Renewed call to retract BMJ articles that mislead on statin effects

Exaggerated claims about side-effect rates with statin therapy are having an adverse impact on the health of patients who are at high risk of heart attacks and strokes, and are causing confusion and anxiety among the wider public.

Due to concerns about the harm caused to public health by misleading claims about the effects of statins in two papers published by the British Medical Journal (BMJ) in October 2013, a group of senior doctors and scientists asked the Committee on Publication Ethics (COPE) to take action to rectify the situation by recommending retraction of these papers.

A number of specific concerns were raised in the letter sent to COPE in October 2014 about failures of integrity in the BMJ’s editorial processes related to the two papers (see detail in left-hand column).

All of these concerns were supported by detailed supporting evidence. In January 2015, the Chair of COPE responded, but she did not address the specific issues raised about editorial integrity at the BMJ and the need to correct the scientific record in order to protect patients and the public. She was, therefore, asked again in February 2015 to consider each of the specific concerns.

Following a response from the BMJ, a summary of the issues was sent to COPE in June 2015 by the group of senior doctors and scientists. In September 2015, the Chair of COPE wrote to say that – despite having adjudicated on the matter that January – she had now decided that another member of COPE should be handling the matter to ensure that the process would be conducted independently.

By March 2016, there had still been no decision by COPE and it was drawn to the Chair’s attention that, following the BMJ’s failure to deal properly with the misleading claims that it had published in October 2013, the authors of those papers were repeating the claims in the UK media. In particular, it was noted that one of them had stated on the national BBC Radio 4 Today programme, related to the side-effect rate that had been published in the BMJ:

“That 20% figure was corrected, absolutely right, but actually – I will be honest with you now – it was actually probably an underestimate. In fact, the side effects in terms of side effects that interfere with the quality of life may well be higher.”

In response, the Chair of COPE advised that the matter was being handed to someone else because the first person who she had asked to take it on also did not feel able to do so.

Eventually, after 18 months, COPE wrote in April 2016 to state that it felt that the BMJ had taken steps to correct its mistakes. Again, however, COPE did not address any of the issues that had been raised about the failures of editorial integrity at the BMJ or about the impact on public health of the misleading claims that the BMJ had published about the rate of side-effects with statins.

Consequently, COPE was asked to have its Ombudsman adjudicate specifically on whether COPE’s Code of Conduct has been contravened by the BMJ and, if so, to decide what steps should be taken to rectify the situation. For example, by advising the BMJ that it should retract the two statin papers that it published in October 2013 (or face sanctions), COPE would publicly discredit the misleading claims and help prevent further harm to patients.

In July 2016, COPE refused to allow its Ombudsman to consider this complaint, stating that it “is a charitable membership organization and not a regulatory authority. In accordance with its charitable objects, it is only able to review complaints where concerns are identified about a journal’s process.”

Given the emerging evidence of an adverse impact on public health caused by the confusion about the safety of statin therapy that followed the BMJ’s misrepresentation of the evidence, and COPE’s failure to deal with this matter properly, it has been decided that it is now appropriate to put the material that was submitted to COPE into the public domain.

Medical journals have a duty to ensure that the scientific record is accurate and to correct it properly (i.e. fully, rapidly and prominently) when published errors have the potential to harm public health. COPE was originally set up with that as part of its remit. However, as discussed in a commentary by the editor of The Lancet, it has become clear that there is a need for a process to arbitrate in such circumstances that is independent of the journals themselves.