“Accident and Emergency: Performance Update”
Agenda item 2 is the continuation of evidence taking on the section 23 report, “Accident and Emergency: Performance Update”. I welcome Paul Gray, who is the director general of health and social care and the chief executive of NHS Scotland; John Connaghan, who is the director for health workforce and performance; John Matheson, who is the director of finance, e-health and pharmaceuticals; Professor Jason Leitch, who is the clinical director of the quality unit; and Dr Aileen Keel, who is the Scottish Government’s acting chief medical officer.
I do not know whether Mr Gray or any of his colleagues wishes to make an opening statement.
I will make a brief statement, if that is acceptable.
I appreciate being given the opportunity to appear before the committee today. We take accident and emergency performance very seriously, which is why I have asked a number of senior colleagues to accompany me to bring their expertise to bear on different aspects of the issue.
The helpful evidence that the committee has taken from the Auditor General and—last week—from NHS Scotland senior leaders and clinicians has emphasised the complexity of the system within which unscheduled care operates. That complexity is not unique to Scotland. Unscheduled care performance was affected during winter 2012-13 in other parts of the United Kingdom and in similar health systems across the world.
Our approach in Scotland to tackling the issues related to unscheduled care is set within our overall vision that, by 2020, more people will be living longer, healthier lives at home or in a homely setting, so we want to do all that we can to ensure that, when people attend A and E departments, they get the right care, from the right person, within the standards that we set. That already happens in many cases, but we want it to happen consistently.
Sometimes it will be better for people to get the care that they need elsewhere—for example, in a minor injuries unit, via an out-of-hours primary care service or through telephone advice. Again, that is happening in some cases, but there is best practice that we can spread further, which will provide improved outcomes for patients and reduce costs.
I thought that it would be helpful to comment briefly on the phrase “A and E waits”. What we are measuring is progress against the target that, by September of this year, 95 per cent of patients will be seen and, as appropriate, treated or discharged within four hours of arrival at A and E. We are not measuring whether patients wait for four hours; we are measuring whether they get out of A and E, with all clinically appropriate actions taken, within four hours.
As I have made clear in earlier correspondence, I welcome the recommendations in Audit Scotland’s report, which was published in May of this year. We are progressing those through our unscheduled care action plan, which is supported by the local unscheduled care action plans that the boards prepare annually.
I will briefly mention some of the key actions that have been taken in the first year of the action plan. We have established the flow programme to improve the way in which patients move through the system and to cut out unnecessary delays. We have recruited an additional 18 emergency department consultants, put in additional bed capacity and issued signposting guidance to help direct patients to the most appropriate treatment point. In addition, we have a number of new initiatives to prevent frail elderly patients from going into hospital unnecessarily in NHS Forth Valley and NHS Ayrshire and Arran; we have introduced discharge hubs in NHS Fife, NHS Lothian and NHS Ayrshire and Arran; and we have invested in theatres in NHS Grampian, beds in NHS Lothian and staffing in NHS Lanarkshire.
Over the period November 2013 to March 2014, NHS Scotland recorded a performance level of 93.1 per cent for patients being discharged or admitted within four hours, which compares with 91.4 per cent over the same period in the previous year, and the figure of 94 per cent has been quoted in relation to published data for June. However, I fully accept that we are not at the standard that we have set, and I want to ensure that patients who attend A and E can leave A and E safely within that standard. That is the commitment that we have made and, despite the complexities, it is the one that we are continuing to strive for.
We have also reduced significantly the number of people who wait for more than eight and 12 hours to be discharged or admitted. We want to eliminate that, as far as possible; we do not believe that people should have to wait that long to be admitted or discharged. That should happen only in very few cases. Fewer than 1 per cent of all patients remained in A and E for longer than eight hours, but we owe it to patients to make further improvements, where we can.
I assure the committee that we are well aware that the context in which we are seeking to deliver the commitments is challenging. I am not here to provide a set of emollient statements about how it is all absolutely fine; there are places where it is not.
We have an ageing population, increases in the number of patients presenting with more than one condition—often referred to as multimorbidity—and recruitment pressures. Those issues are not unique to Scotland; nevertheless, we are committed to doing all that we can for the people who are served by NHS Scotland to provide timely treatment so that they experience safe, person-centred and effective care and enjoy good health outcomes.
I am happy to answer the committee’s questions. If we do not have the data immediately to hand, we will undertake to provide it as quickly as possible. I know that you have had a lot of detailed information; we want to ensure that any responses that we give in that context are accurate, so if we do not have the data today, we will provide it as soon as we can.
Thank you for allowing me to make a statement.
Thank you very much.
You recognised in your opening statement that you are not yet meeting the 98 per cent standard or target. However, you also talked about milestones. Did you reach the 95 per cent performance milestone in September?
As you will appreciate, convener, the data will not be published until, I think, the end of December. Is that right?
No—it is published two months after the period ends.
Therefore, the data for September will be published in November. Until we have that ratified data, I cannot confirm that we have reached the milestone.
Are you confident that you will reach at least 95 per cent?
On the basis of advice from boards, I think that a number of them will reach 95 per cent. I cannot say that all of them will, and until we have the data I do not want to make a firm statement about that.
Okay.
Dr Martin McKechnie, the new chair of the College of Emergency Medicine Scotland, has acknowledged that the additional £50 million that the Scottish Government invested has helped to “curb a crisis”—I am not sure whether those are his words or the words of the journalist who reported him.
However, Dr McKechnie went on to say:
“We still have unresolved serious issues”.
Do you accept that?
I accept that there is an issue with staffing in emergency medicine, for example, as the committee heard last week in relation to NHS Grampian. I have acknowledged that not everywhere is perfect and I have no difficulty in accepting that in some pockets there are, have been, and will continue to be, difficulties. There was information in the public domain recently about a day when Hairmyres hospital A and E dipped below 70 per cent in relation to the target due to unexpectedly high attendance on that day, which was outwith all the norms. We are not in a situation where every day will be absolutely perfect.
You mention Lanarkshire, and last week we heard evidence from a number of boards, including NHS Lanarkshire. One of the interesting things that I suspect the committee will want to look at more closely is how boards share good practice as well as how problems are identified.
We heard from NHS Tayside about its efforts to ensure that people do not attend A and E unnecessarily when they could be treated elsewhere. We also heard from NHS Lanarkshire that it was quite confident that it could sustain the present A and E configuration, but since then I have seen correspondence that suggests that the general practitioner out-of-hours service in Lanarkshire is having problems. I think that there are five units in Lanarkshire and a couple of them had to be closed.
In areas where the GP out-of-hours service is unable to cope, we heard concern that the public would make their own decisions. Mary Scanlon identified a number of areas in the country where it seemed that the public were attending A and E because they felt that that was probably the easiest and quickest way to receive a service. In areas such as Lanarkshire, if the GP out-of-hours service is unable to cope because of a lack of staff, does that not place a huge burden on an already-overstretched A and E service?
09:45
If a particular aspect of the service in any board stops working for a period of time, the demand goes elsewhere. Ms Scanlon made some important points about the choices that individuals make based on their perception of where they are most likely to get a service.
In the national health service, we never—nor should we—refuse to provide a service. If someone cannot get a service from a GP out of hours, they have the choice of phoning NHS 24, attending accident and emergency or, in extreme cases, phoning an ambulance. Those choices will remain available, and the pressure will displace from the unavailable service to the available services. That is a fact of the way in which the NHS operates in Scotland.
