Jane Armitage and colleagues, led by senior author Rory Collins, (Nov 5, p 2237)
1 and Richard Horton (Nov 5, p 2237)
2 appear to believe that retraction of an article from
The BMJ will end the debate about statins and primary prevention. Even were there grounds for retraction, I fear they would be disappointed. Questions about the evidence base for statins continue to emerge from many quarters: how strong is the evidence, how large is the benefit for individuals at lowest risk of heart disease, how well did the trials record common minor side-effects, how representative were the trials of women and the elderly, what was the effect of active run-in periods and composite endpoints, how does taking a statin affect a person's diet and exercise patterns, why is there a discrepancy between the real-life experience of muscle pain and what was reported in the trials, why have the data for harms not yet been given the same levels of scrutiny as the data for benefits, and is cholesterol a reliable surrogate endpoint to guide prevention of cardiovascular disease?
3, 4
So despite Horton and Collins and colleagues wanting to shut down the discussion and award themselves the final word, the debate about statins in primary prevention is alive and kicking. It is a debate that needs to be resolved as thoughtfully, objectively, and openly as possible, and not by eminence-based narrative reviews,
5 however extensive, based on meta-analysis of data that only Collins, his fellow trialists, and industry sponsors have seen. This absence of independence and transparency is not unusual in medicine—indeed it is sadly still very much the norm. Why then should statins be singled out in calls for independent review of the raw data?
6 Because they are already the most widely prescribed class of drug in high-income and middle-income countries, and because proposals for even wider prescription have sparked justifiable controversy. As the reassessment of the evidence on oseltamivir showed, when unabridged internal clinical study reports containing anonymised patient-level data are scrutinised by fresh expert eyes, new information and uncertainties emerge to guide decisions and inform future research.
7 Collins and Horton would do most to resolve this damaging controversy by adding their voices to calls for independent review of the data for statins.
Their attacks on Virginia Barbour and the Committee on Publication Ethics (COPE) are misinformed and based on inaccurate and selective reporting. Barbour's delay in recusing herself from handling a complaint against
The BMJ was in part due to a change in personnel at
The BMJ, which created a conflict of interest partway through the process. COPE investigated the complaint to the full extent of its remit, more quickly than reported in
The Lancet, and with due care to ensure independence.
COPE's conclusion was emphatic: that
The BMJ had taken “extraordinary steps” and “acted appropriately by completing an internal investigation and audit to a high standard, and promoted transparency by making information on the process publicly available”.
Comparisons between the statin saga and the measles, mumps, and rubella vaccine scare also do not serve Horton or Collins well.
The BMJ's article in question was not a poorly done and fraudulent piece of research
8 but an expert reanalysis and commentary;
9 after it was published it was not the journal or the authors but Collins who launched the media scare when he went to the
press despite repeated invitations to air his concerns in
The BMJ. The BMJ did not attempt to cover up concerns,
10leading to years of delay, but corrected the article within months and referred the decision about retraction to a panel of experts.
Retraction Watch called the panel's report “the most detailed justification for a journal's decision not to retract a paper that we've seen in a long time, perhaps ever”.
I fully support efforts to ensure that everyone involved in creating and publishing medical knowledge can be held accountable. I also support proposals for a radical rethink about how the
evidence base is built and used. In this effort, the lessons learnt by all parties involved in the statins saga could play an important part.
I was Chair of COPE from 2002 to 2005 and have helped to draft its original code of conduct for editors published in 2004. I commissioned and edited Brian Deer's BMJ series on the measles, mumps, and rubella vaccine scare. The BMJ runs campaigns on overtreatment, open clinical trial data, and access to the clinical trial data for statins. The BMJ receives a proportion of its income from pharmaceutical advertising and sponsorship.
