18
Examination of witnesses
Dr Neil Modha, Ben Richardson and Mr Gary McAllister.
Q92 The Chair: Welcome to the second session this afternoon of the special inquiry into the integration of primary and community care. We are delighted to have with us Dr Neil Modha, chair of the data workstream for the Fuller stocktake report. He is also a GP partner. We have Ben Richardson, the managing partner of Carnell Farrar, and Gary McAllister, the chief technology officer for OneLondon. Thank you very much indeed for coming to see us this afternoon.
As you know, we will take it in turns to ask you questions, possibly directed at all of you and possibly specifically. We have a lot of questions to get through, so I would appreciate succinct questions and answers. I will start with the first question. This inquiry is focused on integration with primary and community care. Dr Modha, how can better access to patient data and medical records facilitate that integration?
Dr Neil Modha: Thank you, Chair. As you said, I have several roles, the most important of which is being a GP. In that role information is key. Knowing someone’s background, knowing what blood tests they have had and knowing what investigations they have had is very important. A lot of the time when time or efforts are wasted is when my patients have been to a different sector, whether that be the hospital, mental health trust or community trust, and that information is not there. The most important thing to my patients is their healthcare. The ability to communicate across sectors is as important as the ability to communicate within a sector.
Ben Richardson: I have three quick examples. First, we heard this morning on the BBC about vaccines and that, in many cases, we do not even know whether someone has had a vaccine or not. It makes it very difficult to be proactive in preventing disease if that is the case. Secondly, in chronic disease management such as diabetes, it is critical to understand where people are managing their diabetes so that we can help best manage that. Lastly, many people in hospital now need to be discharged. If we do not have information about each of those patients and exactly their condition, it is hard to make suitable arrangements for them to be discharged.
Gary McAllister: It is important to say that sharing information is transformational. We have issues where three key systems exist, across primary, community and national. Those systems do not lead to easy access to and sharing of information. Dr Modha mentioned prescriptions, which is one of the issues we have. For continuity of care, it is vital to share allergy and other safety care planning and documentation about individuals that gets captured in different care settings. Not sharing that information in the past has introduced scenarios of harm, so, from a safety point of view, sharing information is incredibly important.
The Chair: Thank you very much for starting us off so well. We now move to Baroness Shephard’s question.
Q93 Baroness Shephard of Northwold: Thank you, Chair. It is interesting that all our witnesses have said how enthusiastic they are about data sharing. The committee would like to know what the technological, physical, legal, institutional—and, I would add, ethical—barriers are to data sharing? What is holding this up? Is it all those things, or is it the sheer difficulty, or is it perhaps all of them?
Ben Richardson: The focus on primary care and community care is very interesting. The most important disconnect between those two is their very different systems. Primary care is a very transparent system with a clear specification of the offer, a clear contract, excellent coverage of systems and good data. Community care is almost the opposite. It is a very opaque system. There is no universal offer for community care. There is no universal contract. Coverage is not 100% in electronic records, so in that context it is very difficult to integrate primary care and community care.
There is an urgent need to focus specifically on community care. That is then compounded by variation in the platforms and electronic patient records. Information governance is poorly understood. Properly understood information governance could be a springboard for enabling the sharing of data, but it remains poorly understood, so the safe answer is always to say no.
Lastly, a major theme is that capabilities in this area are not the strongest in the NHS and social care, as has been well rehearsed.
The Chair: Tactfully put.
Gary McAllister: We could probably have an hour on this one on its own. We have had quite a challenge when it comes to information governance. The protection of information itself has always been a barrier. Embedded in the 2012 Health and Social Care Act was that we should share information for direct care. That message has not translated itself to the point of care very well.
Coming back to Ben’s points about contractual obligation, I am the chief technology officer for London. The technical infrastructure that you need to share information across boundaries is variable at best in the standards and the adoption of standards. The law on the implementation of standards does not enforce the ability for information to be shared easily. We tend to get bogged down in managing suppliers and contracts with suppliers when it comes to actually sharing the information itself.