We are looking at the Audit Scotland report on accident and emergency services today. We could not ask Audit Scotland what was happening with the Scottish Ambulance Service and NHS 24, and why two out of three people presenting at accident and emergency self-refer, but I feel that we can ask you.
The overall increase in the number of patients over the past four years is 50,000. However, if we drill down slightly, we see that at Ninewells hospital the number is down by 46,000, while in Aberdeen it is up by nearly 63,000. The number for Edinburgh is up by 112,000 and for Glasgow it is up by 85,000.
What are you doing to find out why there is a huge increase in self-referral? My point is that we no longer have an accident and emergency service—we have a 24/7 open door to the NHS. Patients are—quite rightly—now saying, “This is where I choose to go”, perhaps because they are getting from one service what they are not getting elsewhere.
If that continues at the same rate, we will no longer have an accident and emergency service. It seems that GPs are doing less and less, and patients are voting with their feet to go to A and E. What are you doing about NHS 24, GP referral and the two thirds—66 per cent—of patients in A and E who are self-referring? Is the service really an accident and emergency service any more?
I have a number of comments, and I will bring in colleagues to speak about some of the wider work that we are doing.
First, I am very reluctant to criticise a patient—I am not suggesting that you are doing so, Ms Scanlon—for making a choice that may not be the best one for them. I will bring in colleagues to discuss the NHS 24 campaign that we will be running over the winter to help people understand what the most appropriate routes to treatment might be. We are seeking to educate the public on what would be best for them. In some cases, going to A and E is not the best option, but people may believe that it is the only option available to them.
I have been speaking to NHS 24 and the Scottish Ambulance Service about what more we can do to help the public understand where they are most likely to get the best outcomes. They may very well get a good outcome from going to accident and emergency, but they may have got a quicker outcome by going through another route. They may feel that they ought to go to accident and emergency, for the want of quick advice that might have given them sufficient reassurance to enable them to wait until the following day when they could go to a GP.
Those things are all possible. As I discussed with colleagues yesterday, the data flagged up to me that we are not yet very good at collecting information from patients about why they made the choice that they made. We know that they made the choice, but we are not always sure why.
I took the time to go round with the Ambulance Service and to spend time in the NHS 24 control centre. With regard to the Ambulance Service in particular, it was clear to me that some patients called for an ambulance because they were afraid. They had legitimate reasons to be afraid, but if we had provided a different source of assurance and advice, their anxiety levels may well have gone down and we would have been able to provide a service to them in a different way.
The same happens in A and E, but for the want of data, I will not make an absolute judgment on why some people turn up at A and E and why others do not. You are right to say that some presentations to A and E could be dealt with elsewhere. That is why we are doing what we are doing, for example, in minor injuries units, which allow people with a minor injury to get a different source of advice and treatment if necessary.
Aileen Keel might want to say something at this point—
We have the data here in the report. At Ninewells hospital, 50 per cent of patients at A and E self-refer; at Aberdeen royal infirmary, the figure is 74 per cent. For other hospitals, such as Hairmyres, the figure is more than 80 per cent.
Why is it that 50 per cent of patients at Ninewells, which we welcome as a beacon of good practice, are self-referring, whereas at Aberdeen royal infirmary the figure is 75 per cent? I picked out Aberdeen just because its representatives appeared before the committee last week. The number of self-referring presentations at accident and emergency at Aberdeen royal infirmary is 25 per cent higher than at Ninewells. Why is that?
I am not saying that we do not have the data. What we do not have is the underlying information that tells us why the patient made that choice.
It may be that patients in Aberdeen made the choice because some facilities are not available in Aberdeen, whereas those facilities may be available in Dundee, for example, but we have not asked the patients so we do not know. My point is that I am drawing from the data the fact that we need a better understanding not just of the facilities that are available, but of why patients make the choices that they make. That is what I am keen to pursue as we move forward. Dr Keel may be able to add something on that.
That is right—we do not understand why people make those choices, and we need to get a better understanding of that. There is a graph in the Audit Scotland report—I am struggling to find it just now—that refers to usage of A and E departments and minor injuries units by board. It is very interesting, because it is clear—
Is it exhibit 5 on page 14?
Thank you, Ms Scanlon.
I do not know whether that is the one that you were referring to—
It is exhibit 4.
Exhibit 4—okay.
We were talking yesterday about why there is an enormous variation between boards in use of minor injuries units. Those units are there on a board’s territory, but it is clear that patients are choosing more to go to A and E. We need to gather a bit more intelligence on the issue and begin to better understand why those choices are being made.
Okay—well, if you do not understand it, we are not going to get much further on that. I was hoping that we would have a more holistic picture from the health service. It would be wrong to ask Audit Scotland about the Ambulance Service and other areas when it looked only at A and E. The question of what is happening with GPs, NHS 24 and the Ambulance Service underlies the Audit Scotland report, which is why I am keen to understand all of that. Until we understand that, we will not know why two out of three patients self-refer. They are doing that for a reason, and we have to respect that reason.
Yes, indeed. I will bring in Professor Leitch on that point. However, just to be clear, I accept the point that we need to understand better why patients do what they do. Our approach so far has not been to collect much data on that, but I believe that we ought to do so because that will allow us to modify what we are doing.
Good morning. Ms Scanlon made a couple of points. She referred to exhibit 3 and exhibit 5. I will deal first with exhibit 3, which covers the differences in attendance between 2008-09 and 2012-13. There are two before/after data points, but we need to see the trend. There are some interesting things about the data—
It is all that we have.
Indeed.
In exhibit 3, Glasgow royal infirmary appears to have the second highest increase. That is because Stobhill hospital closed its A and E department during that time and Glasgow royal infirmary absorbed all its patients.
The Victoria infirmary, where I did most of my training and most of my surgery, has the second biggest drop in attendance—it says that it reduced attendance by 61,000. That is because the hospital opened a minor injuries unit, which took all the people whom it used to count as A and E attendances. The figure does not give you the overall trend, which, for the whole country, is 150,000 people a month, and is roughly stable.
That does not deal with the points about where people are going and holistic care. I completely agree that we should give people the most appropriate care in the right place at the right time.
The data in exhibit 5 is—how best to describe it?—weak. I do not think that we code that particularly well in the national health service. Let us take the two hospitals that you used as examples. The rates of self-referral to Ninewells and Aberdeen royal infirmary are given as 50 and 75 per cent respectively. The table also says that 27 per cent of people come to Ninewells in a 999 ambulance, as opposed to 7.2 per cent for Aberdeen royal infirmary. That cannot be true; cases are being coded differently. We are talking about two major district general hospitals, one of which says that a third of its patients come in ambulances and one of which says that 7 per cent come in ambulances. The data is not coded well by the health service.
The issue came up with Audit Scotland last week. It is difficult for us to do the job when we do not have accurate, comparable data. I can only work with what is in front of me. Audit Scotland said that the data are not comparable. I will leave it to the convener to pick up on that—I am only the deputy convener.
Professor Leitch said that the data is weak, but that is not Audit Scotland’s fault. Audit Scotland can compile only what is presented to it. If the data is weak, that is the fault of the NHS and the health department. It is not Audit Scotland’s responsibility. Why has the NHS, collectively, not sorted out the issue of weak data, to enable Audit Scotland to do an effective job on behalf of the public?
Because the data are not routinely collected for publication and so are not subject to the standards and strictures that would apply if they were.
In the interests of transparency, let me say that I discussed exhibit 5 with colleagues yesterday and made it clear that when I see chief executives this afternoon I will say that, given that the data that we have tells us that there are differential approaches to collecting it, we must improve in that regard.
