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.

Depression and personality disorders most common diagnoses in psychiatric patients requesting euthanasia

28 Jul, 15 | by Richard Sands, Managing Editor

 

Most common diagnoses among those requesting help to die, due to unbearable suffering

Depression and personality disorders are the most common diagnoses among Belgian psychiatric patients requesting help to die, on the grounds of unbearable suffering, finds research published in BMJ Open today.

Drugs, given either by mouth or administered intravenously, are used to perform euthanasia in Belgium, where the practice has been legal since 2002.

The researchers wanted to find out if there were any discernible patterns in requests for euthanasia among mentally ill patients in Belgium in a bid to inform recommendations for future research.

So they tracked requests for help to die, made by patients receiving treatment for psychiatric problems in outpatient clinics in the Dutch speaking part of Belgium between 2007 and 2011, and followed up to the end of 2012.

During this period, 77 women and 23 men asked for euthanasia on the grounds of unbearable suffering associated with mental illness. Their average age was 47, but this ranged from 21 to 80.

Most (91) of the patients had been referred for counselling. Seventy three had been deemed medically unfit to work, and 59 were living alone.

Ninety had more than one mental health issue, with depression (58 patients) the most frequent diagnosis, followed by personality disorder (50).

Thirty eight patients required further tests and/or treatment, 13 of whom were specifically tested for autistic spectrum disorders. Twelve were subsequently diagnosed with Asperger’s syndrome, a form of high functioning autism.

In all, 48 of the requests were accepted, and 35 carried out. Among the remaining 13, eight cancelled or delayed the procedure on the grounds that simply having the option gave them enough peace of mind to continue living.

By December 2012, 43 of the patients had died, including six who had taken their own lives. Among this group, one patient committed suicide because she found the approvals process too long, while another did so because her family had objected to euthanasia. A third woman killed herself after a spell in a psychiatric ward.

Another had died as a result of palliative sedation by the end of 2012, and one had died of the eating disorder, anorexia nervosa.

Thirty patients died surrounded by family/friends, and in a serene and positive atmosphere, “which would have been impossible to attain in the case of unassisted traumatic suicide,” note the authors.

In 2010 and 2011, 2086 patients died by euthanasia in Belgium, accounting for 1% of all deaths during that period, with those who were not terminally ill making up less than 10% of the total.

But as yet, there is no consensus on what constitutes ‘unbearable suffering,’ nor are there any guidelines in Belgium on how best to deal with requests for help to die from those who are mentally ill, say the researchers.

“Taking into account the ongoing fierce ethical debates, it is essential to develop such guidelines, and translate them into clear and detailed protocols that can be applied in practice,” they conclude.

A video abstract for this manuscript is available to view here: https://www.youtube.com/watch?v=jd1V3DQ_xKE

‘Successful ageing’ linked to harmful drinking among over 50s

23 Jul, 15 | by fpearson

The over 50s who are ‘successful agers’—healthy, active, sociable, and well off—are more at risk of harmful drinking than their less successful peers, concludes research published in the online journal BMJ Open.

Harmful drinking is a “middle class phenomenon” which may be a hidden health and social problem in otherwise successful older people, warn the researchers, who call for explicit guidelines on alcohol consumption for this group.

They base their findings on more than 9000 responses to the two most recent waves (2008-9 and 2010-11) of the English Longitudinal Survey of Ageing (ELSA)—a long term study of a representative sample of those aged 50 and above living independently at home in England.

Given that alcohol consumption is on the rise among older people in England, the researchers wanted to tease out the social and economic factors associated with harmful drinking, and changing patterns of consumption over time.

They used national guidance to define increasing risk of harmful drinking at 22-50 weekly units for men and 15 to 35 weekly units for women; and higher risk, at more than 50 and more than 35 weekly units, respectively, for men and women.

Survey participants were asked about a range of potentially influential factors: income; educational attainment; self reported health; whether they smoked; diet; physical activity levels; whether they felt lonely or depressed; ethnic background; marital status; caring responsibilities; religious beliefs; employment status; and social engagement (civic participation, networks of friends, cultural activities).

Analysis of the responses showed that the risk of harmful drinking peaked for men in their early 60s and then gradually tailed off, whereas for women risky drinking fell in tandem with age.

These patterns suggest that the current group of over 50s may be carrying on levels of higher consumption developed in their younger years, in later life, say the researchers.