Moving forward—I am sure we will come to this later—we need to work out how to give patients better access to their own records so that they can control the sharing themselves. At the moment, one of the barriers is that the information is too siloed. Information sits in far too many systems across too many parts of the health sector. We probably need to think about that.
Those are just a few of the many barriers that we have to face when it comes to sharing information.
Baroness Shephard of Northwold: I think we can all see the difficulties. Who is reluctant to share? Is it care homes? Is it patients themselves, who have been brought up to think hard about patient confidentiality? Where is the real reluctance coming from, or is it something that has to be plugged away at, bit by bit?
Dr Neil Modha: From my perspective, whenever I talk to my patients about sharing data, the vast majority of them say, “If it helps my care, I’m very happy for my information to be shared”. For example, we are working with the local community pharmacy. In that work, we are trying to make it as easy and safe as possible for people to continue their medicines while having the appropriate checks. From the community pharmacy angle, if they say, “Look, I don’t know whether you have had checks. Are you happy with me to liaise with your GP or organise prescriptions so that you don’t have to go several times a month?”, the answer has always been yes. There has never been the answer no to that question.
Within services we expect to be able to share data, but there is the role of the data controller. One thing that I was trying to see if I could help with, when I was chair of the Fuller stocktake data and information governance workstream, was that if there is a clinical mistake, we are all protected by Crown indemnity. I would hope that if, for example, there was an assurance that if we shared data that we believed to be in the best interests of people but that was found to be not the right thing to do, there would be some underwriting of that centrally. Things like that would help to give people the confidence that as long as they were doing it for the right, principled reasons it would protect people’s ability to share data.
Ben Richardson: If you focus on community providers, typically you do it not with an individual GP but with a larger organisation, and you have a data protection officer, who is probably five levels down in the organisation. For them, the safe answer is always to say no. No one ever got into any trouble for saying no to an information governance request, and that is one of the biggest issues. It was alluded to before; it is not clear what the duty to share actually means.
Gary McAllister: I also want to pick up the breadth of primary care. Primary care is not just about general practice; it is about optometry, dentistry and some private services that stretch beyond that. Some of those disciplines are not digital. They are very variable in their application and in the delivery of digital record-keeping. Until we create equilibrium in digitalisation across primary care services beyond general practice, we will struggle to share information. There are some very good clinical use cases, certainly when it comes to optometry. Individuals are losing their sight because their records get lost and their referrals do not get expedited. We talked about safety at the start. It is really important that these individuals are managed well.
Q94 Baroness Finlay of Llandaff: Mr McAllister, you mentioned a previous health and care Act. In the latest one, we got agreement for there to be a single unique identifier for children, which would probably be the NHS number, yet we are not seeing that happening. There is a real worry that safeguarding is the way anything gets triggered—in other words, after the problem has arisen. Do you see a role for having a unique child identifier? I am linking back to Mr Richardson’s comments about what we heard on the “Today” programme about immunisations.
Gary McAllister: The way the process should work in practice is that when a child is born, they are registered—through the hospital, the acute provider or whoever attends the birth of the baby—and immediately get an NHS number. That should obviously then stay with an individual throughout their life for immunisation and postnatal care and beyond into adulthood.
If that is not working, there are issues with the process that we need to look at. There are challenges with our national infrastructure, which hopefully we will pick up later. We need to retain the NHS number. It is the gold dust of the NHS, in my opinion. I say that, because when it comes to record sharing, and information sharing generally, having a single identifier that we can trust for an individual allows the records to be shared in a trusted way. If we do not have that, it becomes some rather complex, fuzzy matching of an individual’s demographic, which is unsafe. For an individual, from childhood and beyond, it definitely gives us faith for the future.
Baroness Finlay of Llandaff: With something like that it has to go into schools as well, as we heard today, and it has to link across social care. The NHS number has to be transferred to other sectors. How do you view that happening?
Gary McAllister: This is not uniform across London, but certainly in parts of London. My work in London in my previous life before NHS England was at Guy’s and St Thomas’ leading the shared care record programme for south-east London. We introduced the NHS number into social care in Southwark and Lambeth. That was transformational. We then had the trusted identifier. We put in the national infrastructure that allowed the trusted identifier to be retained in social care.
