Audit Committee, 05 Dec 2006

Meeting date: Tuesday, December 5, 2006

Official Report 262KB pdf

---

“Informed to care: Managing IT to deliver information in the NHS in Scotland”

Under item 2, the Auditor General will present a briefing on his report "Informed to care: Managing IT to deliver information in the NHS in Scotland".

Good morning. I am sure that the committee agrees that there is great potential in the health service for information technology to provide clinical and management information quickly and securely, but that that requires substantial investment and that the whole process must be managed well. Against that background, I published on 23 November a report on information management and technology—which I will refer to as IM&T from now on—in the national health service in Scotland. The report covers the arrangements for delivering information through IT to meet the future needs of the NHS and examines national arrangements in three areas: leadership; involving the stakeholders who use the information; and programme and project management.

The Scottish Executive Health Department's strategy, "Delivering for Health", makes it quite clear that a more corporate approach is now required. That will represent a significant cultural shift in how IT is managed in the NHS, which I acknowledge will take time to plan and implement, but at the time of our review the Health Department did not have fully developed arrangements in place to demonstrate that leadership, stakeholder involvement and project management meet internationally recognised good practice standards. We concluded that there was scope for improvement in each of those areas. Throughout the report, we give examples of good practice and best practice, which we hope will be of some assistance to the department in implementing the strategy.

Many of the problems stemmed from fragmentation as a result of the previous trust and board freedoms to procure and implement local solutions prior to the move to single-system working. The department has taken steps to improve its management arrangements, but at this stage we cannot say whether the changes will be sufficient to address the issues that are raised in the report.

Leadership was the first key area that was considered. Clear leadership is vital so that people know who is in charge and where accountabilities and responsibilities lie. Leaders must ensure that there is a clear line of sight from national health service policy and business strategy to an information strategy and an associated IT strategy. Given that the national budget for IT in 2006-07 is £100 million and that the budget is scheduled to rise to £140 million in 2007-08, it is vital that NHS funding arrangements for new and on-going programmes and projects are robust.

In 2004, the department published a strategy that focused on e-health. Exhibit 3 on page 9 of the report lists all the current IT projects. We encourage the department to ensure that, in future, the IM&T strategy is clear about overall information requirements across the full range of stakeholders, which includes clinicians, managers, planners and policy makers, and that it provides a sound basis for developing implementation plans at both national and local level.

The situation is constantly changing, but it was not clear at the time of our audit who was accountable for directing IM&T strategy development and implementation. In particular, the roles of the Health Department and NHS National Services Scotland needed to be clarified and agreement was still needed on the balance between national standards and the freedom for local boards to implement local solutions. Tensions still exist between national and local priorities, and boards occasionally opt in or out of national systems depending on local circumstances. The examples that are mentioned in the report may reflect the fact that the NHS in Scotland is in transition. We found that the overall strategy must be revised to reflect the full range of information needs and recent policy initiatives. We were told that the department recently considered a paper on bringing about the convergence of national strategies and local policies and plans.

Good governance is part of good leadership. The report comments on the governance arrangements and identifies areas of significant weakness. The department recognises that the arrangements need to improve in line with the good practice that we outlined in appendix 3 of the report. Since the completion of the fieldwork for our report, the department has announced a new organisational structure for managing IM&T, which is outlined in exhibit 9 on page 16. That structure is in line with good practice, but at this stage we do not know in detail how far it has been implemented or how effective it will be.

Ensuring that the benefits of IM&T are delivered requires good programme and project management so that systems are delivered to specification, on time and within budget. People must be sure at the outset about what benefits are expected and they must know what success will look like once the systems are implemented. The report says that the department does not have key performance indicators to monitor the implementation of the strategy. A number of information initiatives are under way in the department, which exhibit 10 on page 17 summarises. I recommend that the department consider exhibit 11 on page 18, which suggests a set of performance indicators that might be typical of an IM&T strategy.

The report also suggests that the department review the funding arrangements for IM&T. For example, we would encourage more widespread use of business cases before projects are committed and the rigorous adoption of the gateway approach, in which funds are released on a phased basis, depending on the achievement of certain specified outcomes.

