The Guardian has reported this weekend that:
Thousands of patients die every year after being admitted to hospital at the weekend, partly because too few senior doctors are on duty and back-up medical services are not available, dramatic new research shows.
About 11,000 more patients a year die within 30 days of going into hospital if they are admitted for treatment between Friday and Monday than do those who arrive on a Tuesday, Wednesday or Thursday.
My wife is a GP: she questions some of the stats used in this report, and some of the logic. I am going to accept them at face value because they are the political narrative of the moment and because no one really disputes the stats that deaths at weekends are worse: they question why. And no one really doubts the problem could be addressed by having hospitals better staffed at weekends (although not just by doctors, but from top to bottom since hospitals only function as teams). So why does that not happen? There are three obvious reasons.
The first is we do not have enough doctors. We could train more, of course, but that takes time. Between now and then (which is a long lead time) without poaching doctors from other countries the only way to have more doctors at the weekend is to have fewer in the week: it's not as though (and I know this is true) they are sitting doing nothing from Monday to Friday. But since that could potentially increase deaths overall by reducing the general level of healthcare because it would mean that you never have an optimal team at work that is no solution to this problem.
Second, you can poach staff from overseas: the classic NHS solution, done for decades. It has worked. There is a problem though: the UK's anti-migrant rhetoric is making this country a less hospitable place to work. We desperately need people to come here to work and send out the message that they are not wanted. As a policy for helping the people of this country get the health care they need this one ranks in the category marked 'dire'.
But there is a third reason, and that is that the government has said that this problem must be solved at the same time as the NHS faces a massive real round of spending cuts in the face of rising real demand for its services and an escalating cost base as what is medically possible expands.This is the issue I want to focus on.
I am not sure what the cost of fully funding the NHS for seven days a week might be: it is significant in my opinion and impossible within current budgets. Given the NHS as a whole costs £140 billion and two-thirds is labour costs let's assume doing this properly might cost £5 billion a year. Many would think that significant in a time of supposed austerity when the supposed deficit will be £69 billion this year. I would argue we have simply got that process of deciding significance wrong. There are other ways of estimating, I would argue.
The first is to say what would be the cost of saving these lives. You could argue that this was £454,000 each, but that would be absurd: that would assume that no one else seen by better-staffed hospitals at weekends got a benefit from doing so, and 98.7% of all people seen at hospitals do not die as a result of admissions. So, it ould be absurd to allocate all the cost to those whose deaths have been prevented.
But, actually, there's then the fact that the cost would not, anyway, be £5 billion because overall and on average at least £1.8 billion of what was spent would come back in tax straight away, and because others would feel better off as well because there was more spending then those others would also spend more, so increasing the tax revenue again. It's likely that the £5 billion cost could effectively be halved in this way (it may be a bit more or less, I am just illustrating a point here).
So the cost, even if allocated to those whose lives are saved is now down to £227,000 each.
And then there's the fact that in practice this additional cost will not be paid: it will be borrowed. Instead of running a deficit of £69 billion we could run a deficit of £71.5 billion (that's and increase of £5 billion spent less £2.5 billion tax back).
That has an interest cost. The current price of 15 year borrowing is about 2.5% for the UK government. And that's about the average borrowing period right now. So the annual cost of this is £62.5 million.
That's £5,681 per life saved. You could argue that should be priced over 15 years: that (very crudely indeed, and to overstate things) would be £85,000.
But then you have to consider four other matters. The first is that government debt has never been repaid, on average. It rolls forever. So we're very unlikely to ever repay the sum borrowed to pay to save these lives. People have always wanted to but UK government debt and there is no reason to think that will ever change. So, we'll actually never pay the £227,000 per person whose life is saved: it will simply sit on the government balance sheet for ever. So we can ignore that.
The second is that if we could create real inflation, as is the government's plan, then whilst it is true that nominal interest rates would rise the real cost would fall: we're actually paying too much for borrowing right now because we do not have inflation. Until recently when inflation was taken into account there was almost no real interest cost to government borrowing because the interest paid was close to the fall in the real value of the debt because of inflation. With PQE in play creating the inflation we need this could happen again: real interest costs could fall dramatically. Mark Carney says 1% is the most likely.