Certain factors were linked to a heightened or lowered risk of harmful drinking.

Income was associated with a higher risk, but only among women, while smoking, higher educational attainment, and good health were all linked to heightened risk in both sexes.

Higher risk of harmful drinking was not linked to feelings of loneliness or depression, but it was more likely among men living on their own, including those who were separated/divorced. And it was more common among men of white ethnicity.

Caring responsibilities lowered the probability of being at higher risk among women, but religious belief did not—for either sex.

Employment status did not seem to be a significant factor, but women who had retired were more likely to be at higher risk.

When the researchers looked at changes in alcohol consumption between the two waves of the survey, they found that among women, loneliness, younger age, and higher income were all associated with the likelihood of becoming a higher risk drinker by 2010-11. A healthy diet seemed to lessen the risk.

Among men, these transition patterns were similar, except that caring responsibilities, loneliness, older age and lower income increased the likelihood of no longer drinking at risky levels by wave 2 of the survey.

“We can sketch—at the risk of much simplification—the problem of harmful drinking among people aged 50 or over in England as a middle class phenomenon: people in better health, higher income, with higher educational attainment and socially more active are more likely to drink at harmful levels,” write the researchers.

“Our findings suggest that harmful drinking in later life is more prevalent among people who exhibit a lifestyle associated with affluence and with a ‘successful’ ageing process,” they add.

“Harmful drinking may then be a hidden health and social problem in otherwise successful older people,” they warn, concluding: “Consequently, and based on our results, we recommend the explicit incorporation of alcohol drinking levels and patterns into the successful ageing paradigm.”

Drinking alcohol while pregnant is common in UK, Ireland, and Australasia

6 Jul, 15 | by fpearson

Drinking alcohol while pregnant is common, ranging from 20% to 80% among those questioned in the UK, Ireland, Australia and New Zealand, reveals a study of almost 18,000 women published in the online journal BMJ Open.

Women across all social strata drank during pregnancy, the findings showed. But expectant mums were significantly more likely to be drinkers if they were also smokers.

The researchers base their findings on an analysis of data from three studies: The Growing up in Ireland (GUI) study; the Screening for Pregnancy Endpoints (SCOPE) study; and the Pregnancy Risk Assessment Monitoring System (PRAMS).

The studies variously assessed the amount and type of alcohol drunk before and during pregnancy and involved 17,244 women who delivered live babies in the UK, Ireland, Australia and New Zealand.

The researchers mined the content to gauge the prevalence of, and the factors associated with, drinking alcohol during pregnancy.

Their analysis indicated a high prevalence of drinking, including binge drinking, among mums to be. The prevalence of drinking alcohol ranged from 20% to 80% in Ireland, and from 40% to 80% in the UK, Australia, and New Zealand.

Ireland emerged as the country with the highest rates of drinking, both before (90%) and during (82%) pregnancy, and of binge drinking, before (59%) and during (45%) pregnancy, based on estimates from the SCOPE study. But the exact prevalence could be far lower than that as estimates of drinking during pregnancy from the PRAMS and GUI studies were substantially lower (20-46%), with only 3% of women reporting binge drinking in PRAMS, caution the researchers.

The amount of alcohol drunk varied across the three studies. Between 15% and 70% of the women said that they had drunk 1-2 units a week during the first three months (trimester) of their pregnancy. But the number of reported units dropped substantially in all countries between the first and second trimester, as did binge drinking.

The findings indicated that the prevalence of drinking while pregnant was generally evident across all social strata, but several factors were associated with a heightened or lowered risk of alcohol consumption.

Compared with white women, those of other ethnicities were less likely to drink alcohol while pregnant, while younger women (30-39) were also less likely to do so than older women.

A higher level of education, having other children, and being overweight/obese were also associated with a lower risk of drinking while pregnant.

But the strongest and most consistent predictor of a heightened risk of drinking alcohol during pregnancy across all three studies was smoking. Smokers were 17-50% more likely to drink while pregnant.

The researchers point out that most clinical and government guidelines advise women to stop drinking during pregnancy.

But they write: “Alcohol use during pregnancy is highly prevalent, and evidence from this cross-cohort and cross-country comparison shows that gestational alcohol exposure may occur in over 75% of pregnancies in the UK and Ireland.”