I am not hiding behind the data. It was given to Audit Scotland. No criticism of Audit Scotland is intended or implied in the presentation that we are making to the committee. We did what we could do with what we had available. If we are asked for new information, which we do not routinely collect for publication, the information will generally be of a lower standard.
I understand that. However, we are not talking about something new; we are talking about something that is done on a semi-regular basis. The report is a performance update, not the initial report. You have known about the issue for a considerable time. Why did it take until yesterday for you to raise the issue and why is it taking until this afternoon for you to say that there is a serious problem with statistics?
There is not a serious problem with published statistics—
So the data is not weak.
The data is weak, but these are not—
And that is not serious?
If there was a serious problem with routinely published statistics, that would be a very significant issue. If we are asked for something on an ad hoc basis, we do our best to provide it, as we did in this case.
I am not suggesting that you are saying this, but if the committee would prefer us to stick to data that is routinely published and subject to the quality controls in that regard, we will do so, but I do not think that that would be a service to the committee. I am simply telling you that I am taking up the issue with chief executives this afternoon because we have a meeting with them. The issue was clear to us some months ago, we have been working at it, and I have got to the point at which I want to speak to chief executives about it.
10:00
It would not be appropriate for the committee to tell you which sets of data to collect and not collect. That is not our responsibility. Our responsibility is to analyse and comment on reports that are produced by Audit Scotland on behalf of the Auditor General. Audit Scotland asks you for information, so you clearly need to have a discussion with it.
However, I am surprised that you say that this report—which is a performance update—contains a set of statistics that are not routinely collected but which appear to me to be part and parcel of a continuing observation of the performance of accident and emergency. We can explore later with the Auditor General whether that set of statistics is unusual and, therefore, subject to the weaknesses that you have described or whether it has been collected for some time, in which case it seems surprising that the problem is being addressed only now.
I will ask you one other thing before I—
I have one more question to ask, when you are ready.
It is a slightly different issue on the Ambulance Service and accident and emergency, but you go first.
I had only two questions, the second of which concerned the point that was raised by the medical director from NHS Grampian about the NHS Scotland resource allocation committee figures and, previously, the Arbuthnott formula. I do not have the figures in front of me, but I remember them and they were accurate. Per capita, Ninewells hospital gets around £1,945 and Aberdeen royal infirmary gets £1,500. Therefore, every person in NHS Grampian is funded at nearly £500 less than a patient in Tayside.
Is Aberdeen being punished for being the oil capital? I remember Arbuthnott 1, Arbuthnott 2 and all that, but is the NRAC formula appropriate? Are we really funding Grampian appropriately to provide the service, given that we can easily criticise its performance?
I will bring in John Matheson on that point. However, to be absolutely clear, since you have asked the direct question, NHS Grampian is not being punished for anything.
Why does it receive £500 less funding per person?
Mr Matheson will explain the formula.
The basis of the formula is the population of the individual health board areas. It is then adjusted for age and sex—
The figures that I gave are per head of population.
The formula is based on population and then adjusted for age and sex, morbidity and life circumstances. An excess cost index is then brought in to recognise remoteness and rurality.
The formula is dynamic; it is continually under review—we have just reviewed the remoteness and rurality—so what the results tell you is that the population of NHS Grampian overall makes less demand on the healthcare service than the population of the other parts of the country.
We recognise that not all boards are at NRAC parity, and NHS Grampian is one of the boards that is below parity. We have an agreed way forward to bring it and the other boards that are below parity to within 1 per cent of parity by the start of 2016-17.
The difference that Ms Scanlon highlighted is driven by the formula, which was agreed across the NHS and is under continuous review to ensure that it is appropriate and up to date.
This is a complex and complicated issue, and it is one for separate discussion at another time. No doubt we can come back to it if Audit Scotland produces a report on it.
From what Mary Scanlon and Mr Gray said, a question comes to mind about the connection between the different services. You have a target of a four-hour wait, to which you aspire and which you admit will be challenging. I have an inquiry from a constituent about the Ambulance Service—no doubt others will have similar inquiries. The woman had to wait seven hours for the ambulance to arrive. When she got to accident and emergency, the clock started for the four hours but, potentially, it was 11 hours from her reporting an issue to her being through the system. Is that acceptable?
Without knowing the detail of the individual case, convener—
I would not expect you to know the details. I am talking about the concept of a four-hour target when the reality could be an 11-hour wait.
I would not like to draw too many conclusions from an individual case—in which the wait for the ambulance sounds long—but one of the important points about how we are trying to help the public better understand what we do is that, particularly in serious cases, if an ambulance arrives with qualified clinicians, the definitive care to the patient starts when the ambulance arrives, not when the person gets to A and E.
There is still something of a mental model in the minds of the public—for which I accept responsibility—that the job of the ambulance is to pick up the person and take them to A and E as fast as possible. In fact, definitive care is delivered at the roadside or in the patient’s home. That is the life-saving care that is often delivered. The decision by the qualified ambulance practitioners or paramedics to take the patient to hospital is informed by their assessment of the patient’s clinical condition. It is right to have a target for the Ambulance Service to arrive, based on the category of the call, and that we then have a target relating to performance at A and E.
Of the cases that I was privileged to be part of during my short time going round with the Ambulance Service, only one patient out of five was taken to A and E. In other cases, care was delivered in the person’s home or by the roadside and there was no requirement for them to go to A and E. I do not think that we can join the two together.
The committee has come up against data collection deficiencies across the board again and again. The report in front of us is almost entirely data driven, and the conclusions that we take from it are entirely dependent on the quality of the information that is provided. It is a concern that there are inconsistencies across the service that make it difficult for us to do comparisons or to draw the conclusions that we might need to draw. I am sure that you will be addressing that—I hope that you will.
As I said in response to the convener, there are standards for published data, which must be upheld and maintained. We seek to be transparent. If we hold information, we will give it, with the caveat that it might be partial or incomplete. We are not in the business of withholding what we know, even if it tells us that we need to get better.
It was said previously that the Scottish Government was encouraging NHS boards to make use of emergency departments and the emergency medical workload tool. How widely used is that tool, and what conclusions have been drawn from its use?
I ask Dr Keel to help me out with this, as I am not familiar with that level of detail.
The short answer is that I do not know how widely the tool is used. We are certainly promoting it as a means of measuring workload—not just the volume of patients coming through the door, but the case mix and the severity of the conditions from which those patients are suffering. John Connaghan can probably say a bit more about how widely the tool is being implemented.
The tool is still under development. We have reached the later stages of its development, and we have piloted it in a number of boards. The plan is that we will start to roll out the tool on a national basis in 2015, as early as we can.
It is a different kind of workload tool compared with what we have used in the past. Scotland leads the way in this area, from a UK perspective, and most of the tools that we have developed have been more devoted to nursing staff. The accident and emergency tool covers all the staff who work there, including doctors, allied health professionals and nurses; that is why the development phase is taking a little longer than we would have liked, but we have done a lot of work on it. As I said, the tool is breaking ground in a UK context and the plan is for us to roll it out in 2015. I do not have the roll-out schedule to hand, but I can supply more information if it is needed.
It has been highlighted that a substantial number of people are self-referring to A and E; however accurate the figures are, the number is substantial. As we have heard, there could have been alternative solutions for many of those patients. A lot of them are signposted to other services—for example, primary care and so forth. How confident are we that there is sufficient capacity within those other areas, such as primary care, to deal with the patients who are signposted on?
