Select Committee on Science and Technology Minutes of Evidence

Examination of Witnesses (Questions 640-652)




  640. In respect of integrating the health systems in the broader extent, how does, for example, DEFRA fit into your plans both in terms of disease monitoring and also with research?
  (Dr Catchpole) I have had a series of meetings with Ruth Leesons, who has recently taken on the role of heading up DEFRA's new epidemiologists surveillance unit. We have had a number of meetings and propose more to look at the ways in which we can integrate our respective surveillance activities. I am sure you are aware of the veterinary surveillance strategy which talks about working not only with data coming from the veterinary field, the VLAs, but also looking at integrating that with human surveillance data. We are very keen to take that agenda forward.

Lord Turnberg

  641. Part of my question relates to how what you are doing in your strategy relates to what is happening nationally and internationally. I am particularly interested in what relationships you have with Scotland and Wales.
  (Sir John Pattison) It is fair to say that the national programme that I have been talking about is a national programme for England. Nevertheless, obviously at the borders—and south and north Wales would be one and the Scottish border would be another—there is the movement of patients and the movement of microbes across the border in this context, they are no great respecters of that. We will work to ensure that we can manage the flow of information across those borders although it will have to be on the basis of ensuring that our systems are interoperable rather than in any sense taking over.

  642. So they are aware of what you are doing and you are aware of what they are doing?
  (Sir John Pattison) Yes.

Lord Oxburgh

  643. Sir John, would you care to repeat for the record what I heard you state in another forum recently about your rigorous insistence on interface compatibilities between systems operating within your new IT world?
  (Sir John Pattison) Standards have been referred to already this morning and I once said—it has been quoted back to me many times to the point where I wish I had not said it but it is correct and it is important—that this programme is going to be characterised by ruthless standardisation and it is absolutely essential that we do that and two of the key standards have already been mentioned this morning, SNOMED CT terms, which incidentally was a joint development by the American College of Pathologists and the Department of Health in England, and HL7. The key thing is that we are going to ensure that all the systems are interoperable. When we are trying to do something on a scale such as this it is not going to be possible to do it in the time that we would like to do it if we only had a single supplier and a single system, not to mention the fact that we have a tremendous set of legacy systems with which for some time we will have to interoperate. Technology is coming to our assistance here because, for example, the main suppliers of primary care IT systems have themselves got to get and publish APIs, which are application programming interfaces, which will allow us to look at the legacy systems, e.g. EMIS, Torrex and IPS, bring them into a data repository for our purposes and then move the necessary information through the national system.

Lord Turnberg

  644. This is very specifically about data relating to infectious diseases and we have been talking rather generally about how one can extract it, but my question is particularly about the reporting, analysis and display in relation to infectious diseases and I think we have dealt with some of it. To what extent is the IT industry aware of the needs here?
  (Mr Spittle) Specifically from a medical point of view, I am clearly not as aware as my colleagues here. In terms of what is needed from an IT system point of view, however, I think there are no fears in this for us because what we are talking about is data capture, we are talking about data aggregation, we are talking about data analysis, reporting, both the pulling out of the data and also the pushing back of the answers or the alerts or the trigger mechanisms that are needed, because very clearly one of the things you will want to do with an infrastructure is not only be able to pull the data but you actually want to push data down to the absolute point where you want it delivered, whether it is in a laboratory or in a GP's office where he/she is in front of the patient. It is very important that we have a system that can act in real time to do that. That is when I think the practitioners will see the benefit of providing data capture, when something comes back out of that system and it has added value to the data that they have put into that system. What you want to understand particularly with infectious diseases is a lot of the geographical context and we have a lot of experience of geographical information systems and that is something you would then want to alert local people to either with an email or a text on your phone. Any of these things are totally possible. From a technological point of view I am very comfortable that the abilities are there.

  645. Could I ask the same question to Dr Catchpole because I know in the PHLS they have quite a lot of experience and it always seemed to me that what was required was—I hate to use the word—straightforward. We all knew what we wanted to get in and out and how and yet it was always a struggle for a variety of reasons, partly laboratory stuff, reluctance to enter data in the form that was required, perhaps the technology was not the right sort for capturing the data, but a lot of effort was put into it and also the lack of funding from the Department of Health for this particular activity during that time. Thank goodness the HPA will have access to that funding. Dr Catchpole, how do you see the industry helping with all that?
  (Dr Catchpole) It comes back to standards. Just as an example of what we have been doing within the Public Laboratory Service, the mainstay of our surveillance is capturing data from laboratories and we have the CoSurv system. Once the data is in CoSurv the laboratory can then flow the data to the local public health departments and to CDSC. We have a challenge to interface that with over 26 different types of laboratory system and even laboratory systems provided by the same supplier can have different coding structures and data structures within them. There is a desperate need for standards. There has to be an open discussion, we have to share with the suppliers what our requirements are and ideally that should be through the NHS Information Standards Boards and so we have that written in as a clearly stated standard. We are well advanced in discussing that with the two major system providers. We very much hope the money that we have earmarked to take this forward will be available within the PHLS and available next year with the HPS to do the same thing. So I think what we would look for from industry is to come to the table and listen and talk to us about standards. When we have talked to providers it has not always been very high up on the priority list. It is not perceived that there is enormous commercial gain in delivering some of these outputs. I think it is really about ensuring that industry is willing to come and share with us and talk about standards.

