Health Committee - Minutes of EvidenceHC 401

Back to Report

Oral Evidence

Taken before the Health Committee

on Tuesday 18 June 2013

Members present:

Mr Stephen Dorrell (Chair)

Andrew George

Grahame M. Morris

Mr Virendra Sharma

________________

Examination of Witnesses

Witnesses: Dr Karen Lowton, Senior Lecturer in Ageing and Health, Institute of Gerontology, King’s College London, Professor Alan Maynard OBE, Professor of Health Economics, Department of Health Sciences, University of York, Dr Sue Roberts CBE, Chair, Year of Care Partnerships, and Dr Judith Smith, Head of Policy, Nuffield Trust, gave evidence.

Q1 Chair: Good morning. Thank you for joining us. We are slightly sparsely attended this morning. A couple of our colleagues, we believe, are coming and are expected during the course of the session. I apologise that we are a bit of a thin attendance. We have more witnesses than we have members of the Committee, at least as we start this morning.

Could I ask you to open the session by introducing yourselves briefly and saying where you come from? Let us start with Dr Lowton.

Dr Lowton: Hello, my name is Dr Karen Lowton. I am a senior lecturer in ageing and health at the Institute of Gerontology at King’s College London.

Professor Maynard: I am Alan Maynard. I am an economist from the University of York. I have spent 30 years involved in NHS management. I am currently chair of Vale of York CCG.

Dr Roberts: I am Dr Sue Roberts. I am currently chair of Year of Care Partnerships, which is an NHS organisation housed in Northumbria Healthcare Trust. I have been a consultant physician for 30 years and was national clinical director for diabetes up until 2008.

Dr Smith: Good morning. I am Dr Judith Smith. I am director of policy at the Nuffield Trust, which is an independent charitable health research foundation based here in London. I have been involved in health services, research into primary care, chronic conditions and commissioning for nearly 20 years.

Q2 Chair: Thank you very much. As you know, this is the first evidence session in an inquiry on management of longterm conditions that the Committee has launched. I would like to open the discussion by looking at the structure of services required to deliver effective care for people with longterm conditions. You will know that there is a very strong sense in the health and care community that, if we are to deliver better quality and more economic care to people with longterm conditions, we are only going to do that if there are more integrated services between health and social care and more emphasis on prevention, early intervention and communitybased services rather than acute services. I start with that as a prevailing orthodoxy.

I would like to begin, if I may-in particular, because Professor Maynard in his evidence to us has suggested the evidence base for this conclusion is not as strong as it might be-by exploring the extent of that consensus. First of all, is it an accurate characterisation of the consensus? Secondly, to what extent do you agree with it? Thirdly, is the agreement or otherwise based on an opinion or evidence? Can I start with Professor Maynard?

Professor Maynard: Thank you. Being ancient, I find it quite fascinating to look at the history of this because, if you look at the first "redisorganisation" of the NHS in 1974, there was a document from McKinsey and the people over in Brunel management. The first objective of that document, published by the DHSS in 1972, was to integrate all the services together. Here we are, 40 years later, and we have still not integrated them. Therefore, one asks the question: what is the incentive structure that prevents the better merging and integration of these services?

Of course, one then looks round for evidence. One wants evidence about the benefits to patients, and in many cases that is not well charted in the sense that ideally one would like randomised controlled trials, but there are few of those. If we do not have randomised controlled trials, we want quasi-experiments that tell us what the effects are. In many cases, we simply do not know whether various elements of community care work or do not work. But we also need information about cost. Being a boring economist, one just says one would like to see the cost of these new diabetes or COPD innovations, because if you decide to fund some sort of experiment you are committing resources and denying other patients care. Therefore, you have an obligation to look very carefully at the cost consequences of this.

Briefly and in summary, Chairman, if you look at the effectiveness evidence, it is very poor and patchy. If you look at the costeffectiveness evidence, in many cases it is simply absent. Therefore, for a lot of the espousal of community care and argument to improve the way in which services are coordinated to the likes of me, with longterm care problems, it is really very poor. In a CCG, you are faced with the problem of how to invest resources. You look for evidence, and it is not there. Essentially, everyone else in the NHS is doing the same thing, looking for evidence, not having it, and so you experiment. The tradition in the NHS for the last 40 years has been to experiment but not to evaluate. Therefore, we do not have that information.

Q3 Chair: Before throwing it to the other witnesses, can we be clear? Is that a comment on community care or is it a comment on more integrated health and community services?

Professor Maynard: It is a comment on both. We do not know what works in community care or quite how to integrate things. We run on a set of views that community care is good. But for instance it may well be that A and E is the best way of reducing admissions to hospitals. A RAND paper has just come out arguing that, if you look at integrated care, primary care does not have the capacity to do a very detailed workup. Therefore, sending patients to A and E and getting them to go more to A and E is a good way of working them up and denying them hospital admissions. So there are a lot of prejudices and views here, but we do not have that firm evidence base on which to proceed.

Q4 Chair: There must be something there that other witnesses would like to comment on. Shall I start with Dr Smith and then move along?

Dr Smith: Thank you, yes. Looking back over a similar period of time to that which Professor Maynard has suggested, we have to bear in mind that a lot of care has changed. There is a lot more care for people with chronic conditions that now takes place in, particularly, the primary care general practice setting that would not have done 20 or 30 years ago. For example, I am sure Dr Roberts will have things to say about diabetes care in that respect. We also see that a significantly greater number of people are living longer with a complex range, often, of longterm conditions, so the whole issue of coordination of care-what is often talked about as integration of care-is becoming ever more important.

Of course we see admissions continuing to rise, including those that would seem to be amenable to good care in the community. That has led to many experiments-often quite small scale-of trying to shift that curve, that rise in admissions. I agree with Professor Maynard that we lack good, longterm robust studies of the overall effect of those experiments. Far too often, they are expected to report too soon and are asked to look at measures such as cost and activity. The majority of those studies that we have carried out at the Nuffield Trust tend to have very disappointing results because we are often expected to report very early in the process. Also, we find that those interventions have often been quite poorly designed and there has not been careful thought given to the outcomes that are expected. The one conclusion I would reach on a look at the evidence that we have is that it is stronger in terms of patient experience and outcomes in relation to having care that is outside of hospital-that is, in the community-but much weaker in terms of the economic evidence. So it partly begs the question as to why we want to shift care.

Q5 Chair: Can I push back? For example, when both McKinsey, whose name has already been mentioned, and also the people from Torbay-Dr Wollaston, a member of the Committee, sends her apologies this morning-say of the Torbay structure of services that it has reduced demand on the local acute hospital services, is that simply not right? That appears to be in conflict with the proposition that both you and Professor Maynard advance, that there is no evidence.

Dr Smith: There is evidence but there are issues about the strength of that evidence. When it comes to looking at changes in emergency admissions, an initial look at that evidence will often suggest a reduction, but, when there are matched controlled studies that happen to look at a population elsewhere that is equivalent-say, for the sake of argument-to Torbay, you will often find there has been a similar change in that population, which begs the question that there may be other factors that have led to that change. It is often very difficult to attribute a change in admissions to the specific intervention. As I say, one of the problems is that we do not study for a long enough period of time.

Q6 Chair: What would be long enough?

Dr Smith: In the work that we have been doing at the Nuffield Trust, we suggest that we need studies of at least five years. Our sense often is that there is a tendency within NHS management to be, first of all, perhaps overly optimistic at the start and set what turn out to be unrealistic objectives, but also to want results very quickly to prove progress at a point at which it is just not realistic. We know from a much wider body of evidence about bringing about largescale change in healthcare that it can take several years to get new systems and ways of working up and running before you then start to see changes happen in terms of services and patient outcomes.

Q7 Chair: Can we bring in Dr Roberts and Dr Lowton?

Dr Roberts: Your summary at the beginning about what the consensus is was absolutely spot on. What is needed is to look within that consensus and unpack it a little bit, particularly where it comes from. It comes from the idea that longterm conditions are just the same as all the other conditions that can be cured and palliated over the years, but they are fundamentally different in terms of management. While technological and drug interventions are obviously part of the problem-that is why people are living longer-in fact the management of these conditions is in the hands of the individual; they make most of the decisions every day that are going to affect their outcome. We need to reframe or reconsider our thinking and build up what we do from the fundamental need of the patient. We need to think of integration, which is key to this but in a much wider way, of selfmanagement with clinical care, integration of the person with the professional-all sorts of other modalities of integration-and form must follow function. So we are indeed going to need a number of the forms of integration that you have alluded to, but we have to be careful when we put them in that they do in fact support what patients need and not traditional thinking.

Q8 Chair: But it is odd, is it not, given the strength of the professional and managerial view in support of the basic analysis you offer, that the research base-the evidence base for the proposition you advance-is regularly reported to be so weak?

