Technical Review Committee (the federal advisory committee that recommends journals for inclusion in MEDLINE).
Despite these efforts, insufficient action has been taken.
Time after time, Nutrients claims its editorial office and ethics committee “checked the articles themselves to confirm that they contained ethics statements that are in accordance with the journal policies.”
Similarly, MEDLINE “has independently verified that all articles you have brought to its attention include statements indicating that the studies were reviewed and approved by institutional animal care and use committees.”
But ethics-committee approval is not sufficient to meet the journal’s stated ethical guidelines. We believe journals should take their ethical guidelines seriously, just as NLM should live up to its vetting function.
The continued retort by the National Institutes of Health is that we should direct our concerns to the publisher, which we have already done repeatedly to little apparent effect.
As open-access, pay-to-publish journals become increasingly prevalent, we need to consider the financial motives of these publishers. MDPI, which publishes 420 journals, charges thousands of dollars for each published article, creating a deterrent to meticulously screen submissions for evidence of ethical misconduct.
In order to sustain research integrity, MEDLINE should make its reevaluation process transparent. Setting up a system for scientists to see why and how often journals are reevaluated and creating a platform where journals can be submitted for reevaluation could help boost confidence in the process.
As scientists, we are often left leaning on the trust of bibliographic databases to know if a publication is trustworthy. If MEDLINE fails to enforce its rules, it is not only the journals that lose their credibility; MEDLINE’s credibility suffers in the process as does that of the body of scientific literature upon which we depend.
If editors, publishers, and the NLM are not willing to take corrective action, we are left wondering whom the scientific community can depend on to uphold scientific accountability. Failing to abide by ethical standards in the literature is, in our opinion, leading to a research-integrity crisis.
