You don't need to be signed in to read BMJ Blogs, but you can register here to receive updates about other BMJ products and services via our site.

Welcome to the BMJ Open blog. BMJ Open is an open access journal, dedicated to publishing medical research from all disciplines and therapeutic areas.

Find out more about the journal here.

We will be updating the blog with news about the journal, highly accessed papers, press coverage, events and matters of interest in the open access and publishing world, and anything else that catches our eye.

Editors’ Picks 2015: Part Three

6 Jan, 16 | by Ed Sucksmith

Combating Dissemination Bias in Clinical Research: Recommendations


Tip of the iceberg: Dissemination bias includes publication bias, where published papers that are exposed to readers are biased in favour of particular characteristics (e.g. positive findings). Papers with other characteristics (e.g. negative findings) are not published and, as a result, remain out of view from readers.









This week we have been highlighting some of our favourite articles from 2015, which were selected by our staff editors in December. Today’s pick looks at what can be done to combat dissemination bias in clinical research.

Dissemination bias is a problem that extends beyond clinical trials to include all clinical research. It is estimated that 50% of research findings are not published at all or only present partial information, which means healthcare professionals and other stakeholders can’t make informed decisions. The focus of our third editors’ pick, by Joerg Meerpohl and colleagues from the OPEN project consortium, was to address how this problem could be resolved. The research group drafted a comprehensive set of evidence-informed recommendations for combating dissemination bias that was discussed by a variety of external experts and stakeholders. These recommendations included the need to raise awareness among stakeholders, e.g., patients participating in clinical trials who may need to be informed that their participation only contributes to scientific knowledge and progress if the results are published or at least publicly available. Other important recommendations included the need to promote trial registration and posting of results and to support initiatives that help trials to be easily searched and identified across registries.

A large number of more specific recommendations were proposed that targeted particular stakeholder groups. For journal editors/ publishers, these included (1) the consideration of research papers for publication regardless of the direction of the findings; (2) making trial registration a compulsory pre-requisite for publication; (3) checking manuscripts next to study protocols to identify selective reporting and; (4) checking for redundant publication of results.

Combating dissemination bias in medical research is an important issue for BMJ Open and its editors. We welcome these proposals and further studies addressing this vital issue.

Editors’ Picks 2015: Part Two

5 Jan, 16 | by Ed Sucksmith

Clinical Trial Transparency: disclosures for new drugs below ethical and legal standards


In December we surveyed our staff editors, who oversee and manage the peer review process, to tell us about their favourite articles from 2015. Part 2 looks at clinical trial transparency for new drugs. 

The ability of medical and healthcare professionals to select the most effective and appropriate treatments for patients depends on an unbiased and scientifically sound evidence base. Clinical trials, particularly randomized controlled trials, are the most rigorous methods for investigating the effectiveness of drugs and other treatments. If clinical trials are not registered or only selectively disseminated then the medical evidence is distorted and clinicians are unable to make informed decisions about which treatments are best for patients. As a result patients may be harmed by taking inferior treatments and participants in clinical trials are not sufficiently protected because they may be unknowingly recruited into a failed experiment that has already been conducted previously but has not been publicly disseminated.

In our second editors’ pick, Jennifer Miller and colleagues examined the levels of clinical trial registration, reporting and publication rates for new drugs approved by the Food and Drug Administration (FDA) in 2012. Out of a total of 318 clinical trials reviewed, 57% were registered per drug on average. 56% of trials were published whilst 65% of trials were either published or had reported results. The authors concluded that disclosures for new drugs frequently fell below the legal requirements and did not meet the ethical standard that results of clinical research should be publicly accessible.

Along these lines, BMJ Open supports editorial policies and procedures that improve clinical trial transparency. These include: (1) our commitment to making all papers accessible to the public; (2) compulsory prospective registration of all clinical trials submitted to the journal in a WHO or ICJME approved registry; (3) the publication of clinical trial protocols and; (4) the publication of negative findings. BMJ is also a founding member of the AllTrials campaign, an initiative specifically set up to improve clinical trial transparency by calling for all past and present clinical trials to be registered and their results reported.

