The Lancet and the New England Journal of Medicine have retracted two COVID-19 papers because of questions regarding the data used in the studies. The papers were both previously the subject of expressions of concern.
The now-retracted Lancet paper had reported that the antimalarial drugs hydroxychloroquine and chloroquine may increase the risk of death among COVID-19 patients, while the now-retracted NEJM paper noted that though cardiovascular disease increases someone's risk of dying from COVID-19, ACE inhibitors did not increase that risk.
Both studies relied on a database run by Surgisphere, which said it had detailed data on about 100,000 COVID-19 patients from 1,200 hospitals around the world, but as the New York Times noted earlier this week, clinicians and medical researchers have raised concerns about the data it houses.
The authors of the Lancet study who were not associated with Surgisphere noted in the expression of concern that they would be seeking an independent audit of the data. However, in the retraction notice, they wrote that Surgisphere would not transfer the full dataset to its independent reviewers, citing client agreements and confidentiality. Because of this, the Lancet notes in a statement that three of the four authors — the fourth author being Surgisphere chief executive Sapan Desai — said they "can no longer vouch for the veracity of the primary data sources."
The NEJM retraction notice similarly says that the authors, this time including Desai, could not "validate the primary data sources" and requested a retraction.