'The NHS should get a high-quality service'

Both the bidding process and the overall vision of the £2.3bn National Programme are coming under increasing scrutiny. Richard Granger, director general of NHS IT, spoke exclusively to Computing about how things currently stand and what still needs to be worked out.

The procurement is running to a very tight timescale. Is everything still on course?

Everything is on track. The things we said would be completed by mid-September have been completed and we'll be making some announcements about the achievements of some recent milestones in due course.

We are continuing to expand the team, in particular we now have a chief programme officer, and we have further strengthened clinical involvement in the programme with two days a week from Professor Peter Hutton, who is the president of the Academy of Royal Medical Colleges. Peter is one of a number of senior clinicians involved in the programme. As we go into development, testing, deployment planning, training and process re-engineering requirements, he will take a leading role.

Has the withdrawal of Lockheed Martin earlier this month affected the progress of the procurement?

Not at all. Bidders withdraw form large procurements for a variety of reasons.

I don't want to breach confidences with this particular ex-bidder regarding my opinions as to why they withdrew. Suffice to say we have every confidence in the quality of the competition and having two very strong bidders on the two competitions where Lockheed were bidding doesn't affect that in any way. In fact it provides the procurement team with the opportunity to spend more time with the other bidders.

Historically there have been a number of occasions where large global organisations find themselves in a situation where their UK subsidiaries have not had the full authorisation of the global body for particular procurements.

I have a sufficiency of real factual data on the market appetite for our terms and conditions to entirely refute assertions that the arrangements we are seeking to secure are inappropriate or not going to be acceptable to the supplier community.

It has been reported that Lockheed felt that the deal was commercially impossible because they did not stand to make any money back for three or four years.

That is absolutely not the case. We think the NHS should get a high-quality service and we think we should pay for things once it is demonstrable that they work. So we think they should be built and, following acceptance, we should then acquire those assets. So there is no circumstance in which the return for the prospective suppliers would be negative 36 to 48 months after the execution of these agreements. That is just plain wrong and it is ill-informed.

We are requiring contractors to achieve the best that came out of Private Finance Initiative (PFI) arrangements in terms of bearing the completion risks and on-going performance risk. But we recognise fully that it should be rewarded. We want to acquire the assets, once they work, and then pay for service management. That is the structure we set out in [the procurement strategy published in] January and we haven't departed from that. I am confident that we've taken the best from PFI and are putting together arrangements with prospective suppliers which make sense for both parties and are clear, fair and equitable.

There is some tension between the desire on the buyer's part for stringent terms and conditions, and the danger that suppliers will then not bid for the deal.

This is getting back to some realism from both parties about the degree to which pre-award of contract specification can occur, and the basis on which post-award of contract variation can occur.

We are seeking to strike a balance there because clearly if one goes into an exhaustive specification process it presents a risk of an elongated timescale and typically the underlying asset that's being procured being at least partially life-expired by the time it's in production.

So we want to strike a balance there, and by the use of open-book pricing arrangements and pre-award of contract specification of unit costs for goods and services that are likely to come under variation notices, we are hoping we can have a sufficiency of transparency between both parties.

There is some worry that suppliers are loath to express their concerns about either the contracts themselves or the overall vision of the programme, for fear of jeopardising their position in the competition.

In terms of communciation with suppliers much mischief has been made from the misinterpretation of a couple of things I said in December last year. We have a full and open dialogue with suppliers and we are in a negotiation to secure the best deal for the NHS. That means suppliers get to make a reasonable profit, take a reasonable level of risk and deliver world-class solutions for the NHS.

Inevitably there will be give and take on both sides as part of that. One of the contractual instruments we are using, which was also used on the London Congestion Charging scheme, is the use of delay deductions to ensure the honesty and clarity of the delivery commitments which are being made by prospective suppliers during the procurement phase.

In asking suppliers to confirm their confidence with interim and key milestones and substantiate that with the potential for delay deductions, we believe there is a degree of honesty present in their representations which is good for the NHS.

We are also putting all the major solutions through a proof of solution episode so that representations made by prospective suppliers are born out by scrutiny of their solutions. Again that is not something that has always been done in the past but it was something which was done on Congestion Charging.

After the contracts have been awarded there will be another series of testing and proving episodes which culminate in the use of the software in a fair representation of a production environment prior to getting into incremental deployment.

What link has been made between the systems making up the National Programme and the clinical targets faced by frontline NHS staff?