Coming back to my previous points about equilibrium in standards, that work needs to be in place in social care, in community and in all other sectors so that the single identifier allows us to trust the information and the record across sectors. Equally, we need a national infrastructure that allows certain parts of the core record to be available centrally.
Ben Richardson: On the vaccine point, because vaccines for the most part are delivered by the NHS, in most cases the vaccination is actually recorded in the record. The programme this morning talked about that. How do you make sure that you secure consent from parents and facilitate access? Why are we still doing that on paper instead of using digital means?
Q95 Baroness Tyler of Enfield: This is very much a continuation of the question that Baroness Shephard asked. To what extent are barriers to data sharing, which we have been talking about since we started this inquiry, and certainly today, the key limiting factor for health integration?
I have a specific question for Dr Modha about the linking together of data between community pharmacists and GPs. I recently had an y excellent experience going to a pharmacy and inquiring about a medication. They took me to one side, to a consultation room, and took my blood pressure to check that it was okay to use the medication. I thought that was fantastic. They then said, “We’ve got this for your doctor”. I said, “Will it automatically go through?” They said, “Oh no, here’s a piece of paper. You’ll have to hand it to your doctor’s receptionist”. It fell at that very last hurdle. What can be done to get all of that linked up?
Dr Neil Modha: I will answer specifically on that point and leave my learned colleagues to answer the other one. Our practice, the Thistlemoor Medical Centre in Peterborough, has 30,000 patients, 80% of whom do not speak English well enough to have a consultation. There are massive communication challenges among the most deprived deciles in the country. We work with Graham Young Pharmacy, which is next door to us. We use SystmOne as our computer system; the infrastructure is used by a company called SystmOne. The pharmacy is one of the first pilots in the eastern region to use SystmOne. Just having the same computer system is half the battle.
The other part of the battle is making sure that the information governance, the information sharing and all those things are put in place. If I sit there with my GP hat on at my GP terminal, there are certain things I can do. If I sit in the pharmacy with a pharmacy log-in, there are other things that can be done. Making sure that we are not only on the same system but that we have as much benefit from both those lenses, while thinking about the different jobs that both of us need to do and being able to communicate seamlessly across them, is vital.
For example, the pharmacy consultation scheme is trying to take activity away from general practice to community pharmacies. If someone rings up with a cough, a cold or conjunctivitis, instead of speaking to someone in the GP practice they speak to the pharmacist. Ninety per cent of those stay in the pharmacy rather than coming back to general practice. With excellent communication, that allows us to say to the patient, “Are you happy for us to liaise with your GP? I can book you directly into their service”.
You can imagine the difference for the person. They have contacted a GP surgery and have been treated by a pharmacy. For 90% of them, that is enough. The 10% for whom it is not enough get directly booked back into the general practice surgery. That level of communication is game-changing for an individual.
Q96 Lord Kakkar: In your region, ICS or whatever, how many different systems are there? You have a common one with your pharmacy next door, but how many systems does your ICS tolerate?
Dr Neil Modha: In our ICS we have about two or three different systems. In our locality—in Peterborough—we are all on the same system. The community service provider in Peterborough, who provides both community care and mental health care, also uses that system. That paints a vision that it is all perfect. It is still not perfect. For example, there are still problems requesting a swab or issuing prescriptions and things like that. It is also the same system as the out-of-hours provider. If a patient of mine is seen in that community sphere by community services or mental health services or they go to the out-of-hours provider, I can see their record.
Lord Kakkar: How has that happened? How have you, magnificently, been able to get the same system across all these elements of provision in community and primary care?
Dr Neil Modha: The honest answer is that incentives were put in. They were not financial incentives. Network infrastructure incentives, for example, were put in quite early, before the ICS structures and before even the CCG structures—for which I was the accountable officer, although I cannot take the credit. Before that, in the PCT days, I think there was a move to try to encourage people to be on the same system, because that would allow system sharing. In our area, the whole of Peterborough has pretty much been on the same system for about eight or nine years, and it is now 100%.