Dr Roberts: I do not think it is a surprise at all in a sense. To go back, the first thing is that when there is no evidence it does not mean that it is not effective. It just means there is no evidence. People who resist change, for whatever reason, often use the lack of evidence as a reason to do so: "The problem is too pressing for us to rely on that." The reason that I do not think it is a surprise is that all of us, I suspect, come from the same scientific sort of background, which is that you take an element and you reduce it to something you can experiment on because that is the easy way-to put it in a test tube and get results.

The totality of international evidence about longterm conditions says that this is a complex intervention. It can be reduced to three components. Basically, it is improving structure and becoming more proactive, which is probably the necessary but not sufficient element, and the other two that make it sufficient are involving people in their own care and a partnership approach between these two. So we endlessly design better research around one or other component and it is not going to work because this is complex. We are not going to get results and see the benefit of it because we have to look at the totality. I am all in favour of what both of my colleagues have said, which is that we fundamentally need more research-massively more research-in these experiments, but the experiments have to be the wholesystem experiments that are going to make a change, not small parts of it where we are not going to see a difference.

Q9 Chair: Okay. I will come to Dr Lowton in a moment, but Professor Maynard was shaking his head vigorously when you said what you need is more research. I thought that is what you were after.

Professor Maynard: What I would like is better research. We still have the issue that we are pouring a considerable amount of public resources into research, but one then has to ask the question: why is it not producing answers to these questions? Over the last 10 years-and further back with Cochrane and goodness knows what-we have put a lot of money into research, yet we are still at a really rather primeval level with regard to evidence. We also have to be pretty critical about how research funding is done. If you look at, shall we say, Iain Chalmers and other people, they would argue, like I would argue, that it is a question not of more funding for research but of better research. We teach people how to do good research and then they do outrageous things, bias their results and generally drive mere academics up the wall. So the issue is not necessarily more in the sense of more money but rather much betterdesigned research articulated to answer these questions.

Q10 Chair: Do you disagree with that, Dr Roberts?

Dr Roberts: I think that the need is to have better interventions so that the appropriate research can be applied. What we do not have is the appropriate interventions. We have some suggestions.

Q11 Chair: We will come on to those. Dr Lowton.

Dr Lowton: I want to raise the issue of new ageing populations in terms of the longterm conditions debate. There are growing groups of people with rare, very complex conditions, who, as they age successfully through medical interventions and changes in social attitudes to people with rare conditions and disabilities, are becoming greater in number, and yet we do not, as my colleagues here say, have an evidence base, in terms of counting the numbers of people who are growing older with complex comorbidities, knowing which services they should fit into. Most people are treated in specialist units that cannot be moved into the community because you need a concentration of expert care and specialist equipment, but we do not know how their care can be integrated within the community so that they are supported to selfmanage often very difficult complex conditions. Also, clinicians caring for them often do not know what is going to happen next because many of these cohorts are the oldest; there is no cohort ahead of them. One of the problems is: how do you manage evidencebased medicine and how do you commission if you do not know what is going to happen next with these particular patients?

Chair: Thank you very much. That was an interesting opening of the discussion.

Q12 Mr Sharma: Proposals for service redesign are around "ambulatory care sensitive conditions", conditions appropriate for treatment in the primary or community care settings. My question is in two stages. Is primary and community care the most appropriate setting for the treatment of all longterm conditions? Where do acute services need to change to support better the needs of people with longterm conditions who still need acute care? Does it make sense?

Dr Roberts: It makes sense completely as a question and I think the answer is clear: primary care is, or should be, the default position for this. It has a unique opportunity to, in a sense, act as the hub between some of the things that I am advocating within better services and a better community, a more supportive community for people to live their lives, but it also is the hub and the link into specialist care. The design of the services that we have been testing would enable the right care to get to the right person at the right time, which it does not necessarily do. To make that possible it does require some changes in primary care. But it also requires changes in specialist care to make sure that their special expertise is available in many more places than it is now. We have a very rigid system in which we have specialists attached to buildings and hospitals. We need to decouple our thinking about this critical issue. After all, I am a specialist and have been for 30 years so I am very keen that people should get the benefit of my knowledge and expertise, but I do not think it necessarily needs to be done in any particular building and in any particular way. That is something that we should be really looking at seriously and looking at the incentives, things like consultant job plans, to enable specialist expertise to be much more widely available. But primary care should be the hub of this.

Dr Smith: British primary care has two really important factors that are often regarded internationally as being very special. They hold the key to some of this. The first one is that our general practices have a registered list, a population for whom they are responsible. The second is that general practice is supposed to have a role in coordinating people’s care wherever they go in the wider health system. In relation to the first one about the registered list, I do not think at the moment that general practice always has the time and capacity to make the most of that resource, of the registered list. In fact, the Nuffield Trust, together with the National Association of Primary Care, recently published a report called "Reclaiming a population health perspective", arguing that general practice could use that registered list to be doing work much further upstream, to be looking at the demography, the epidemiology indeed, of that population and saying something like, "We have 2,000 smokers in our practice. What should we be doing about that?", or, "We have 2,500 people whose bodymass index is above 30. What should we be doing?" So actually getting upstream, which is critical in terms of longterm conditions, much more could be done there. Certainly, there are innovative general practices and general practitioners who are starting to think about how they could start to take more of a population health perspective with their teams.

On the second point about general practice as the coordinator of care in the wider system, again, there are practices that are starting to think about ways in which they could take that further. They are often doing that by operating perhaps more at scale, now starting to work in networks or federations of general practices. Because of that scale, they have the opportunity to perhaps contract for the sort of specialist expertise that Dr Roberts was talking about and have that much more directly available to them locally and design and put into practice some different approaches to care. As to those two factors about the population health role of general practice and the coordinating role, there is much we could learn from some innovative examples that are out there.

Professor Maynard: Yes. We have known that there is a relationship between GP and hospital admissions for about 20 years. I remember Clive Smee, who was chief economic adviser to the Department of Health, arguing this point 20 years ago, and Jarman argues the same sort of thing. What we seem to have done is relatively deprive primary care as the demand has gone up. So there are some really major policy issues here, ie "You could quite easily advocate more funding for primary care," but what are you going to cut to do that?

The other little point I would make is about hospitals. Hospitals can be dreary places-he said, as a patient-but they do benefit from economies of scale. That is a significant plus. If you take things out into the community and you have, shall we say, nurses going round in classic community care, rather than patients going to the diabetes nurse in the hospital, you may have differences in productivity, cost and effectiveness. So one should not necessarily dismiss hospitals as something that we need to reduce. It is rather like the mergers literature on hospitals. We seem to be merging hospitals, and we have again known since the time that the Chairman was Secretary of State-we were publishing things about it-that there is no evidence that hospital mergers reduce costs or improve quality, but we continue to do it. It is a sort of rhetoric-

Chair: If you make it smaller-

Mr Sharma: It is debatable.

Professor Maynard: I think the questions you are raising about primary care, where you put your resources, raise this issue of how you shift resources within a constant budget. That is the problem with this community care advocacy. The risk is that we will invest in community care and you will make the quality of my care more expensive but we will not shrink the hospitals and the total cost rises. So you have to ask the question: is community care a complement or a substitute? What time horizon is there for ensuring that they are substitutes?

Q13 Grahame M. Morris: Can I ask a supplementary there of Professor Maynard with regard to the lack of evidence and what seems to be established logic about it being cost-effective? What about the international evidence? As a Committee, we visited Sweden and looked at some of their investments in communitybased services. I found it difficult to understand how integrated teams travelling relatively long distances, visiting patients in their own homes, would be more costeffective than bringing them into a single centre where they are all housed in a multidisciplinary group under one roof, but the evidence does seem to support that being a wise use of resources. It has had the consequential effect of reducing the number of hospitals, the number of beds in hospitals and improving outcomes, which seems remarkable, but that was their experience based on their evidence. Had you looked at that?

Professor Maynard: I know something about the Swedish system. I see what is argued about them. Some of my colleagues in my clinical commissioning group have been over in Sweden recently. I am sceptical because, again, the issue that Judith was talking about was: what is the design, what are the controls and what is the cost data? I am a mere economist and I am not interested in evidencebased medicine; I am interested in economicsbased medicine because what is clinically effective may not be costeffective and what is costeffective isn’t always clinically effective. One has to get into these designs not just the effectiveness for improving my outcomes as a patient but also the costing implications and the shifting of resources.

Q14 Chair: You also have to get into the design, do you not, a sense of the value you are delivering for the cost, whether it is the objective clinical outcome or the patient’s experience on the outcome achieved as well: the patient’s view about the outcome?

Professor Maynard: Yes, but in the treatment I had I was distressed, let alone not satisfied, but they kept me alive, so I am grateful. The issue I always end up thinking about is, "That was a pretty horrible process-that chemotherapy was not very nice-but at least it gave me the outcome I wanted": that I could stagger through your door this morning.

Q15 Chair: So you are capable of working out that the personal cost, in terms of discomfort, was worth the benefit. But people with longterm conditions requiring them to travel to a remote hospital for a relatively less technical intervention than the one you personally experienced, where that same intervention could be delivered more conveniently to them, is something that is a cost without a benefit, is it not?

