Main content

Hospital admissions and the 'weekend effect', Peyronie's disease

Politicians have quoted research claiming that people admitted to hospital at the weekend are more likely to die, and Dr Mark Porter reviews the evidence for a seven-day NHS.

Dr Mark Porter unpicks the science behind the so called 'weekend effect'. Politicians have quoted research claiming that people are 20% more likely to die of a stroke at the weekend, while another much cited study finds 11,000 more deaths in people admitted at the weekend. But how valid are these figures and the research that generated them? Dr Margaret McCartney reviews the stroke data that has been criticised by experts as being out of date. While Mark Porter talks to Editor of the BMJ, Fiona Godlee, who published the 11,000 figure but is concerned about the political use of the findings. And discusses the study with lead author Nick Freemantle, plus Consultant Surgeon Sam Nashef who is sceptical about the results.

Available now

28 minutes

Programme Transcript - Inside Health

Downloaded from www.bbc.co.uk/radio4 

THE ATTACHED TRANSCRIPT WAS TYPED FROM A RECORDING AND NOT COPIED FROM AN ORIGINAL SCRIPT.  BECAUSE OF THE RISK OF MISHEARING AND THE DIFFICULTY IN SOME CASES OF IDENTIFYING INDIVIDUAL SPEAKERS, THE BBC CANNOT VOUCH FOR ITS COMPLETE ACCURACY.

 

 

INSIDE HEALTH

 

Programme 2. - Hospital admissions and the 'weekend effect', Peyronie's disease

 

TX:  19.01.16  2100-2130

 

PRESENTER:  MARK PORTER

 

PRODUCER:  ERIKA WRIGHT

 

 

Porter

Everyone looks forward to the weekend. Right?  Well, not if you are unwell and need to be admitted to hospital, thanks to the spectre of the so called “weekend effect” which, if you believe the politicians, means you could be forgiven for thinking you may never come out again!

 

This week it’s all about strokes and even the Prime Minister joined the fray:

 

Clip – David Cameron

I don’t want to blind you with statistics but if you have a stroke at the weekend you are 20% more likely to die.  And so it is important to try and build a more seven day NHS.

 

Porter

But strokes are not the only condition implicated in the weekend effect, they’re just the tip of a much larger iceberg, according to Secretary of State for Health Jeremy Hunt.

 

Clip – Jeremy Hunt

For England we are absolutely determined to eliminate the weekend effect which sees 11,000 excess deaths every year as a result of inadequate cover at weekends and we don’t want that to continue.

 

Porter

That was last October and the ensuing debate in the media over the last few months has only added to public concern, not least because much of it has been muddled and confusing.

 

So, how valid are these figures, and the research that generated them?  It’s an important question because if the weekend effect is that pronounced then the public are right to worry, and things need to change. And if weekends are not that dangerous then why is the research being used to justify government plans to alter junior doctors’ contracts to make it easier for hospitals to rota more of them at the weekend?

 

Well to answer the question we need to unpick the science behind the effect in order to clarify what it does and what it doesn’t tell us. And to help us do that we’ll be joined by the Editor of the BMJ, along with the lead author of the “weekend effect” paper and a sceptical surgeon.

 

But first Inside Health’s Dr Margaret McCartney is in our Glasgow studio. Margaret, let’s start with the claim that you are 20% more likely to die from a stroke at the weekend. Where has that figure come from?

 

McCartney

This was from a study published in the journal Public Library of Science last year.  And on the face of it the numbers are correct.  The problem is the numbers are now very much outdated.  Their study looked at people who’d had strokes in the UK between 2004 and 2012 and the problem is that over that time, and even now, stroke care in this country has been revolutionised – we have these things now called stroke units, dedicated hospital wards where people with stroke are admitted immediately to, they’re given very rapid scans, thrombolysis – clot busting drugs if they’re needed – rapid physiotherapy, rapid speech and language assessments, swallow assessments.  Care of stroke patients in this country really has changed dramatically over the last 15 years or so and I think it is really folly to start to think that we have a problem now that has not been changed because I do think that we cannot compare what’s happening now to what was happening then.

 

Porter

So what you’re saying is that going back a decade or more ago – I mean being admitted to an expert stroke unit that worked a seven day service was the exception rather than the rule but actually today it’s the other way round – most people get the full Monty thrown at them.

