News blog

‘Misreading’ of data led to errors in statin papers

The BMJ is modifying, and is considering whether to retract, articles that questioned whether many patients should be given cholesterol-lowering statin drugs. The articles made a critical statement about the rate of side effects that were based on a “misreading” of another study, according to the journal’s editor-in-chief Fiona Godlee.

In October 2013,  John Abramson, a medical researcher at Harvard University in Cambridge, Massachusetts, and his colleagues stated in a review article that around one-fifth (18–20%) of patients on statins experienced side effects. The authors’ main thrust was to re-analyse data by the Cholesterol Treatment Trialists’ (CTT) Collaboration, and they cite evidence that there are no benefits when statins are given to people at low risk of cardiovascular disease — a controversial issue in medicine. But they also noted the 18% side-effects figure. That had come from an observational study that said statin-related events were documented for 17.4% of adults taking the drugs at Brigham and Women’s Hospital and Massachusetts General Hospital in Boston.

But in an editorial published today, Godlee says the side-effects statement was a mistake. It is being withdrawn from the article and from an opinion piece, published at the same time, that also used the figure. It was based on a “misreading” of the observational study, says the editorial (although the conclusion on the lack of benefits stands).

“The true incidence of adverse events from use of statins in people at low risk of cardiovascular disease continues to be disputed. Data compiled by the CTT Collaboration show that rates of adverse effects are similar in the active and the placebo arms in trials of statins,” writes Godlee. She adds, “This editorial aims to alert readers, the media, and the public to the withdrawal of these statements so that patients who could benefit from statins are not wrongly deterred from starting or continuing treatment because of exaggerated concerns over side effects.”

Godlee says that her journal was alerted to the error by Rory Collins, an epidemiologist at the University of Oxford, UK, and head of the team whose data Abramson re-analysed. Collins wishes both articles to be withdrawn, but the BMJ says this is not necessary, as the error was not the main thrust of the articles. The journal has convened an independent panel to look into whether the articles should be retracted.

Comments

  1. Report this comment

    Enzai Du said:

    Academic Cheating in Research

    In some fields such as biology, physics and chemistry, the deliberate fraud by researchers can be easily detected by repeating the experiment. However, in the fields such as ecology or climate change science, it is really not easy to detect the academic cheating. The experimental results can not be repeated successfully in many cases. Therefore, it easy to modify or chose datasets to sopport the conclusions but it is difficult to find out the cheating.

    It is really a challenge to consider this issue since accumulating cheating publications can be very destructive to the science.

Comments are closed.