Editors’ Picks 2015: Part One

4 Jan, 16 | by Ed Sucksmith

2015 was another exciting year for BMJ Open, with over 1500 articles published. Since launching in 2011, we have published over 4000 articles, and we look forward to February, when the journal will celebrate its fifth anniversary.

In December we surveyed our staff editors, who oversee and manage the peer review process, to tell us about their favourite articles from 2015. We managed to narrow the list down to seven editors’ picks, and we will be presenting the results over the next few days.

Peer Review Models: which work the best?

Peer review_P1

Whilst peer review is entrenched as the ‘gold standard’ used by medical journals to assess the quality and validity of submitted research, many different models of peer review are used and surprisingly little research has been carried out to examine just how effective these models are. In a study by Maria Kowalczuk and colleagues from the Open Access Publisher BioMed Central, the quality of reviewer reports were analysed and compared between two journals that had different peer review models, namely ‘open’ peer review and ‘single-blind’ peer review. They also compared the quality of reviewer reports within a single journal that had switched from open to single-blind peer review. The authors found that the quality of reviewer reports were 5% higher on average in the open peer review journal compared to the single-blind review journal, a difference that was statistically significant. However, no differences in report quality were found in the journal that had operated under both peer review models. The findings suggest that open peer review may lead to higher quality reviewer reports on average, or at the very least does not lead to a decline in report quality as some people have feared it would. The authors also found differences in the recommendations of author-suggested and non-author-suggested reviewers, with author-suggested reviewers recommending acceptance significantly more frequently than non-author-suggested reviewers. This suggests that caution should be taken by editors when making decisions based on author-suggested reviews alone.

BMJ Open operates a fully open peer review process where authors and reviewers are made aware of each other’s identity. We also publish the reviewer reports and authors’ responses along with the accepted manuscript. By prioritising transparency in the process, we believe our model opens up the ‘black box’ of peer review. It also gives authors greater incentive to produce high quality reports and makes reviewers more accountable.


The importance of reviewers – 2015

23 Dec, 15 | by Emma Gray


The peer review process is central to scholarly research, a critical part of the publishing process and a method of quality control for the scientific community. While peer review can seem like a daunting, never-ending task at times, without it journals would not survive and continue to publish the important, accurate findings they do today.

At BMJ Open, we rely on our large bank of reviewers to help us ensure that the papers we publish are useful and of good quality. A good review can often require a great deal of work, and the continued success of the journal in the past year would not have been possible without the help of reviewers who provided us with their time, expertise and detailed comments. Our transparent, open peer review process is part of our aim to provide a home to sound medical research, making as much research available to the scientific community as possible while still upholding the BMJ’s core standards for research conduct.

It is important to recognise the contribution that reviewers make to the journal, and the essential nature of their work. BMJ Open would like to thank all those reviewers who have worked with us this year – we hope that you will continue to work with us in the year ahead! We are pleased to be able to continue with our reviewer discount of 25% for reviewers who submit manuscripts to us within 12 months of completing their review, and we would like to remind you that CME points are available for those who submit detailed comments within the given timeframe. Existing reviewers for the journal are encouraged to update their profiles on our submission system with their areas of expertise, so we can more easily allocate you appropriate papers – and, of course, we always encourage new reviewers to come forward and join our growing bank of referees.

We look forward to seeing what the next 12 months will hold for BMJ Open. With our fifth birthday on 23rd February 2016, we hope that the journal will continue to go from strength to strength, with the help of our reviewers and authors.

Certain antidepressants linked to heightened risk of mania and bipolar disorder

16 Dec, 15 | by Emma Gray

Strongest association found for SSRIs and venlafaxine

Taking certain antidepressants for depression is linked to a heightened risk of subsequent mania and bipolar disorder, reveals research published in the online journal BMJ Open.

The strongest association seemed to be for serotonin reuptake inhibitors, or SSRIs for short, and the dual action antidepressant venlafaxine, the analysis indicated.

The researchers base their findings on the anonymised medical records of more than 21,000 adults in receipt of treatment for major (unipolar) depression between 2006 and 2013 at a large provider of inpatient and community mental healthcare in London.

The research team looked at subsequent diagnoses of bipolar disorder or mania following an original diagnosis of unipolar depression, so-called because it lacks the ‘highs’ typical of bipolar disorder.