 

McCartney

The figures would bear that out.  So in 2002 a minority of people were being admitted into stroke units, now it’s very much the exception that if you’re not – and again treatments have changed too – in 2008 a minority of people, just 1.8% of people, got clot busting drugs; in 2014 that was 12% and these drugs are not used for everyone, only certain circumstances.  But you can see a huge change in that as well.  Now it’s very interesting – this level of really fantastic care has been provided within the current system of contracting doctors to the NHS, there’s no extra contract that’s going on there.  And we know, as well, there was an association between nursing staffing ratios rather than consultant ward rounds, in terms of mortality rates, again and stroke units are there to protect that, they’re trying to get the staff ratio up, so they’ve got more staff to look after and give hands on care to patients.

 

Porter

So just to be clear, this study does show an increase in death rates amongst people who are in care at the weekend but it’s outdated is what you’re saying.

 

McCartney

It’s outdated plus it’s association not causation, so we’ve talked about this before on the programme, there does seem to be a link, but what does that link mean?  And the other problem is that different kinds of people might be coming in at the weekend, people with more severe strokes, people with less severe changes on their brain scans might be being picked up more often during the week for example.  It’s really important that people know that there is care available 24/7 in the NHS for emergency situations such as stroke.  I have come across people who are afraid to go into hospital because they’re afraid that if they go in on a Saturday or a Sunday they have got an increased risk of dying.

 

Porter

Thank you very much Margaret. And a group of stroke experts have said very much the same in an open letter stating that recent advances in stroke care have made that 20% weekend effect a thing of the past. And they’ve urged patients not to be put off seeking help on Saturday or Sunday as rapid treatment is key to a successful recovery.

 

So that’s stroke,  but what about the other more established claim that reduced staffing is behind thousands of extra deaths from myriad other causes among people admitted over the weekend.

 

Clip – Jeremy Hunt

What we do need to change are the excessive overtime rates that are paid at weekends that give hospitals a disincentive to roster as many doctors as they need at weekends which lead to those 11,000 excess deaths.

 

Porter

That 11,000 figure comes from a paper published in the British Medical Journal last autumn, but there has been widespread concern that its findings have been misrepresented. A concern shared by the Editor of the BMJ, Fiona Godlee, prompting her to write to the Secretary of State.

 

Godlee

I wrote the letter because I became increasingly cross with the fact that Jeremy Hunt was saying that a BMJ report had found that excess deaths, increased deaths, avoidable deaths were happening at the weekend because there weren’t enough doctors on duty.  And the paper that we published did find excess deaths from people admitted at the weekend compared with those admitted in the week but there was no way that that paper could show what was causing that excess of deaths.  And certainly no way that you could jump to the conclusion that it was because there weren’t enough doctors on duty, there are many factors that could contribute to that finding.

 

Porter

And that’s what you said in your letter?

 

Godlee

And that’s what I said in my letter.

 

Porter

May I ask what reply you had?

 

Godlee

I had a perfectly civil reply about a few weeks later saying that his comments had been of a more general nature, wanting to draw attention to the problem of weekend effect care being less good than in the weekdays and he didn’t really address the exact issue which was the misuse of the statistics.

 

Clip – Jeremy Hunt

The British Medical Journal did an independent study that said that we have 11,000 excess deaths because we do not staff our hospitals properly at weekends.

 

Porter

Because this is the problem, as far as I can see, just as a casual observer, you hear time and time again that the weekend effect has almost become synonymous of the fact that there aren’t enough doctors there and therefore you’re more at risk, that’s what people understand – the general public understand by the weekend effect.  How do you feel about that?

 

Godlee

I think it is wrong to draw that conclusion, doctors already work at weekends, there are many other reasons including patients being sicker when they’re admitted to hospital at weekends.  So I think it’s entirely wrong to highlight this one aspect of need to improve provision.

 

Porter

Where did this study first come from?

 

Godlee

Bruce Keogh, who is the Medical Director of the NHS, has been looking at this problem for some time.  Evidence of excess weekend deaths emerges on a number of fronts around the world and he wanted to look at whether it was happening in the UK.  A study that was published in 2012 by Bruce Keogh and some colleagues found, yes, an excess deaths in people admitted at weekends and these patients are more at risk of dying in the subsequent 30 days.  More recently he was asked, and wanted to anyway, update that analysis, so they looked at more recent data, the same team, published those findings in the BMJ…

 

Porter

You said Bruce Keogh was asked, was that by the Department of Health?