However, most of these women consumed alcohol at very low levels and the number of pregnant women who drank heavily in the three studies was small, they say.Nevertheless, given that the risks of light drinking are not fully known, the most sensible option is not to drink alcohol during pregnancy, they add.

“Since most women who consume alcohol do so at lower levels where the offspring growth and development effects are less well understood [than at higher levels], the widespread consumption of even low levels of alcohol during pregnancy is a significant public health concern,” they conclude.

Weekend screen time linked to poorer bone health in teen boys

10 Jun, 15 | by flee

Weekend screen time is linked to poorer teen bone health—but only in boys, reveals research published in BMJ Open.

The apparent lack of impact of leisure screen time on teen girls’ bone health may be explained by their different body fat distribution, suggest the researchers.

They base their findings on participants in the Tromsø Fit Futures Study in Norway, which involved 961 of the region’s 15-17 year old school pupils in 2010-11 (first wave) and 688 (66%) of this original group two years later in 2012-13 (second wave).

At both time periods, the teens were quizzed in detail about their lifestyles, including how much time they spent on their computers or watching TV/DVDs at the weekend and outside of school hours during the week; how much they smoked and drank; and what they ate, collected by food frequency questionnaires to gauge calcium and soft drink intake—factors known to affect bone mineral density.

They were also asked about their average weekly levels of physical activity in the preceding year, which was graded into sedentary; at least 4 hours walking, cycling or formal exercise; at least 4 hours of recreational sports; and hard training/competitive sport several times a week.

The bone mineral density was assessed at the hip, top of the thigh bone (femoral neck), and the whole skeleton, and their vitamin D level was measured from blood samples. Height and weight measurements (BMI) were also taken.

The analyses showed that boys spent more time in front of any screen than girls, averaging around 5 hours a day at the weekend and just under 4 hours during the week. The equivalent figures for girls were 4 hours at weekends and just over 3 hours during the week.

While more time spent in front of a screen at the weekend was linked to lower levels of physical activity, one in five girls and one in four boys, who whiled away more than 4 hours on Saturdays and Sundays on screen time, also said they clocked up more than 4 hours a week on hard training or competitive sports.

Lower bone mineral density was linked to weekend screen time, but was only significant among boys, among whom bone mineral density was lower at all the sites tested. Among girls, the positive association was only evident for the femoral neck.

After taking account of potentially influential factors, such as age, the degree of sexual maturity, and weekday screen time, the association strengthened for boys.

Two to 4, or more than 6 hours, in front of a screen were linked to statistically significant reductions of bone mineral density at the femoral neck compared with boys clocking up fewer than 2 hours of screen time daily at the weekend.

But boys who spent 4 to 6 hours in front of a screen tended to have higher than expected bone mineral density levels.

The opposite was true of girls among whom 4-6 hours of weekend screen time daily was associated with higher bone mineral density, even though they took less exercise than those who said they spent less time in front of a screen.

All these trends persisted when the assessments were repeated after two years.

“These conflicting results may be related to different factors, as the relationship between fat and bone varies with age and hormones,” say the researchers.

This is an observational study so no definitive conclusions can be drawn about cause and effect.

But the researchers conclude: “Our study suggests persisting associations of screen based sedentary activities on bone health in adolescence. This detrimental association should therefore be regarded as of public health importance.”

Dissemination bias, biguanide and clinician self-management: Most read articles in May

4 Jun, 15 | by flee

This month’s Top 10 Most Read includes a selection of new entries; including Meerpohl et al’s article on evidence-informed recommendations to reduce dissemination bias in clinical research. Mudge et al’s article on clinicians view on their role in self-management approaches and Anholm et al’s paper on the therapy of biguanide in patients with coronary artery disease and newly diagnosed type-2 diabetes were popular additions to this month’s most read articles.

 

Rank Author(s) Title
1 Anholm et al. Adding liraglutide to the backbone therapy of biguanide in patients with coronary artery disease and newly diagnosed type-2 diabetes (the AddHope2 study): a randomised controlled study protocol
2 Parand et al. The role of hospital managers in quality and patient safety: a systematic review
3 Branas et al. The impact of economic austerity and prosperity events on suicide in Greece: a 30-year interrupted time-series analysis
4 Smith et al. Why do smokers try to quit without medication or counselling? A qualitative study with ex-smokers
5 Meerpohl et al. Evidence-informed recommendations to reduce dissemination bias in clinical research: conclusions from the OPEN (Overcome failure to Publish nEgative fiNdings) project based on an international consensus meeting
6 Mudge et al. Who is in control? Clinicians view on their role in self-management approaches: a qualitative metasynthesis
7 Hysing et al. Sleep and use of electronic devices in adolescence: results from a large population-based study
8 Deane et al. Priority setting partnership to identify the top 10 research priorities for the management of Parkinson’s disease
9 Kato et al. Diagnosed diabetes and premature death among middle-aged Japanese: results from a large-scale population-based cohort study in Japan (JPHC study)
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.

