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.

Social stigma, patient knowledge and bed sharing: Most read articles in November

6 Dec, 13 | by bdevaney

The most-read article in November was Snow et al’s article on what happens when patients have more knowledge than their doctors. Werb et al‘s article on international government surveillance systems continues to be popular. Other newly-published papers in the top ten include Rajmil et al‘s look at the impact of the economic crisis on children’s health and Browne et al’s study of the social stigma around type 2 diabetes.

Rank Author(s) Title
1 Snow et al What happens when patients know more than their doctors? Experiences of health interactions after diabetes patient education: a qualitative patient-led study
2 Carpenter et al Bed sharing when parents do not smoke: is there a risk of SIDS? An individual level analysis of five major case–control studies
3 Werb et al The temporal relationship between drug supply indicators: an audit of international government surveillance systems
4 Ford et al Current treatments in diabetic macular oedema: systematic review and meta-analysis
5 Doyle et al A systematic review of evidence on the links between patient experience and clinical safety and effectiveness
6 Wakefield et al Introduction effects of the Australian plain packaging policy on adult smokers: a cross-sectional study
7 Rajmil et al Impact of the economic crisis on children’s health in Catalonia: a before–after approach
8 Browne et al ‘I call it the blame and shame disease’: a qualitative study about perceptions of social stigma surrounding type 2 diabetes
9 Nguyen et al Statin treatment and risk of recurrent venous thromboembolism: a nationwide cohort study
10 Taylor et al Prevalence and incidence rates of autism in the UK: time trend from 2004–2010 in children aged 8 years


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

C-section rate for private patients double that of publicly funded patients

25 Nov, 13 | by bdevaney

Differences in medical or obstetric risks don’t fully explain disparity.

The rate of scheduled caesarean sections among private patients is around double that of publicly funded patients, indicates a study of more than 30,000 women in Ireland, published in BMJ Open.

Differences in the medical and obstetric risks between the two groups don’t fully explain this disparity, say the researchers, who looked at the impact of social, medical and obstetric factors on mode of delivery among women booked for privately or publicly funded care in the same hospital.

The researchers were particularly keen to see if funding source made any difference to operative deliveries, as it’s not clear whether private care within a publicly funded setting prompts higher rates of costly interventions, they say.

Both private and public healthcare have been offered in Irish hospitals for decades at a ratio of 80 (public): 20 (private). And recent changes to UK health policy mean that hospitals in England will now be able to bump up the proportion of income they generate from private healthcare to 49%.

The researchers analysed the method of delivery for just over 30,000 women with singleton pregnancies, in a large urban maternity hospital in Ireland between January 2008 and July 2011.

The hospital delivers between 8000 and 9000 women every year, with obstetric care provided by around 14 consultants and 16 trainees. Out of the total, 24,574 women were publicly funded and 5479 were private patients.

Private patients were more likely to be older, more affluent and better educated, and to be Irish than publicly funded patients. And they were less likely to be single, childless, have an unplanned pregnancy, or to have booked late for obstetric care.

And while they were less likely to have a medical disorder, they were more likely to have had fertility treatment, recurrent miscarriage, or experienced a previous stillbirth or infant death.

The analysis showed that compared with publicly funded patients, private patients were more likely to have a C-section or surgical vaginal delivery—vacuum or forceps.

But the greatest disparity was in the rate of planned C-sections, which was around twice as high among the private patients, particularly for mums who had given birth before, and by C-section.

The differences remained after taking account of medical (including age), obstetric, and social differences between the two groups.

A common argument advanced for planned C-section is that the procedure does less damage to the pelvic floor than a vaginal birth, say the authors, but while private patients requested more C-sections, very few such requests were made.

“We found the differences observed in relation to operative deliveries were not explained by higher rates of medical or obstetric complications among private patients,” write the authors, although older age and higher income may have played their part, they suggest.

But the findings raise important questions about equity, both in terms of use of resources and choice, and whether that choice really is in the best interests of the woman concerned, say the authors.

“Healthcare systems that include public and private patients need to reflect on the potential for disparate rates of intervention and the implications in terms of equity, resource use, and income generation,” they comment.

NHS 111 increases ambulance and urgent and emergency care use

13 Nov, 13 | by bdevaney

Call handling service did not reduce pressures during first year of operation, as intended.

