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 Six

9 Jan, 16 | by Ed Sucksmith

The 2D:4D digit ratio and autism risk: is there an association?


In December we surveyed our staff editors to tell us about their favourite articles from 2015. Our penultimate editors’ pick investigates the relationship between autism risk and a putative marker of testosterone exposure: the 2D: 4D digit ratio.

In spite of large investments in autism research, the underlying causes of autism diagnoses remain largely unknown, although twin and family studies suggest that autism is highly heritable. Perplexingly, autism is more frequently diagnosed in males, with male-to-female ratio estimates ranging from 2:1 to 9:1. A popular theory put forward to explain the sex differences behind autism diagnoses is the ‘Extreme Male Brain theory’, which proposes that people with autism are exhibiting an exaggerated profile of the male cognitive profile. A proposed physiological mechanism underlying this theory is the effect of prenatal sex steroids (such as testosterone) on the developing brain. The index to ring finger ratio (2D:4D digit ratio) is believed to be a proxy for fetal testosterone exposure; males have a lower 2D:4D digit ratio on average than females.

In this study by Anna Guyatt and colleagues, the 2D:4D digit ratio was examined in 6015 children from a British birth cohort known as The Avon Longitudinal Study of Parents and Children. The 2D:4D digit ratio was then tested for association with autistic traits in the general population and with autism diagnoses. Analysis of the data did not indicate a significant association between 2D:4D digit ratio and autism diagnoses or high autistic traits in the general population for either sex. Thus, the data did not support the extreme male brain theory of autism.

The study is just one of many published in BMJ Open in 2015 reporting negative findings. We believe that a study should be published irrespective of the direction of the findings providing that the study meets our publication criteria. Publishing negative findings has the added benefit of combating the damaging effects of dissemination bias in medical research (see Editors’ Picks 2015: Part Three)

Editors’ Picks 2015: Part Five

8 Jan, 16 | by Ed Sucksmith

How well reported are interventions in systematic reviews?


In December we surveyed our staff editors to tell us about their favourite articles from 2015. Part 5 focuses on another important editorial issue: the quality of reporting in systematic reviews.

Clinicians and healthcare professionals often rely on systematic reviews and meta-analyses for information on how effective interventions are, as they do not have the time to read through the literature on each individual trial that is published. But how useful are systematic reviews for clinicians and are the interventions reported in sufficient detail for clinicians to implement them in clinical practice? Interventions should also be reported in sufficient detail so that other researchers could independently replicate the interventions in future studies.

In our fifth editors’ pick, Tammy Hoffman and colleagues selected a random sample of systematic reviews of non-pharmacological interventions for stroke, and examined the completeness of reporting of the interventions in the reviews. Using an established reporting instrument called the Template for Intervention Description and Replication (TIDieR) checklist, the authors found that a number of checklist items were incompletely reported in the reviews, including: (1) modifications to the intervention; (2) a description of intervention adherence/ fidelity; (3) a description of the materials used in the intervention; (4) a description of the procedures used in the intervention and; (5) if there was a plan for the intervention to be personalised or adapted during the study. The incomplete descriptions of the intervention materials and procedures were highlighted as crucial missing elements that have previously been reported as most frequently missing in publications of individual randomized trials.

The authors concluded that better reporting of interventions were needed in systematic reviews if they were going to be effectively used and interpreted by readers. Recommendations are provided for authors of systematic reviews to use the TIDieR checklist as a guide for reporting interventions, and to provide an ‘intervention options table’ that summarises the pros and cons of all usable and feasible interventions reported in the review.

Editors’ Picks 2015: Part Four

7 Jan, 16 | by Ed Sucksmith

Long-term health effects of exercise for the elderly: study protocol


Each day this week we’ve been looking at a published paper from 2015 that sparked the interest of one of our staff editors. Our fourth pick is a study protocol for a randomised controlled trial of exercise training in an elderly population.

BMJ Open supports the publication of study protocols to improve transparency in the ‘research to publication’ process and to protect the wider community against numerous damaging research practices, such as publication bias and HARKing.

In this study protocol, Dorthe Stensvold and colleagues from Norway, Australia and the USA aim to evaluate the effects of 5 years of exercise training on mortality in an elderly population. Whilst previous epidemiological studies have indicated that physical activity is associated with reduced risk of premature death and that changes in fitness predict mortality, it is unclear whether the level of physical activity is the direct cause of favourable health effects rather than other factors associated with a person’s state of health.

The proposed study in just under 1600 participants is considered to be the first randomised controlled trial to assess the effect of exercise training on mortality and morbidity in an elderly population. The protocol currently has 6 citations in Google Scholar, suggesting that the article has been a useful addition to the scholarly literature since it was published in February 2015.

If you are interested in learning more about study protocols then check out BMJ’s recently launched e-learning programme: Research to Publication. The programme includes a free taster module on how to write and publish a study protocol. If you successfully complete the module, then you will receive formal recognition with a certificate from The BMJ’s Editor-in-Chief and UCSF’s associate dean. If you provide this certificate when you submit your study protocol to BMJ Open, then you will receive a 75% reduction on the open access publication fee if your article is accepted.

Antidepressants and risk of mania, medically unexplained symptoms and the association between alcohol and mortality in Alzheimer’s: Most read articles in December

6 Jan, 16 | by Emma Gray

Antidepressants, medically unexplained symptoms, and alcohol consumption in Alzheimer’s patients

File:Glass of white wine.jpg

December saw a lot of new papers entering the Top 10 Most Read, with half of them being from the current issue. Patel et al‘s research paper, complete with video abstract and press release, questions whether antidepressants increase the risk of mania and bipolar disorder in people with depression in the top spot this month. Miller et al and Kristensen et al maintain their places at numbers two and three, with papers on clinical trial registration and the effect of statins on survival. Our fourth most read article examines junior doctors’ knowledge and experiences of managing patients with medically unexplained symptoms via a qualitative study using in-depth interviews. Berntsen et al enter the list at number seven, with a prospective cohort study investigating the association between alcohol consumption and mortality in patients with Alzheimer’s disease, and Jonas et al write about surgery and the placebo response.

Rank Author(s) Title
1 Patel et al. Do antidepressants increase the risk of mania and bipolar disorder in people with depression? A retrospective electronic case register cohort study
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 Yon et al. Junior doctors’ experiences of managing patients with medically unexplained symptoms: a qualitative study
5 Martin-Misener et al. Cost-effectiveness of nurse practitioners in primary and specialised ambulatory care: systematic review
6 Bourne et al. The impact of complaints procedures on the welfare, health and clinical practise of 7926 doctors in the UK: a cross-sectional survey
7 Berntsen et al. Alcohol consumption and mortality in patients with mild Alzheimer’s disease: a prospective cohort study
8 Jonas et al. To what extent are surgery and invasive procedures effective beyond a placebo response? A systematic review with meta-analysis of randomised, sham controlled trials
9 Boaz et al. Does the engagement of clinicians and organisations in research improve healthcare performance: a three-stage review
10 Hysing et al. Sleep and use of electronic devices in adolescence: results from a large population-based study

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

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.