In the past couple of years, we have established a requirement for each board to produce a local unscheduled care action plan. Those local unscheduled care action plans are now in their second year and should take account of the demand and capacity in each part of the system that supports unscheduled care. The matter is very much for local boards to determine, but we have set up that national requirement for boards to make those action plans, which are published on the boards’ websites.
You say that it is up to the local boards to determine their plans. Is there any consistency of approach and are there guidelines that they have to follow?
We issue guidance annually and we continuously refresh that guidance with our partners—for instance, the College of Emergency Medicine—to ensure that it is accurate and up to date. We call that the national unscheduled care action plan.
Most of us were worried by the Audit Scotland report when we first saw it, because it shows that Scotland’s performance against the A and E targets has deteriorated over the past four years. We spent some time trying to work out the main reasons why that might be the case in order to ensure that the matter is being addressed.
Last week, we had a good and frank discussion with colleagues from the NHS in which we touched on staffing, delayed discharge, sustainability issues and other things. One particular issue emerged, starting with a comment by Professor Ferguson, who is an emergency consultant in NHS Grampian. He said:
“We still operate the way that we have always operated. We know that people are more likely to die if they go into hospital at the weekend—there is good evidence to suggest that”.
I followed the matter up with the Cabinet Secretary for Health and Wellbeing, Alex Neil, and he said that there is no evidence to suggest that. However, Dr Dijkhuizen, the medical director at NHS Grampian, said of the international studies that show that there is an issue at weekends:
“I agree with Ken Mackintosh that, because those studies show such a relationship, we should assume that the effects are the same in our country and our organisations. That is why we do studies: to learn in order to know what to focus on.”
When he was questioned again by the convener, Professor Ferguson later suggested:
“What I am saying is that there is international evidence that backs up that that happens. I would surmise from that that we have the same problem in Scotland—otherwise, why would we need the safety programme?” [Official Report, Public Audit Committee, 1 October 2014; c 19, 28, 32.]
What do you make of that, Mr Gray? Do we have a problem with excess mortality at weekends in our A and E departments or in our hospitals generally?
The evidence that we have tends not to support that suggestion. You quote colleagues saying that they “assume”. I ask Professor Leitch to give us some insight into what the data is, what it is telling us and what the international reports say. You ask a fair question, Mr Macintosh, and we anticipated it, so we have prepared an answer.
Mr Macintosh and I have discussed the matter previously and there was a freedom of information request.
Some international studies suggest an increase in mortality at the weekend compared to mortality on weekdays, although they tend not to explain why and they do not adjust for everything that could be adjusted for because it would be very difficult to do that. It could simply be that patients are sicker or more complex, that there is more trauma or that there is more alcohol use on Fridays and Saturdays.
There are two pieces of Scottish evidence. One is the Handel study, “Weekend admissions as an independent predictor of mortality: an analysis of Scottish hospital admissions”, which was quoted in the committee last week. That study does not adjust for admitting diagnosis—so it does not make any decision about why a patient comes in—and it does not adjust for the severity of the diagnosis, so it does not tell us whether a stroke was very bad or very mild.
10:15The Handel report concludes that
“It may also be that emergency departments see a different, more unwell population of patients at weekends, since, in one study which used a biochemical measure of severity, adjustment for this variable rendered the weekend effect insignificant ... This could mean that the effect we observe is actually due to admissions over the weekend comprising a more unwell population of patients, who would suffer a higher rate of mortality regardless of factors that may apply exclusively to the weekend.”
In response to your previous questions, both in the Parliament and to Mr Neil, we asked ISD to examine Scottish data in particular. You have had that FOI response. It examined all deaths from 1 January to 31 December 2012 by specialty and of course there is variation—by day and by specialty. There is constant variation because the mortality rates do not stay the same all the time. ISD says:
“the assumption that mortality is higher for patients admitted at the weekend cannot be backed up by statistical evidence ... The data only took the type of admission into account. To understand this issue fully there are a number of factors ... such as case-mix, age and underlying health issues.”
I am not dismissive of the weekday-weekend mortality literature, but I am passionate about mortality in the whole week. I am passionate about unexpected mortality and what we are doing about it in Scotland’s hospitals.
The committee will not be surprised to hear me use the example of the Scottish patient safety programme—it exists on Mondays and on Saturdays and Sundays. The programme is about sepsis, venous thromboembolism and early warning scoring. The witnesses who appeared before the committee last week used it as an example of trying to fix the whole system all the time. In global terms, it is the best recognised safety programme in the world. It is about reducing mortality every day. I am not rejecting completely the weekday-weekend mortality thing, but I am more focused on reducing unexpected mortality throughout the whole system.
Thank you for that and for following up the issue. If there is no shared acceptance or identification of the problem—if people do not think it exists—it is very difficult to address.
I want to pick up on the points that you made. The figures that you published through the FOI were welcome. I have spoken to several people about this, but I was contacted by Professor Paul Aylin, who is professor of epidemiology and public health, and co-director of the Dr Foster unit at Imperial College London. The Dr Foster unit has been influential in changing health patterns in England and I imagine that that is why the issue matters to its staff. Professor Aylin gave me his views on the information that was published through the FOI:
“the document cited by Mr Neil, which claims to support the fact that there is no excess mortality at weekends in Scotland is inconclusive. The analysis as it stands breaks down the data into individual specialties by day of the week and as such the numbers are just too small to show an effect either way.”
Although you referred to one comment in the BMJ paper, which was authored by Dr Handel and many others, the paper’s conclusion was actually that
“There was a significantly increased probability of death associated with a weekend emergency admission compared with admission on a weekday”.
That is key. The paper did say that “further research should be undertaken”, but it also showed other factors. The study spanned 11 years and showed a decline in mortality over that time. It was quite a positive study; it was not unremittingly negative about what was happening in Scotland—far from it.
There have been other studies. Professor Aylin, along with others, published “Weekend mortality for emergency admissions. A large, multicentre study”, which also showed differences between weekend and weekday admissions. There have also been international studies showing that difference.
My point is that in England, the NHS is taking policy action to address weekend mortality and readmission rates—it recognises the issue and has changed its policy. I am not going so far as to say that we should do that; it could be that the patient safety programme is doing that. I am trying to work out whether we can conclude, as Mr Neil suggests, that we have no problem at all. The figures that you published in response to the FOI request suggest a marked improvement between 2009 and 2012.
The BMJ study and others showed quite a difference—a 40 per cent difference in excess deaths at the weekend, which is significant. If your figures show that there is now no difference, can you point to the policy initiatives that have made the difference? Can you show what is working? That is important to us. If a decrease has happened, we should welcome and celebrate it. We should also work out what caused the beneficial effect. I suggest that the figures that you have published do not prove anything; they do not demonstrate a problem one way or another.
Indeed. That is precisely my point. I do not suggest that the Handel study does not tell the truth. It shows an odds ratio of 1.42 in comparing weekends with weekdays but, by its own admission, it does not adjust for severity of diagnosis. One conclusion that could be drawn is that weekend patients are expected to die, for lack of a better description, because they are sicker than those who are admitted during the week.
We could isolate Tuesdays and spend all our time looking at the data on Tuesdays if we were worried about Tuesdays. I am worried about mortality and the safety of our healthcare system every day, which is why the Scottish patient safety programme and its interventions apply every day. That does not mean that we are not tackling seven-day working and seven-day services, but that is about not mortality but the flow, care and getting people out—reducing delayed discharges. Within seven-day working is, of course, safe and effective person-centred care.