The analysis revealed that the overall yearly risk of a new diagnosis of mania and bipolar disorder between 2006 and 2013 was 1.1% (10.9/1000 patient years).

The peak age for diagnosis was seen among patients aged 26 to 35, among whom the yearly risk was 1.2% (12.3/1000 patient years).

The most commonly prescribed antidepressants were SSRIs (35.5%); mirtazapine (9.4%); venlafaxine (5.6%) and tricyclics (4.7%).

Previous treatment with certain antidepressants was associated with a heightened risk of a subsequent diagnosis of bipolar disorder and/or mania, the yearly risk of which ranged from 1.3% to 1.9% (13.1 to 19.1/1000 patient years).

Further analysis revealed that this heightened risk was particularly associated with treatment with SSRIs and venlafaxine. These drugs were associated with a 34-35% increased risk of being diagnosed with bipolar disorder and/or mania.

These findings held true even after taking account of potentially influential factors.

This is an observational study so no firm conclusions can be drawn about cause and effect, and the researchers point out that their findings may be explained by latent bipolar disorder rather than any effects of drug treatment. Furthermore, they were unable to obtain information on important risk factors.

“However, regardless of underlying diagnosis or aetiology the association of antidepressant therapy with mania demonstrated in the present and previous studies highlights the importance of considering whether an individual who presents with depression could be at high risk of future episodes of mania,” say the researchers.

Pertinent risk factors include a family history of bipolar disorder, a depressive episode with psychotic symptoms, young age at first diagnosis of depression, and depression that is unresponsive to treatment.

“Our findings also highlight an ongoing need to develop better ways to predict future risk of mania in people with no prior history of bipolar disorder who present with an episode of depression,” they write.

In an accompanying videocast, the researchers highlight that the absolute risk of developing bipolar disorder is low and antidepressants are safe and effective treatments for depression and anxiety. Patients should not stop their treatment suddenly as this may result in withdrawal symptoms, they advise.

A video abstract for this manuscript is available to view here:

Bicycling injury hospitalisation rates, clinical trial registration and women in leadership roles: Most read articles in November

9 Dec, 15 | by Emma Gray

Bicycling injuries, clinical trial registration, and the under-representation of women in leadership roles

This month sees a lot of change in the Top 10 Most Read, with many new papers entering the list. At number one this month is Teschke et al‘s assessment of bicycling injury hospitalisation rates with relation to helmet legislation and mode shares, which has generated a lot of interest both on our site and on Twitter. Our second most read article takes a look at clinical trial registration, reporting and publication rates for new drugs approved by the US Food and Drug Administration. Bismark et al examine the reasons for the under-representation of women in medical leadership roles via a qualitative study at number six, and maintaining its position at number five this month is Hysing et al‘s paper on adolescent use of electronic devices in relation to sleep.

Rank Author(s) Title
1 Teschke et al. Bicycling injury hospitalisation rates in Canadian jurisdictions: analyses examining associations with helmet legislation and mode share
2 Miller et al. Clinical trial registration, reporting, publication and FDAAA compliance: a cross-sectional analysis and ranking of new drugs approved by the FDA in 2012
3 Kristensen et al. The effect of statins on average survival in randomised trials, an analysis of end point postponement
4 Tricco et al. Comparative safety and effectiveness of long-acting inhaled agents for treating chronic obstructive pulmonary disease: a systematic review and network meta-analysis
5 Hysing et al. Sleep and use of electronic devices in adolescence: results from a large population-based study
6 Bismark et al. Reasons and remedies for under-representation of women in medical leadership roles: a qualitative study from Australia
7 Bove et al. Efficacy of a minimal home-based psychoeducative intervention versus usual care for managing anxiety and dyspnoea in patients with severe chronic obstructive pulmonary disease: a randomised controlled trial protocol
8 Deane et al. Priority setting partnership to identify the top 10 research priorities for the management of Parkinson’s disease
9 Chaibi et al. Chiropractic spinal manipulative therapy for migraine: a study protocol of a single-blinded placebo-controlled randomised clinical trial
10 Hubble et al. Trunk muscle exercises as a means of improving postural stability in people with Parkinson’s disease: a protocol for a randomised controlled trial

Most read figures are based on pdf downloads and full text views. Abstract views are excluded.