Lord Oxburgh

  646. One aspect which is different about medical information, it has analogies but not exact analogies with other areas, is this question of patient confidentiality. I wonder to what extent the industry has taken that on board? Perhaps the question of patient confidentiality becomes overemphasised through a touching faith in the security of the paper records, but it is important and many of the problems about using information in the way that we have described are not serious provided this information is not attributable or readily traceable to a particular patient.
  (Mr Spittle) We are very familiar with that. It is not only in medical operations, it is in all public-type operations. It is amazing how much security you can achieve if you strip somebody's name and address off the information, which you can do very simply with information and send it as strings and then it gets re-combined at the end point or you can encrypt or you can have public and private keys. I do not think that is a technological challenge at all.

  647. Sir John, this is being built into your system no doubt?
  (Sir John Pattison) Indeed. I have a number of points to make. The first is that we really need to push the use of the NHS number which would help us greatly in having a unique identifier which we can encrypt, etcetera. Secondly, the whole question of confidentiality we are going to have to solve at a national level once and in a sense build it in as part of the infrastructure. It is very difficult if well people sit round a table and try to have far-ranging philosophical discussions about the ethics and philosophy of confidentiality. So we have taken the approach of going out to consultation with a suite of documents, an NHS code of practice, a patient's charter, asking the question supposing this was how it was, what do you think? The period of consultation has just finished, we have had over 2,000 responses and they are in the process of being analysed in terms of how one might need to adapt either the NHS code of practice or the patient's charter. In terms of the focus groups and the patient groups that we engage in that process, there are no real surprises in what they said and some areas of their health or potential ill health they are very sensitive about and HIV Aids and other sexually transmitted disease is one area which is going to impact directly on this. So we are going to have to treat that differently along with certain aspects of mental health and certain aspects relating to pregnancy and its termination. There we have developed a concept of a virtual sealed envelope which is essentially part of the system, where a patient can choose to put some information into it that can only be accessed with the permission of that patient. That raises some issues for us in terms of the surveillance of communicable diseases, but in a sense PIAG has helped with that, the Patient Information Advisory Group, allowing that sort of information to be accessed for surveillance purposes, although in the case of HIV Aids anonymously.

  648. Would it be allowed to count the number of sealed envelopes?
  (Dr Catchpole) I chair a group for the Department looking at defining a data standard for sexual health surveillance and that was an issue that came up and I am not sure about that, the question if it is in a sealed envelope can one share that information anonymously and will there be a mechanism to allow that to happen.

Chairman: We have answered quite a lot of the questions and I am just wondering if we can end up combining questions 6 and 7 for Lord Oxburgh.

Lord Oxburgh

  649. The PHLS IT strategy and the development of CoSurv and so on I am sure have been taken into account by the NHS in its approach to these things. I do not know whether Dr Catchpole or Sir John would care to comment on this if there are issues that have not already arisen?
  (Dr Catchpole) I think the main points I was going to make have largely been made. The implementation of CoSurv has been hugely hampered by the lack of standards. We need to recognise that implementing IT systems is not about just putting in boxes and wires, it is the training for the user, it is the actual time taken to implement new systems and it is about having good project managers. I do not think I have anything that has not already been said to add. The single key message really is we need standards, we need them rapidly and we need to recognise the true costs that that includes, the training and the human elements of the implementation of IT projects.

Lord Oxburgh: I think question 7 has already been answered very clearly.

Chairman: Baroness Finlay has raised a number of points, but I am wondering if she would like to emphasise any of those?