Professor Maynard: Yes. There are obvious tradeoffs there about how you treat different conditions. If you have kidney disease, you are going to be looking at your creatinine, which needs a blood test, and you can have a telephone call with your renal physician. If you are looking at some form of cancer, you are looking at your paraprotein and whatever, and you could do a lot of that at distance. But, basically, if you are getting into complications, then clearly you want to be there. So it is horses for courses.

Dr Roberts: We are in danger of looking at longterm conditions as acute conditions that always need treatment and things being done to people. The international evidence that the redesign has been built on would suggest that we need to be doing things with people that help people live in their communities. Those people who now have complications and have an acute episode on the basis of their longterm condition do use a lot of expensive resources, but they are very much smaller in terms of the total numbers of people we are dealing with. Unless we turn the tap off and prevent those complications coming and unless we help people to contribute-and people want to contribute, so this is not just a costcutting episode-to their own care, and there are now ways that enable them to do that, we should be thinking about the services provided right across, not just when they get their need for acute intervention, or we will lose the plot.

Dr Smith: That is absolutely right. Going back to my point about primary care and the potential for population health management there-which is absolutely building on what Sue was just saying-I agree that there are economies of scale in hospitals, and I want to add that I think there are economies of scale to be had in primary care practices working in a much more networked and federated manner, getting beyond, in a sense, small corner shops into something more organised and at scale, yet preserving the ability of people to access their local services.

To give one example of that and how we could extend things further, pharmacy is an area of primary care that already delivers some services to people with longterm conditions but could do much more in terms of both screening and helping people manage both use of medications and also in providing advice and support. That needs to be done thoughtfully in terms of a network, with sharing of information and records with general practice and so on. We could build up that base of support, advice and proactive work within the primary and community setting. We have many of the professional skills there but we are not always making the most of them.

Chair: I interrupted Virendra.

Q16 Mr Sharma: Do you want to answer?

Dr Lowton: I agree with Dr Smith on the idea of networks. We have cancer networks now that appear to work very effectively where you have specialist care in specialist tertiary units, and then throughout the network at primary level, at acute care level, you have people trained to know exactly what to do. So you could have specialist care in a specialist hospital and perhaps more routine chemotherapy delivered much more locally. There are many conditions, certainly within the new ageing populations, that would benefit from not always going to tertiary care. We have just been following up the first British recipients of childhood liver transplants who are nearly 30 years on and they still attend Addenbrooke’s hospital on a sixmonthly or annual basis for a checkup. Many of those participants were saying how much they would prefer to have a consultation by Skype. To have to completely rely on tertiary care for a 10minute checkup is quite difficult when they are travelling from all over the country.

Chair: Thank you.

Q17 Mr Sharma: Dr Roberts, the Year of Care approach was trialled with diabetes patients. How adaptable is it to the treatment of other conditions? Also, what conditions would not be appropriate and has any assessment-although mentioned briefly-been made of the likely savings per patient from the Year of Care?

Dr Roberts: The Year of Care programme used diabetes as an exemplar, and diabetes is often used in longterm condition work because it affects every organ of the body-every organ has a bit of the NHS-so if you can get it right for diabetes you can get it right for other conditions. It was a feasibility study, so it did not make out that it was a randomised controlled trial. The purpose of it was to see if it was possible in UK situations, particularly in primary care, to work in new ways, ways that did involve the person, with diabetes in this case, in really being supported to get the skills, knowledge and confidence they needed. So it was a feasibility trial: could we do all the things that the patients’ organisations had been listing that they would like to have done?

We worked intensively with three very different populations, with primary care, to work out how to do it, and we successfully showed how to do it. We measured quite a few impacts of what we did. This was an uncontrolled trial so it was just what we observed there and it was extremely helpful in delivering the outcomes that the patients wanted. It improved experience and they perceived that they were getting more support for behaviour change. The thing that was astoundingly good was that it was a better way of working for staff. That was the thing that enabled this to embed and to carry on.

After that study-and we did very little economic cost working, measurement, for a variety of reasons-we were able to show that, within primary care overall, it was cost neutral to make the changes that were required, but it was not a serious study at all.

We then took the principles of it, which I have outlined in some of the papers that I have given you, produced a prototype and scaled it up to 12 other sites to show that it was transferable, reproducible, and solid. Then we took the principles and have transferred it to lung disease, COPD, to health checks, to cardiovascular disease, and, latterly, we are working with integrated health and social care teams with people with multimorbidity to show that that also is very effective and feasible for them.

What is needed now is for communities of practices to be doing this whole scale so it can be properly evaluated using appropriate methodology so that we can see a bit more about the economics and the impact. But it is feasible, highly popular with the people who do it and a better way of working. That is why we are commending it not just for diabetes but for the majority of longterm conditions.

Q18 Chair: Can I probe you a bit further? You say "the majority". You started with diabetes because that is an obvious case. There must be longterm conditions-possibly some of the new forms of longterm conditions-where the Year of Care is more difficult, more of a challenge. Are there areas you focus on where it is particularly appropriate?

Dr Roberts: The principles of what I am advocating are a very simple approach, which is that you identify the population that you are concentrating on from your register, or from a risk stratification tool in some way. You then invite them, essentially, to sit down and have a discussion that is forward looking about both how they would like their care to be managed and how they would like to contribute to this. The critical issue is that you have, in some way, to prepare them to be more involved in the consultation than they normally would. They need to know their results, their tests, things that they do not know at the moment. You need to find ways to help them to be involved. Then you have a discussion about, "How would you like your care to be managed?", and you are building around their own goals and contributing your expertise so that it becomes a meeting of two sorts of experts. Out of that comes an action plan.

For the majority of people with longterm conditions, that action plan will be things that they can do for themselves, particularly in the community if it is better or more systematically resourced. But, if they are somebody with either a very complex medical condition, or indeed some of the ones that Karen is talking about, quite a lot of the action plan might be things that the service does. So you have this combination of what the service will do and what you will do. That, if I set it out in terms of principles like that, is applicable for everyone. It is a mindset, "Who are we talking to? What sort of discussion can we have? How can we plan based around your goals and objectives?", and then, "What is the action that comes out of it?"

Q19 Chair: The key point about a Year of Care is that it is different for each-in theory, at least-individual patient.

Dr Roberts: That is the point. It is a common approach so that everyone can understand it. Public Health England, NHS England and Health Education England could all begin to put their priorities on to this model. It is a common approach and ends up with a personalised solution for each of the 15 million people.

Q20 Chair: Then, asking the economist question, how do you price 15 million individual personalised years of care?

Dr Roberts: Well, I am going to ask him.

Professor Maynard: Thank you for that. This is the issue. Here we have an obviously wellintentioned and nicely designed intervention. There are lots of them about. This one is very attractive, but it has to be accepted that it is a pilot and you really need to do a proper evaluation. The timing of how far you go with pilots before you do evaluation is important. Our problem is that we have lots of ingenious doctors-there are a couple here-always thinking of new things to do, how to change A and E, how to change community care, how to do neighbourhood care teams. It is always happening. But I stand there and say, "I demand evidence. Get used to thinking critically, please. Evaluate."

Q21 Chair: But there is a prior question before you even look at evidence about how Dr Roberts’s idea works in practice. You have to have some theory about how resources can be attached to a personalised Year of Care model-

Professor Maynard: Indeed.

Q22 Chair:-moving away from tariffs for standardised procedures towards resource allocation per patient. You have to have a theory about how you are going to do it in order to have a theory to test in research.

Professor Maynard: Yes.

Q23 Chair: How far down that road have we got?

Dr Roberts: Just as a contribution to this, we have not got nearly as far as we should. You will be aware of social income bonds, the concept that a number of people put in money and carry some of the risk around putting in this new intervention-for the reasons that Alan says, you have to have it in there-and then they will, after a period of time, reap the benefits, if such there are. There are a number of initiatives round the country, and one that I am aware of-because it is in my local patch-called "ways to wellness" is about putting in social income bonds for the sort of things that Judith was talking about across the community. They have begun to do some of the economics and finances on this and they have a model for the whole population saving-and that is the thing-and then they divide it by the number of people and get an artificial figure for what it will be per person. They come up with around £150 to £200 saving per person. Some of this and these models are beginning to be thought of practically and it might be useful for you to see their thinking.

Q24 Chair: It would be interesting. The whole concept of a Year of Care is another one of these concepts that is much talked about. How you attach resources to a Year of Care seems to me to be a critical question. You acknowledge that we are not as far down the road as we need to be, but we ought to be having a look at anybody who is out in front, it seems to me, and it would be interesting.