US pedestrian wheelchair users third more likely to die in car crashes than general public

19 Nov, 15 | by Emma Gray

And men’s risk is five times higher than women’s, particularly among 50 to 64 year olds

Pedestrian wheelchair users in the US are a third more likely to be killed in road traffic collisions than the general public, finds research published in the online journal BMJ Open.

And men’s risk is five times higher than women’s, the findings show.

Every year nearly 5000 pedestrians are killed and another 76,000 injured in road traffic collisions on public roads in the US.

But it is not known what the risk of death is among pedestrian wheelchair users, and whether this is higher than that of the general population who don’t use wheelchairs.

The researchers therefore used a technique that mines the overlap between two independent but incomplete data sources (capture-recapture) to estimate the total number of pedestrian deaths caused by car crashes between 2006 and 2012.

These were national news stories on car crash fatalities published on the LexisNexis US newspaper database; and the National Highway Traffic Safety Administration’s Fatality Analysis Reporting System (FARS), which is based on police reports of road traffic collisions on US roads.

Based on these data, the researchers calculated that approximately 528 pedestrians using wheelchairs were killed in road traffic collisions in the US between 2006 and 2012.  This equates to a risk of death for a pedestrian wheelchair user that is more than a third higher than for American pedestrians in general.

Furthermore, the risk of a car crash death was over five times higher for men in wheelchairs than for women, particularly among men aged 50 to 64, the analysis showed.

The figures are likely to be an underestimate of the true number of deaths because mobility devices were not always differentiated from recreational scooters and mopeds in news stories.

Around half (47.5%) of the fatal crashes occurred at intersections, and in almost four out of 10 (39%) of these cases, traffic flow was not controlled.

Many fatal crashes occurred at locations with inadequate pedestrian facilities.  For example, crosswalks were unavailable at the site in one in five fatal crashes.

The police reports also indicated that drivers frequently failed to give way to wheelchair users, and that the wheelchair user was not sufficiently visible in 15% of the incidents.

it is important—and often required by the Americans with Disabilities Act—that there should be well-designed curb cuts, ramps, and sidewalks to enable people who use wheelchairs to safely traverse roads, say the researchers.

“[Our] findings underscore the need for policy-makers and planners to fully incorporate disability accommodations into pedestrian infrastructure and for persons who use wheelchairs—and others with disabilities—to remain a salient population when road safety interventions are designed,” they write.

BMJ launches new Chinese publishing portal: Q&A with David Wang and Huili Chen

17 Nov, 15 | by aaldcroft


Recently BMJ launched a publishing portal intended to help Chinese authors publish in BMJ’s portfolio of more than 60 journals, including BMJ Open. To help understand the motivations surrounding the portal, along with some of its content, we asked BMJ China’s Business Development Manager Huili Chen and the Deputy Editor of BMJ’s new journal Stroke and Vascular Neurology, David Wang, a few questions.

Q: David, how do you think the portal might benefit Stroke and Vascular Neurology?

DW: As the newest journal of the BMJ family and the official English journal for the Chinese Stroke AssociationSVN would like to be a successful journal in the area of cerebrovascular disease. To be successful requires quality papers. Chinese scholars are always seeking journals for their work and SVN hopefully can benefit from that need. The new portal is the only such kind of service from a major publisher that offers the authors a “one stop shopping” convenience. Authors can easily find the journal to submit their work in one portal. SVN can certainly benefit from the portal since Chinese authors can enjoy the wonderful services provided by BMJ publishing, easily identify our journal, and send in their work.

Q: What information on the portal do you think will be most useful to potential authors in SVN?

DW: SVN would like to provide a user-friendly and state-of-art platform for the authors to submit their manuscripts and track their submission. BMJ portal is the right place to get started. It offers quality information and instructions to potential authors on how to complete the submission. As a new journal that has just been launched, our potential authors can find themselves at ease when they submit their work through the portal.

Q: Huili, what was the motivation behind the portal?