 

Godlee

My understanding it was asked by Simon Stevens, who’s the Chief Executive of NHS England, that distinction may seem unnecessary to make but the NHS England is an independent organisation that works closely with the Department of Health.  As a paper it went through our standard procedures, it was peer reviewed, edited, published, press released, we’re providing the peer review comments so that people can look at those and we’ve just had a request for the underlying statistical analysis code, which we will certainly ask the authors to provide.

 

Porter

Did you anticipate back then, when you decided to publish this study, that it was going to create quite the storm that it has?

 

Godlee

I think we knew that the political environment was such that it would be very hotly debated and could well become a bit of a political football.

 

Porter

Did you anticipate though back then how they were going to take this leap and blame this weekend effect on staffing levels for instance?  Because that wasn’t clear from the paper, that’s not the conclusions that the authors themselves drew.

 

Godlee

That’s not the conclusion that the authors draw and I don’t think we anticipated exactly that spin on the paper.  We thought that it would become a discussion about what needed to happen and I wish that discussion had been much more sophisticated and open and less heated.  I think we have to acknowledge that there is this weekend effect, it’s been found in study after study in the UK and in other countries.  What distresses me is that in the UK there was a very, very good clinician led, patient involved, process, ongoing, started five, six years ago, to try to address this problem, looking at the evidence, what was needed, consultants were on board, the Royal Colleges were on boards, all the professions were beginning to work together to think what shall we do.  All of this was happening and was going on extremely well.  Bruce Keogh the Medical Director of the NHS was driving it and my sense is that this politicisation of the issue has derailed that process and hijacked it to the detriment of the NHS and to patients.

 

Porter

So you see the process not being hastened by political intervention but actually it may have stalled?

 

Godlee

I think political intervention has damaged this process.

 

Porter

BMJ Editor Fiona Godlee who remains unhappy about the way the research into the weekend effect is being used.

 

But what about the quality of the study itself? Nick Freemantle is Professor of Clinical Epidemiology and Biostatistics at University College London and the lead author of the BMJ paper. And Sam Nashef is a consultant heart surgeon at Papworth Hospital with a special interest in risk.

 

Nashef

Can I just say that I fully support seven day working and I believe that all services should be available to everybody seven days a week as long as they’re funded properly.  So I don’t have a personal axe to grind.  But when I read the paper I was struck by the number of limitations in this paper that could actually modify any conclusions you could take from it.  It looked at every single hospital admission and the vast majority of hospital admissions, not only do they not die, they’re not at risk of dying.  When you go to hospital to have a minor procedure or a dressing change you’re not going to die from that.  So looking at all hospital admissions in order to detect differences in the death rate, it’s a bit like trying to measure the size of a brussel sprout by measuring every conceivable vegetable on the market.  So I think it was the wrong population to study.

 

Porter

Looking at this sort of headline – the figure that came out of it was this 11,000 extra deaths attributable to the so-called weekend effect, if you were admitted over the weekend.  What you’re saying is that you don’t believe it.

 

Nashef

No of course I don’t believe it.  The first thing that was found was that there was a slightly higher risk of death for patients admitted at weekends.  But then we also know that the patients admitted at weekends had a much greater proportion of emergency admissions.  Now these are the patients who are truly sick and who are more likely to die.  Now it’s true that the – Nick and his colleagues – did try to adjust for that by applying a risk model but I know quite a lot about risk models and I’m not aware of any risk model that can predict with any sort of accuracy the likely rate of death from an ingrown toenail and a heart attack, I mean I just don’t think that such a model exists and if it does I would like to look at it and analyse it properly and see how robust it is.  So I’m not entirely sure that the risk adjustment was sufficient to eliminate the difference between a group that come in at weekends, many of whom are emergencies and really sick and a group who come in during the week for just a test or a minor procedure.  I think the two populations are different, we’re comparing apples and oranges here.

 

Porter

Nick, two criticisms there.  One is that you were looking at too big a picture to work out what was happening in this small group – the brussel sprouts on the plate let’s say.  The other thing is that this 11,000 figure it’s an educated gestimate.