Dr Smith: What Dr Roberts has been talking about there with the Year of Care represents a different philosophical and cultural approach to care. It is about working with people rather than the health service doing things to them. So much of our health service is organised on that basis. Even a lot of the way that we pay general practice at the moment is about paying them to have people in to do certain checks and screening, or whatever it is. That is what is critical. As Professor Maynard was saying, if we are working to design really good experiments on this kind of different philosophical approach-which I think all of us would instinctively support: it feels like the way we would like to work, with professionals to help us manage our conditions-we need the researchers working closely with the practitioners and managers from the outset to design the intervention, the measures and the hopedfor outcomes. I would add into that, "With patients and users involved in that process." So then, right from the outset, you have something that is a really good largescale and quite longitudinal study. It needs those researchers involved from the outset. It is about that bringing together of the rigour with the cultural change. This is complicated stuff, but it does not mean we should not be trying to do it.

You are absolutely right that the payment and economics, in some ways, do drive so much behaviour, and, at the outset of that sort of experiment, I would be suggesting we need to be experimenting with some different incentives and payments as part of that, whether it is for the network of primary care practitioners or of primary care practitioners with specialists, whatever that may be.

Chair: Given that the process started in 1974, I am looking forward to the next step.

Mr Sharma: You are familiar with that.

Q25 Grahame M. Morris: May I move on now to effective commissioning? To continue the points that Dr Roberts was making regarding the treatment of diabetes as an example, I have seen in your written submission your concerns about the fragmentation of commissioning and some of the other barriers to effective commissioning, given the new architecture that we have, with clinical commissioning groups. Do you want to elaborate on that for the sake of the record?

Dr Roberts: Yes. It is one of the two big barriers. One is healthcare professional attitudes and the other one we found, as we did our feasibility work, was about the issue of fragmented commissioning. In a sense, the new structure has not been designed around this integrated approach to longterm condition management. There has not been-and there still is not really-a longterm condition strategy for England.

The components of care that the practices and the CCGs discovered as we were looking at the Year of Care programme were three. One was that there had to be change within primary care and, as you know, that is commissioned from NHS England and the local area teams. Second is the aspect that Judith was talking about, which is much better community support, healthy living centres, exercise on prescription and social prescribing. In a sense, that falls more within the health and wellbeing boards and working with the local authorities, and that is very positive. The third thing is of course-the bit that I have said we need to play down but it is critical-that people do need acute intervention, the best medicines and the best traditional clinical care, and that is the bit that is commissioned by the clinical commissioning groups.

So the three big groups of intervention that people with longterm conditions need and need to have joined up are, in the new world, led by three different commissioning mechanisms, NHS England for primary care, health and wellbeing boards for the community services and CCGs for specialist services. Obviously, I know everyone is trying to work to merge, meld and do the best that they can, but our sites found that this was not helpful and it would be good if there was an overall view for how longterm conditions should be commissioned.

Q26 Grahame M. Morris: In relation to the quality and outcomes framework, thinking about the current architecture of commissioning, is that problematic in terms of inhibiting better integration in commissioning services for longterm conditions?

Dr Roberts: I would be very scared about talking about this if your GP member was here, but, to comment as a nonGP, the impression as we go and work with primary care is that QOF has had quite a negative effect for longterm conditions within practices. It emphasised the single disease and the tick box. For instance, highly common musculoskeletal conditions are not really QOF conditions and get lost. QOF has undermined the capacity for primary care to work in a really rounded generalist way. It has moved most of the care to nurses, some of whom do this very well, but we have discovered none of them has consultation skills in any form in their training. They are trained in different ways to do different things. It has been unhelpful in a number of respects.

Q27 Grahame M. Morris: Thank you. I do not know if any of the other witnesses have any strong thoughts on barriers to effective commissioning. I did particularly want to ask Dr Smith about the evidence from the study that Nuffield had done, but if anyone would like to contribute-

Dr Smith: Before that, in response to what Dr Roberts was saying, I think clinical commissioning groups are well placed to commission for longterm conditions. Why do I think that? When we look back over 20 years’ evidence of different forms of primary careled commissioning, what we find time and again is that what they are really interested in is longterm condition care in primary and community services, not the secondary specialist care, and there is certainly no evidence that they can have an impact on the hospital side of care. That is where I say they will naturally want to focus their attention and where they can have a real impact.

The other thing is that we will be publishing shortly some research that we have done in partnership with the King’s Fund, where we have been looking at a set of clinical commissioning groups and the extent to which they see themselves as responsible for primary care services and quality. Quite a number of them do see that as part of their responsibility and feel that they have to be able to influence primary care and the sort of services it delivers if they are going to have an impact on longterm conditions and, indeed, urgent care. So, while there are some structural issues in the current reforms that could make that difficult, I think that thoughtful, innovative clinical commissioning groups could push ahead and really start to have an impact on longterm conditions and primary care. It is about looking for the opportunities rather than getting too preoccupied with structural barriers.

Q28 Grahame M. Morris: Without getting tied up with structures and so on, would it be advantageous if the health and wellbeing boards were the commissioners in terms of particularly the management of longterm conditions and commissioning preventativetype approaches to health? I am thinking of evolving policy in other parties.

Chair: That was first proposed by this Committee.

Dr Smith: Evolution would be significant there because many of us would quake at the thought of another reorganisation of commissioning. One of the profound problems that commissioning has faced over the past two decades or more has been the regular reorganisation or, as Professor Maynard always rightly puts it, "redisorganisation" of commissioning.

The other caution I would sound here goes back to research evidence. The research evidence base on what has always been called joint commissioning, or partnership working, is not terribly strong in terms of outcomes for users and carers. In fact, there is a new, very thoughtful good study from the university of Birmingham that came out in February-I think it was a National Institute for Health Research study-of what are considered leadingedge examples of joint commissioning but suggesting that it is often in the area of achieving outcomes that those sorts of arrangements struggle. It goes back to this-that they might be quite positive for those people working in them and how they feel about working within those arrangements, but, in terms of the actual outcome, the evidence is equivocal. So we have to be cautious about assuming that that is an automatic solution.

Q29 Grahame M. Morris: Professor Maynard, you wanted to say something there. I want to come back to you about the evidence, but-

Professor Maynard: You want to go back.

Q30 Grahame M. Morris: You mentioned the university of Birmingham study, but could you say something about the Nuffield’s own study into Calderdale, Somerset and the Wirral? What evidence did you discover there about moves towards integrated commissioning, that that is a good model that could be applied elsewhere?

Dr Smith: The study to which you refer was focusing on the commissioning of care for people with longterm conditions in those three geographical areas. You talk about integrated commissioning, but what is interesting about that study is that the integration we saw, more significantly, was between the commissioners and the providers. We saw very collaborative work going on between what at the time were primary care trusts and practicebased commissioners. Whether it was the hospitals or mental health organisations and so on, the collaboration was happening there.

Q31 Grahame M. Morris: Rather than between health and social care.

Dr Smith: Yes. There were some with health and social care, particularly in the area of dementia. The example I remember there is the work being done in the Wirral around commissioning new forms of services for people with dementia, particularly around early diagnosis and treatment. But, as I say, more significant in our study was the integration we saw with providers, and we can perhaps come back to some of the issues it raises.

The final thing I would say that we saw in that study-and it was carried out between 2010 and 2012, which was clearly covering a period of very major restructuring and transition-was that, when we talked to and interviewed colleagues from local government in particular, they were finding it difficult because of all the changes to the NHS structures and relationships, so joint commissioning arrangements were being, in a sense, undermined and challenged. The other thing to point out is, of course, that local government was going through significant reductions in funding and people were changing on that side as well, so it was a difficult situation to be doing commissioning, albeit people were doing so.

Grahame M. Morris: Thank you.

Professor Maynard: If I can come in on the whole issue of commissioning, the difficulty as you try and develop community initiatives, and hopefully evaluate them, how do you take resources out of the providers? So if you put diabetes care into the community, the risk is that the resources are still going to stay in the hospital. They have this hopper of bodies that they can treat, and they pull down the hopper and get payment by results. The incentive system maintains the income of the hospital, which is a bit frustrating when you are trying to economise and improve quality. The risk is that you do your community stuff and improve the quality, but you are not really doing much in terms of saving resources because, basically, we are price takers. So if you are a CCG, we cannot fix the price, which is a bit strange. If you are Marks & Spencer buying knickers, you would want to adjust the price.

We are also quality takers in the sense that we are not really quite sure what we are buying. We do our best to find out and assure ourselves about what we are buying, but the quality measures are quite poor. That inequality in bargaining power between providers and the commissioners is a continuing problem. As Judith says, it is jolly good if you can get good collaboration and openness in your providers, but they are being crunched up by Monitor to make sure they hit their rate of returns and make a profit, and we, as a CCG, are also trying to make sure we balance. Commissioning is quite difficult.

The final point I would make is to agree with Judith. CCGs do have scope to begin to look at the quality of primary care and work with them through the GP forums. That is beginning to develop, but, of course, the GPs are extremely sensitive at the moment because, if you are trying to, shall we say, cut down outpatient followups, that is affecting the GPs’ work load. GPs are very fragile about anything that increases their work load at the moment, because, as the Nuffield evidence shows, GPs are working a damn sight harder: the consultation rates have gone up quite sharply and some of them are quite near the end of their tether.