The call handling service NHS 111 increased the use of ambulance and urgent and emergency care services during its first year of operation, shows a detailed evaluation published in BMJ Open.

This is despite the fact that NHS 111 was set up with the intention of relieving pressure on these services by ensuring that patients are directed to care that is appropriate for their level of need, say the researchers.

They were commissioned by the Department of Health in England to evaluate the use and impact of NHS 111, a new 24/7 telephone service for non-emergency healthcare problems, which was rolled out nationwide in March-April this year.

The service is staffed by trained call handlers, who are not clinicians, but who have back-up from nurse advisors. They triage calls to other services or home care, using an algorithm based assessment system, NHS Pathways.

The researchers analysed 36 months of routine data – to include the first year that NHS 111 was up and running (2010-11), and the two years before it started (2008-10) – in the four designated nationally representative pilot sites, and in three comparable sites, covering a total population of 3.6 million.

The routine data analysed covered the use of emergency ambulance calls and incidents; attendances at emergency care departments; contacts with out of hours urgent care services; and calls to the telephone triage service NHS Direct.

During its first year of operation at the four pilot sites, over 400,000 calls were made to NHS 111, just over 277,000 of which were triaged using NHS Pathways. Of the triaged calls, 28% were referred to a nurse for clinical advice, and over half were judged to need primary or urgent care services.

The number of calls to NHS Direct fell by almost 20% during the study period, which is not unexpected, says lead author Janette Turner, in an accompanying podcast, as NHS 111 is intended to replace this service.

But compared with the three other sites, NHS 111 did not reduce the overall number of emergency ambulance calls, attendances at emergency care, or use of urgent care services at the four pilot sites during the study period.

And contrary to expectations, NHS 111 did not reduce emergency ambulance incidents, which rose by just under 3% – equivalent to an extra 24 incidents per 1000 triaged calls, each month.

The authors estimate that this could translate into an additional 14,500 call-outs for an ambulance service attending 500,000 incidents a year. Nor did it curb overall emergency and urgent care activity, which rose by between 5-12% per month at each site.

The researchers point out that a triage system designed to be used by handlers without clinical expertise will inevitably err on the side of caution and might offer less flexibility than one designed for clinicians. In the accompanying podcast, Ms Turner suggests that this issue may resolve as the service matures and the handlers gain more confidence.

And the provision of a single point of telephone service that quickly guides people needing urgent care advice to the most appropriate service “is sensible,” given the public’s confusion about which service to access, they say.

NHS 111 also offers an easily remembered number, with an emphasis on fast triage and smooth transfer to ‘the right service first time,’ all of which patients say they want, they add.

But the researchers say their findings raise other key questions, including the potential increase in volume of callers prompted by the planned closure of NHS Direct, and therefore the service’s sustainability, and whether NHS 111 is actually creating rather than curbing demand.

And they suggest: “It is probably unrealistic to expect any one service, such as NHS 111, to do everything, and real improvements may only be gained when a series of coordinated measures designed to increase efficiency across all services are implemented.”

The findings are discussed further in the accompanying podcast here.

Government surveillance systems, cleaning agents and autism: Most read articles in October

6 Nov, 13 | by bdevaney

The most-read article in October was  Werb et al‘s article on international government surveillance systems. Carpenter et al’s article on SIDS and bed sharing remained popular for the third month in a row.  Other newly-published papers in the top ten include Vandenplas et al‘s clinical insight on the relationship between asthma and cleaning agents, and Kalil et al’s review of hospital-acquired pneumonia.

Rank Author(s) Title
1 Werb et al The temporal relationship between drug supply indicators: an audit of international government surveillance systems
2 Johansson et al Common psychosocial stressors in middle-aged women related to longstanding distress and increased risk of Alzheimer’s disease: a 38-year longitudinal population study
3 Carpenter et al Bed sharing when parents do not smoke: is there a risk of SIDS? An individual level analysis of five major case–control studies
4 Taylor et al Prevalence and incidence rates of autism in the UK: time trend from 2004–2010 in children aged 8 years
5 Doyle et al A systematic review of evidence on the links between patient experience and clinical safety and effectiveness
6 Vandenplas et al Asthma related to cleaning agents: a clinical insight
7 Nybergh et al Psychometric properties of the WHO Violence Against Women instrument in a male population-based sample in Sweden
8 Ford et al Current treatments in diabetic macular oedema: systematic review and meta-analysis
9 Kalil et al Treatment of hospital-acquired pneumonia with linezolid or vancomycin: a systematic review and meta-analysis
10 MacCallum et al Patient safety and estimation of renal function in patients prescribed new oral anticoagulants for stroke prevention in atrial fibrillation: a cross-sectional study


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

Large medical research funders committed to open access publishing

24 Oct, 13 | by bdevaney

But smaller medical charities, although keen, fear impact on budgets and funded researchers.

