In the wake of large-scale retraction scandals, we urge scientific publishers to be more proactive in stamping out fake peer-reviewing practices. They should work with editors, authors and research institutes to implement an effective system of precautions and penalties.

Fraudulent peer review can arise when editors rely on authors' recommended reviewers. These names are often genuine but have a false e-mail address that enables the authors to write a favourable review of their own paper.

Springer Nature, also the publisher of Nature, this year retracted 107 papers from one of its journals on the basis of fake peer review (see T. Stigbrand Tumor Biol. doi.org/b7gg; 2017). Two years ago, it retracted 64 articles in 10 of its journals on similar grounds (see Nature doi.org/b7gh; 2015).

In our view, retracting such papers is not enough. Editors need to double-check the authenticity of potential reviewers and insist that authors provide academic identities for their suggested reviewers, including institutional e-mail addresses, ORCID identifiers and Scopus Author IDs.

Journals should confidentially share each other's databases of falsified reviewers' details and of offending authors. Publishers could then reject new submissions from those authors for a set period. National academic committees and research institutes might consider revoking research funding for such authors and demoting them.