ORCE rented as a miracle cure for covid-19, an antimalarial called hydroxychloroquine has rarely made the headlines since the start of the pandemic. It was hoped that it might find new use as therapy in patients who are not well with the new coronavirus. But in the past few weeks, a scientific picture has emerged from a treatment that doesn’t seem to help patients at all and could even cause harm.
That it helps seems clear now: this is not the case. As far as harm is concerned, however, it turns out that the scientific literature can be misleading. On June 4, Lancet, a respected medical journal, retracted a high-profile article published just a month ago. This had suggested that hydroxychloroquine and its analogue, chloroquine, actually increased the mortality rate in hospitals when taken by people with covid-19. This has led the World Health Organization to suspend testing of the drug. It also caused considerable concern to patients and those who were enrolled in other such trials.
Towards the end of May, however, scientists began to question the reliability of the data that had been used. the New England Journal of Medicine had withdrawn a separate document on hypertension medication in covid-19 which was based on data from the same company, Surgisphere, which provided the data set for the Lancet article.
Chicago-based Surgisphere said in the Lancet paper that 671 hospitals on six continents had provided data. The dataset is said to contain nearly 100,000 detailed patient records. On June 2, Lancet says an independent data audit is underway and writes that “serious scientific issues” have been brought to his attention. The editor of New England Journal of Medicine expressed a similar concern and said the authors were asked to provide evidence that their data was reliable.
The Economist could not contact Surgisphere for a comment. All content has been removed from its website. However, the firm’s website said its mission was to harness the power of data analytics to “improve the lives of as many people as possible.” The company said it used machine learning, artificial intelligence and big data to help hospitals make better decisions. Science, a leading academic journal, contacted the newspaper’s authors for comment on June 8, but they did not respond.
Many wonder, more broadly, what could go wrong to the point of causing two retractions of articles in well-known medical journals. There are calls to the Lancet publish comments received on the article during the peer review process.
Ironically, concerns about the rush to publish science during the pandemic have focused on pre-publications. These documents are published online without independent review. However, two reputable peer-reviewed journals found themselves in difficulty. It remains to be seen whether this incident will change the balance of power in scientific publishing.
Editor’s Note: Part of our covid-19 coverage is free for readers of The economist today, our daily newsletter. For more stories and our pandemic tracker, check out our coronavirus hub
This article appeared in the Science & Technology section of the print edition under the title “Testing times”
Reuse this contentThe Trust project