Gary McAllister: Dr Modha has answered the question very well. I want to give you a different view as to how it works in other parts of the country. You are probably talking about half a dozen systems at the core of the infrastructure in London, across all the care settings or sectors.
As an example, the community pharmacy case study that Dr Modha raised would be a very different state of affairs in part of our community. You would end up with a GP using a system that would talk to the referral system, which would talk to the community system. That might or might not talk to an acute provider system, which would then talk to the shared care record. That leads to a lot of frustrated colleagues. They have to use all these systems at the same time in order to get a clear picture of what is going on.
I think the work that has been done in Peterborough is ideal, but the outcome as to what systems and services are provided and delivered generally depends on the organisational structures of an ICS, and the collaborative nature of people and individuals in our health service.
Ben Richardson: Integrated care requires people to have the mobile number or telephone number of whoever the patient is, and for that to be shared. In order to achieve that, ideally everyone would be using the same system. In many cases that is difficult. The transition takes years, and it costs millions, or perhaps hundreds of millions, across the country to achieve that. If we do not resource that, we need to make sure that we do not just have information sharing; the issue for community pharmacies and GPs is that they will need to both read and write to the same record, otherwise it is just generating paper tickets. That is the challenge that needs to be solved.
Q97 Lord Watts: I apologise for asking similar questions, but this is the one area that we are all struggling to understand. How is patient data currently stored and shared within primary and community care? Should it, and could it, be done differently quickly? If you wanted to have a system built in your area, what would the difficulties be in finding one system that everyone agreed to and getting the funding that would be required to actually do that? Can you give us some indication?
Gary McAllister: London is in the process of following the convergence strategy that NHS England has been promoting for quite some time. All the ICSs, in the round, are generally converging in some way, shape or form. The risk we have with IT systems is that they are incredibly complicated. Moving data from one system to another when you procure a new system to cover many sectors is highly complicated and takes several years.
I will talk about south-east London, because that is where I come from. As an example, a large teaching hospital in south-east London has procured a new system for half of the integrated care system at a cost of £250 million. That will take up to five years to configure and implement. It will require significant infrastructure, from the basics of having the right PCs in place right the way through to ensuring that medical devices talk to it effectively. There is safety in that as well. You end up with really complicated programmes of work to get to a converged platform for an ICS, and it will take a long time.
Coming back to the question of whether it can be done quickly, there is no easy way to solve the problem, unfortunately. My personal opinion is that we probably need to revisit what a good national architecture or a national platform looks like and how we get a core record for an individual available again. That will enable levels of autonomy within ICSs, which then means that if we are feeding a single national record, at least the core of my being, if you like, can port with me.
As we have heard from colleagues here, we can talk about community prescribing as an example, but I do not really have a choice about where I pick up my prescription. It is going to be a pharmacy in the locality or one that is tied or bound to my doctor. Is that really 21st-century health and care delivery? It is not, is it? We should be in a position where my prescriptions belong to me. I should have a choice where I go to access my prescription, and get it dispensed at a pharmacy of my choice. That choice challenge is something that we probably need to resolve more generally. I am sorry that there is no easy way out when it comes to money and the resolution.
Lord Watts: It is possible to get a prescription delivered to your home these days. It is starting to move in that direction. The NHS app, which a lot of people have, including me, already contains a lot of information that you can share. Is there any mechanism for feeding more information into someone’s personal NHS record, so that when they go to the doctor they can give the doctor some information? I did it on Friday. I had had a blood test, and I was able to give the information on the blood test results to the doctor. That was helpful to him. Is there any way, even if it is not perfect, we could build on something like that?
Dr Neil Modha: I think the tension is in having the golden ticket—the panacea of a record—that has everything I would want as a GP and that Gary would want as someone at a more strategic level. That is really complicated and hard.