Large medical research funding bodies are fully committed to open access publishing. But although smaller charitable funders back the principle, they worry about the impact open access will have on their budgets and their funded researchers, reveals a qualitative study published in BMJ Open.

The findings come as Open Access Week (October 21-27), a global annual event to promote open access as the norm in scholarly publishing and research, celebrates its seventh year in business.

The premise of open access is that it provides free, immediate, online access to the results of scholarly research, and ideally, the right to use and reuse those results as desired, rather than restricting access behind a subscription pay wall. The UK leads the world on open access.

Semi-structured interviews about open access policies and preferences were carried out with 12 employees at 10 UK biomedical research funding bodies in the Spring of this year in the public, charitable, and commercial sectors.

The results showed that all three sectors back open access, but that while public and charitable funders have clear policies on it, commercial sector funders don’t tend to.

Most funders are happy to support the “gold” route, whereby a study’s publication is paid for out of a research grant to cover the journal’s costs. This is the model used by BMJ Open and the other open access titles published by BMJ.

But funders said that not all publishers have embraced open access; many are resisting it and are either reluctant to move away from traditional publishing or are using it purely as a means to boost income.

Open access also seemed to be part of a funder’s overall mission – that is, improving health, healthcare, patient outcomes, and patients’ lives – although for many of the charitable funders, this had to be balanced against other types of public communication and support for people living with medical conditions.

But the way in which a funder gets its income influences its attitude to open access. Charities, which rely on public donations, had to be able to justify their costs and make the best use of available cash for all their competing objectives.

But research funders across the board were worried about the escalating costs of open access as the gold route becomes more mainstream and the cost of publishing shifts from institutions to funders.

This is of particular concern to the smaller charitable funders, who have historically not covered off these costs in their overheads, and fear they will need to stump up the additional costs required.

Other key findings:
• Funders find it hard to monitor compliance with open access, as a condition of funding; compliance rates vary enormously.
• Sanctions for breaching funding conditions are not harsh in the belief that it is more important to encourage researchers to comply rather than to punish them when they don’t.
• Research funders are turning their attention to other priorities for publication, including data sharing, research protocols, and negative results, which are not always favoured by more traditional outlets.

Sharmila Nebhrajani, Chief Executive of the Association of Medical Research Charities commented: “The public choose to donate to medical research charities to fund research into new treatments and cures, and they want their money to make a difference. To [do this], charities want everyone to be able to learn from their research findings, which is why they support the principle of open access.”

But this involved upfront costs, which were not always easy for charities to justify because of competing funding needs, she said.

“To make the biggest difference for their patients and supporters, funding open access may not be the best decision for all charities right now. We need to create an environment where it can be,” she said.

“This means prices that are clear and competitive so that charities can get the best deal and show how they are investing their donations to make a difference,” she explained.

The findings are discussed further in the accompanying podcast here.

Open access: the research funders’ perspective

24 Oct, 13 | by Richard Sands, Managing Editor


Anyone following progress on open access since the Finch Review report last year might think that the UK’s direction of travel is pretty much set. Since the Government and then Research Councils UK (RCUK) backed Gold open access, the same policy endorsed by the Wellcome Trust, it seemed likely that this would become much more prevalent for UK academics wishing to publish their work. Once a funder throws both its weight and its money behind a particular course of action, researchers have little reason not to comply.

But for medical research, the Medical Research Council (MRC) and the Wellcome Trust are not the whole story. Certainly, they are major funders and are widely recognised for leading the development of open access policy. But they are part of a wider community of charities, government bodies and commercial organisations who fund medical research. Many UK researchers receive funding from these organisations and, when those researchers come to publish their findings, they must comply with their funders’ rules.