Baroness Finlay of Llandaff

  650. Thank you. One thing that goes through my mind is you have spoken a lot about standards. Who is actually going to define these standards and who is going to be responsible for reviewing them as new clinical scenarios emerge? We have spoken about three that have emerged in the last 20 years already. What is being built into the system to make them sufficiently adaptable to those standards being added on?
  (Sir John Pattison) Each of the standards is not something that is static. If you take SNOMED CT, there will be a first release and that will be the standard of the day, but there will be a commitment to developing those standards to reflect clinical practice. We have an information standards board which has three sub-committees and it requires standards to be presented to it and for those standards to come with a sponsor and the role of the sponsor is to be the person who ensures that that standard, once it is accepted, is further developed in the light of national and international developers. It is so important that Professor Martin Service, who is a clinician in Portsmouth, the Chairman of our Information Standards Board, has agreed and his university have released him on secondment for three days a week in order that he may spend more time on the information standards for the NHS and you will be pleased to know that one of the other days a week he still does clinical practice in order to be firmly rooted in the reality of what is required. His consideration of standards includes not just the technicalities of it but an important element which he calls implementability, that is you can have a wonderful standard but it is simply impossible to implement in clinical practice. On the negative side, I will admit to you that I and others in the Department of Health and the NHS need to co-ordinate the development of standards and data sets much more than we are at the moment and I fear that we have a proliferation of standards and data sets that all focus down and impact on single individuals in their clinical practice to the point where it is becoming almost impossible for them to comply and we must change that.

  651. The development and the commercialisation depends on being able to sell something new. How are we going to ensure that the industry is going to take a long-term look at these developments rather than a relatively short-term look and come along and want to be inputting the new system and back to the commercialisations behind upgrades because every time the system changes clinicians have to adapt to that change and just change in itself is hard?
  (Sir John Pattison) It is our view from where we are sitting that we have had a productive dialogue in the last 12 or 15 months, particularly with the advent of Richard Granger, with the suppliers who are coming with this new investment for the longer term and who are concerned to cover their costs for us to recognise that they must do some further development because otherwise in a few years' time it will be out-of-date and as a commercial company they make a modest profit and we do recognise that, though we know that we can deal with suppliers on a much more rational basis than we have in the past. An example of that which is already in place is the NHS used to buy for 92 per cent of its use Microsoft licences by raising 34,000 invoices independently up and down the land. We converted that into a single national invoice for those licences with the upgrades that will come with it for a three-year period and saved £50 million for the NHS. So I think our relationships with the supply community are improving.
  (Mr Spittle) I would like to answer on behalf of IBM on this one. IBM is totally committed to open standards, that is to say standards that are in public bodies and if you talk about messaging and data sets for instance, we support XML, which is the common lingua-franca for messaging going forward. All our products are built to open standards and we do that because fundamentally we believe in having open standards because systems do need to talk to each other. Nobody is going to buy every piece of their kit from one company. Companies increasingly need to talk to other companies and other suppliers and other customers. It makes absolutely no sense in a heterogeneous world to have anything other than open standards, you have to have open standards. For our customers, it gives our customers a confidence that when the next release of this thing comes out we are building to standards that are known and they will not have to then only buy from us or somebody else, they are not locked in, it is a degree of future proofing that is needed. From our perspective open standards is a company-wide commitment.

Lord Oxburgh

  652. Could we come back to sentinel practices for a moment because I would like to get Sir John's view on this. Last week we were in Birmingham and we had a presentation from one of the people who runs a sentinel practice and this was a scheme which is run by the research unit of the Royal College of General Practitioners. Looking at the distribution of those 70 or so practices, it was not very well distributed geographically. Much of the east of England, for example there was one in East Anglia, the south-west, was probably similarly bereft and much of Wales. Do you anticipate this kind of surveillance being superseded by the plans that you have in mind in drawing information in real time from all sorts of practices or do you see a continuing need for a selected group of practices which will actually operate slightly differently from others insofar as they will send off for laboratory investigation a variety of samples which would not be done in the run of the mill circumstances, which can give early advice on the presence of particular pathogens locally? The comment was made to us that the density and distribution of those practices at present was not such as to give useful local information in most cases but it could be used to give national information. Would you like to elaborate on how you see that fitting into the future?
  (Sir John Pattison) My personal view is that I see it as the second of your alternatives, that we are putting in place a national system which will deliver some of the general and country-wide surveillance that is required, but I cannot myself envisage there ever being a situation in which you could supplant the need for some collections that are going to go into some issues in more depth either as a standing order as it were or in response to some events that we might only guess at in the future. Is that how Dr Catchpole sees it?
  (Dr Catchpole) Yes. There is a continued role for sentinel practices particularly for the reason you stated and I am sure that Dr Fleming would have said that we need to have sentinel practices where specimens are collected perhaps more to serve the public health surveillance function than immediate clinical care and I think there will always continue to be that purpose.

Chairman: Gentlemen, thank you very much indeed. I have just one or two points. I wonder if Sir John could provide us with the proportion of NHS spend versus PHLS spend in IT and information services for the country and, if possible, the relationship of this to other spending that is going on in IT, for example in banking and large retail outlets and the international health service. We will write to you on that. If there is anything you feel we have not touched upon or has not been made as clear as you would like it to be we would very much welcome input from you. You will get a transcript of what has gone on and you will have the opportunity to correct any factual errors there. It just remains for me to thank you for coming along.

previous page contents next page

House of Lords home page Parliament home page House of Commons home page search page enquiries index

© Parliamentary copyright 2003