Q32 Chair: It does rather beg the question whether, rethinking the way services are delivered within primary care and between primary care, social care and the rest of healthcare, you might not end up with a scenario that was more rewarding for both the GP and the patient because it used the GP’s skills more effectively than many of the consultations GPs currently undertake.

Professor Maynard: That is so, and, being the usual provocative economist, as you begin to move from corner-shop GPs to Tesco Express, you might begin to get much greater use of specialist nurses, with the reservations that Sue has mentioned. But can you use specialist nurses as substitutes for GPs as opposed to complements to GPs? That, again, has a research base that we can argue about.

Q33 Chair: It was one of the questions that were prompted in my mind by Dr Smith’s earlier suggestions around NHS general practice, that its two strengths were the coordinating function, provided we can do that on a 24/7 responsibility basis, but, more importantly, the single list. I wonder-and you were talking about networks and developing scale in primary care-whether the concept of a single list still means anything if you are talking about a much larger scale of practice in primary care or whether the natural consequence of scale in primary care is more specialism in primary care so that the individual GPs do not have a list that covers maternity and elderly care, for example.

Dr Smith: For me, having the population basis to primary care is really important even as it moves, as I think it will, to a greater scale. What that population base-we can call it the list-ends up meaning might be different. If it is a population of 50,000 that a particular large merged partnership or a federation is responsible for, that gives them the basis for what we were talking about, for doing preventative work, proactive work, for thinking about the sorts of services and all of that.

As to the point about the connection each of us has with an individual practitioner, whether that is the GP or the nurse, a lot of people would argue that that has been to quite a large extent broken down anyway. That needs building back into the new thinking that people are starting to do about primary care because there are some people for whom that connection with the particular practitioner is particularly important, and, given the topic of the conversation this morning, people with longterm conditions, particularly complex longterm conditions, will very much need that person who they relate to, who coordinates their care. It may be less important if you are someone who is much more rarely using primary care, but the point is that you can have that continuity and, in a sense, the list works for you in that way when you need it.

Q34 Chair: Are you not saying that for some people continuity is important and for other people it is not?

Dr Smith: Yes-

Q35 Chair: I am not sure what the concept of a list associated with the practice means if the list is 50,000 people. We are all registered with the NHS because we all pay for it, but beyond that I am not sure the concept of a populationbased list means anything to the individual. It certainly does to the use of resources for public health purposes.

Dr Smith: It goes back to what they do with that population. If they have that list and they-I do not know-know that someone smokes or weighs too much or they want to work with them in a certain way to do something about that, it is going to mean something to someone when the practice proactively gets in touch with them and offers them some interventions and support. But for other people it might just-

Q36 Chair: The practice, if it has a population of 50,000, is indistinguishable from just the NHS. It is not a personal general practice any more, is it?

Dr Smith: That is the challenge for the scaledup-

Q37 Chair: It is not necessarily wrong for that, let me say.

Dr Smith: No. The challenge for these scaledup models-some of them now exist as merged partnerships, some are called superpartnerships, others are federations and networks-is how they work at scale but yet retain the personal connection with people and provide that support. This goes back to general practice and primary care becoming more complex because of what we are talking about this morning: people with longterm conditions living for much longer but with a different range of needs. In a sense, how primary care responds to that is absolutely about dealing with both the personal on the one hand but also the population approach on the other. People are working that out. It is not the majority, but there are groups of primary care practitioners who are working to do this. There are many others who would like to, if they had the time and the space. I absolutely support what Alan says, that many of them are at the end of their tether and do not have the head space or the time to think through these new approaches.

Dr Roberts: It might be worth remembering that 65% of work in general practice is already longterm conditions and I accept that there may be more that comes there, but it is not managed at all in terms of being efficient or effective. People are just coming in and having a small amount of their issues dealt with in a most inefficient way. The advantage of a list, whether it is with the practice or a group of practices, is that you can then start to organise this in a way that we can see, from the impact of our work, that it then begins to make a difference . We have not got the data because one of the tragedies is that there is very little data on how practices organise within themselves. So you cannot show at the moment, sadly, that if you spend slightly longer with a person and plan for their next period of care-and then they do not come back-that you have made good use even of time resources let alone work force resources. But the list should help that to be organised.

Q38 Andrew George: First of all, my apologies. As with a lot of the members, it seems that this morning there is a serious problem of diary clashes. If the area of questions that I raise repeats areas that may have been covered before, I apologise.

In the evidence you presented, Professor Maynard, you question the received wisdom about-which I will broadly paraphrase-the approach that discourages the use of acute admissions when acute episodes occur with longterm conditions. The accepted mantra is that of course we should avoid the acute episode happening in the first place and that if an acute episode were to happen we should avoid unnecessary admissions and achieve the most rapid discharge possible from acute hospitals.

You seem to be questioning that. To what extent do you think that all of the discourse around this is using honest language? Often, commissioners talk in terms of the worst place for a patient to be is in an acute hospital-if that is the case, it seems that senior Royals are put at great risk by going into acute hospitals and staying there for quite long periods of time-which concerns me. Is it that it is medically risky for them to go there, or is it that we are being dishonest and really talking about the efficient use of resources and that we want to save money and achieve greater efficiencies by avoiding unnecessary admissions and achieving earlier discharges?

Professor Maynard: The first point is that it is statistically proven that the NHS is more risky than bungee jumping. The error rates are quite high. They may kill you, maim you and give you a nice infection, and that is in the public and the private sector, so it is not an ideological point. You can see that healthcare is risky, so you obviously want to get them into the right position. The position we were arguing for earlier on was essentially that we have not got a robust evidence base that says, "Does it work? Does it maintain the quality and length of my life at a reasonable cost?"-its costeffectiveness. We do not have that evidence base. Therefore, there are many fashions like, "Community care is good."

What is community care? We have to break it down and look at different groups and different interventions. But, even when you do that, you still have the problem of evidence, and it is poor. We like to denigrate A and E and say it is all collapsing and that hospitals are even more dangerous. They are clearly appropriate places at times during the pattern of care. If you are having an acute episode, what you want is a good diagnosis in the hospital and exploitation of the economies of scale in terms of the diagnostics. They can throw you into an MRI scanner and work out what is going on.

We tend to overemphasise and generalise. The role some of us have been talking about this morning is to say, "Stop, we really must move from pilots,"-which are all fine and dandy-"and evaluate in a very systematic way." Judith was saying yes, in about five years and I would agree with that. You need to have a mechanism of reining in impatient politicians, like your good selves, who want instant answers. Sadly, the world does not work like that. If we are going to get a good evaluation, we all have to be patient because the outcome may be that it does not work, in which case we can save a lot of resources that might be wasted by speedy implementation.

Q39 Andrew George: I will be interested in the views of others, but, if that is the case, it seems that a lot of hospitals at the acute end, at least, are operating on the basis of red alert. In other words, they have pruned their inpatient beds to such an extent that a lot of the theoretical-like approach that you are taking does not allow them the latitude to bring patients in and to give them the time for diagnostics and to review their condition before a managed discharge. Is that not true?

Professor Maynard: That is the risk. One has to be very cautious both about reducing the number of beds and about mergers because, again, you are back to, "Where is the evidence?" We have known for about 20 years that mergers do not work, as we mentioned earlier on, so we have to be very cautious and ask all the time, "Where is the evidence?", and really say, "If you are going to do something like change the composition of your A and E and put more GPs into A and E, where is the evidence that that works?" GPs are damned expensive. They may be effective at controlling the trivial demand going into A and E-perhaps 20% and 30% of it is trivial and the GP could manage it and keep it out from the specialist people-but one all the time is saying, "Where is the evidence, please?" and you can put up a sign. We do not know. "Everyone is doing it," we might say, but we are not sure whether what they are doing is costeffective.

Q40 Andrew George: Most of the assessments of those approaches appear to be on the immediate judgment of how many patients of those GPs in the receiving units manage to avoid being admitted to hospital. There never seems to be a followthrough as to how many then are readmitted to hospital because of a misdiagnosis. Is there any work of that nature going on?

Professor Maynard: I am not aware of it; maybe Judith is. It is a classic example of you put a policy in and the basis on which you continue or reverse it, which is often the case in the NHS, and you say, "Stick the GPs in and then the GPs out." The bases of those decisions are usually quite arbitrary and financially driven.

Q41 Andrew George: Is this a fair analysis? In other words, is the mantra wrong? It seems to be accepted mantra that would avoid unnecessary admissions. No one believes that there should be unnecessary admissions, but this kind of approach-the early discharge to often threadbare community or primary care services-may not necessarily be working. Is that fair?