Third, you have to realise that if we cancelled the debt - which is what QE has done to one-quarter of all government debt at present, the real interest rate could fall to no more than 0.5% at present with the option available of it being 0%. Now I am not saying we would ever cancel all government debt: that would be unwise simply because people really want to own it, and I think they should have that chance (which is a good reason for not clearing the deficit) but the chance that QE will not be used again over the next 15 years is remote in the extreme.
So now let's assume one-quarter of the debt has no real cost and 75% of it has a 1% real cost having allowed for inflation (which Mark Carney thinks is fair) and price this at a simple rate over 15 years, which is the longest period likely to be needed once the time value of money is taken into account, and we come to cost of £25,568 per life saved.
Now there's the fourth point: that interest will be taxed. Let's assume it is only taxed at 20% (which is low) but that still cuts the cost to £20,454.
Now let's assume that only quarter of the cost of the NHS being open seven days a week can really be attributed to lives saved: the rest being down to other conditions better treated, reduced claims for negligence at weekends, greater efficiency because people do not need to take time off work to get appointments and so on. In that case, the real cost using these assumptions per life saved may be about £5,100.
Now, I stress, I've simplified things.
I stress, you cannot keep making these assumptions that everything can be paid for in this way, but that's not for financial reasons; that's because we'd run out of people to employ long before we ran out of money creation ability.
And, I stress, the additional cost of work may be higher or lower than I have shown and the tax recovery likewise a bit different.
But the assumptions on interest rates, QE and loans not being repaid are all good approximations to reality.
And so the question is whether or not a life saved is worth £5,100, spread over time?
And whether it's worth trying to close the deficit to save that sum.
Thanks for reading this post.
You can share this post on social media of your choice by clicking these icons:
You can subscribe to this blog's daily email here.
And if you would like to support this blog you can, here:
A useful analysis. However (and I understand the political narrative argument) deaths were not shown to be worse at the weekend. There is an increased 30 day mortality of those WHO WERE ADMITTED at the weekend. Not a large increase. Relative risk of 16% for those admitted on a Sunday, compared to a Wednesday, where absolute mortality for the group analysed is (approximately – I don’t have the figure) 1.3%. The number for absolute risk change is very small. The difference between absolute and relative risk change is vital. Did the report actually state which day of the week they died on?
Also, I don’t think there is agreement on the cause for this change. It may be to do with staffing, it may not. This is a correlation from which cause cannot be deduced. I know as a GP that patients can be admitted because of a lack of community services, especially hard to access at the weekend. Some of these people should have been nursed or cared for in other ways at home. Some were at the end of their life.
From this point on the debate is largely drawn from one’s hat.
I know that point
I said I am assuming that the data to is right to suggest that if it is it is also soluble
I won’t comment on the problem of lack of available resources in hospitals at weekend. But of course you would EXPECT more fatalities in patients admitted at weekends because most weekend admissions come via A & E and are suffering from life-threatening conditions: heart attacks, strokes, severe trauma (car crashes) etc. You simply don’t get people admitted at the weekend for routine matters such as ingrowing toenails, straightforward appendectomies, hernia repairs etc.
I’m afraid journalists are not that good at statistics . . .
On your blog before I asked how you would get the amount of the tax gap that is recoverable above £20bn (out of £120bn) and you said you would start off by improving the data, modelling what goes wrong, and then you would work out what to do. Before we start trying to find the money for extra doctors we need to follow a similar approach and conduct research to make clear how many deaths are the result of understaffing.
The Guardian article doesn’t say how many of the 11,000 deaths are due to too few staff. As the article says, some of the deaths are because people admitted at weekends are more likely to be sicker. For example, routine surgery takes place on weekdays so weekend surgery tends to be emergency surgery, which has a higher risk of death. No amount of extra doctors will change this. The article gives no indication of the proportion of deaths caused by understaffing so it could be a very small or a very large proportion of the 11,000. We need to know this before formulating policy.
If you go to the original British Medical Journal paper that the Guardian article is based on it says that “It is not possible to ascertain the extent to which these excess deaths may be preventable; to assume that they are avoidable would be rash and misleading.” And later on: “Our analysis shows that we need to determine exactly which services need to be improved at the weekend to tackle the increased risk of mortality”.
http://www.bmj.com/content/351/bmj.h4596
I said I knew all that at the start of the article
I pointed out I had been well advised on it
And I said I was tackling the data as if it was true nonetheless
Sorry: you are arguing with a straw man
If you agree with Simon’s point and the BMJs about what can be inferred from the data is it not politically partisan to credit Osbourne’s ideology with the deaths in your article headline?