However, the technology exists in people’s internal systems, like my SystmOne and the hospital’s eTrack system. Sometimes the tension is in trying to solve it in the way that we want one thing that communicates as well as it should do, and that we would all expect on the NHS, versus making the best of what we already have, and how we make that as accessible as possible.
I had the pleasure of being here a bit earlier, listening to the last session. I think empowering people to have their own information and data is a really good way to do that. It creates other situations and things that we need to factor in, but the best situation is people having their own information in a way that is easily understandable but protects individuals from finding out from an app that they have cancer and things like that. With appropriate safeguards and levels of detail, empowering individuals to have data is a good way forward.
Ben Richardson: If the ideal, articulated by a clinician, is to have everyone on the same system, although that has some practicality issues, as outlined by Gary, probably the only thing that is a bit of a shortcut to that is integrating the data; the systems stay where they are, but you seek to integrate the data in a way that is shared, or shareable securely, with those with a budget and a purpose for accessing it, and secure so that it is not accessed inappropriately.
That can be done much faster. It was 13 years ago that I worked with colleagues in north-west London to create the whole system’s integrated care dataset. I think 462 separate data-sharing agreements were signed in a six-month period and created one of the oldest integrated datasets in the country. Naively, in retrospect, I thought that was the answer and that it could be copied and rolled out to many other places. There are other pockets that have pursued it more recently, including Manchester, Wessex and Cumbria in the north-east, but they remain the exception as opposed to the rule. A different route from saying we need to have all the same systems is to make sure that we invest the resource in creating shared datasets so that they can be accessed.
Lord Watts: I think you are saying that that should be a priority, that it already exists and that there is no reason why it should not be rolled out, because it would not prejudice any future developments but would make them easier. I do not want to put words into your mouth.
Ben Richardson: I would agree with that.
Gary McAllister: It is important to note that the infrastructure required to aggregate those datasets would need a level of investment that probably is not there at the moment. A lot of work has been done on creating datasets that can support aggregation and on a standard to support the aggregation of datasets. There is no carte blanche approach to doing it though, unfortunately.
The Chair: Your word of warning is that that would cost money. Is that right?
Gary McAllister: You would need to invest in it, yes.
The Chair: It would cost money.
Ben Richardson: I would not suggest that it is cost-free. I think it would be less costly than ripping out all the systems and making sure everyone moved to the same one.
The Chair: Thank you very much.
Q98 Baroness Wyld: I am afraid I have the age-old question, and you have touched on this. How should the health sector strike the correct balance between data sharing and data security? You have each already touched on this, but I wondered if you had a bit more to say.
Ben Richardson: I would be happy to jump in and start. On the point about an obligation to share data, I think “to share data” should be clear. It is a tricky topic and slightly anorak. One of the troubles that we get ourselves into is that it is formed, by statute on the one hand and GDPR and common law on the other, by what people expect. If people say, “We don’t share your data”, that is the expectation. If you do not abide by that, there will be difficulty.
If, universally, every NHS institution said, “We share your data to improve outcomes for your health and for research”, it would create an expectation throughout the system that that is going to happen. I think we would find that patients are in favour of that communication and engagement, as Dr Modha has already talked about, but we need to be consistent about it.
Baroness Wyld: Dr Modha, would you comment on this? I do not know how much patients worry about it—genuinely. You said that they generally say, “Yes, do please share my data if it’s in my interests”.
Dr Neil Modha: Absolutely, especially when any of us has our health at risk. If something serious is happening, we would expect that the services looking after us would share information to care for us better. I have experienced, through the lens of my practice of 30,000, my federation of 300,000 or the CCG level of a million people, that people hold their data and information very seriously, and they worry about how information is shared generally.
The best way into it is to ask people and to give them the ability to opt out. If people fundamentally disagree with sharing information or data, we should respect their wishes. I still think that should be an informed consent, a bit like if someone declines a vaccine. We need to have a conversation to explain why we want to give them a vaccine or why we want to share their data, and what the advantages are. If that person then declines, they should have the ability to do that. However, as Ben has just said, I think the default position should be that we share information and data. By doing that, we can care for people better.