We have the policy initiatives that the English have. I know of nothing in England that is specifically on the safety of care at weekends in comparison with weekdays. I know that the English have been up to look at the Scottish patient safety programme and that, if they can, they will launch 15 such programmes in the regions of NHS England. I have seen nothing in England to show that the English are doing something special at weekends to reduce mortality in comparison with levels during the week.
I would advise against specific interventions to deal with safety on a weekend day that we would not use on a Tuesday. I know of nowhere where such interventions apply. That does not mean that we do not want to increase the use of diagnostics at the weekend to improve the flow, to increase the use of pharmacy and to do all the other activities that are part of our seven-day-working service. However, the thousands of people who worked in the national health service last Saturday and Sunday already believe that it is a seven-day service, so we should not pretend that we are not already working in a 24/7 environment.
I do not want to spend too long on the issue. I agree that we should not draw the wrong conclusions from the BMJ article. As you said in relation to evidence in Audit Scotland’s report, evidence reveals problems, but we do not want to draw the wrong conclusions.
I am trying to work out whether we accept that mortality at weekends differs from that on weekdays. Professor Ferguson—a consultant who works in NHS Grampian and who gave evidence last week—believes that there is a difference.
The Handel study suggests that that is true.
The study suggests that that is true. It says:
“Particularly influential to policies has been the report by Dr Foster on an increased hospital mortality in the UK at weekends, which has been linked to a reduced cover by senior doctors at weekends.”
That is a separate report.
Foster does not have Scottish data.
That is exactly the case.
Foster has only English data.
There are different reports, from which different conclusions might be drawn.
Indeed.
At this stage, I am not suggesting even that we draw conclusions. I am just trying to work out whether we accept or believe that there is a problem of increased mortality at weekends. In his answer to me, Alex Neil said that there is no problem at weekends. He accused me of scaremongering, but I was not scaremongering in the slightest; my questions arose from a constituent’s case and I was trying to work out whether that was an individual situation or typical of what happens at weekends. I was slightly worried by what struck me as complacency on his behalf. If he believes that there is no problem, based on a survey that is not peer reviewed and is statistically inconclusive, according to Professor Aylin, that worries me. It seems that this is the one study that proves to Alex Neil that there is no problem at weekends.
Would it be possible for you to provide exactly the same evidence as Professor Aylin, Dr Foster and other medics including Dr Handel and all the ones at the BMJ? They studied evidence over 11 years up to 2009. Would you be able to provide the same evidence, breaking down, for example, not just weekend and weekday admissions but elective admissions. Basically, could you provide something comparable with the BMJ paper so that we could actually make the comparison?
Hanlon has the Scottish data. He has as much as is available to all the countries. The difficulty is that nobody measures severity of diagnosis so nobody knows how sick the patients are when they arrive. Hanlon is a very good researcher. If he had had severity-of-diagnosis data, he would have adjusted for it. He has not done that not because he forgot but because the data are unavailable in all our countries. We do not have a neat measure of how sick people are when they come to A and E.
You should remember that we have done safety across the whole nation, with the safety programme and early warning scoring operating every day; the assumption is that that safety system is in place on Saturday and Sunday, just as it is on Monday. The system is not perfect, and sepsis care and infection care are not perfect. However, my focus in leading the safety programme and the focus of the hundreds of people who are doing that work is in making those things better every day. That needs attention on a Saturday, but it also needs attention on a Tuesday.
I am not being critical of the safety programme.
I understand that.
I do not think that anybody is—far from it. It is just a question of trying to work out whether the safety programme by itself is going to address the weekend issue and whether the issue at weekends is a lack of cover.
This is not a political issue. It is a reflection of society and the five-day week. It is not a reflection of the political Government of the day, but it has to be addressed by the Government of the day.
The patient safety programme addresses patient safety. It does not address the issue of weekend working and whether there is a problem or not.
That is correct. The work around seven-day services addresses staffing, diagnostics and all the other elements. I am confident that patient safety is not affected more at weekends than it is on weekdays. However, the seven-day working process is about making the system and the service better, and not just about making the service safer.
So, you have a seven-day programme despite the fact that you do not think there is necessarily a problem with—
The seven-day programme is not about making it safer. It is about improving the flow through the system and making delayed discharges better.
Traditionally, in my job, it was more difficult to discharge on a Saturday than on a Thursday or Friday. The family may have been perfectly happy to have the patient home on a Saturday, but doing so was more difficult because diagnostics and pharmacy services were not available. We are fixing that element of seven-day working. It was very unusual to do scheduled surgery on a Saturday, but now it is becoming more usual to do day surgery on a Saturday. That is what the seven-day working process is about.
Those are all good things.
Do you accept—this is the key thing—that a number of studies in Scotland, the UK and internationally have all suggested there is an issue at weekends? Do you accept that that is the case and do you believe that it applies in Scotland or not?
I believe that there are a number of studies that suggest that mortality is higher at the weekend than it is during the week, and I think that that may well be true in Scotland. What I do not accept is that that is a patient safety problem. I think that it is a severity-of-illness problem.
Dr Keel wants to comment.
I was just going to say the same. The evidence, such as it is, is deficient because we do not have the case mix or severity-of-illness scores for patients who come in at weekends. As Jason Leitch does, I think that they are probably a sicker cohort of patients. As he said, seven-day working is about trying to speed up the patient journey through hospitals because we know that the longer people stay in hospital, the more likely they are to get a healthcare-acquired infection. Patients do not want to be in hospital unnecessarily. The idea of sustainable seven-day services is to improve access to routine diagnostics at weekends and get patients discharged at weekends, rather than having to wait until the next week to get those investigations.
Until we have studies that look at the case mix of patients coming in at the weekend, compared with those coming in Monday to Friday, we will not know the answer to Mr Macintosh’s question about whether there is a problem. The data indicate that more patients are dying at the weekend, but do not tell us why.
10:30
Professor Leitch said much the same thing. Some of it might be down to sicker patients coming in, as you suggested. The nub of it is, can you say with certainty that there is no increase in mortality rates at the weekend over weekdays?
I can say quite the opposite. There is an increase in mortality rates at the weekends, compared with the weekdays.
So why did the Cabinet Secretary for Health and Wellbeing say with certainty that there was not that problem?
He was referencing the fact that it is not a patient safety issue. I can also show that there may be higher mortality rates on a Tuesday than on a Thursday. There is variation according to case mix. There is no systematic safety problem at weekends, compared to weekdays, that causes excess mortality.
So the answer that the cabinet secretary gave referred only to patient safety and not to the rates? We can check that.
What he actually said was:
“the programme is probably a major contributing factor to why the mortality rate at weekends is no higher than it is during the week.”—[Official Report, 12 March 2014; c 28811.]
You have just said, Professor Leitch, that the mortality rate at weekends is higher than it is during the week.
The Handel study, which is the best study that we have, found higher mortality rates at the weekend. More people die on a Saturday and a Sunday. My premise is that that is not to do with safety but that it is to do with case mix.
That is not what was said. You are saying that there is a higher mortality rate at the weekend. The cabinet secretary said that there is not. That is something that we need to explore further, so I think that we should move on now.
I appreciate the candid and thoughtful responses that we are receiving this morning, and I was grateful to Paul Gray for laying out in his correspondence the areas where key actions and improvements had taken place.
I would like to move on from the issues of data or statistics that Ken Macintosh was dealing with and get to the core of what we as a committee are trying to do, which is to find a positive way forward with regard to flow through hospitals. Before I do that, however, I want to comment on Mr Macintosh’s rather sweeping statement in his opening remarks that in recent years performance in A and E had deteriorated. From the figures for 2012-13 that Mr Gray has provided to us in his correspondence, it seems to me that waits of over four hours have reduced by 19.2 per cent and that waits of over 12 hours have reduced by 66.4 per cent. Mr Gray, can you confirm that I have got those figures right, and can you comment generally on what you believe to be the overall performance in A and E?