HC: There were a few driving elements. Firstly, BMJ and The BMJ have strong reputations in China, but we’ve not had the chance to promote the journals in a systematic way, directly to authors. Secondly, there is a lot of misinformation about publishing in China, much of it taking the form of comments on websites. We wanted to communicate the correct information, and a portal seemed a good way to do this. Finally, several publishers have initiatives in China, and we want to be competitive and demonstrate that BMJ values work from Chinese authors. The portal seemed like a good starting point.

DW: As China is increasingly becoming a powerhouse in biomedical research, there are greater needs needs for Chinese scholars to publish their work. However, Chinese authors often struggle in finding the best journal for their papers. BMJ has over 60 journals–and growing–but some of the journals are not well-known to Chinese scholars. By having such a portal, it helps to educate the authors about all the journals published by BMJ and offer them a very convenient way to submit their work. Such ease of use will certainly benefit new journals, such as SVN, since authors will recognise that they can submit their work just as easily as they could to other BMJ journals.

Q: Who do you expect will use the portal?

HC: The portal was designed for all Chinese speaking doctors and medical researchers who have an interest in publishing papers in international journals. The portal is available internationally, but we expect that most of the users will be based in China.

DW: The portal is in Chinese. Therefore, any authors who reads Chinese can find it easy to use. It is certainly not limited to the scholars in China but anywhere in the world who understand Chinese

Q: What information is available on the portal?

HC: There is an overview of BMJ’s journal portfolio, with each journal having a page containing key information intended to help authors decide whether the journal is suitable for their work. There is also a page outlining The BMJ, highlighting its criteria for research articles, article requirements, and peer review process. There are regular updates on research articles published in The BMJ and Open Access articles published in all journals, highlighting strong articles from Chinese researchers. We have recently added a Chinese blog by editors from the Journal of Medical Genetics–this can be done for any journal that is interested.

In addition there is an author center where authors can read about journals’ editorial policies and publishing processes. We have included resources that authors might find useful, such as an introduction to the Research to Publication e-learning course. Finally, we highlight Chinese experts who are editors and board members on BMJ journals

Q: What impact do you hope the portal will have?

HC: Ideally, we’d like all medical researchers and doctors to know about the portal and use it whenever they want to publish a new piece of work. We want it to be a tool to establish loyalty and trust. It will, of course, take time to accomplish, but with BMJ’s growing portfolio of journals, we think the portal is an important starting point.

The effect of statins, obesity-related behaviours in Europe and the accuracy of Wikipedia content: Most read articles in October

9 Nov, 15 | by Emma Gray

Statins, obesity-related behaviours, and the accuracy of Wikipedia


by Panoramix303

Lovastatin, from Wikipedia

Several papers have entered the Top 10 Most Read this month, including Kristensen et al‘s systematic review on statins and survival in randomised trials, which tops the list. In our tenth most read article, Azer et al assess the accuracy of content of Wikipedia articles on cardiovascular diseases, and at number nine, Lakerveld et al provide the first results of the SPOTLIGHT survey of obesity-related behaviours and body mass index in adults in Europe. Previously featured papers on the list include Deane et al‘s identification of the top priorities for the management of Parkinson’s disease and Dall’Ora et al‘s study on the association of long shifts for nurses and job dissatisfaction.

Rank Author(s) Title
1 Kristensen et al. The effect of statins on average survival in randomised trials, an analysis of end point postponement
2 Deane et al. Priority setting partnership to identify the top 10 research priorities for the management of Parkinson’s disease
3 Hubble et al. Trunk muscle exercises as a means of improving postural stability in people with Parkinson’s disease: a protocol for a randomised controlled trial
4 Dall’Ora et al. Association of 12 h shifts and nurses’ job satisfaction, burnout and intention to leave: findings from a cross-sectional study of 12 European countries
5 Hysing et al. Sleep and use of electronic devices in adolescence: results from a large population-based study
6 Bernacki et al. Development of the Serious Illness Care Program: a randomised controlled trial of a palliative care communication intervention
7 Plouvier et al. Transitions in Parkinson’s disease in primary care: protocol of a longitudinal mixed methods study
8 Lakerveld et al. Obesity-related behaviours and BMI in five urban regions across Europe: sampling design and results from the SPOTLIGHT cross-sectional survey
9 Kowalczuk et al. Retrospective analysis of the quality of reports by author-suggested and non-author-suggested reviewers in journals operating on open or single-blind peer review models
10 Azer et al. Accuracy and readability of cardiovascular entries on Wikipedia: are they reliable learning resources for medical students?