The second major area that we considered was how the users of information are involved in planning and delivery. Information technology is not an end in itself—it is a tool to support the health service in providing good patient care. To achieve that, it is essential to get and keep stakeholders such as clinicians and managers on board. We suggest that more formal and rigorous processes for involving stakeholders are needed. The department is taking action to improve matters in that area.

We found examples of good practice in programme and project management, which was our third theme, but more must be done. For example, our case studies show that programme and project management skills vary throughout the NHS in Scotland. In particular, the skill level in local boards generally needs to improve. Another example is the need for a more rigorous approach to identify the anticipated benefits of investing in projects. At a later stage, an evaluation should be carried out of whether the benefits are being achieved. One example that is mentioned in the report is GPASS—the general practice administration system for Scotland—which has involved significant investment under a top-slicing funding arrangement for a number of years, even though a significant number of general practitioners are not convinced that the investment continues to represent good value for money.

Given the importance of investing in IM&T in the health service, I have asked Audit Scotland to keep the implementation of the strategy under review and I intend to keep an open mind on whether a further report might be appropriate in due course to assess progress. As ever, my colleagues and I are happy to attempt to answer any questions that the committee may have.

You mentioned GPs' view that GPASS does not provide value for money. On what basis did they make that observation?

GPASS is not one of the four projects that we considered in detail, but we are aware of a recent review that the Health Department has undertaken of how GPASS is operating. Rhona Jack may be able to give an indication of the current state of play on that.

GPASS has been a source of anxiety for GPs since about 2001. Several reports on it have been produced, such as that by Professor Lewis Ritchie. The most recent one was an independent piece of work that the Scottish Executive Health Department commissioned from a company called Deloitte. In essence, Deloitte concluded that the current product does not meet current clinical requirements, although the GPASS-clinical module should help if it is successfully delivered. However, Deloitte says that, even if that module is delivered successfully, GPASS will not meet the long-term requirements of GPs and other allied health professionals who work in the community, such as physiotherapists. The Deloitte report recommends that the Scottish health service go for a more commercial approach, using at least one commercial supplier.

Do we have a guarantee that 100 per cent of GPs will buy in to such an approach? My understanding is that part of the problem with GPASS is that some GPs refuse to use it.

That is certainly a problem. GPs are independent and they have rights—

So do patients.

Indeed. Under their contract, GPs have rights to select a suitable supplier for their information technology. It will take considerable effort to get GPs, as key stakeholders, on board and to keep them there, but it will be well worth it. However, there is no guarantee.

On page 23 of the report, we attempt to capture some of the concerns about GPASS. Central to them is the concern about how functional the system is in meeting clinical needs. There are some technical issues to do with the length of records, which can mean that inappropriate information is recorded and used. The concerns are real. This is an example of the importance of involving stakeholders at the outset to ensure that clinical needs are addressed when systems are being designed. It also shows the importance of ensuring that a good business plan is in place that assesses the business needs against the technical functionality of such systems before they are implemented and of carrying out a regular review against the standards as the project is rolled out.

The British Medical Association's GP sub-committee, which represents at least some GPs, has said that it is very keen to work with the Health Department. Everybody acknowledges that if different systems proliferate throughout the country, that will have a cost and help nobody.

I do not want to sound sceptical or a complete luddite but, given the problems that you have just outlined in even attempting to sign up GPs to something as simple as GPASS, is it really possible for us to design an IM&T system that encompasses all the facets of the national health service to which people will be signed up and which will deliver by improving patient care and meeting the needs of people who do the job, at an affordable rate? We continue to put vast sums of money into IT, but all we seem to do is decide that one system will not meet needs and move on to the next one.

The short answer to your question whether it is possible to design a strategy is yes. Many of the programmes and projects that we outline in appendix 2 are progressing well and will achieve their purpose. The problem is that a fully integrated and up-to-date strategy does not yet exist to explain how all the projects fit together into a strategy that is directed primarily at improving the delivery of health care at the front end.

