The Lancet Bolsters Standards for Research Based on Real-World Data

Prominent medical journal The Lancet has tightened its publishing requirements for research papers in response to an embarrassing retraction of a study based on questionable data from multiple trials. The journal said it will now require stronger data reviews for papers based on large, real-world datasets.

The changes were spurred by the journal’s retraction of a hydroxychloroquine study in June over data concerns. The trial’s authors had called for an independent review after questions were raised about their study’s data and analyses, but the data analytics company used in the trial, Surgisphere, controversially refused to provide the full dataset for evaluation. The authors requested the retraction themselves after they found they could “no longer vouch for the veracity” of their primary data sources (DID, June 5).

Under the journal’s new policies, papers that use large, real-world datasets will be required to have more than one author directly access and verify the reported data, and they must name all authors that have done so. For research articles written through an academic and commercial partnership, one of the authors named as having accessed and verified data must belong to the academic team. Additionally, all authors will have to confirm, through signature, that they had full access to the data reported in the article and must accept responsibility for their submission.

The journal has also implemented stronger peer-review requirements for papers based on large, real-world datasets, vowing to make sure that at least one peer reviewer is knowledgeable about the dataset’s details “and can understand and comment on its strengths and limitations in relation to the research question being addressed.”

Additionally, The Lancet will task its editors with ensuring that an expert in data science conducts a review, and said it “will explicitly ask reviewers if they have concerns about research integrity or publication ethics regarding the manuscript they are reviewing.”

All research papers are now required to include a data-sharing statement outlining what data will be shared, whether the study protocol will be shared, when data will become available and how the data will be shared, the journal added, noting to trial investigators that “editors will take data-sharing statements into account when making editorial decisions” from now on. — James Miessler