That’s why BMJ commissioned a small qualitative study of ten medical research funders, designed to explore their approach to open access and communicating research findings. The study included the MRC and the Wellcome Trust, but also sought the views of the National Institute for Health Research (NIHR), charities of all sizes, and some pharmaceutical companies who invest in research. The aim was to understand the different policies and priorities of all these funders, in order to ensure that when a researcher submits to one of BMJ’s 50+ journals, they are offered a publication route that allows them to be compliant with their funder’s requirements.

The full findings of the study are available in an article that’s just been published in BMJ Open, or you can see a brief digest on the BMJ blog. If you read either, you’ll see that open access is welcomed by all of the funders in the study, but the route to implementation is causing them some concerns. There’s also a podcast with the report’s author, available here.

Meeting the costs, monitoring the move, supporting researchers as they transition from one model to another, and ensuring compliance, were issues for most of the funders we spoke to.

This is interesting because it really echoes some work BMJ undertook earlier this year with university staff. Back in January 2013, we ran a series of workshops for librarians, looking at some of the issues that they were facing when trying to implement RCUK’s open access mandate. We heard many of the same messages about money, monitoring, compliance and researcher support. These conversations informed the development of our open access membership model, which is designed to support institutions in meeting some of the challenges they described.

The key message that we heard from both funders and institutions is that researchers must remain in control of how and where they publish their work. Although they might want to see price become more of a factor in that decision, they are very uncomfortable with any suggestion that they should exert direct control over a researcher’s choice of journal. So, the onus is on publishers to help by ensuring that, whichever journal the researcher chooses, they can comply with any conditions imposed on their grant or employment contract. And that’s what we’re trying to do at BMJ.

New cases of autism in UK have levelled off after five-fold surge during 1990s

16 Oct, 13 | by flee

The number of newly diagnosed cases of autism has levelled off in the UK after a five-fold surge during the 1990s, finds research published in BMJ Open.

The findings differ from widely publicised results issued by the US Centers for Disease Control and Prevention (CDC) last year, which reported a 78% increase in the prevalence of the condition in eight year old children between 2004 and 2008 in the US.

Prompted by these data, which found that one in every 88 eight year old children in the US had been diagnosed with an autistic spectrum disorder in or before 2008, the authors wanted to find out if there were comparable rates in the UK.

They used entries into the General Practice Research Database (GPRD), which contains around three million anonymised active patient records from over 300 representative general practices in the UK – equivalent to 5% of the UK population.

Data from practices enrolled from 1990, when the GPRD was set up, were used to calculate the annual prevalence (number of people living with the condition) and the annual incidence (number of newly diagnosed cases) of autistic spectrum disorders among eight year olds, all of whom were born after 1996.

Annual prevalence rates for 2004-2010 were calculated by dividing the number of eight year olds diagnosed as autistic in that or any previous year, by the number of eight year olds enrolled in the database for each year.

Annual incidence rates were calculated by dividing the number of eight year olds who had been newly diagnosed with autism between 2004 and 2010 by the number of eight year olds enrolled into the database for each of those years.

The results showed that the annual prevalence and incidence of autism did not materially change over the entire study period, for either boys or girls.

The annual prevalence of autistic spectrum disorders was estimated at 3.8 per 1000 boys and 0.8 per 1000 girls, while the annual incidence was estimated at 1.2 per 1000 boys  (1190 in total) and 0.2 per 1000 girls (217 in total).

Girls were about 75% less likely to be diagnosed with an autism spectrum disorder as boys.

The UK prevalence of about 4/1000 children is substantially lower than the equivalent US figure of about 11/1000 children in 2008, which was reported in 2012.

“The large difference between countries is closely similar to differences in rates reported for children diagnosed and treated for attention deficit hyperactivity disorder in the two countries,” the authors point out.

Their previously published research, based on the same database, showed that the cumulative incidence of autism among children born in UK between 1988 and 1995 increased continuously by a factor of five during that period.

And they say that both studies provide “compelling evidence that a major rise in incidence rates of autism, recorded in general practice, occurred in the decade of the 1990s but reached a plateau shortly after 2000 and has remained steady through 2010.”

Similar widespread sharp rises in the number of children diagnosed as autistic were also seen in the 1990s in other parts of Europe and North America, they add, making it unlikely that better understanding of the condition or a broadening of the diagnostic criteria alone could have been responsible for these simultaneous large increases.

