Keeping coronavirus numbers straight: JAMA sounds an alarm

As Retraction Watch readers know, reporting on the same data more than once — without notifying editors and readers — is bad for the scientific record and can lead to a retraction. Apparently, in the rush to publish findings about the coronavirus pandemic, some researchers are doing just that.

According to an editorial in JAMA today by editor in chief Howard Bauchner and two deputy editors, Robert Golub and Jody Zylke:

The editors have become aware that some of the patients described in some of these manuscripts, sometimes with overlapping authorship, have been reported in more than 1 submission. This inclusion of the same patients in more than 1 report has not been clearly indicated in the submitted manuscripts. This is of concern and may represent a lapse in ethical standards of scientific reporting.

The editors explain why that’s a particular problem for COVID-19:

Reporting of the same patients in different articles (without clear indication of the duplicate reporting) creates an inaccurate scientific record, may affect the accuracy of subsequent estimates of prevalence of the disease or outcomes, and may preclude valid meta-analyses, unless authors of the meta-analyses are able to obtain individual patient data to ensure that patients are not being counted more than once in any publication.

In other words, if you’re doing a meta-analysis — aka a “study of studies” — to figure out the fatality rate, or how a particular drug works against the virus, but some people are being counted twice, your estimates are going to be off. 

JAMA “has rejected well over 100 brief reports, research letters, and original investigations,” Bauchner told Retraction Watch. Among those manuscripts that were under serious consideration, the journal made inquiries four or five times. In two or three of those cases, “authors did admit that the patients had been enrolled in more than one study,” he said. “Hence the editorial.”

The editorial ends by urging:

all authors of reports related to COVID-19 to clearly identify if any patients in any submitted manuscript have been reported in any previous submissions or publications.

Like Retraction Watch? You can make a tax-deductible contribution to support our work, follow us on Twitter, like us on Facebook, add us to your RSS reader, or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at [email protected].