Dr Smith: As to the desire for care to be in the community-whether that is home- based or in some other community facility-that is probably what the majority of us would want, and I think there is evidence around patient experience and outcomes from a quality perspective that that can be a positive thing. Also, as Alan has suggested, hospitals can be quite dangerous places as well, but also the majority of us, given the choice, would rather be at home.

We have also explored quite a bit the fact that it is on the costeffectiveness side of things that we really do lack evidence about the alternatives, so it partly begs the question of: why is it we want to make the shift? If it is around quality and outcomes, we are probably on a reasonable track, but given the financial context in which the health service is working-one that looks like it is going to remain austere for at least another decade-the costeffectiveness aspect is becoming even more important. What feels to us to be really important is the need for researchers to be working closely with practitioners and managers from the outset to carefully design these different interventions that are used to try and shift care but to put in place proper measures and then have a thorough longterm study, not requiring results after a year-when, at best, probably people have just got some of the governance and other arrangements in place-but over three to five years to see what works or does not.

Q42 Andrew George: This is the last one from me. You would accept my assertion at the beginning, as to such decisions being clinically driven, that it would be better and more honest to say that this approach is one that needs to be explained in resource terms and to use language such as that it is about the efficient use of resources to avoid unnecessary admissions and to achieve earlier discharge. That would be a more honest narrative.

Dr Smith: In resource and quality terms. I would always put the two together. For me, effectiveness is about those two.

Q43 Chair: Dr Roberts, do you want to comment on that?

Dr Roberts: Listening to this, I think the reason that we are in the situation that we are now is because we are just looking at one part of the system-the obvious one, in fact, where we are up against it at the moment-which is emergency admissions to acute hospitals. By only concentrating on those, in a sense, and only accepting that we are going to do two things to people-"Shall we do them there or there?"-we miss the point of how we should be managing the whole population, picking up things that Judith said earlier about the potential for CCGs to manage the whole system of longterm conditions, and move from, "We are going to treat and do things to you," to, "We are going to involve you much more throughout your whole life in how this is going to happen."

Q44 Mr Sharma: Before I put the question, I hope that the present Secretary of State for Health, Professor, takes your point of view on the mergers.

Professor Maynard: I will rely on you to tell him.

Chair: He will be here in a couple of weeks. You can ask him then.

Q45 Mr Sharma: Dr Roberts, in a typical CCG, what sort of changes would be required to successfully implement a delivery system for care of longterm conditions based on the Year of Care approach? What are the essential prerequisites for establishing this delivery system in a local care area? Where would you start?

Dr Roberts: Gosh.

Q46 Mr Sharma: You can say where you would start and go up from there.

Dr Roberts: Okay. If a new place wants to start doing this-a clinical commissioning group-they have to have a clear understanding of the principles that I outlined before, of what we are talking about and what in fact the delivery system is so that they can see it. Therefore, it would be enormously useful if the centre would say, without prescribing-I am very keen that there should be not prescribing that one must do things-"This is an extremely good way to work and you should start designing your services around that." I think a clearly articulated narrative is needed. Then there needs to be some local support for this.

There are two elements that are going to have to be done. Primary care needs support and training: it is not just what you do differently in primary care, but how you do it. That is why it needs to be quality assured, and, by that, I mean training that would be the same everywhere that gives the impact that we expect. That has to be available for primary care. That training needs to address attitudes. We know at the moment that 60% of GPs do not endorse people being independent decision makers. Training that challenges these points of view is the starting place. You are not going to change your ways if you still think people are here just for you to do things to them. So it has to incorporate challenging attitudes, helping people to get new skills, to support people and coach them to be better able to solve problems themselves.

Then, it helps them to reorganise. We have this house model and say, "How would you build your house in your practice to enable this to come about?" Commissioners need to commission that sort of training. I would like to see that happen locally, but then I would like to see groups of CCGs, or groups in primary care that are doing this, come together for mutual learning as to how it worked and what the barriers were. There are a couple of barriers. There are some barriers to do with IT, and it would be great if there was national support for overcoming them. There are also barriers to do with lack of consistent metrics and it would be very helpful if there were some national approach to that. Then there are the incentive barriers that I dealt with before in that this approach is not part of the core primary care contract and it would be helpful if it was. Have I answered anything you expected?

Q47 Mr Sharma: I am sorry for the long question, but I am sure that you answered.

Dr Roberts: Is there still an element I missed?

Mr Sharma: Thank you very much.

Dr Roberts: Thank you for asking.

Q48 Mr Sharma: How successfully can the delivery system model address multimorbidities by treating the patients, not the disease?

Dr Roberts: We as clinicians see them, in a sense, as a series of organs and conditions, but people see their symptoms and how this affects their life. If they have more than one condition-for instance, a heart condition-that may make it less easy to exercise. If you have diabetes, it would be very good if you exercised. So it is a question of pulling this together or people cannot make sense of that in their daily life.

I believe there is about to be some work published from Glasgow, from very disadvantaged populations, that shows that when you interview people with multiple longterm conditions-who, if you observed them, you would think would really be quite burdened by their condition-they do not talk about their conditions at all; they talk about living their life and the difficulties that they have on a daily basis. They do not talk about conditions. The approach that I am advocating, the delivery system, is one that takes that point of view and asks them first, "Tell us about life. What is important for you?" The bit that we are all working on is how all of us professionals manage to cope with our views about wanting our condition to be got in there and dealt with properly. The approach that people have in practice worked out is how you pull together all these little bits of medical information and enable them to be put in the pot, but it is a pot that is led by the life of the person who is the centre of all of this.

Q49 Chair: That approach presumably has particular application to people with what you describe as the new longterm conditions.

Dr Lowton: Definitely, absolutely. I was just thinking, as Sue was speaking, that the adults who had liver transplants in childhood certainly told us how well they were, how healthy they were. When we asked them about the co-morbidities they were experiencing, they included cystic fibrosis, kidney disease, heart disease, rickets and a wide range of really quite complex conditions, such as cystic fibrosisrelated diabetes. But we also need to be mindful that many people with longterm conditions cannot be selfmanaging and independent as though they were an island. There is a lot of family and unpaid carer support that underpins and enables these people with longterm conditions to live productive lives. My slight concern is that by focusing on the patient’s independence and selfmanagement we are at risk of ignoring the huge family support that goes into allowing people to grow older with their longterm conditions.

The other thing is how we would manage the uncertainty that a lot of these people with longterm conditions face. Certainly, through our research with people with cystic fibrosis, one of the common questions in their minds is, "When will I die? How much longer have I got?" One of our respondents with a childhood liver transplant who had cystic fibrosis too said to our researchers, "It’s all right. I am working on the cystic fibrosis life expectancy of 40 and I am not quite sure how the liver transplant plays in with that." Paediatric services are very good at providing familycentred care with psychological support, but once young adults leave paediatric services, they are cast adrift. We need to pay attention to that psychosocial support because what are we treating people for? People want to grow up, to work and to have relationships. They often need support from health services to achieve that.

Q50 Chair: You mean from health and social care services.

Dr Lowton: And social care services, indeed.

Q51 Chair: That brings us back to one of Grahame’s questions about the role of the health and wellbeing boards. Without wanting to go round another element of "redisorganisation", the development of integrated commissioning across institutional divides is surely central to the delivery of any meaningful Year of Care but also reinforces the difficulty of attaching resources to it, to which Professor Maynard says, "Show me the evidence." Unfortunately, you will never make any progress in life at all if you never do anything unless there is a precedent. I think it was Ralph Waldo Emerson who said that it is "the hobgoblin of little minds."

Dr Lowton: The only thing I want to raise in relation to things like social support or social work is that, in specialist centres, social workers are highly experienced in knowing the conditions of their patient population, but often, when people with new longterm conditions are trying to live in the community and being assessed for disability, the disability assessors have absolutely no idea about what their condition involves, that, yes, some days they may be very well but the next day they may be unable to get out of bed. This becomes a problem when people try to get the benefits they are entitled to and assessors do not really understand the conditions that they are living with.

Chair: Okay.

Q52 Mr Sharma: What changes to existing delivery systems are necessary for personalised budgets for longterm conditions to work effectively? That was to anyone.

Dr Smith: I was looking at Dr Roberts.

Mr Sharma: The question is for anybody who takes the initiative to answer.

Dr Roberts: At the risk of repeating myself, the Year of Care team are in fact providing all the training and support for teams to do personal health budgets. I say that because this delivery system model is perfect for personal health budgets. At the moment, as you probably know, it is only being advocated for people who are in receipt of continuing care, but even when it gets extended beyond that, so you have your group of people, they are supported to define what is important to them in life. Patients have a discussion with somebody who is trained in this care planning process, and then, instead of being referred to things, they have a pot of money which they are supported to spend on different things. It is a process that, particularly with this culture and attitude change for professionals, has helped the current personal health budget proposal enormously.