Note the premise of the article
It will be even more difficult to find reliable data as more services are outsourced and the internal market grows like topsy. Lord prior and cohorts will be looking at charging as opposed to free at the point of delivery, though he says he would prefer it to be tax payer funded.
I am a retired community nurse and our service covered 7 days a week, we did rota system. If we had concern for a patient, always had access to GP who would willingly visit, Christmas Day or not. We know that the health andd social care act 2012 served the death knell for the NHS and to continually malign the frontline carers in the service is to lower the morale further .
What this government say and what they do is quite duplicitous, can I say that, How the hell did the country watch this happen and not shout loud for OUR NHS. Family members and friends still work in the service and are pushed to the limit. Deliberately and chronically underfunded because the agenda has always been to privatise since Mrs Thatcher and was it Keith Joseph who went through the service with a fine tooth comb in order to find it wanting, they did not. Sorry if I have gone astray again. Sentence structure gone awol again, never mind.
So if the absolute risk is 1.3% (from Polk above) and the relative risk (from the Guardian) 12.5% for dying on Saturday or Sunday, then that means instead of 13 people dying during the week, it will be 15 at the week-end. For every 1,000 admissions.
The Guardian further states that these extra deaths are mainly in cancer, or heart and stroke patients.
Now, it would seem to me that that is the starting point to improve services, if there is indeed a problem. That would cost a lot less than £5bn.
The Guardian further suggests that the Health Secretary wants to unilaterally change the contract of consultants, so that the now work seven days instead of 5, as a rule. That would make it less attractive to work for the NHS, and the NHS could lose further staff. That is already a huge problem, doctors leaving the NHS because it is not an attractive place to work.
Interestingly, the data used just a few months ago, suggested 6,000 extra deaths on week-ends. It is a political football. Now, all of a sudden, it is 11,000 extra deaths.
From the Daily Mail on 16th of July:
“Speaking on Radio 4’s Today programme this morning, Mr Hunt said: ‘What is causing those avoidable deaths? A lack of senior consultant cover at weekends is one of the critical points.”
http://www.dailymail.co.uk/news/article-3163057/Top-doctors-told-work-weekends-Health-Secretary-takes-touch-consultants-halt-scandal-high-death-rates.html
But, let us say £5bn additional funding were made availablle. That could employ about 100,000 extra staff, taking NHS staff from 1.6million to 1.7 million.
Where are we going to get 100,000 extra doctors and nurses from, even if the money was there?
The real constraint is skilled staff, and not money.
I think I said what you concluded
And I noted it could be solved
But only with money
Of course the problem is one and the same. Short periods of unemployment for any individual may well be an acceptable, and even a good thing, if it frees up some time to learn new skills.
However, longer periods of unemployment have the opposite effect. The country’s skill base diminishes and we all become poorer as a result.
So the priority for any government should be to reduce wastage by keeping unemployment as low as reasonably possible and encouraging workers to use any periods of unemployment to update their skill levels. Just the opposite of what happens now, in fact, with our public debt fetish overriding more sensible considerations.
At the local NHS trust hospital:
There is 7-day consultant-led emergency heart service.
There is 7-day consultant-led stroke service.
MRI is available 7-days a week.
X-ray is available 7-days a week.
Haematology/pathology services are available 24/7/365
The problem at the last election was that neither Labour nor the Tories were any different on the question of the deficit. Arguably Labour was even worse. Labour’s May promise to “cut the deficit every year” ignored economic reality. The difference between what the Government spends and what it gets back in taxes is simply what is saved in the economy by the users of the currency. ie the pound sterling. That saving could be you and I putting some money into National Savings certificates or it could be the Chinese or Germans not wanting to spend all the currency they earned selling us stuff.
That saving is the source of government deficit and debt. The savers end up owning government debt by choice in other words. It’s the same story in all other countries too. All the world’s National Debts total to something like $60 trillion, or the equivalent in different currencies. We don’t owe that to Mars! We owe it to all those who have chosen to save in different currencies.