In quite a number of areas, design and project management could be improved so that the risks that are associated with those complex projects are managed well. In a sense, the report is interim. It comes at a time when the department has changed direction towards a more strategic approach. For that reason, we put quite a lot of effort into identifying examples of good practice and best practice standards that we think the health service should follow more rigorously to achieve the objective of managing those projects well and managing the risks out.

Do the ability and leadership to make the changes that will make it possible to design such a strategy exist?

One issue on which we commented at length when we produced the report—the fieldwork was done until spring this year—was that we had some difficulty identifying clearly who was in charge of the overall strategy and direction. More recently, the department has introduced a new management structure. We describe that in exhibit 9 on page 16, which shows the new post of director of e-health and several structures to support that. That is in line with best practice, but the structure is not yet fully implemented, so it is too early to say how effective it will be. For that reason, I will ask Audit Scotland to continue to monitor developments.

That goes back to your initial question about whether the implication is almost that some super-duper single IT system will answer all the questions that everybody in the health service has. The answer is that the health service is huge and very complex. What is required is a series of different systems that support individual elements, contribute to the whole and, when brought together, start to answer some of the complex questions, such as what, if we invest £X million of additional moneys, the productivity gain will be. That is when the power of being able to bring together in one place different systems that cover activity, quality, outcomes and cost is felt, as it offers the hope of answering such questions, which, when people such as members ask them at the moment, are difficult to answer.

We all want to know whether anyone has a firm grip on what is obviously a complex situation. It is never easy to integrate IT systems, especially large and dispersed systems. You mention that what is required is a

"significant cultural shift in the way in which IT is managed"

and that it

"will take time to plan and implement."

Can you give us an idea of the timescale and of how the systems will be integrated?

What we mean by that comment is that the strategy published in 2004 gave a clear commitment that there would be more strategic corporate direction of the major systems needed for a modern and efficient health service. That strategy is barely two years old, so the Health Department and the health service are at the transition stage.

What we mean by a cultural shift is that, before the strategy was published, the emphasis was on health boards and trusts developing systems that met local needs in the context of their local business plans. What has changed is that there is recognition of the importance of the health service in Scotland as a whole having a core strategy that covers all the major information requirements. The strategy must ensure that certain core systems are designed that are relevant across the whole of the health service and that those are applied by individual boards. A transition is now taking place, because a number of important systems have not necessarily been adopted by all health boards. There is a complex and mixed pattern. The culture of change is moving towards recognition of the importance of balancing strategic information requirements across the whole of the health service with the need for discretion at a local level to develop systems that meet local business needs.

Given the complexity of the matter, there will be local and national gravitational pulls, so strong direction and a clear strategy are required. The timetable is important. Secondly, there are always spiralling costs when we deal with IT. It is concerning to learn that the NHS in Scotland does not know exactly how much it spends on information management and technology overall. Why is that the case? What monitoring systems are in place to give us an idea of the costs?

Can Rhona Jack help to answer that, please?

It comes back to the fragmentation that we have mentioned and the local freedoms that existed. There has been spend on IM&T at national level and at local level. They were recorded in different ways, so it was not possible to find out how much was being spent. The Health Department conducted a one-off exercise to try to do that, but people allocated costs to different headings and so on, so it was difficult to identify how much was being spent on IM&T at local level. As a result of that, the Health Department has committed, as part of the new governance arrangements, to consider how the funding arrangements are put in place so that in future it can specify the initial and on-going costs of different projects.

Does the Health Department need that complex financial information, or could it bypass it in planning a system that works?

The difficulty is that you want to make a decision based not only on the purchase of a system, but on its on-going costs. An issue is at what point you say, "Here is the cut-off. We will not invest in that system any more because it is being overtaken by other systems." An example of that is GPASS, which has received substantial investment over the years. At what point is it appropriate to pull the plug, as it were, and say, "No. Now is the time to move on and invest in something else"? Without appropriate financial information, it becomes difficult to make such decisions rationally.