Dr Smith: I will add there as to personal budgets that, when we think about commissioning, there is a spectrum right through from national commissioning of specialised services at one end to personal budgets at the other end of the continuum. There is not a onesizefitsall for commissioning. For some services, the personal budget will be right, for some it is the national level and for others the CCG and so on. That is important to remember. We have had some useful studies of personal budgets, some encouraging results in terms of impact on quality of life and certainly some cautious positive messages about the costeffectiveness side of things. That is certainly my reading of the evaluations.

The other thing to bear in mind is the cost of running those approaches. Also, they are not going to be right for everyone-the point that Karen was making about people having different needs. Some people are going to be much more dependent or need to be tied in much more closely with statutory or other services. For someone with dementia-and they may not have family and carers locally who can advocate for them-it may not be appropriate. But for others-particularly, it has been found, in the area of physical disability and perhaps younger people as well with physical disability-it could be highly appropriate. It feels to me as though they are a manifestation of the shift from the doing things to people to giving people more of a role in defining and managing their own care. They are not the overall solution. They are part of it and will work as part of the spectrum.

Finally, for the CCGs-who have quite a restricted pot of management resource-understanding how to use personal budgets, alongside all the other kinds of challenges they face with A and E and general longterm conditions, commissioning and what they are going to do about primary care, there will just be practical questions about how much time they can focus on developing that approach.

Q53 Chair: Would Professor Maynard like to comment on that from a CCG perspective?

Professor Maynard: Yes, a random sample of one. It is very difficult. The agenda my own CCG has is very ambitious. We want to change this, that and the other. We have a staff of about 40 managing a budget of £340 million. We have the CSU, which is a bit parson’s eggish, and there is a real difficulty in sitting there being an old miser saying, "Where is the evidence? Slow down. Aren’t you going to evaluate all this?" They clearly want to drown me regularly! Seriously, it is a very difficult challenge, given the parsimony of the management budget that CCGs have. You have GPs coming along who are bright and intelligent and have been running their own businesses, but it is a very difficult thing when you are used to running a small, shall we say, GP business of 20,000 or 30,000. When you innovate there, the risk of loss is quite small. When you come to a CCG and you begin to innovate, the risk of loss is many millions. Therefore, it is a very steep learning curve in terms of them beginning to understand governance issues, which Judith was talking about, and generally thinking about how they can operate in this environment. It is very difficult for them, particularly when they are pulled apart because they are GPs and they are worried about their contract as well.

Q54 Grahame M. Morris: While we are on the issue of personalised budgets, there is a bit of an urban legend, isn’t there, that the individual, particularly in relation to longterm care, would prefer to be treated in the community, in primary care? Thinking about if I had a condition and I had to choose between a specialist consultant in whatever-let us say dermatology-or seeing a GP with a special interest or a specialist nurse, why would I choose the communitybased option? Is there any evidence to suggest that people would, given the choice, be more inclined to place their personal care budget in the community rather than with the hospital?

Dr Roberts: I do not know if I am picking you up right, but there are only certain things that come within the remit of personal health budgets. In fact standard clinical care-drugs and so on, and hospital care-is not within it. It is the care that you need to live effectively, the sort of social care and care in the community. Sometimes, people choose to replace a physiotherapist coming in with going out to an aromatherapy session. For them, the social gathering of that is better for them and they get benefit. It is that sort of option, as I understand it, that we are talking about.

Q55 Grahame M. Morris: I understand that, but as a mechanism, even if it is a blunt instrument, to drive change, to transfer resources from a hospital setting to a community setting, do you think personalised healthcare budgets are part of that or do you see that as potentially a problem?

Dr Roberts: This is second hand, but the team I work with have been working with people and there is evaluation, as Judith has said. It is a huge burden for local CCGs, but, in terms of the greater order of things, the profound thing it has done is help professionals understand that people make their own decisions that are right for them and that there is a whole range of things that may be beneficial that are outwith what we used to just do to people, "You will have this and you will have that." So it is a cultural change that I think has been quite important. I am personally not sure, even without the money, that rolling it out to everybody is absolutely the best use of resources. But it has been profoundly useful culturally.

Q56 Mr Sharma: Patientcentred care by definition puts the needs of the patient first. What evidence is there that patients with longterm conditions would welcome a shift to primary and communitybased treatment for their own conditions? How can patients be convinced that a move away from acute sector treatment will bring benefits for them? First, are they convinced and what benefits do they have?

Dr Lowton: For people newly ageing with longterm conditions there are certain groups who would not want to shift to primary or community care-for example, adults with cystic fibrosis and adults with congenital heart disease-primarily because the specialist expertise is still within tertiary centres. In the research that we have done, participants have told us that they will deliberately bypass a GP because they do not perceive a GP to have any understanding of why they are going to consult. Often, GPs would put everything, for example, down to cystic fibrosis, when in one case it was actually appendicitis and by the time this young woman had got admitted to hospital it was a burst appendix.

On the other hand, there are reports from our participants of very good GP care where a GP has taken a long time to learn that patient’s condition over their whole lifetime. I was thinking earlier, when we were talking about Tesco Express GPs with large lists, that there are GPs with very good expertise in these rare conditions, who perhaps could be identified for other patients in the community so that they could become their GP. Also, we have people growing older with Down’s syndrome, with cerebral palsy. I think they would very much welcome a shift to primary and community care that actually cares for them once they become adult. Again, there are people with expertise in, for example, dementia in Down’s syndrome, who can properly respond to their needs and also the needs of their carers.

Q57 Andrew George: In relation to that answer, taking the example of the cystic fibrosis patient-and I can understand that the rarer the condition the more likely it is that the patient would feel more reassured to see a consultant in the acute sector rather than chance it, as it were, with their local GP who may dabble, as they see it-what are the cost implications? What does it mean in terms of the GP’s own budget, in terms of what they pay for the care of their patient? What impediments are there, or what disincentives or incentives are there, for the GP to refer or not to refer that patient in those circumstances? Do you know?

Dr Lowton: I am not an expert in cost implications, but I would say, from a commonsense perspective, that if the patient can get themselves to the right person at the right time then, ultimately, their care will cost less because it is identified early. Some of the cystic fibrosis centres work very well through email, text and phone calls with specialist nurses. It becomes almost like community care in that patients can email and say, "I have this or the other symptom. Do I need to come to see you? What do I need to do about it?" So there are mechanisms for keeping people at bay, if you like.

Q58 Andrew George: This is not a criticism, but, if you cannot answer the question about what you understand to be the incentives and the cost drivers within the system in relation to referrals from GP to secondary care, can any of the other panellists explain what they understand to be the nature of the accounting system that applies cost to a patient referral that may or may not be managed at the surgery level in primary care but may be referred to secondary care in the case of the cystic fibrosis patient who is referred on a regular basis perhaps? I do not know.

Dr Smith: We are effectively talking about commissioning and it is going to be about how-taking that as an example-care for people with cystic fibrosis is commissioned. That is going to come back, I would guess, with that condition, to very much a discussion between NHS England and clinical commissioners.

The other thing I would say in relation to that as well is that, going back to what we were talking about earlier about the scaling up of primary care, whether that is through networks, federations or larger partnerships, that is where there is potential to have GPs or nurses with special interests and skills who can serve that larger population of 50,000, 70,000 or whatever. One could imagine within that a small team for-I do not know-say, a population of 100,000, a network of practices, who take a particular interest in people with Down’s syndrome and their primary and community services. But, ultimately, all of that does need to be, as I say, put in place. It is going to need to be commissioned and goes back to what we were talking about earlier, the role of the CCG and commissioning for longterm conditions, both what it ensures is in place within primary care as well as how it spends its money on specialist services and indeed works with NHS England to spend money on more specialised care, which is another complexity that the clinical commissioning groups are having to handle.

Q59 Andrew George: Just so I am absolutely clear-this is a final attempt and it may simply be me exposing my ignorance of how this process works-a patient with cystic fibrosis presents with an acute episode, goes to see a GP and the GP then refers to the acute sector at no cost to the GP. There is no financial incentive or disincentive implication in terms of that particular referral to the GP, the GP’s budget or, indeed, who is paying for that particular referral. I want to be clear where the cost consideration lies: who makes that budgetary decision, or is it entirely clinically?

Professor Maynard: If the GP refers, the resource consequences for the GP are minimal. There is an element in the QOF which says "We will look at your referral rates," but that is quite blunt and novel and has relatively limited effect. For most GPs, we do not have distributions of things like referral rates. We do not have distributions of blood tests or xrays. Managing GPs, even if it was our role as a CCG, is extraordinarily difficult because it is rather a datafree environment. But it is going to improve. We are going to get better resources there.

Chair: Good.

Q60 Andrew George: So it may become more budget driven and less clinically driven.

Professor Maynard: It may do. Some of the leaders of the profession are now talking about wanting everyone to be salaried. So they are beginning to think about what should happen. Aneurin Bevan blundered: he should have nationalised them in 1948.

Chair: Grahame.

Q61 Grahame M. Morris: I completely agree with that sentiment, I might say.