Most read figures are based on pdf downloads and full text views. Abstract views are excluded.

OA Journals and Wikipedia: Open for collaboration

16 Oct, 15 | by aaldcroft

The theme of this year’s Open Access week is Open for Collaboration, with the aim of highlighting “the ways in which collaboration both inspires and advances the Open Access movement”. Recently BMJ Open published an article by Samy Azer and colleagues investigating whether articles in Wikipedia relating to cardiovascular disease were accurate enough to function as a suitable learning resource for medical students. The conclusions point to the limitations of Wikipedia, making us wonder, “How can scientists, editors, and non-expert readers work together to improve our open knowledge resources?”

Evolving connections

One of the key advantages of Open Access is that it enables an endless opportunity to collaborate. Openly available knowledge that can be connected via hyperlinks means that all participants in the Open Access community are free to work together to create a richer, more meaningful information landscape. One of the great achievements of the connected, online environment is, of course, Wikipedia. Currently the seventh most visited site on the internet, Wikipedia makes knowledge previously only available to those with the means to purchase it, available to anyone with an internet connection. But what is perhaps even more revolutionary is that the articles are not necessarily written by experts, but by anyone who has the desire to participate in the project. Known as Wikipedians, they are the driving force, and they strive to make Wikipedia more accurate, consistent, and complete. 

When considering Open Access collaboration, the origins of Wikipedia are noteworthy. Before Wikipedia, there was Nupedia–an abandoned project involving the founders of Wikipedia, Jimmy Wales and Larry Sanger, with Sanger acting as Editor-in-Chief. Like Wikipedia, the knowledge in Nupedia was openly available, but unlike Wikipedia, the knowledge was written by experts and underwent peer review prior to publication. In fact, the seven-stage editorial process of Nupedia closely resembles that used in BMJ Open and other scientific journals, with peer review as a crucial element:

  1. Assignment
  2. Finding a lead reviewer
  3. Lead review
  4. Open review
  5. Lead copyediting
  6. Open copyediting
  7. Final approval and markup

Ultimately, the strict control over quality made Nupedia unsustainable. Authors and reviewers often reached an impasse, and only a small number of articles were ever published. It did, however, blaze the trail for Wikipedia. Initially viewed by experts as unreliable, it was not long before there was evidence indicating that the accuracy of information in Wikipedia was as trustworthy as that in traditional encyclopedias.

Putting it all together

The evidence, however, also reveals the limits of Wikipedia. The study by Azer et al, for example, demonstrates that while Wikipedia presents an accurate overall picture of cardiovascular diseases, more detailed information, such as the pathophysiology, signs and symptoms, diagnostic approach, and treatment, are incomplete. Similar conclusions have been reached in studies investigating content on pathology, osteosarcoma, and drugs. Many caution that Wikipedia should not be used as a substitute for peer-reviewed material written by subject matter experts. These finding are, of course, not surprising. For a broad overview–or as a way to settle family disputes–Wikipedia is brilliant, but as a resource for expert medical knowledge, readers need to be more critical.

Despite the limitations, Wikipedia should not be sold short. Wikipedia offers a number of advantages over traditional learning resources, such as textbooks, because it enables self-directed learning and active engagement with knowledge–critical skills in today’s digitized world. For someone new to an area of study, Wikipedia is the perfect entry point.

So how can we put it all together?

We can’t expect Wikipedians to do everything. Instead, it is important for all of us–scientists, journal publishers, and readers–to collaborate. Experts–and particularly authors–should ensure that when work that is close to them is cited in Wikipedia, it is discussed accurately–and if it’s not, then they have a responsibility to edit it (and are free to do so). Similarly, non-experts whose interest is sparked by an entry in Wikipedia should have the opportunity to read the primary material, ideally in a source that is Open Access. For all parties involved, we must accept that knowledge will always be a work in progress and that what we can achieve through open collaboration is much more than we could achieve alone.