 

Freemantle

Our starting point was to produce a statistical model which predicting the likelihood that a person who was admitted to hospital died within 30 days of that admission.  We used lots of pieces of information that describe the risk that that person faces.  The model is doing a fantastic job of identifying who’s at risk of dying.  People die, it’s a sad truth, so we’re not at this stage attributing any cause for that, we’re just saying can we identify those likely to die.  Now it’s a common misconception amongst people who are not statisticians to think that we need to have everybody who’s the same in a statistical model.  Actually we do better if we have a range of characteristics and we include people who are at low risk and people who at high risk.  So if we’d only included people at very high risk it might have been very difficult to disentangle the risk between them, plus we had very low risk people and very high risk people, we can do a straightforward job to divide things between them.  Survival analyses are not driven by the people who are in the model at risk, they are driven by the events that occur.  So the idea that appeared from clinicians commenting on the paper that somehow we’d diluted the effect by including people at very low risk of dying, who didn’t subsequently die, simply isn’t the case.  The reason we wanted to include all patients is that 19% of the admissions that end in death are not emergency admissions.  Now one of the things that has disturbed me about this paper, which we really tried to write it as clearly as possible, is that we have a clearly identified increase in risk for a Friday, for example – being in hospital on a Friday…

 

Nashef

And a Monday.

 

Freemantle

… is at increased risk.  Friday’s a 2% increase in risk, which is strongly statistically significant and nobody’s discussed it at all.

 

Nashef

Okay, may I ask another question then about it?

 

Freemantle

Please do.

 

Nashef

Can you remember the median length of stay for all patients – it was one day, one day, so we’re looking at some several million, 15 million admissions, the median length of stay for them was one day, the median length…

 

Freemantle

And your point is?

 

Nashef

… my point is the median length of stay for those who died was nine days, which means that if you died chances are you would have been nine days is more than seven and therefore you would have been in hospital at a weekend anyway.

 

Freemantle

And your point is?

 

Nashef

My point is I don’t see how this can actually be used to say that the figures that you found in the title of the paper are a case for expanded seven day services because – I hugely admire the work that you’ve done but because of all the issues I have raised I do not see this can be seamlessly made into a case for seven day services and unfortunately it has been.

 

Freemantle

I can’t comment on the use that others may have put our paper to but I can comment on our intentions in publishing it, my intentions in publishing it.  We published this as an analysis paper in the BMJ, it is not a full scientific paper, we deliberately published it with an idea of contributing to the on-going debate around the recognised weekend effect in the NHS and in other healthcare systems.  You will see that there’s a question mark at the end of the title that says – A case for expanded seven day services?  The intention here was to have a useful and constructive debate.  You raise a point about the minimum – the median length of stay being short.  Yes the median length of stay is short, the median length of stay can be short for two reasons.  It can be short because people are well and they’re sent home, it can be short because people die.  I’m afraid the idea that most people didn’t spend very long in hospital, and those were also mostly the ones who didn’t die, the idea that that somehow undermines the massive effect that we found associated with weekend admission is frankly denying what is an obvious truth – there is…

 

Nashef

On the contrary – I’m afraid – I’m afraid that is probably one of the most important points that I would make in that the majority of patients stayed one day.  These are patients coming in for elective simple stuff and the way the NHS is structured at the moment they don’t come in at weekends, so we’ve taken the safe bunch out of the weekends.  And it comes back to we are comparing two different groups with risk adjustments which in my opinion – and I accept that you perhaps disagree with this – but in my opinion your risk adjustment is not robust enough to distinguish between the differences between these groups.

 

Freemantle

Well if we are to discard our analysis on the basis of an inadequate risk adjustment I’m afraid I have to discard pretty much every piece of work that I’ve ever been involved in before this and actually pretty much every piece of work that’s ever crossed my desk as a referee, as an editorial advisor, as a senior statistician and probably every guideline in the land.  This is predicting with incredible precision who is going to die and who isn’t going to die.

 

Nashef

I think patients who are admitted at weekends have a higher chance of dying and I think the reason for that is because of the kind of patient that they are.  I don’t think we have proven that it is due to the service that the NHS provides.  It may be but I don’t think…

 

Porter

But that’s not what this study was trying to do.

