Better evidence

Patients are being let down by serious flaws in the creation, dissemination, and implementation of medical research. Too many research studies are poorly designed or executed. Too much of the resulting research evidence is withheld or disseminated only piecemeal. As the volume of clinical research has grown the quality of evidence has often worsened, which has compromised medicine’s ability to provide affordable, effective, high value care for patients. There are many problems. In our editorial launching the manifesto we describe the following:

• Results from half of all trials are never published, and positive results are twice as likely to be published as negative ones.

• 85% of research spending currently goes to waste.

• Over four fifths (86%) of a sample of Cochrane reviews did not include data on the main harm outcome.

• A systematic review of 39 studies found no robust studies evaluating shared decision making strategies.

• The drug industry has been fined for criminal behaviour and civil infringements, but little happens to prevent such problems occurring again.

• Despite repeated calls to prohibit or limit conflicts of interests among authors and sponsors of clinical guidelines, the problem persists.

• A third of scientists report questionable research practices, including data mining for statistically significant effects, selective reporting of outcomes, switching outcomes, publication bias, protocol deviations, and concealing conflicts of interest.

• More than one in 10 authors and reviewers has firsthand knowledge of inappropriate adjustment, alteration, or fabrication of data. More than one in 20 admits having lied in authorship statements.

EBM Live
With the University of Oxford’s Centre for Evidence Based Medicine we campaign for better evidence, through a yearly conference designed to debate the current status and future directions of research.

Evidence manifesto
The team behind EBM Live is working on “The Manifesto: better evidence for better healthcare.” Listen to the projects leaders discuss its aims or read its editorial.

External leadership
BMJ editors take part in advocacy and outreach to promote transparent and accurate reporting of research evidence. We help to develop and publish reporting guidelines such as those produced by the EQUATOR network.

Data sharing
The BMJ has led the way in calling for and promoting the sharing of clinical trial data. As of July 2015, The BMJ only considers for publication those clinical trials where the authors commit to making the relevant anonymised patient level data available on reasonable request. We also support the RIAT (Restoring Invisible and Abandoned Trials) initiative which “posits that when original investigators and sponsors abandon their trials, either by not pursuing publication or by refusing to correct demonstrable errors in a trial publication, third parties that have obtained the underlying trial data are free to publish and correct the record.”

Design and conduct of research:
The BMJ publishes articles on Research Methods and Reporting, with a free archive going back to 2008, and Statistics Notes, going back to 1994. This editorial explains why. We have also developed a distance learning programme, Research to Publication, to help medical schools and researchers do and disseminate their health research better.

Patient partnership
The BMJ has led the way in promoting the involvement of patients and the public in medical research and education. We require Patient and Public Involvement statements on all research. Our research articles are reviewed by patients as well as by medical experts. We campaign for patients, carers, and their advocates to play a greater role in medical research, and education. Further details on our patient partnership campaign can be found here.

Better use of evidence
Rapid Recommendations is a collaboration with MAGIC (a non-profit research and innovation programme). It is a new way of synthesising and translating evidence for practice quickly, reliably, and in user friendly formats. For each project the authors include patients and practising clinicians, and we aim to meet the Institute of Medicine’s criteria for trustworthy guidelines. Each set of recommendations is typically triggered by new or forgotten evidence that might change practice.

Taking a stand on financial interests
The BMJ requires that its most clinically applicable education articles must be written by authors without relevant financial ties to industry. By "industry" we mean companies producing drugs, medical foods, nutraceuticals, devices, apps, or tests; medical education companies; or other companies with a financial or reputational interest in the topic of the article. For transparency we also publish each year our own revenues from pharmaceutical companies. You can read more about this in our editorial on the topic.

Campaigned for statins data
The BMJ calls for anonymised individual patient data from the clinical trials of statins to be made available for independent scrutiny. This was sparked by a 2013 debate on the balance of benefit and harm of statins in asymptomatic, lower risk populations. The quest for the data remains ongoing. Read the detail here.

What has The BMJ done to promote better evidence?

Campaigned for data on influenza drugs
Tamiflu and Relenza: A four year battle supported by The BMJ for access to more than 100,000 pages of clinical study reports to support independent scrutiny of the data for a Cochrane review of the drugs was ultimately successful. The campaign spurred changes in data transparency policies in Europe and abroad.