Since when has it ever been possible for any UK government to tell you or I or the Chinese that we shouldn’t save our money in pounds?
Anyone who thinks that is economically illiterate. Crazy even. These kind of pledges makes no sense whatsoever. They are on a par with wanting to legally redefine the value of PI to be some more convenient and rational number as some tried to do in Indiana in 1897.
It all changed when the NHS felt you needed a degree to be a nurse.
This isn’t something new its been going on for the past 15 years.
This is not a new issue and has been reported before, so to blame it on the current govt. seems to be a stretch. As you correctly point out hospitals are a care team with numerous support departments, ICU for example may operate 24/7/365, but if they need radiology at 3am Sunday it’s going to take longer than at 3pm Monday. You acknowledge that we can’t just fix this tomorrow as there aren’t the resources available to do so, but then seem to think that throwing money at the problem is the answer or was this some sort of thought experiment.
Let’s also not forget that in some areas money isn’t the answer, increasing the wages for nurses in special care baby units won’t necessarily increase the staffing as not everyone can cope with the emotional demands of this job, my wife worked in ICU for many years and often commented on the burn out rate and staff who only stayed for short periods, sometimes it’s not a job you do just for the money.
An ICU unit has ALL the pathology and radiography services it needs irrespective of time.
Ironically; getting X-rayed at 3PM on a Monday may well be harder than at 3AM on a Sunday.
If there is an A&E there must be X-Ray/blood and heart services available, in the majority of places 7-day.
Even more ironically; NHS IC units are also used by private hospitals……since IC is much better manned/womanned, and they are highly trained. As such, they are an expensive facility to run: Too expensive for the majority of private hospitals.
Dear Repliers,
The point is not whether the figures are accurate or not, or whether the solution is the need for hospital consultants to work 7 days a week . The point is Jeremy Hunt wishes to make a political point, casting aspirations on Consultants and suggesting that they refuse to work w/e’s As clearly you all are only too well aware, emergency care has Consultant cover now, and a minimum of the other services available, although some will only be on call over night and therefore delay is possible. HUNT wants to force a different contract on Consultants and to try and make the public think they are overpaid and work shy.
Richard’s point was that IF the problem is as presented, then more money needs to be made available to increase staffing and it would be perfectly possible to access more money. There has been a deadly silence on that front . This government wants the NHS to fail and has been doing a very good job in trying to undermine the pubic’s view of it. Nearly every “fact” that they use to denigrate it can be rubbished, but the papers, including sadly the Guardian just seem more than willing to lap it up. They no longer seem to employ journalists able or willing to investigate information they are fed.
Thanks
Appreciated
And definitely agree with the last point
And:
https://flipchartfairytales.wordpress.com/2015/08/28/the-healthy-ageing-challenge/
What about opportunity cost?
£5bn promoting sites like the one below and policies based on the research they promote (Eat more veggies and beans and less meat, fat and sugar) would save many more lives, if promotion was successful.
Worth watching the annual presentations, Dr Greger is quite funny.
http://nutritionfacts.org/
I’m loath to say Osborne’s deficit fetishism is built on anything but lies.
Seeing as the IMF only “realised” that austerity was deflationary at the ZIRB in 2012, which was when government spending continued it’s upward trajectory.
No, to call Osborne a blatant liar is perhaps too strong. Opportunist, wingnut, whiplash fetist, perhaps, but he does pick his sophistry.
As regards PQE, clearly a case for it at the zero-interest bound.
Is there a single economist who argues for austerity these days?
What are other EU countries doing (eg Italy)?
Quite possibly history will show that Osborne was the most political chancellor we have had [so far].
I wonder if he will survive the leap to PM?
Worth watching:
Sell Off the abolition of your NHS.
http://selloff.org.uk/nhs/default.html
2 big problems:
1. PFI
2. Internal ‘market’
The arithmetic of this fails at the point where we get the cost per life down from 227k to 5.7k by effectively calculating that the cost is just the interest. But why stop there? Why not do this trick a second time with the interest payments, by again borrowing or printing the money to cover the interest and never repaying it. That brings the cost down to £143 per life.