 

Nashef

Yes exactly, yeah.  And by all means we should look a little bit further into why patients who are admitted at weekends have a higher mortality rate.  But I don’t think that this paper actually identifies the cause.

 

Freemantle

Sam, can I check with you because before – I would agree with you the paper doesn’t identify the cause beyond identifying it as being weekend.  Now could I make a comment about what you’re asking for – the model that looks at specific groups of patients?  Well actually that’s exactly what our model does – it looks at many, many groups of patients but then it aggregates a result on top of that.  So a clear limitation of our model is it doesn’t tell us within a particular specialty or within a particular diagnosis whether there’s a weekend effect, it tells us the overall effect for the NHS.  But what you’re asking for is in the engine room of our study we have all of those strata of different diagnostic groups.

 

Porter

Nick, there’s been a lot of debate about this paper in the media, I mean the concern has been that this is an independent study, and we heard Fiona Godlee describe it as an independent study, but it was commissioned by NHS England and Bruce Keogh and at least one of the other authors works for NHS England, was that an issue for you?

 

Freemantle

Our paper says that the article arose from a request by Bruce Keogh to update our previous analyses.  It says NF wrote the first draft of the paper and the truth is that the first draft of the paper is the one that contains the results.  So that was all put together independently beyond the request from Sir Bruce to update this analysis.  But the last time we published almost identical results we received almost no attention at all.

 

Porter

Don’t you find that interesting, I mean that interests me a lot, I mean 2012 I think was your initial study, basically the same conclusion…

 

Freemantle

Well you’d go further – the same conclusion.

 

Porter

The same conclusion.  And very little coverage.  But now that seven day NHS is much more on the political agenda it’s got a lot more coverage and I suppose that’s been the concern amongst people that it’s been a politically convenient study, not suggesting that that’s your motivation at all but that’s the way it’s been used.

 

Freemantle

I believe our paper was very clear.  If we have misled people in the words that we’ve used I apologise and we’ll do our best to clear it up.  But what people feed back to me is actually what we wrote was very clear.  We said that we couldn’t attribute a cause to this effect but it was a very large effect.

 

Porter

Professor Nick Freemantle and Consultant Surgeon Sam Nashef . And there is a link to the original BMJ paper on the Inside Health page of the Radio 4 website.

 

Coming up in a moment we answer a listener’s query about an embarrassing problem, so if you’ve tuned in hoping to catch the item on blood pressure – how low should we go? – that I trailed last week then I am sorry but we couldn’t fit it in today but we will be airing it soon.  I can, however, offer you some compensation in the form of Claudia Hammond who has popped in to tell us about the All In The Mind Awards.  Claudia there’s still time to enter?

 

Hammond

There is just time, the closing date is midnight on January 31st, so there is just time for people to enter if they want to nominate somebody who has helped them with their mental health problem.  So it can be an individual who’s helped them, it could be a friend or family member, it could be a professional – it could be a psychologist, a GP, a nurse, a social worker – anyone who’s helped them.  Or it could be a group or an organisation that’s helped them, somewhere they went to meet a group of people and this made the real difference to their mental health problem.

 

Porter

And how do they enter?

 

Hammond

They can enter by going to the All in the Mind page of the Radio 4 website and a panel of judges will be picking the winners and there’ll be a ceremony in June.

 

Porter

Claudia, thank you very much.

 

Now back to our listener with that embarrassing problem:

 

Listener

I noticed that there was a bend in my penis when erect and that it was tender when flaccid.  During intercourse it would range from uncomfortable at best to painful at worst.  Obviously the psychological issue is a factor and it plays on my mind quite considerably.

 

Porter

It sounds like our listener has Peyronie’s disease, named after the 18th Century French surgeon who first described it. Peyronie’s can develop at any time in adulthood but is most common in middle aged men, affecting as many one in 20 to some degree.

 

Rowland Rees is Consultant Urological Surgeon at University Hospital Southampton where he sees around a hundred men a year with the condition.

 

Rees

Men generally are quite reticent to come forward with this type of condition or to discuss it but usually when there’s a bend that prevents intercourse then that will instigate some sort of medical attention.  But actually there’s lots of symptoms that bring people to the doctor, particularly in the younger crowd that would maybe notice a lump in the penis, painful erections, penile curvature, penile shortening and erectile dysfunction are all signs and symptoms of this condition.