Many UK patients with gonorrhoea prescribed outdated antibiotics

28 May, 15 | by flee

Many UK patients with gonorrhoea are being prescribed antibiotics that are no longer recommended for treating the infection by their family doctor (GP), reveals research published in BMJ Open.

This failure to keep abreast of national clinical guidance is of concern, given the global threat of antibiotic resistance, say the researchers.

They base their findings on an analysis of electronic health records entered anonymously into the Clinical Practice Research Datalink— a large database containing the health records of around 5.5 million patients registered with 680 general practices around the UK—as well as information from anonymous monitoring of sexually transmitted infections in England.

They looked particularly at how doctors in general practice had treated the two most commonly diagnosed bacterial sexually transmitted infections in England, Chlamydia and gonorrhoea, between 2000 and 2011.

GPs diagnosed an estimated 193,000 people with Chlamydia and nearly 17,000 with gonorrhoea during this period, accounting for between 9% and16% of all Chlamydia cases and between 6% and 9% of all gonorrhoea cases in England.

The number of diagnoses GPs made for Chlamydia rose substantially from 22.8/100,000 of the population in 2000 to 29.3/100,000 of the population in 2011. And the proportion of patients treated for this infection rose from around six in every 10 (60%) to almost eight out of 10 (78%).

Most (90%) were prescribed an antibiotic recommended in national clinical guidance. But this was not the case for gonorrhoea.

The number of diagnoses fluctuated between 3.2 to 2.4/100,000 of the population, while the proportion treated ranged between a third (just under 33%) and just over half (54%).

Despite being discontinued as a recommended treatment for the infection in 2005, ciprofloxacin continued to be prescribed. This antibiotic accounted for more than four out of 10 prescriptions (42%) in 2007, and one in five in 2011.

The bacterium that causes gonorrhoea, Neisseria gonorrhoeae, is adept at developing resistance to the antibiotics used to treat it, and the evidence from other research shows that over a third of gonorrhoea infections treated at sexual health clinics were resistant to ciprofloxacin, for example, while up to one in five cases may be resistant to penicillin.

The researchers conclude that GPs make an important contribution to the diagnosis and treatment of bacterial sexually transmitted infections, but while most patients with Chlamydia are treated appropriately, “significant numbers” of those infected with gonorrhoea are not.

“Treatment of infections with reduced susceptibility or resistance to the prescribed therapy may inadvertently facilitate onward transmission and risks infection complications,” they write.

“Antimicrobial resistance in gonorrhoea is a global problem and may become an issue for Chlamydia in future,” they warn. “Practitioners should be alert to the likelihood of revisions to national treatment guidelines and of treatment failure in their patients.”

Oldest old less likely to be investigated or aggressively treated after surgery

25 May, 15 | by flee

Patients aged 80 and above are significantly less likely to be investigated or aggressively treated after surgery than their younger counterparts, reveals a national audit of hospital deaths, published in BMJ Open.

This is despite the fact that the oldest old have higher rates of trauma and multiple underlying conditions on admission, say the Australian researchers.

Care in the oldest old may be less aggressive, or scaled down because the outcome is expected to be poor or treatment considered futile, they say. Perceived future quality of life issues may also be a factor.

The researchers assessed data from a national audit of deaths after surgical procedures in every specialty carried out between 2009 and 2012 in 111 public and 61 private hospitals across Australia. This included one in five private hospitals and virtually all public teaching hospitals (99%).

In all, nearly 20,000 surgical patients died in hospital. Around half of these deaths were excluded because of data issues or because they occurred in people classified as brain deaths, or in those younger than 17 or in those who had had no surgery.

Some 11,201 were included in the final analysis. This group was divided into three age bands: 17-64; 65-79; and those aged 80 and above.