Gary McAllister: Colleagues have covered this very well from a pragmatic point of view. I suppose the challenge is the awareness of making sure that patients know specifically what is being shared and what infrastructure is required for them to opt out, and that they understand that they can opt out. We have had a lot of challenges in public trust over the last five to 10 years, which I do not think have helped us. Reinvigorating public communication around trusted sharing of their information is important, as well as making sure that they know what is being shared.
We had a citizens engagement forum. The challenge that we faced, certainly from a London perspective, was that the expectations were there. People just expected their information to be shared for direct care, and they were very happy with that. The challenge we find now comes when you get into secondary use and research. It is having trust in the use of their data in industry, outside direct care situations.
It is not necessarily a security issue. From a technological point of view, we have well-established security protocols and mechanisms in place for keeping information secure. As long as people are audited and managed appropriately, it is fine. It comes back to communication. In some cases we do not have a real understanding of the law on information governance in certain scenarios. Research is a very typical example: at what point do you allow the re-identification of a record if it is in the patient’s best interest when it comes to research?
We have found, through our consultation work, that patients are happy as long as they are being treated. We can share their information as much as possible. If you are using it with industry, which then results in new treatments, there seems to be a bit of a grey area in how you securely re-identify an individual to benefit them moving forwards.
Q99 Lord Altrincham: We understand the advantages of common systems. Clearly that is very central to our inquiry. Can we go back to what you were commenting on just now? How can we improve interoperability between systems, maybe at modest cost, knowing full well that there are multiple different systems still in place? I fully understand that the idea of that was not so, but we are where we are. How can we make a difference? It is very central to us in the inquiry.
Gary McAllister: Setting some standards would be helpful to start with. We are in a bit of a transitional period in the management of our standard of standards, if you like, in the NHS. There are international standards such as the fast healthcare interoperability resources, which are internationally renowned now. They are Internet First standards, which have been part of NHS policy for quite some time. If we get the supplier community to adopt those consistently, we could do a lot very quickly.
Equally, we talk about read and write access to records. At the moment, we have progressed to being able to read records pretty well. We have been very pragmatic in London in just being able to consume what we can to share information at the point of care. It is substandard in the use of standards. That is the best way to describe it. If we could get uniformity of standards that we can publish nationally and create an incentivisation programme to get suppliers to adopt those standards, I think we could do a lot of exciting things very quickly.
Ben Richardson: I completely agree with that. Going further, if we could be clear about practical guidelines for information governance and what sharing of data is required, we could avoid the grey area. The NHS could say, “Here is a series of cases and the answers”. For instance, we have been blocked for weeks on a specific project because people thought that sharing aggregate information, which is not personal and therefore there is no information governance, needed IG approval. It is aggregate information about a population. That is just a misunderstanding and a prevention of information that would be useful about how to manage, in that case, the urgent care system. That is an example of something you can clarify by saying, “Here is what data sharing means and what we expect of all organisations”.
That is something that can be communicated within NHS organisations and, crucially, as we keep coming back to, with patients. If you write down the right things, patients would be happy to go along with that—or, rather, the guidelines that are written should be ones that patients are happy with, as opposed to ones that they are not happy with.
Dr Neil Modha: It is about making practical examples of that. We set the standards and then we give examples of what good looks like. An example of what good looks like could be that a person is seen by a hospital cardiologist, and the hospital cardiologist has the ability to issue their medicine at a pharmacy of their choice. Rather than a letter going back to a GP who picks up the letter two weeks later and then issues the medicine, how do we set the standards and give tangible examples of what people would expect the health service to be able to do at the moment? That would be a good first step.
Gary McAllister: We come back to the barriers. We still do not have a national agreement for the sharing of information, which is a real barrier to access across all sectors. The first thing every ICS needs to do, as soon as they create an information-sharing programme, is to get approval from every data controller in the ICS—there will be hundreds—to enable them to share information and put it in an information-sharing repository so that it is legally sound and okay to share data. That has to stop. We have to have a national information-sharing agreement that, for direct care especially, enables information to be shared.