I will pick up on an issue that Andrew Welsh touched on. You mentioned that tensions exist with boards opting in and out of certain strategies. Does that happen because boards believe that that strategy or IM&T is not what they require, or is it because of financial pressures in the sense that opting in has a price tag and they calculate that it is not worth the investment?

There are several factors that help to explain the situation, some of which are implied by your question.

One factor is that some of the systems have been developed over several years, so they have not had applied to them the same business planning discipline that is being applied to some of the bigger schemes that have been developed more recently. Boards have the opportunity to decide whether to opt in or out of some of the systems that have been around for a few years. We give some examples on page 12 of the report. The Scottish care information store is at different stages of adoption among NHS boards. There are different versions of it in operation and boards can decide whether to take them up. One of the problems with that project is that it does not have a business case with a benefits plan that indicates what it is meant to achieve.

A second feature that explains some of the tensions is that some key systems have been developed from the bottom up. In other words, a health board—or a group of boards—has developed to meet its own business needs a system that is thought to have some value nationally. The accident and emergency system is an example of that. Some boards were allowed to opt out of the national system for accident and emergency when the existing system meant having national standards. That opting out might have an impact on the overall rigour of the system and on the cost of developing such systems. It is fair to say that different factors are at play.

It might be significant that the department will fund the development of key software for the whole of Scotland, but the cost of implementation will fall to be met by health boards through local business planning. Health boards will consider their priorities and the pressures within their limited budgets. For reasons that might make perfect sense at a local level, boards might decide to opt out of a system or to go in a different direction.

Those are examples of what we mean when we talk about the NHS's development of a coherent IT strategy being in transition. Rhona Jack might have more to say about it.

When a local health board is under financial pressure, it is difficult for it to be seen to be investing in systems, particularly if they are not patient-based systems. For example, we looked at BPI—best procurement implementation—which is basically a supply system. If a board is under real pressure and having to close wards, why would it invest in a supply system? The answer is that 30 per cent of costs are made up of supplies and services and the whole purpose of that system is to provide better quality supplies and services and to achieve some of the efficient government targets. The board might need to spend to save, but it could be extremely difficult to handle that locally if it were under pressure.

Surely the current fragmentation must mean that there is an inevitable lack of communication between existing systems, which is a major problem given that people will be comfortable with the existing systems. Who will drive through to get co-ordination, and at what cost?

The short answer is that the Health Department has a new structure in place for managing IM&T projects. The structure is outlined in exhibit 9. Although it is not yet fully in place, we believe that it is a very positive move. It should offer the prospect of better integration, which is what you seem to be, quite rightly, suggesting.

Or leadership towards a common goal.

Indeed.

I have just one short question. Will there ever be a comprehensive, cost-effective system that allows people to opt out and go somewhere else?

The short answer to that, rather like that to the previous question, is yes, it should certainly be possible. Such a system requires a clear strategy that links the information requirements of the whole health service to the strategy of the NHS in Scotland. It also requires business propositions for developing IT that support those information requirements. And it must recognise that alongside the comprehensive system there might well be a need for a locally developed system to meet local needs—although a core of activity and system support should be common throughout the health service.

You say that there needs to be a core. Unless we have that, it will be difficult to bring it all together to provide the comprehensive patient information that is necessary to deliver the service.

That is true.

Although that is true, the good news is that people are seeing the benefits of going for a co-ordinated corporate approach rather than everybody going their own way. There might well be situations in which it is appropriate to tailor systems, but every time someone tailors something to their specific needs, a cost is attached—so they ought to be able to justify why that cost is being incurred. That kind of thing takes time, which is why stakeholder engagement is crucial.

The Auditor General mentioned that he is planning a follow-up study on which he will report. Can you give us an idea of the timescale for that?

I find that difficult to do at this stage, although it is unlikely to happen in 2007. It will be appropriate to review how the strategy develops and take a judgment at the back end of next year as to whether it would be appropriate to report again and to what timescale.

The committee will discuss under agenda item 7 how it will proceed after the briefing on the report "Informed to Care—Managing IT to deliver information in the NHS in Scotland". I thank Rhona Jack for her help on item 2.