The researchers looked at the relationship between age, preoperative factors, such as multiple underlying conditions or trauma, and eight indicators of postoperative care.

These indicators were: fluid balance problems; return to theatre; unplanned intensive care unit admission; treatment in an intensive care unit; clinical issues; postoperative complications; the presence of infection at death; and whether, with hindsight, the surgeon would have managed the patient differently.

The average age of those who died soon after a surgical procedure was 78, and just under 44% (4892) were aged 80 and above. Of those who died in hospital, most had been admitted as emergencies (83.4%). Nearly half (45%) had an incapacitating and life threatening disease on admission.

The oldest old had higher rates of admission as a result of trauma or other emergency than either of the two other age groups. But they were treated differently, receiving lower levels of aggressive and expensive treatment.

They had around half the rate of unplanned returns to theatre (11.2%) of those aged 65-79 (20.2%). They were also less likely to have unplanned admissions to intensive care (16.3% vs 24%) and less likely to be treated in intensive care (59.7% vs 76.7%).

On average, the most elderly spent fewer days in hospital than those aged between 65 and 79 (9 vs 11 days) but more than the youngest (8 days).  And they were less likely to have postoperative complications diagnosed and reported than 65-79 year olds.

Those aged between 65 and 79 also had a higher prevalence of cancer, which may have explained their higher rates of admission to intensive care, a move “that is often rationalised due to high operational costs whatever the age group,” they write.

Surgical care for older patients tends to be complex, because of functional, physiological, psychological, and social factors, which are likely to influence decisions about their hospital care, say the researchers.

None the less, they point out that the oldest old had the lowest rate of diagnosed postoperative complications of all the age groups, despite virtually all of them having multiple underlying conditions, which are usually associated with a higher risk of problems arising after surgery.

“Our data suggest that there may be a culture of less intensive investigation, monitoring and possible failure to intervene in the elderly group,” they say, adding that the costs of surgery may rise more slowly than expected if older people continue to be given less aggressive postoperative care.

Should we stop talking about ‘negative’ results?

22 May, 15 | by Richard Sands, Managing Editor

 

BMJ Open recently published the results of a major EU-funded project (OPEN) investigating the problem of dissemination bias. Also know as publication bias, this is the distortion of the evidence base caused by selective or non-reporting of results. The authors concluded:

‘Despite various recent examples of dissemination bias and several initiatives to reduce it, the problem of dissemination bias has not been resolved.’

Many journals, including BMJ Open, have peer review processes that don’t judge importance, novelty or impact, making publishing so-called negative studies easier.

Still, the perception remains that authors find it harder to publish these studies or remain reluctant to do so. If this is true, how important is the terminology we use? Would more ‘negative’ results be published if we called them something else?

The term ‘negative result’ is used widely.  The EU OPEN project used it (OPEN Stands for ‘Overcome failure to Publish nEgative findings’). It appears in journal titles, such as the Journal of Negative Results in Biomedicine. There’s a PLOS negative results collection. Professor Stephen Curry from Imperial College wrote about the importance of negative results for UK newspaper The Guardian.

BMJ Open encourages researchers to publish their studies, especially trials, regardless of results. It’s a premise of the AllTrials campaign, initiated by BMJ. In this spirit, we recently promoted this paper on Twitter. In it, Andrew Wilson and colleagues at the University of East Anglia report a randomised controlled study of maintenance sessions following pulmonary rehabilitation in patients with chronic obstructive pulmonary disease. They concluded “We do not recommend that our maintenance programme is adopted.”

To promote the paper and our willingness to consider this kind of study, we tweeted the paper with the comment that it was great that the authors were publishing their RCT ‘despite [a] negative result’.

The feedback from researchers was interesting. “Don’t talk about ‘negative results’ in trials” said Ruairidh Milne (@ruairidhm). “Best not to talk about ‘negative’ and ‘positive’ results in trials” said Shaun Treweek (@shauntreweek); “An important result rather than negative.” Andrew Cook (@ajcook) commented “I meet people who think trials which show now [sic] effect are a waste – calling them negative doesn’t help arguing how useful they are”.

All good points and we haven’t even covered that there are different types of ‘negative’ result (papers showing evidence of the absence of an effect or papers reporting an absence of evidence can both be labelled ‘negative’).

So – should we be thinking about what we call studies like these? Is the term ‘negative result’ just too… negative? Could it be influencing authors (or editors) unduly and if so what should we be calling these studies?