Given the apparent sudden halt in the rise in rates from early 2000 onwards – at least in the UK – the “actual cause of the dramatic rise in the 1990s remains a mystery,” they write, emphasising that the suggestion that it might be linked to the MMR vaccine has  been conclusively ruled out.

Journal policy on research funded by the tobacco industry

15 Oct, 13 | by Richard Sands, Managing Editor


This article is being copublished as an editorial in the BMJ, Heart, and Thorax and as a blog in BMJ Open. Full references, competing interests, provenance and citation details can be read here.

Fiona Godlee editor in chief [1], Ruth Malone editor, Tobacco Control [2], Adam Timmis current/outgoing editor, Heart [2], Catherine Otto incoming editor, Heart [2], Andy Bush coeditor, Thorax [2], Ian Pavord coeditor, Thorax [2], Trish Groves editor in chief, BMJ Open [2]
[1]BMJ, London WC1H 9JR, UK; [2]BMJ Journals, London WC1H 9JR, UK

As editors of the BMJ, Heart, Thorax, and BMJ Open we have decided that the journals will no longer consider for publication any study that is partly or wholly funded by the tobacco industry. Our new policy is consistent with those of other journals including PLoS Medicine, PLoS One, PLoS Biology[1]; Journal of Health Psychology[2]; journals published by the American Thoracic Society[3]; and BMJ’s own Tobacco Control.[4]

Critics may argue—as many did when journals stopped publishing cigarette adverts—that publishing such research does not constitute endorsing its findings and that, as long as funding sources are fully disclosed, readers can consider that information and make up their own minds about the quality of the work. Peer review should prevail, goes this line of thinking: it’s not the editor’s job to make these kinds of judgments. However, this view ignores the growing body of evidence that biases and research misconduct are often impossible to detect,[5] and that the source of funding can influence the outcomes of studies in invisible ways.[6, 7]

Underlying all the activity of peer review, editing, and publishing is the assumption that medical journals exist for the purpose of advancing knowledge that can be used to promote health and reduce disease. But the deputy editor of JAMA, Drummond Rennie, who has perhaps studied the process of scientific publishing longer than anyone, has written about what he calls “little murders.” These are deceptive publication practices that are “destructive of the delicate web of trust between colleagues that keeps the whole enterprise functioning and afloat.”[8] The editor’s job, observes Rennie, is to “try to separate the insufferable behaviors of mere jerks from the illegal actions of bona fide crooks.”

The tobacco industry, far from advancing knowledge, has used research to deliberately produce ignorance and to advance its ultimate goal of selling its deadly products while shoring up its damaged legitimacy.[9] We now know, from extensive research drawing on the tobacco industry’s own internal documents, that for decades the industry sought to create both scientific and popular ignorance or “doubt.” At first this doubt related to the fact that smoking caused lung cancer; later, it related to the harmful effects of secondhand smoke on non-smokers and the true effects of using so called light or reduced tar cigarettes on smokers’ health.[9-12] Journals unwittingly played a role in producing and sustaining this ignorance.9

Some who work within public health and who buy the notion of “harm reduction” argue that the companies that now produce modified cigarette products and non-cigarette tobacco products, including electronic nicotine delivery devices (e-cigarettes), are different from the tobacco industry of old, or that the tobacco industry has changed. For “hardened” cigarette smokers who can’t or won’t quit cigarettes,[13] the argument goes, new tobacco products could represent potential public health gains, and company sponsored research may be the first to identify those gains.

But one fact remains unassailably true: the same few multinational tobacco companies continue to dominate the market globally and, as smaller companies develop promising products, they are quickly acquired by the larger ones. However promising any other products might be, tobacco companies are still in the business of marketing cigarettes. As US federal court judge Gladys Kessler pointed out in her judgment in the case of US Department of Justice versus Philip Morris et al, the egregious behaviour of these companies is continuing and is likely to continue into the future.[14] And just this summer documents leaked from one company showed a concerted campaign to “ensure that PP [plain packaging of tobacco products, bearing health warnings but only minimal branding] is not adopted in the UK.”[15] The tobacco industry has not changed in any fundamental way, and the cigarette—the single most deadly consumer product ever made—remains widely available and aggressively marketed.