Over the winter period between November 2013 and March 2014, there was a 66 per cent reduction in patients remaining in A and E for more than 12 hours, and less than 1 per cent of all patients remained in A and E for longer than eight hours. That is the information that I have to hand.
Performance in A and E, which was 91.4 per cent in the previous year, is now up to 94 per cent. Indications from a number of boards—this is their data, not the published data—are that a number of them are continuing to meet the 95 per cent target. As I said in response to the convener, I am not certain that all boards will meet that target, but the trajectory is in the right direction.
We are treating 1.5 million patients a year and the vast majority of those patients get treatment within the time that we said they would get it. The target is 95 per cent because there are some people for whom it will not be clinically appropriate to have them out of A and E within four hours. We are talking about 1 to 2 per cent of patients who are not being seen and discharged or admitted within the time set.
A and E staff—consultants, trainees, nurses, other professionals, administrators—are working under high pressure, and I do not think that it is the committee’s intention to undermine that work in any way. Indeed, at the previous evidence session that I attended, the convener was quick to assure everyone that the committee did not intend to undermine the work of NHS staff. I appreciate that point and want to make it again.
It is quite interesting to look at where Scotland has been, in relation to the other home countries and those further abroad. For a considerable time, Scotland’s performance has been better than that of Northern Ireland or Wales and pretty comparable with that of England—in fact, it has been almost the same.
Scotland’s position on median waits is the best in the UK. It is roughly 10 per cent better than England’s and has been considerably better than that of Northern Ireland and Wales over recent years.
The committee might be interested in a study published in June 2014 by the Canadian Government that looked internationally at best practice in waiting times, particularly in A and E. Using the phrase “Imagine a land where”, it highlighted Scotland and then compared Scotland’s performance with Canada’s. It showed Scotland in a relatively positive light and is a good read.
Mary Scanlon asked about how individuals present themselves, which is, as you have accepted, something that we need to understand better. Where in Scotland is best practice taking place? From what we saw last week, one of those areas seems to be Tayside. Now that Tayside has reached that performance level, the job not only for boards but the centre of the organisation—in other words, Government—is to ensure that others can achieve the same high performance rates. How can we use the Tayside experience and other good practice to get other boards up to the same level of outcomes and help the people of Scotland?
Before I ask John Connaghan and Professor Leitch to provide more detail, I will give you an example to ensure that we do not give you just a series of generalised propositions. Andrew Russell, the medical director of NHS Tayside, has been to NHS Grampian to assist the development of its processes and protocols, including those for A and E, precisely because we believe that there are good lessons that Grampian can learn.
I want to say a few words about NHS Tayside and then make some comments about flow. Professor Leitch might also want to add to my update about flow.
A practical example of some of the good practice in Tayside is signposting, which addresses some of the points that Ms Scanlon raised about self-referrals. For some time now, Tayside has operated a relatively good signposting system; we took a look at it and, earlier this year, issued signposting guidance to boards.
Can you tell us for the public record what signposting is so that we can make it more visible?
Signposting is directing the patient to the most appropriate point of treatment, which could be an out-of-hours referral back to the GP or treatment in accident and emergency. Signposting is clearly important, because it gets the patient to the most appropriate treatment.
As I have said, we used the Tayside experience in guidance to NHS boards that was issued earlier this year. We are now reviewing how that has gone down and it is quite likely that, in the very near future, we will issue refreshed guidance based on the first six or nine months’ experience of its roll-out.
You used an important word—“flow”—and I will explain what I mean by that in a minute. However, I do not want to give you the impression that the flow programme for Scotland is a recent invention; flow has been addressed for a considerable number of years. Committee members might remember Audit Scotland reports on day-case surgery, which is one element of promoting better flow, because the more we move people from an in-patient setting to a day-case setting, the better the flow through in-patient beds and throughout the hospital resource.
Last year, in the course of our consideration of the national unscheduled care action plan, we established a national flow programme. We are piloting new techniques in four boards, and we have imported from the Institute for Healthcare Optimization the best international experience and advice on how to set up the programme. Particularly in NHS Forth Valley, we are at a fairly advanced stage of assessing how we can promote better flow.
There are three main components to the flow programme, the first of which is better utilisation of operating theatres. The second is smoothing the elective programme. As Professor Leitch said, when we look at elective care—that is, non-emergency in-patient care—we find differences between, for example, Mondays and Tuesdays and Fridays, Saturdays and Sundays. Smoothing out those differences will give boards a much better chance of being able to cope with unexpected peaks in demand for unscheduled care. Smoothing electives is important.
The third component is managing some of the natural variation in unscheduled care. An example of the kind of thing that we want to look at and promote is discharge time of day. When we profile how hospitals discharge patients, we find that far too many patients are discharged late in the day. If we could shift the curve and have more discharges much earlier in the day, we could ease the congestion that we sometimes see in some hospitals.
Those are a couple of practical examples. Professor Leitch will talk about other aspects of flow.
I will be brief. The sharing of best practice across systems the size of our NHS is a big challenge, and a global challenge is to find ways of sharing with everyone what is going well such as Lanarkshire’s hospital-at-home service, which is probably the best in the country, or NHS Tayside’s signposting system, which, again, is probably the best in the country.
We have a number of ways of doing that. We do it through improvement programmes, using improvement science; we have the safety programme, the early years collaborative and the person-centred care programme; and we have learning systems that create the opportunity for practitioners, in particular, to share best practice.
The quality and efficiency support team—or QuEST—which is part of John Connaghan’s organisation, applies the same method to efficiency and productivity, bringing people together to share data and best practice and sending people on visits. For example, Bill Morrison, the A and E consultant in Tayside, is regularly in other A and E departments, sharing what Tayside has done on signposting.
NHS Lanarkshire has started to use public advertising. Its nurse director is on the back of a number of buses—not literally, of course; she is on a poster—telling people about the most appropriate person to engage with, particularly over the winter.
I commend John Connaghan’s comments on the flow programme. Professor Litvak, who has worked principally in the United States, is probably the global expert on hospital flow, and we have engaged his organisation, which is working with Forth Valley first. Two weeks ago, John and I spent a day with the team, which is beginning to do the data crunching on flow through the board’s hospitals. We will then start work on what could be a fairly radical redesign of how we do scheduled care, in particular, and on engaging surgeons on how they might change their weeks. After that, we will spread the approach to another three boards. Having seen what it has achieved elsewhere, I am confident that it will make a significant difference.
The complexity and scale of what you have to deal with are quite mind-boggling.
You mentioned the Lanarkshire hospital-at-home service. We want to ensure that integration of social care and hospital care works better, and I assume that that is what the service is about. Will you say more about how it operates? Am I right in thinking that the service is designed to improve the delayed discharge position, thereby helping with flow and stopping the backlog in hospitals?
10:45
The service is provided by what is called the age specialist service emergency team, or ASSET. Most of us have visited it, and it is a shining light for how to do things. However, I do not want to give you the impression that it is the only such service in Scotland; NHS Ayrshire and Arran, for example, has a very good system. However, the Lanarkshire one—ASSET—fundamentally involves moving hospital care into houses.