On interoperability, even if we implemented the technical standards, which we have just mentioned, at the moment there is no sound national basis for data controllers to share their information willingly. It comes back to Ben’s point about setting the right guide rails and framework for sharing in the first place.
The Chair: Mr Richardson, you said that you were blocked for weeks in doing this. Can we ascertain where the blockage came?
Ben Richardson: In that case it was a misunderstanding about whether information governance applied, and whether it was going to be okay or not. It was resolved through escalation to a higher level, which then said, “That’s not right”, and backed down again, but it resulted in a significant delay.
Even in places that have data-sharing agreements in place, they can vary enormously. In one place, the answer was that any sharing of the data would require going back out to every signatory to the data-sharing agreement. Other places have set it up so that there is a central committee that decides. That is much more efficient; you do not want to ask every single GP practice, “Are you okay if we use it for this instance?” It has delegated that authority to someone to say, “Well, that is sensible. There’s a legitimate purpose and a legal basis, so by all means go ahead”. That is a much more sensible way of doing things.
Q100 Lord Kakkar: So that we understand it correctly, individual general practitioners are data controllers for their lists. Would it be feasible to have a single data controller for the entire NHS body of data? The benefits are clear, but what are the potential risks and how might they be overcome?
Dr Neil Modha: That is absolutely right. We are the data controller for the data that we hold in our clinical system. As a GP we have many different roles, including that one. A lot of us take that role very seriously. For example, when it comes to protecting our patients’ data and making sure that it does not get into the wrong hands, there is a balance to be struck between our duty as a data controller and the ability to share information that allows care to be better. You can imagine that, if you lined 100 GPs up and asked them their opinion about this, they would have different opinions. In many people’s systems—in ICSs and CCGs before them—there were often people who acted as a data protection officer and you could talk to them.
To go back to what I was trying to achieve when I was chairing part of the Fuller stocktake, it was almost to set the precedent that we are the data controller and we have responsibilities, but that, as you said, across NHS England there would be the ability to say, “We are the data controllers of information across NHS England. GPs are part of our system, or community providers or mental health providers. As long as they follow the following principles and are not trying to sell people’s data but are doing it with a trusted provider within the NHS family, those are the kind of conditions in which, if something were found to go wrong, we would underwrite that position”, in the same way that we are underwritten in our clinical liability with Crown indemnity.
Lord Kakkar: Is that happening?
Dr Neil Modha: When I had conversations as part of the stocktake, there were conversations with NHS England and in ICSs. I think there is a tension between the NHS overall underwriting this and systems. For example, NHS England could say to systems, “You should really underwrite this, because it’s happening within big geographies, the 42 ICSs of the country”. From my point of view, it is still unresolved. That might be something that the committee could follow up.
Q101 Baroness Armstrong of Hill Top: You have already talked about patient access to their own record. We can all see the benefit of patients being more certain about what is happening in their care pathway, or whatever, and helping them to navigate that pathway. Can you say a bit more about how you think that can be progressed, and how we get it across the board? We hear the odd, good example, but it is really making sure that people have that access, because it is not uniform.
Dr Neil Modha: There are things in general practice, for example, where patients have access to a lot of their information. As I said, I fundamentally think it is a good idea for people to have that information and data. There are things like the NHS app that was mentioned earlier. Our clinical systems have access portals as well. For example, SystmOne has SystmOne online, which allows prescription ordering, and there are apps to access doctors. People can go online and do that. On progressing it, there is already quite a lot of movement in that area. When I talk about it with a general practice lens, I would say that there are good systems and processes in line already to further that.
Again, I had the fortune of hearing some of the earlier conversation. It is difficult. I remember about a month ago someone asked me, “Why is my blood venomous?” I said, “No, that is a venous sample. It is not a venomous foot blood sample”. Over time, people will get used to these words. Although, obviously, as a doctor it was interesting that someone raised that, we can understand it. At least they are taking their information seriously. The more people feel responsible, the better.