Editors of BMJ journals are committed to integrity in scientific publishing and to ensuring that—as far as possible—their journals publish honest work that advances knowledge about health and disease. Back in 2003, the editor of the BMJ defended publication of a study with tobacco industry funding saying “The BMJ is passionately antitobacco, but we are also passionately prodebate and proscience. A ban would be antiscience.”[16] But it is time to cease supporting the now discredited notion that tobacco industry funded research is just like any other research. Refusing to publish research funded by the tobacco industry affirms our fundamental commitment not to allow our journals to be used in the service of an industry that continues to perpetuate the most deadly disease epidemic of our times.



War Veterans, OTC Medications and Dementia: Most read articles in September

4 Oct, 13 | by flee

The most-read article in September was once again, Carpenter et al‘s article on Bed sharing and the risk of SIDS. Wakefield et al‘s study of Australian packaging policy stayed in the top ten for another month.  Other newly-published papers in the top ten include Brabers et al‘s survey investigating consumers’ attitudes towards over the counter medication.

Rank Author(s) Title
1 Carpenter et al Bed sharing when parents do not smoke: is there a risk of SIDS? An individual level analysis of five major case–control studies
2 Billings et al Choosing a model to predict hospital admission: an observational study of new variants of predictive models for case finding
3 McBride et al The mortality and cancer experience of New Zealand Vietnam war veterans: a cohort study
4 Shaw et al The work of commissioning: a multisite case study of healthcare commissioning in England’s NHS
5 Hope et al Prevalence of, and risk factors for, HIV, hepatitis B and C infections among men who inject image and performance enhancing drugs: a cross-sectional study
6 Ford et al Current treatments in diabetic macular oedema: systematic review and meta-analysis
7 Brabers et al Where to buy OTC medications? A cross-sectional survey investigating consumers’ confidence in over-the-counter (OTC) skills and their attitudes towards the availability of OTC painkillers
8 Griffiths et al How active are our children? Findings from the Millennium Cohort Study
9 Wakefield et al Introduction effects of the Australian plain packaging policy on adult smokers: a cross-sectional study
10 Gao et al Effects of centrally acting ACE inhibitors on the rate of cognitive decline in dementia


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

Plentiful mid-life stress linked to heightened risk of dementia in late life

30 Sep, 13 | by flee

Coping with a lot of stress in middle age may boost the risk of developing dementia in late life – at least among women – suggests research published in BMJ Open.

The response to common life events may trigger long lasting physiological changes in the brain, say the authors.

They base their findings on 800 Swedish women whose mental health and wellbeing was formally tracked over a period of almost 40 years as part of the larger Prospective Population Study of Women in Gothenburg, Sweden, which started in 1968.

The women, who were all born in 1914, 1918, 1922 and 1930, underwent a battery of neuropsychiatric tests and examinations in 1968, when they were in their late 30s, mid 40s, and 50s, and then again in 1974, 1980, 1992, 2000 and 2005.

At their initial assessment, the women were quizzed about the psychological impact on them of 18 common stressors, such as divorce, widowhood, serious illness or death of a child, mental illness or alcoholism in a close family member, personal or partner’s unemployment, and poor social support.

How many symptoms of distress, such as irritability, fear, and sleep disturbances, and how often they had experienced these in the preceding five years, were noted at every assessment.

In 1968 one in four of the women had experienced at least one stressful event;  a similar proportion (23%) had experienced at least two, while one in five had experienced at least three, and 16% four or more. The most commonly reported stressor was mental illness in a close family member.

During the monitoring period, 425 of the women died (at the average age of 79). Between 1968 and 2006, around one in five (19%, 153) developed dementia, 104 of whom developed Alzheimer’s disease.

On average, it took 29 years for dementia to develop, with 78 the average age at which the condition was diagnosed.

The number of stressors reported by the women was associated with longstanding symptoms of distress at all of the time points assessed, irrespective of the year of birth.

And the number of stressors reported in 1968 was associated with a 21% heightened risk of developing Alzheimer’s disease and a 15% heightened risk of developing any type of dementia, the analysis showed.

The findings held true even after taking account of factors likely to influence the results, including a family history of mental health problems.

The authors emphasise that further research is needed to confirm the results of their study, and to look at whether stress management and behavioural therapy might help.

But they suggest that “stress may cause a number of physiological reactions in the central nervous, endocrine, immune and cardiovascular systems,” and point to other studies showing that stress can cause structural and functional damage to the brain and promote inflammation.

Furthermore, research has also shown that stress hormones can remain at high levels many years after experiencing a traumatic event.