A friend of mine is a carer for his elderly wife, who is very frail, with multiple morbidities. In my old world of hospital work, she would have been in hospital for a long time, but she never is. Instead, she is cared for at home, where intravenous fluids and antibiotics can be administered. Doctors and nurses can visit, and every morning, there are virtual ward rounds at a certain location during which each of the patients is discussed. Nurses will then go out to those people. I am astonished at how well sick people can be looked after in their own homes. That is certainly a big change.
I might make a mistake here, because I cannot quite remember the statistic, but I think that according to ASSET’s most recent data it has reduced over-75 admissions from 70 to 11 per cent. You might want to put brackets around that—I will get you the real figures later—but the system represents a fairly radical approach to the way in which we deliver care. Lots of people have visited the service, and lots of people are using it.
I should point out, however, that contexts are different. Inverness is different from Motherwell, and we cannot just take this sort of system to Inverness or to the Western Isles. It will need adjustment. In any case, people are increasingly using it; indeed, NHS Lothian is very interested in investing in it.
You are right to point out that it requires the integration of health and social care. Not all the people who visit the patients’ houses are national health service employees; they are also social workers and care workers, but the badges that they wear become less relevant to the family.
That helps our understanding. The system stops older people going into hospital in the first place. Given that some older people are still causing delayed discharges—I am not going to use the term that has been used in the past—can you give us a general feel for how the integrated social care work that has been going on through the relevant legislation will help to improve things over the next few years? That will help A and E, because—if I have got this right—more beds will be available and people will be able to get in there a bit quicker.
When an elderly person with multiple morbidities goes into hospital, their case is probably of a category that is more likely to become a delayed discharge. If the lady mentioned by Professor Leitch went into hospital, the difficulties associated with her getting out might be more profound than they would be in an ordinary case. If we can prevent older people from going into hospital in the first place, the likelihood of their becoming a delayed discharge will be reduced.
At the other end of things with regard to the integration of health and social care, we are bringing together provision by local authorities, the third sector and the health service. At the moment, people are waiting for care packages, which holds things up. What can we do to make the process for getting a care package slicker? Is there anything that the NHS can contribute to the development of care packages?
It is not just a matter of saying what the NHS’s job is and what the local authority’s job is, and never the twain shall meet. We need to have a conversation that brings together the people who are developing the care package to ensure that they understand better, from a health service point of view, what the individual actually needs and to ensure that we avoid making a mechanistic assumption in all cases that someone cannot leave hospital until they have X.
I can give you a very simple example. A hindrance to a person’s being able to leave hospital was that they had to be able to go up three steps. The person in question might live in a bungalow, but the standardised approach meant that, until they could go up three steps, they could not leave hospital. If my mother was in hospital—mercifully, she is not—and was asked to go up three steps, she would probably never get out. The point is that we need to make these conversations happen so that we take away any misunderstanding—however well meant things might be—between the various aspects of the care provision.
I know that this discussion is about accident and emergency, but there is a 75 per cent correlation between delayed discharge and increased pressure on accident and emergency. In many pockets of Scotland, there is a straightforward lack of care home places. One of the discussions that therefore has to happen—and which is happening—between the NHS and local authorities is about what we can do to provide more step-down facilities and how we can ensure a sufficiency of care home places. The committee will be aware that Glasgow City Council was not able to let a contract for care homes because of the economic conditions and the differential between what the council was prepared to pay and what the market wanted.
The integration joint boards and the chief officers of those boards will work with the health service and local authorities to seek to address a number of issues, but market conditions are also involved. I do not want to leave that point out.
I recall from my time as a council leader—I am sure that the convener will, too—that councils sometimes withdraw from the market and that, in those circumstances, the private sector is left to deal with the market in its own way, so sometimes a wee bit of regulation is needed.
I have taken up a fair bit of time, convener, so I thank you for your forbearance, and I thank the officials for answering my questions.
I will stick with the issue of what we can do to improve things. Earlier, I referred to Dr Martin McKechnie. He said that
“there was no problem recruiting young doctors to the first years of emergency medicine training in Scotland, but they were not completing the course to become senior doctors or consultants.”
Last week, we heard from Mr Thakore from Tayside about his concern that medical students are being asked to specialise very early, sometimes before they have even completed their courses, I think. He said that that is prejudicial not just to accident and emergency but to their training. Are you looking at that issue?
We are certainly keen to ensure that such flexibilities as can be made available are made available. I saw the point to which you refer. Dr Keel can say a bit more about that.
The current trainee doctor recruitment system, which is called modernising medical careers, was introduced in about 2006-07. The aim was to better match the number of trainees to the number of expected consultant jobs at the end of their training. Doctors were recruited to what was called run-through training, in which they were set on a career specialty course very early in their career, after they had done their house jobs. As I said, the main aim was to better match the number of doctors that we were training to the number of available consultant or GP posts, but the aim was also to shorten the length of time that it took to train a specialist. In fact, the average length of specialty training is still between eight and nine years, so it has not got much shorter.
Some members may be aware of the Greenaway review, which was published a few months ago and which looks at the shape of postgraduate medical training, recognising the new world that we all inhabit. Much of what we have talked about today is set in the context of an ageing population, more people having more than one health condition, or multimorbidities, and the question whether we have the medical workforce that is best able to deal with the new population of patients. The conclusion probably is that we do not have that at the moment and that we need to roll back a bit from the subspecialisation to produce more generalist doctors who are better able to cope with the whole patient and to deal with their multimorbidities.
Therefore, we are in a transition period, because we are beginning to explore how we might implement the Greenaway review across the UK. That would offer benefits not only to the NHS in providing a more flexible medical workforce, but to the trainee doctors in that they would be recruited to broad-based training schemes involving groups of conditions—such as women and children’s health—the training would bridge primary and secondary care, and there would be more opportunities for them to opt out of one particular course of training if they thought that it would not suit them. Therefore, there would be more flexibility in the workforce that we produced for the NHS and more flexibility for doctors, because they would not be locked into a specialist route.
We have a UK steering group, which, it so happens, is chaired by somebody from the Scottish Government. A number of stakeholder events have been held throughout the UK. We will gather the views from those and make a decision about how the Greenaway report should be implemented in Scotland. That will not happen overnight, but the aim is a better, more flexible, more generalist-trained medical workforce.
The message from what Bruce Crawford said was to recognise that performance in A and E has significantly improved over the years rather than deteriorated. I am less concerned about whether we achieve the 95 per cent target because, only a number of years ago, performance was 84 per cent and that was hailed as fantastic. Statistics can tell us different things and we can use them in different ways. As a member of the Public Audit Committee, I am more concerned to hear from witnesses that there are systems and processes in place to continually improve and address the issues that arise from time to time.
I am encouraged by many of the things that I have heard during this and previous meetings. I am particularly encouraged by what we heard from the NHS Tayside representatives who came last week. Bruce Crawford raised their issue—they talked about signposting and trying to deal with patients as they arrive to send them to the appropriate care route. I sincerely hope that such lessons are being learned and shared with the rest of the boards.
Can you say with any confidence that we will get to the 95 per cent? Do we have to? You said, Mr Gray, that it might not be appropriate for some patients to be pulled out of the system within four hours. I was quite struck by that. As politicians, we will react to the performance figure when you release it regardless of whether you achieve the 95 per cent target, but I would like to hear your view of that as health professionals. Are we going in the right direction? Are we improving the service? Will we reach the 95 per cent target? Do we really have to achieve it?
First of all, I should say that the health professionals who are with me today are Dr Keel and Professor Leitch—I have other professionals from other disciplines with me, too—so I will ask them to comment.