Increasing the value of health research, e-cigarettes and the role of hospital managers in patient safety: Most read articles in April

7 May, 15 | by flee

April’s Top 10 Most Read includes a selection of new entries; including a bibliometric analysis on increasing the value of health research in the WHO African Region by Uthman et al and an article on non-health decision-makers’ use of evidence on the social determinants of health written by McGill et al. Parand et al’s systematic review on the role of hospital managers in quality and patient safety has made it’s way back up to the top of the table.

Rank Author(s) Title
1 Parand et al. The role of hospital managers in quality and patient safety: a systematic review
2 Uthman et al. Increasing the value of health research in the WHO African Region beyond 2015—reflecting on the past, celebrating the present and building the future: a bibliometric analysis
3 Parsons et al. What the public knows and wants to know about medicines research and development: a survey of the general public in six European countries
4 Moore et al. Electronic-cigarette use among young people in Wales: evidence from two cross-sectional surveys
5 Deane et al. Priority setting partnership to identify the top 10 research priorities for the management of Parkinson’s disease
6 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
7 Hysing et al. Sleep and use of electronic devices in adolescence: results from a large population-based study
8 Cripton et al. Severity of urban cycling injuries and the relationship with personal, trip, route and crash characteristics: analyses using four severity metrics
9 Rachiotis et al. What has happened to suicides during the Greek economic crisis? Findings from an ecological study of suicides and their determinants
10 McGill et al. Trading quality for relevance: non-health decision-makers’ use of evidence on the social determinants of health

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

UK doctors unlikely to be able to repay student loans

20 Apr, 15 | by fpearson

UK doctors are unlikely to be able to repay their student loans over the course of their working lives, amassing debts of more than £80,000 by the time they graduate, in some cases, finds research published in the online journal BMJ Open.

What’s more, there are clear gender differences in the amount of cash required to service these debts, the analysis shows, with women paying more in interest, despite earning less than men.

The researchers base their findings on the average earnings of 4286 doctors working more than 30 hours a week, who had taken part in national Labour Force Surveys between 1997 and 2014.

Annual tuition fees amount to £9000 for English students attending UK universities, and most medical degrees take five years to complete.

Since 2012, students have been able to borrow the annual fees and get a maintenance loan to cover living costs from the Student Loan Company.

Repayments are charged at an annual interest rate of 3% plus annual inflation rate (Retail Price Index) and based on 9% of salary earned above £21,000 gross income. Debts are automatically written off after 30 years, irrespective of the sums outstanding.

A medical student graduating in 2014 would have clocked up a debt of £40,000 for tuition fees.
If maintenance loans are factored in, this would add £24,000 for a student living at home; an additional £30,000 for a student living away from home; and £42,000 for a student living away from home in London, amounting to a total of between £64,000 and £82,000 by the time of graduation.

The survey responses showed that average full time salaries rose with age, but then gradually fell after the age of 55, with a wide gap in earnings starting to emerge between men and women from the age of 30 onwards.

At the age of 55, male doctors earned 35% more than their female colleagues, which was mainly attributable to hourly wage rates rather than the number of hours worked.

The researchers used the average age-salary profiles, projected future repayments, and cumulative debt levels to calculate the total sums required to service the loans.

For those borrowing against tuition fees alone, full time male doctors would have to stump up £57,303 to clear their debts over 20 years, while their female colleagues would need to find just short of £62,000 over a period of 26 years.

When maintenance loans were factored in, the researchers calculated that the total sums to be repaid added up to £75,786 for an initial debt above £46,000 for women doctors and £110,644 for an initial debt above £65,145 for their male colleagues.

For initial debts below £50,000, women repay more, despite earning less, because their debt lasts longer and accrues more interest. But for initial debts above £50,000, men repay more because their average yearly salaries are higher.

The researchers point out for those able to repay their loans within the 30 year timeframe, a higher yearly salary is advantageous because an early repayment means less interest. But for those unable to pay off their loans, a lower salary is advantageous, because it means lower yearly repayments and the writing off of the residual debt.

“It seems reasonable that these repayment variations may actually exist across many graduate careers in the UK,” write the researchers. “It is also apparent that at the current level of fees, even small changes in the student loan contract will have substantial implications for lifetime wealth across different income groups, across male and female graduates, and on the sustainability of the student loans system.”