We touched on this a little earlier. If we think of the health economics arguments, as well as the output, support, managing longterm conditions in the community rather than in a hospitalbased setting, where appropriate, what do we need to do in terms of training for GPs and other staff? What do we need to address? Do we need to have more GPs with special interests? Do we need to have more specialist nurses? What are the training implications that we as a Committee need to highlight in our report for the Department if we go down this route?

Professor Maynard: Can I jump in and say-because I am an economist, aren’t I?-that these people have to manage budgets? These people have to manage resources, and their management of resources affects me as a patient and everyone in the community. Yet we teach them little in the way of statistics, economics and the issues of evidence. Therefore, I would argue that we should beef up the undergraduate and postgraduate training of GPs and the consultants because they become clinical directors with a clinical budget. The poor souls are thrown into this with very little expertise. They are bright and they pick it up, but it should be part of the revalidation process that they are made aware of these sorts of things.

Q62 Chair: So they should attend an economics course as part of revalidation.

Professor Maynard: That is absolutely essential, Stephen. The doctors will have other views.

Q63 Grahame M. Morris: You have made a valid point there, that in seven years of medical school training there is nothing in terms of commissioning in that, as far as I am aware. I was thinking about specialised training in order to treat the longterm conditions in dermatology, diabetes and so on. It is valid that they should have some understanding of the health economics of it and commissioning as well, but, in terms of being able to deliver the service from a clinician’s point of view, how do we need to change that?

Dr Smith: My take on this would be that, when you look at it in the international context, general practice training in this country has moved a long way. Medical students spend time in primary care from very early on now as undergraduates, and a significant body of their training does happen in a community in a general practice setting-something that is certainly not the case in many other countries. So we have achieved a lot there.

I know that colleagues in medical and nursing education are thinking carefully about this, but almost the next frontier in this area for us, though, is the specialists and where they are trained. My sense would be that that is still often too focused in the hospital setting. The implication of a lot of what we have been saying this morning has been about specialists working differently with networks of general practice or in some of the ways Karen described in terms of linking quite differently with patients and their families over a much longer period. So it is perhaps about some of the specialist training and where that happens. It would be nice to think that, if we are going to scale up general practice, have more general practice networks, larger partnerships and so on, that that could become a training ground for nurses, doctors and indeed other health professionals who are looking for a more specialist career as well as those who are going to pursue a generalist route.

Q64 Grahame M. Morris: Might I ask Dr Roberts, though I know time is short, how does a GP become someone with a special interest in whatever it is? How much training is involved before they are a GP with a special interest in whatever, dermatology or-

Dr Roberts: Some of those criteria were laid out about 10 years ago, with the Colleges for different specialties, laying out what would be needed and periods of mentorship with specialists and things like that. Before we assume that it is making GPs into specialists, though, which is sometimes of use, particularly around technical things like taking off lumps and bumps and ENT-there are some technical things that seem to have worked very well-they have not worked that well for all longterm conditions, and I would pick up some of the things that Judith mentioned about not saying, "How many specialists do we need?", but, "How are we going to get the specialist knowledge into primary care?" It may be that it is more important to concentrate on specialists working differently because they have been trained-they really know all the specialty-than getting the GPs to learn a bit of the specialty. So it is horses for courses, I think.

Q65 Grahame M. Morris: How long would it take for a GP to be trained as a GP with a special interest rather than a specialist? Is there a set period where a GP would-

Dr Roberts: I would have to go back to the things that were in those documents from about 10 years ago, but essentially it is a mentorship and an apprenticeship process locally. It might involve going on a masters course in the biology of that particular condition, but it is basically sitting in and being mentored locally by a specialist. I would have to go back.

Q66 Grahame M. Morris: How long would that take?

Dr Roberts: I think it would be an ongoing process over two or three years, but I am now making it up as I go along, so I need to look back at the documents.

Q67 Grahame M. Morris: Maybe you could send us a note, if you don’t mind.

Dr Roberts: I will have a look back.

Grahame M. Morris: Thank you.

Q68 Chair: I think we have covered the ground that we wanted to cover. I would like to close by bringing us back to the question of evidence and the focus of how we move this argument forward, because even Professor Maynard, who has played the role of selfappointed sceptic in the panel this morning, has not seriously argued with the proposition that what is required is a more patientfocused-putting words into your mouth-service, a service that is better able to listen to the priorities of patients, respond to those and to deliver a joinedup service. It is very hard to argue against that.

If it is to be anything other than common sense, supported by evidence of how to turn those words into reality, the question, it seems to me-and, given that Professor Maynard started off by saying this has been going on since 1974, it is a fairly urgent question-is how to equip ourselves with robust evidence about which models, and there is no perfect solution here, deliver better outcomes at better value than the alternatives. Are there any suggestions you would offer, picking up Grahame’s question-"What should we put in our report?", in simple terms-about how we move this argument forward so that we do not sit here in five years’ time and agree that there is no evidence?

Professor Maynard: We have already talked about trying to make clinicians more economically literate. There is also a big challenge of making managers, whether they are clinical or nonclinical, more literate in these skills. I am finding, in a CCG, that the challenge in saying, "Where is the evidence?", trying to get people to evaluate, works because they are listening, maybe because they are GPs and they can tolerate my intolerance. I found it much more difficult in a hospital setting, which was much more run by finance.

The management of knowledge is quite important. One of the things I mentioned in the evidence was that there are people who supply that knowledge-the academics-and they should be encouraged and coerced to be much better communicators with people in the service and people designing healthcare systems. The reward system for academics is, "Publish in The Dandy and The Beano and you will get a professorship." It is not quite like that, but you know what I mean. The reward system does not encourage you to disseminate and talk to MPs and the community. So there is a supply side issue there in the creation of new knowledge.

Then there is the demand side. Even if the researchers are good communicators, "I have found that this particular intervention for diabetes looks quite costeffective in the community," even if they identify that, they still have to convince the managerial people in the NHS, and they simply do not have the skills to understand evidence. That gives you a challenge about training in leadership, career training for senior managers. They are all bright and able people, but if you say, "Have you looked at the Cochrane collaboration evidence on that?", they will just pass on and will not do it. So I think there is a nice challenge for perhaps the Select Committee to focus on training of these bright people who come in at the bottom, who are something of an elite coming into NHS leadership, and making sure that they are much better equipped and that they interface much more with clinicians. That would be it for my prejudices for now.

Dr Roberts: Can I build on that, particularly academics working with deliverers? What is needed is recognition that they each contribute different things, so that deliverers can design interventions but then they need to be appropriately evaluated. There are currently in the literature coming out three or four examples of interventions in selfmanagement support that are said not to work, but when all the delivery people look at them they say, "We would not have expected them to work. They are not competent interventions." So that excellent research, perfect stuff-

Professor Maynard: Published in the BMJ.

Dr Roberts: Exactly. It was completely wasted because they were evaluating something that you would not have expected to work in the first place. My view is that you have to take something that is credible, that is built on the best chance that it will work, designed by people who design things and do them, but have the evaluators in right at the beginning working on this so that what comes out is a good use of that evaluation resource.

Q69 Chair: Without wishing to be excessively bloodthirsty, as to the people who launch pilots that were clearly never going to work, presumably there should be some career penalty for that.

Dr Roberts: The ones I am thinking of were designed by the academics, though, because they could see that that was something that would be useful to research.

Q70 Chair: Yes. Even academics can be subject to a career penalty.

Dr Smith: I support all that has been said, but I would add to it. There is a responsibility on the system, for which I guess I am meaning the Department of Health and NHS England in particular, in terms of how at that level leadership is given to what we are talking about, so that there is proper support and leadership given to the idea of, say, a fiveyear programme of experimentation around particular interventions. We are talking about longterm conditions, and this is indeed longterm change, longterm work. It is hard, it takes time and it is not a quick fix. Quite a significant cultural change needs to happen at those senior levels in terms of saying, "It is okay for this to be a fiveyear programme," and yes, absolutely, that the researchers, the managers and the clinicians should be required to work together to design the intervention, design the measures and see it through. Indeed, if it is found not to work, it might be better to stop the intervention earlier as well, so it is not just a question of waiting for five years. It is about the readiness to say-and I guess it almost becomes like NICE guidance-"That is not appropriate to carry out," or, "Yes, that is highly effective. We will get on and roll it out." There is a central leadership question that needs wrapping round this to enable particularly managers in boards to continue on with what would be, in a sense, a countercultural approach.

Q71 Chair: That is presumably particularly relevant to the group of people that you raise.

Dr Lowton: Yes. Being an academic, I am feeling thoroughly bashed.

Professor Maynard: We survive.

Dr Lowton: There are developments. For example, I work at King’s College London and we have King’s Health Partners, which is an academic health sciences centre. The university works very closely with partner foundation trusts, particularly, for example, in the area of stroke care where we have academics working with clinicians and managers to try to make sure that this research is going to end up benefiting patients.

Chair: Thank you very much. On that note, thank you for your attention. Thank you very much indeed.