My view—I emphasise that it is my view—is that, when we say that we are going to do something, we should make a determined effort to do so. A target is set to be challenging; it is not set to be simple. I could just say that we should achieve a figure of 90 per cent and then we would be achieving the target all the time, but that does not seem to me to be realistic. For the sake of public confidence, when we say that we are going to do something, we should do it.
I spoke to the lead A and E consultant in the Borders general hospital about whether he thought that 95 per cent was the right number. We could argue whether 94 or 96 per cent is the right number but, in his view, the 95 per cent target gives a sufficient amount of what I call impetus to the system to ensure that people are not left in A and E beyond the point at which it is clinically appropriate for them to be there.
That consultant was equally clear that, in a number of complex and difficult cases, there are no benefits and some disadvantages to taking patients out of A and E if that is the best place for them to receive care. Therefore, having a 100 per cent target would be plain wrong, because it would disadvantage patients and mean that they got worse outcomes.
One could argue about a few percentage points either way but, as a national health service, we have committed to working towards the 95 per cent target. It is important for public confidence that we do so, but we should never at any point allow a target to cut across a safe clinical judgment.
Perhaps Dr Keel and Professor Leitch want to add to that.
11:00
I agree with Willie Coffey’s question—we need constantly to ask whether it is worth driving that extra percentage, but as Mr Gray said, it is clear that emergency medicine consultants think that the four-hour target is a good one, which gives the 5 per cent flexibility for those who need to be in A and E for longer.
In my professional life—this is going back many decades—I can remember patients languishing in A and E for well over 12 hours. They were there the next day when you went back to the department. The amount of improvement that has been achieved by NHS Scotland staff is quite incredible, even if you look back just a few years.
I think that the performance is great and it is clear that consultants—the medical profession—want the target to remain. They do not like all the targets that we have, but they like this one, so we need to stick with it.
I agree that we should strive for the 95 per cent target, but I also accept the premise of Willie Coffey’s question that it does not make a huge difference whether the target is 94, 93 or 96 per cent. To use a target to make simplistic judgments about the quality of services is not the right thing to do. That is one lever that we have to improve the quality of the service that we deliver to the population, and I think that we should keep it. Underneath that, however, the fact that we treat and discharge or admit half of patients within two hours probably says as much about the quality of the service that we deliver as performance against the 95 per cent figure does.
The target is part of a package of things around quality improvement methods, scrutiny and the delivery of the quality service that we should aim for. I emphasise Mr Gray’s point that at no time should the target supersede clinical judgment. If somebody should stay in A and E because they should wait for a surgeon, they should stay and wait for the surgeon. At no time should the target be used to undermine patient safety in any way. I am confident that that does not happen.
I am pretty sure that you will never hear anyone at the Public Audit Committee taking a view like that about targets.
A good example was raised during the previous evidence session—the issue of people being discharged at weekends and queues building up for admissions on Mondays and Tuesdays, which seemed quite an obvious area in which we could win. That could help push up the target, if that is what, collectively, we all want to achieve.
Are people discharged more slowly at the weekend across the NHS? There can be a glut of people arriving on Mondays and Tuesdays because they have waited all weekend to present. What can we do about that? How can we smooth that over across all the boards and push up the target even further?
You make a very good point, which I think I addressed in part when I talked earlier about the flow project. One of the things at the heart of the flow project is how we can better balance the other half of the work: the elective or planned work. In our experience, there is a weighting towards that work being done at the beginning of the week. Mondays and Tuesdays are very popular operating days for surgeons; Thursdays and Fridays are perhaps less popular. As I said earlier, one of the things at the heart of the seven-day project is consideration of how we can utilise the NHS’s entire resource over seven days, to smooth out those peaks and troughs.
Willie Coffey is right that discharging at weekends and discharging earlier in the day helps with the flow. It sounds simple—just discharge the patients—but the patients are often frail and elderly and often need adjustments to be made at home. They often do not have ready-made carers who just happen to be in the family, and they can require extensive drugs on discharge. The bag of drugs is only one element of the discharge process; patients require very clear instruction and education about what will happen with their drugs. It is not always as simple a process as we are led to believe, so we cannot just push them out at 3 o’clock on a Saturday and think that all will be well.
The seven-day project is about making that process better, in conjunction with social care colleagues and those who put in the little doors on patients’ showers so that they can be at home. It is not just healthy people who have had scheduled surgery whom we need to get out on a Saturday and a Sunday.
I will pick up on the debate about mortality at weekends. I was fascinated by the exchange between Ken Macintosh and Professor Leitch. If we look at statistics about anything, we can find a story, can we not? We can ask when people are more likely to be killed in a car accident, for example. There is probably a time and a day when that is more likely.
I understood you to say that the question is whether the figures that were discussed reflect neglect in the system or a lack of resources or management on a particular day, or whether they just reflect a characteristic of the population and general behaviour. I took it from your explanation that the answer is probably the latter, because we are uncertain about the reality and the facts.
People are people; we do not always act uniformly and consistently, and our behaviour varies. Until or unless you have data, analysis and research that pinpoint causes, we will be no further forward.
To be critical about my position, I am not remotely complacent about safety in Scotland’s hospitals. I do not think that anybody could accuse me of that. If anybody has been focused on the safety of our hospital care, I suggest that it is me through our leadership of the safety programme.
I care about the data and about making it better. If I see things in the data, narratives or stories that suggest something different, I will be the first to implement appropriate change.
You have not seen such data yet.
That is correct.
I will go back a question or two to Dr Keel’s answer about difficulties in relation to people going through training. Does the NHS track the destinations of people who have been trained? We hear all sorts of stories about people being trained who have set their minds on a future in research, for instance. How do you encourage people not to follow the popular subjects after training, given that you really want to fill local vacancies?
NHS Education for Scotland is beginning to do such tracking, which has not been commonly done but is becoming more feasible. As for people who train in emergency medicine, it is clear from trainee fill rates in A and E—the numbers of trainee posts that are filled—that the specialty is in difficulty. That is multifactorial. People in that specialty—consultants and trainees—work extremely hard. Younger doctors might not be as keen on the lifestyle choices that must be made to follow an emergency medicine career.
People in that specialty work under enormous pressure, at the hospital’s front door. That cannot be done for a career that will last 30 or 40 years. Increasingly, we must recognise that, as people go through their career and become consultants, they cannot do the sharp-end front-door stuff that they did when they were in their late 20s and early 30s. However, adjusting the system to accommodate that is difficult.
When young doctors enter training, they look at issues such as lifestyle choices. We know that significant—although not vast—numbers of them are choosing to emigrate to Australia or New Zealand. That relates to lifestyle and work patterns in those countries; it is not all about the climate. We are paying attention to all that. On retaining people to work in Scotland, we know that, if someone trains in Scotland and has a good training experience in a specialty, they are more likely to stay in the country.
Role models in the medical workforce are incredibly important to junior doctors. If someone ends up working with a consultant who is enthusiastic about how their career has panned out and about their work, they are much more likely to be enthused and to stick with the training. If—unfortunately—someone ends up with a person who is a bit more burned out and cynical, they will pick up on that and might not stick with the training. It is incredibly important and part of the chief medical officer’s role to ensure that the medical leadership is there to demonstrate to trainees that what they are doing is worth while and a rewarding career, and that they should stick with it.
The position is not unusual to Scotland; it applies everywhere.
Indeed. Emergency medicine has vacancies across the UK—filling roles in the specialty is difficult.
I thank all our witnesses for their contributions. It is clear that the area is challenging and there is no doubting our witnesses’ commitment to improvement.
We will have a short break before hearing from our next set of witnesses.
11:11 Meeting suspended.Next
Section 22 Report