But why stop there? We could repeat a couple more times and bring the cost per life down to pennies, or for practical purposes zero.
And when you have an absurd conclusion, you have to go back to your working and figure out the absurd assumption.
I made real world assumptions
Ones backed by evidence
Roy Lilly’s nhsmanagers.net newsletter:
Let me start by making something plain, absolutely black and white, no arguments:
Any unnecessary death, whilst in the care of the NHS, is unacceptable and must be the subject of rigorous examination to root out why, learn and put whatever is needed in place, to make sure it doesn’t happen again.
Right, got that? Remember it, please.
Question; do 11,000 people die ‘unnecessarily’ within 30 days of admission to hospital at the weekend. The press got very lit-up (more here) about a new BMJ report on the ‘weekend effect’.
The BMJ brain-boxes have revisited work they did a few years ago. There is fresher data, polished up methods of crunching the numbers and more than a hint of bamboozling.
For starters; the BMJ has redefined the weekend as Friday, Saturday, Sunday and Monday. Are they really saying 57% of the week is dodgy?
Before we unpack this report let’s do a simple sum. Eleven thousand deaths mean 211 a week. Assuming there are about 155 Trusts at the sharp end of admissions; we are talking about 1.3 question-mark deaths a week, per Trust.
Before we go any further please reread paragraph two, above.
A simple mortality case note review could get to the bottom of why the person died; error, neglect, poor decisions, reduced staffing. We don’t do that so we don’t know and we will continue to be in the dark until we do.
Please reread my second paragraph, above.
The BMJ report actually says:
Our analysis of 2013-14 data suggests that around 11000 more people die each year within 30 days of admission to hospital on Friday, Saturday, Sunday, or Monday compared with other days of the week (Tuesday, Wednesday, Thursday). It is not possible to ascertain the extent to which these excess deaths may be preventable; to assume that they are avoidable would be rash and misleading.
The BMJ continues:
Appropriate support services in hospitals are usually reduced from late Friday through the weekend, leading to disruption on Monday morning. This could go some way towards explaining our finding of a “weekend effect” extending into Friday and Monday.
Note ‘usually’. The brain-boxes have moved into an evidence free zone. They might be right is all we can say. Oh, and there is ‘this could go some way towards explaining’. Yes it ‘could’. Funny how the big brain boxes can be so pedantic with the number crunching but so cavalier with their conclusions.
An uncharitable observer might say the deductions are being tailored to fit the objective.
The objective? Trying to win a row with the 1% of hospital consultants who opt out of weekend working but have been reduced to bystanders in the debate. The report doesn’t help. It does say, junior doctors might be unsupported (For 57% of the week? Really?) and the analysis speculates:
“… that we need to determine exactly which services need to be improved at the weekend to tackle the increased risk of mortality…”
Please re-read paragraph 2 above.
Patients admitted at the weekend are more likely to be in the highest category of risk of death which might say more about primary and community care (where the Tinker-Man is embroiled in another row with the BMA), than it speaks to secondary care.
So, now what? Make hospitals, working on a Saturday, look like it’s a Wednesday? Good idea but you’ll need imaging, porters, administrators, social services, catering and the whole shebang. Let’s do it. Make the assets sweat.
Cost? Forget the ‘extra £8bn’, that’s just enough to keep the NHS doing what it’s doing now. Seven-day-working cost estimates vary; some say 2% of total income.
Just how much should we spend on evidence based 7-day-working? We could ask NICE. They use a calculation called a QALY; the cost of a year of good health. The NICE price is about £30k.
The sum is simple:
11,000 deaths(?) times £30k = £330m pa.
Is that all we should spend. Will that fix it?
Please reread paragraph 2 above.
In my view the BMJ’s brain boxes have redefined ‘dodgy dossier’ and caused a brouhaha in the press, we can well do without… largely because people writing about it didn’t read it and as far as I could see the BMJ did little to spell out what they are actually saying.
The BMJ work takes us no further forward and does not explain ‘the weekend effect’. Not even if you allow them to stretch the weekend to 57% of the week.
To me this looks political; manipulating public opinion, supporting the Tinker-Man who is struggling with some pretty flaky arguments and in a fruitless row with the BMA.
Is there something going on in hosptials? Obviously. What’s the answer?
Please re-read paragraph two above.