Phony peer review: The more we look, the more we find

An unknown number of published studies have a hidden flaw: The “peers” who supposedly vouched for their publication are phonies.

And the closer publishers look, it seems, the more rotten studies they find.

Now, in the biggest haul to date, publisher Springer has announced that it’s retracting 107 papers from a single journal over concerns that authors had cheated the peer review process — some perhaps unintentionally. (The journal, Tumor Biology, was at the end of last year sold to publisher SAGE.) That amounts to a remarkable 2 percent of all the papers the journal published over the years in question, 2010 to 2016.

It’s not the first time that such fraud has been uncovered. But observers had hoped the problem had been nipped in the bud in 2014 after a group of publishers patched some vulnerabilities in the systems they use to review manuscripts.

Now it seems clear that isn’t the case. And, following what it called a “deeper manual investigation” prompted by the earlier retractions, Springer said it likely will turn up more studies needing retraction in its other journals.

The latest debacle went like this: The journal, using a not-unusual practice, allowed authors to submit the names and email addresses of potential reviewers for their manuscripts.

The names were real, but the emails were bogus, with some apparently controlled by companies paid by the authors to help them get their papers published. One such author told a news outlet in China that his paper had originally been rejected, but that when he hired one of the companies to help, the paper was accepted. Now he’s worried that the retraction will stain his CV.

We’ve cataloged more than 450 retractions of papers for fraudulent peer review over the years. In December 2014, following our coverage in Nature of fake peer review, the Committee on Publication Ethics — a UK-based group that promotes best practices in scholarly publishing — issued a statement decrying “systematic, inappropriate attempts to manipulate the peer review processes of several journals across different publishers.”

One of its member publishers? Springer.

Despite that, 32 articles in the latest batch show submission dates in 2015 — well after the journal should have been aware of the vulnerabilities in its systems. That suggests some unscrupulous players are using techniques we didn’t know about earlier. SAGE, the new publisher, said it has put a number of measures in place — including an overhaul of the journal’s peer review processes and new policies on the use of author-suggested reviewers — to ensure this kind of breach doesn’t happen again.

Springer is taking steps to ensure the same thing. In addition to cleaning its own house, the company is sharing its screening processes with other publishers. Through COPE and another organization, the company has “shared anonymised information about our screening processes which could assist [other publishers] in identifying similarly problematic papers,” says Tamara Welschot, Springer’s director of research integrity and publishing services.

The problem, however, is that by the time publishers round up all the usual suspects and find every case of faked peer review — assuming that’s possible — the clever people who came up with the scam will probably have moved onto another.

Plumbers, in their line of work, are fond of saying that “water finds a way.” Publishers need to stay committed to patching their own leaks.