There have been a number of studies that state very clearly in their conclusions that having the right available information on the spot allows clinicians to focus on the delivery of care rather than the collation and assimilation of assorted and often unavailable data.

A lot of people's visiting either their GP or a consultant will observe important information being available in an unsummarised format on paper. And although we have the largest IP network in the world, much information is still transcribed onto documents which are then transferred in hard copy and may be keyed in repeatedly across different systems.

The programme is about modernisation. It is about information being keyed in once, accurately and then being available to people who have reason to access it. It is about making clinicians' lives easier so they can focus on patients, with the right information being available at the point at which they deliver care.

Government technology programmes often suffer from a disconnect, with the IT project 'tacked on' to the main business of the organisation. What is being done to combat this tendency?

The technologists are not in sole control of what is happening here. The programme board has strong representation from the end user community - the five Strategic Health Authority chief executives leading the clusters for example, people with clinical leadership responsibilities like Professor Peter Hutton, the chief executive of the Modernisation Agency and so on. So we don't have an isolationist IT programme in any way.

Taking ebooking as an example, the project is co-led by somebody from the Modernisation Agency who used to manage a number of hospitals, a clinician, and somebody with a technology programme background. I would need to check the exact numbers but a substantial proportion of the headcount on the programme and the time deployed, probably to the order of 20 or 30 per cent, is from clinicians.

There's also substantial input to the programme from people with frontline NHS management expertise, so we have a team which is synthesising the necessary experience of how things are done at the moment and how people wish to do them in the future, to avoid the type of problems to which I imagine you are referring.

There seems to be some confusion around who is responsible for what at various levels of the programme - local NHS IT departments, existing suppliers, newly-appointed Local Service Providers (LSPs), National Application Service providers, and so on.

That's a fair comment. It is important to recognise our starting point 11 months ago was with a very limited headcount and a very broad brush set of requirements - the strategy document 'Delivering 21st Century IT Support for the NHS'. We are now implementing that strategy and every week that passes provides greater clarity and detail as we drill down into it. As time passes, the detail of the scope of the programme, the interfaces with existing entities both NHS and supplier, and the responsibility of people locally, nationally and in the new suppliers gets clarified.

Should this not be in place already, considering the stage we are at in the procurement?

The NHS is a very large and heterogeneous organisation and there's no sense that one answer would work everywhere. We face some hard choices about the level to which due diligence and responsibility mapping around existing systems and services is undertaken, versus the specification and procurement of the new services that will overlay them. Achievement of the timetable and fulfillment of the requirements set out in the specification means that some of these things are getting sorted out after a time which people would state is ideal. I accept that, but as we move towards the appointment of LSPs we will get more capacity and we will hammer out some of these things.

Do these details need to be decided before the deals are signed, or can they wait to be discussed with the specific suppliers who sign the deals?

As every week passes more of the scope and responsibility around the programme and more of these aspects of ambiguity and uncertainty get resolved. There will always be a raft of things to be resolved and they will change with time. People may be concerned at the moment about the interface between both in-bound suppliers and existing suppliers and between in-bound suppliers and existing IT departments. Those matters will be resolved as we narrow down the number of suppliers and their views and capacity are assessed as part of the procurement process. Then new issues will emerge that will need resolution - around responsibility locally for implementation activities, responsibilities within LSPs and the central team for the support of implementation and so on. This is entirely normal. What is abnormal is the level of interest being expressed in sorting out normal programme problems.

Another key area lacking clarity is around the ownership of data. It has been reported that Lockheed's withdrawal was at least partially because they were to be held responsible for the accuracy of data input by NHS staff.

Let me set the record straight because that is inaccurate and unhelpful to everybody. We are looking for suppliers to take the utmost care with data which is entered and stored on their systems and I think most right-minded people would not want anything other than that.

So we are asking prospective suppliers to sign up to contracts which mandate their performance, without equivocation, around the care and security of patient data. Clearly that data is not entered by them, it is entered by employees of the NHS, and we would not be looking to ask a supplier to warranty the accuracy of data entered by people for whom they are not directly responsible.

But I would hold them responsible for the corruption of data that has been entered. We are seeking intervention rights, and consequences for failure to take seriously and perform against this requirement, commensurate with the importance and sanctity of patient data, and again I think right-minded people would not dispute this. It is interesting that the existing suppliers of large-scale data processing services to the NHS, and other UK government entities, have no problems with that as a requirement because they are used to it. There is nothing unusual here and I'm not looking for anything which is not sensible and in line with the requirements one would expect for a system containing data of this type.