 

Porter

What’s actually happening, what’s the underlying mechanism?

 

Rees

Well it’s not perfectly clear but the current most popular theory is that it’s some sort of minor trauma, so the condition mimics quite closely a penile fracture which is an injury to the penis.  But in Peyronie’s there’s no clear history of trauma but you then get two phases of the disease, you get an inflammatory phase which is painful and then generates eventually some scar tissue or fibrosis and that’s where you get your lump and that’s where you get the lack of elasticity on the concave side of the penis and then you get your penile curvature.

 

Porter

So what’s effectively happening?  This is an organ that’s designed to expand but obviously if you’ve got scar tissue that part of the penis can’t, so the rest of it expands so you get the bend with the scarring on the inside of the curve.

 

Rees

Indeed, so the scarring is always on the side of the concave curvature.

 

Porter

How bad can it be, what’s the worst case you’ve seen?

 

Rees

Well it was a few weeks ago – 120 degree bend, so over 90 degrees…

 

Porter

So it was pointing backwards on itself.

 

Rees

Pointing backwards.  The commonest curvatures are dorsal or upwards but you can get downwards or in any direction actually.  But the commonest range is up to 60 degrees I would say, it’s rarer to have more severe curvatures than that but obviously the more severe it is the more difficulty with intercourse.

 

Porter

What’s the natural history of the disease, if we don’t actually understand what caused it do we know what happens to it if we don’t do anything?

 

Rees

It depends when people come forward with it and if they come forward early then there’s more chance of improvements but overall about 10% of patients will improve, as defined by an improvement in penile curvature…

 

Porter

That’s without any…

 

Rees

Without any treatment at all.  About half will stabilise and about 40% will get worse over time.

 

Porter

So when do you as a surgeon want to get involved in this, what would make you decide that you should be more aggressive than watch and see?

 

Rees

Unfortunately there aren’t any good medical or tablet therapies for this condition.  So although it’s always wise to come early to the doctors and seek advice because then we can counsel and advise accordingly, there’s not a great deal that you can do to modify the early inflammatory phase.  In terms of surgery to correct the curvature then we usually wait a year because as I said the inflammatory phase can last quite a while, maybe up to a year, at which point it stabilises and beyond that time you don’t usually have further curvature, so we tend to wait a year before operating.

 

Porter

And you’re operating to do what – to remove that piece of scar tissue?

 

Rees

No, no, removing the scar tissue’s been shown to damage erectile capacity quite significantly, so there are two main approaches.  One is to shorten or take little pieces out of the opposite side – the convex side – to produce symmetry and that’s called a Nesbit procedure and it’s very straightforward and generally excellent results with that.  When the curvature’s more severe, and again we use 60 degrees as a sort of cut-off point of angulation, then that becomes less of a feasible option and then we have to approach it on the scarred side, which is more difficult an operation with a higher complication rate.  But we tend to incise or cut into the scar tissue and then fill the defect with a graft material of some sort.

 

Porter

For our listeners out there who’ve got a slight bend in their penis but it’s not bothering them, is there any need to do anything about that?

 

Rees

Not really, I mean the treatment is based on symptoms and bothersomeness for the patient and their partner.  So a 20 degree upward curvature, for instance, that rarely causes a problem.  Obviously men may need reassurance, particularly if they can feel a lump, and once they’ve received that reassurance they may not need any further treatment.

 

Porter

If somebody has an operation – the Nesbit procedure – the standard procedure that you would do, you said it’s very successful, can they carry on a normal active love life afterwards?

 

Rees

Yeah absolutely, I mean patients will complain initially of a curvature that prevents intercourse so in terms of defining success really we’re looking at angulations of about 80% of people having had a Nesbit procedure will have a straight penis afterwards on erection and a similar number – higher numbers – are having sexual intercourse again and are satisfied with the outcome.  And very few actually – if it’s done in the right way at the right time – return needing any further treatment.

 

Porter

Surgeon Rowland Rees.

 

Just time to tell you about next week when I will be taking a closer look at the latest development in the world of private medicine – online services that offer easy access to a GP, any time any place, via your smartphone, tablet or computer. And we’d like to hear from you if you have used such a service. So please share your experiences – good or bad – you can tweet me @drmarkporter or email insidehealth@bbc.co.uk.

 

ENDS

Broadcasts