Getting the balance in how much information is there is a really interesting journey for general practice. We have always felt that the clinical record is our clinical record—my GP clinical record. How do you strike the balance in the shift to it being the person’s clinical record and the person’s information and data? Often, when we use the clinical record, we are using it for our governance as well, to document what people did or did not say, or what we might have picked up.
There are certain circumstances, such as safeguarding problems and if we are worried about how a child is being cared for or we are picking up some subtle signs of things that we are concerned about, that we would traditionally document in a record. How do you strike a balance and not put people off from documenting those things? Often, it is clinical suspicion that has built up over time that empowers clinicians to understand that there might be something else going on. How do you do that while balancing the need to be open with people about the information that we are recording? It is a fine balancing act.
Gary McAllister: At the moment, the patient record predominantly sits as the primary record in primary care. The way that t record is structured at this point does not enable it to port very well to a single consolidated point of access for the patient. We heard about the NHS app earlier. Much of the information presented in the app at the moment is directly from primary care. It does not come from the secondary care community or other service providers.
I come back to standards. What we need to enable readier and more supported access to patients are portability standards for the record. Those standards need to feed into something, or feed to primary care, so that there is a maintained single source of truth for an individual that enables consolidation, so that you do not end up with the same drug on your list five times because it has been ported around the system so many times. That will reduce transcription.
There are some good pieces of pilot work going on across the country. We are looking at some work in Leeds on leveraging the European core information standard for the core record, which then enables your record to port with you. It genuinely then becomes yours. You delegate access to it to clinicians, using technology. Moving forward, I would see us looking more towards the European adoption of standards to drive pure portability and standardisation of records to enable that to happen. That requires some national work to get us to a point where there is a national standard for that level of portability. That is where the barriers lie at this point in time. It is ensuring that we have the appropriate level of standardisation to enable access, and then porting it.
Coming back to Dr Modha’s point, once you have standardisation in the record you can safely know what to display to a patient. There is some stuff that you want to talk to them about before they see it, which obviously leads to a better experience of the service.
Ben Richardson: I agree with that. The great potential shown by the NHS app is not just in patient care but in the huge progress in research and allowing research to take place, which has the potential to advantage everyone and the UK economy. In order for that to happen, if we are to enable patients to access things through apps, whether the NHS app or a different one, it comes back to the ability to read and to write back—another anorak topic that needs standards to make it universally consistent as an offer.
Baroness Armstrong of Hill Top: There is lots on this, but I think we have done enough today. That is very helpful, thank you.
Q102 Baroness Finlay of Llandaff: This has been a really interesting session. It is wish list time. What is the one thing that you feel the Government could do, making a change in health data, that would enable better integration of primary and community care?
The Chair: We are looking for one thing.
Dr Neil Modha: I had a list, I am afraid, but the one thing I would do is to introduce some legislation, law or something, that gives us more power over the supplier market or the community. There is HIPAA in America, which is very strict on portability and standardisation of health records to ensure that they do not get locked into vendor systems. It is a rule, a law. You cannot procure or provide a system in the service in America, or any service in America, unless those records can port out of it. We probably need something similar. We spend a lot of time working with industry. It can be very painful.
The Chair: Dr Modha, please leave the list with us. We will not have time to go through it now, unfortunately.
Gary McAllister: As we discussed earlier, it would be probably setting the scene for information sharing, and then underwriting the risk from the individual data controllers of information, with sound governance and principles that everyone could abide by, and the ability for people to opt out of that. If that was underwritten, I think it would change the game for providers.
Ben Richardson: This is a slight cheat. I will say that it is the data-sharing points that we have talked about, but that is not going to be possible in community care without a clearly defined and measured standard of a universal offer for community care. It is impossible to collect and share data otherwise.
The Chair: Thank you very much. I reiterate, please make sure you give us the list. Thank you for those very pointed answers.
I remind you that this was a public session and that there will be a record of it, which you will be able to have a look at. If there is anything else that you think would be of use to the committee, we would be more than delighted to have you share that with us. For the moment, on behalf of all my colleagues, thank you so much for a most interesting session and for your time this afternoon.
